commit 6547139a3d029114d0b83476eed0aa930f3e5afb Author: ModelHub XC Date: Wed May 6 09:44:46 2026 +0800 初始化项目，由ModelHub XC社区提供模型 Model: HCY123902/llama-3-8b-dpo-tw31-beta-1e-0-ift Source: Original Platform diff --git a/.gitattributes b/.gitattributes new file mode 100644 index 0000000..f8fe2af --- /dev/null +++ b/.gitattributes @@ -0,0 +1,37 @@ +*.7z filter=lfs diff=lfs merge=lfs -text +*.arrow filter=lfs diff=lfs merge=lfs -text +*.bin filter=lfs diff=lfs merge=lfs -text +*.bz2 filter=lfs diff=lfs merge=lfs -text +*.ckpt filter=lfs diff=lfs merge=lfs -text +*.ftz filter=lfs diff=lfs merge=lfs -text +*.gz filter=lfs diff=lfs merge=lfs -text +*.h5 filter=lfs diff=lfs merge=lfs -text +*.joblib filter=lfs diff=lfs merge=lfs -text +*.lfs.* filter=lfs diff=lfs merge=lfs -text +*.mlmodel filter=lfs diff=lfs merge=lfs -text +*.model filter=lfs diff=lfs merge=lfs -text +*.msgpack filter=lfs diff=lfs merge=lfs -text +*.npy filter=lfs diff=lfs merge=lfs -text +*.npz filter=lfs diff=lfs merge=lfs -text +*.onnx filter=lfs diff=lfs merge=lfs -text +*.ot filter=lfs diff=lfs merge=lfs -text +*.parquet filter=lfs diff=lfs merge=lfs -text +*.pb filter=lfs diff=lfs merge=lfs -text +*.pickle filter=lfs diff=lfs merge=lfs -text +*.pkl filter=lfs diff=lfs merge=lfs -text +*.pt filter=lfs diff=lfs merge=lfs -text +*.pth filter=lfs diff=lfs merge=lfs -text +*.rar filter=lfs diff=lfs merge=lfs -text +*.safetensors filter=lfs diff=lfs merge=lfs -text +saved_model/**/* filter=lfs diff=lfs merge=lfs -text +*.tar.* filter=lfs diff=lfs merge=lfs -text +*.tar filter=lfs diff=lfs merge=lfs -text +*.tflite filter=lfs diff=lfs merge=lfs -text +*.tgz filter=lfs diff=lfs merge=lfs -text +*.wasm filter=lfs diff=lfs merge=lfs -text +*.xz 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 +tokenizer.json filter=lfs diff=lfs merge=lfs -text +checkpoint-500/tokenizer.json filter=lfs diff=lfs merge=lfs -text diff --git a/README.md b/README.md new file mode 100644 index 0000000..e974135 --- /dev/null +++ b/README.md @@ -0,0 +1,69 @@ +--- +base_model: princeton-nlp/Llama-3-Base-8B-SFT +library_name: transformers +model_name: llama-3-8b-dpo-tw31-beta-1e-0-ift +tags: +- generated_from_trainer +- trl +- dpo +licence: license +--- + +# Model Card for llama-3-8b-dpo-tw31-beta-1e-0-ift + +This model is a fine-tuned version of [princeton-nlp/Llama-3-Base-8B-SFT](https://huggingface.co/princeton-nlp/Llama-3-Base-8B-SFT). +It has been trained using [TRL](https://github.com/huggingface/trl). + +## Quick start + +```python +from transformers import pipeline + +question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?" +generator = pipeline("text-generation", model="HCY123902/llama-3-8b-dpo-tw31-beta-1e-0-ift", device="cuda") +output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0] +print(output["generated_text"]) +``` + +## Training procedure + +[Visualize in Weights & Biases](https://wandb.ai/2320032466hchy/attention_dpo/runs/4nvq1qsy) + + +This model was trained with DPO, a method introduced in [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://huggingface.co/papers/2305.18290). + +### Framework versions + +- TRL: 0.20.0 +- Transformers: 4.54.1 +- Pytorch: 2.7.1+cu128 +- Datasets: 3.6.0 +- Tokenizers: 0.21.1 + +## Citations + +Cite DPO as: + +```bibtex +@inproceedings{rafailov2023direct, + title = {{Direct Preference Optimization: Your Language Model is Secretly a Reward Model}}, + author = {Rafael Rafailov and Archit Sharma and Eric Mitchell and Christopher D. Manning and Stefano Ermon and Chelsea Finn}, + year = 2023, + booktitle = {Advances in Neural Information Processing Systems 36: Annual Conference on Neural Information Processing Systems 2023, NeurIPS 2023, New Orleans, LA, USA, December 10 - 16, 2023}, + url = {http://papers.nips.cc/paper_files/paper/2023/hash/a85b405ed65c6477a4fe8302b5e06ce7-Abstract-Conference.html}, + editor = {Alice Oh and Tristan Naumann and Amir Globerson and Kate Saenko and Moritz Hardt and Sergey Levine}, +} +``` + +Cite TRL as: + +```bibtex +@misc{vonwerra2022trl, + title = {{TRL: Transformer Reinforcement Learning}}, + author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallou{\'e}dec}, + year = 2020, + journal = {GitHub repository}, + publisher = {GitHub}, + howpublished = {\url{https://github.com/huggingface/trl}} +} +``` \ No newline at end of file diff --git a/chat_template.jinja b/chat_template.jinja new file mode 100644 index 0000000..39bd0c9 --- /dev/null +++ b/chat_template.jinja @@ -0,0 +1,5 @@ +{% set loop_messages = messages %}{% for message in loop_messages %}{% set content = '<|start_header_id|>' + message['role'] + '<|end_header_id|> + +'+ message['content'] | trim + '<|eot_id|>' %}{% if loop.index0 == 0 %}{% set content = bos_token + content %}{% endif %}{{ content }}{% endfor %}{% if add_generation_prompt %}{{ '<|start_header_id|>assistant<|end_header_id|> + +' }}{% endif %} \ No newline at end of file diff --git a/checkpoint-1000/chat_template.jinja b/checkpoint-1000/chat_template.jinja new file mode 100644 index 0000000..39bd0c9 --- /dev/null +++ b/checkpoint-1000/chat_template.jinja @@ -0,0 +1,5 @@ +{% set loop_messages = messages %}{% for message in loop_messages %}{% set content = '<|start_header_id|>' + message['role'] + '<|end_header_id|> + +'+ message['content'] | trim + '<|eot_id|>' %}{% if loop.index0 == 0 %}{% set content = bos_token + content %}{% endif %}{{ content }}{% endfor %}{% if add_generation_prompt %}{{ '<|start_header_id|>assistant<|end_header_id|> + +' }}{% endif %} \ No newline at end of file diff --git a/checkpoint-1000/config.json b/checkpoint-1000/config.json new file mode 100644 index 0000000..3f8f5c0 --- /dev/null +++ b/checkpoint-1000/config.json @@ -0,0 +1,29 @@ +{ + "architectures": [ + "LlamaForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "bos_token_id": 128000, + "eos_token_id": 128001, + "head_dim": 128, + "hidden_act": "silu", + "hidden_size": 4096, + "initializer_range": 0.02, + "intermediate_size": 14336, + "max_position_embeddings": 8192, + "mlp_bias": false, + "model_type": "llama", + "num_attention_heads": 32, + "num_hidden_layers": 32, + "num_key_value_heads": 8, + "pretraining_tp": 1, + "rms_norm_eps": 1e-05, + "rope_scaling": null, + "rope_theta": 500000.0, + "tie_word_embeddings": false, + "torch_dtype": "bfloat16", + "transformers_version": "4.54.1", + "use_cache": true, + "vocab_size": 128256 +} diff --git a/checkpoint-1000/generation_config.json b/checkpoint-1000/generation_config.json new file mode 100644 index 0000000..fc3c54a --- /dev/null +++ b/checkpoint-1000/generation_config.json @@ -0,0 +1,6 @@ +{ + "_from_model_config": true, + "bos_token_id": 128000, + "eos_token_id": 128001, + "transformers_version": "4.54.1" +} diff --git a/checkpoint-1000/latest b/checkpoint-1000/latest new file mode 100644 index 0000000..e2d3435 --- /dev/null +++ b/checkpoint-1000/latest @@ -0,0 +1 @@ +global_step1000 \ No newline at end of file diff --git a/checkpoint-1000/model-00001-of-00004.safetensors b/checkpoint-1000/model-00001-of-00004.safetensors new file mode 100644 index 0000000..cc3cf36 --- /dev/null +++ b/checkpoint-1000/model-00001-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:05fd13a63d492f4c95cceeb355969db2e0fe2e3ad7024dd4e1ba276fa6161920 +size 4976698672 diff --git a/checkpoint-1000/model-00002-of-00004.safetensors b/checkpoint-1000/model-00002-of-00004.safetensors new file mode 100644 index 0000000..a31db71 --- /dev/null +++ b/checkpoint-1000/model-00002-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:45407d57ec097d388d8a8aeab818f8d8e336355d3d238b3552b4fd759f7797c2 +size 4999802720 diff --git a/checkpoint-1000/model-00003-of-00004.safetensors b/checkpoint-1000/model-00003-of-00004.safetensors new file mode 100644 index 0000000..ef3aa7a --- /dev/null +++ b/checkpoint-1000/model-00003-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:94e7fcedd9f0e33b44dea6cbc66e809278d0b4ac342bdb2cd38491a469fbfea6 +size 4915916176 diff --git a/checkpoint-1000/model-00004-of-00004.safetensors b/checkpoint-1000/model-00004-of-00004.safetensors new file mode 100644 index 0000000..8465c03 --- /dev/null +++ b/checkpoint-1000/model-00004-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:26dd484512c940e5bb2ea4b459e94840f4be3f3084fbe0f686f63776afd0ff1b +size 1168138808 diff --git a/checkpoint-1000/model.safetensors.index.json b/checkpoint-1000/model.safetensors.index.json new file mode 100644 index 0000000..58a7ef4 --- /dev/null +++ b/checkpoint-1000/model.safetensors.index.json @@ -0,0 +1,299 @@ +{ + "metadata": { + "total_parameters": 266240, + "total_size": 16060522496 + }, + "weight_map": { + "lm_head.weight": "model-00004-of-00004.safetensors", + "model.embed_tokens.weight": "model-00001-of-00004.safetensors", + "model.layers.0.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.0.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.1.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.10.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.10.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.10.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.11.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.11.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.12.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.12.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.13.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.13.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.14.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.14.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.15.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.15.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.16.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.16.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.17.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.17.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.18.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.18.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.19.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.19.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.2.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.2.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.20.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.20.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.20.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.20.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.20.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.20.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.20.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.20.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.20.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.21.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.21.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.21.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.22.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.22.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.23.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.23.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.24.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.24.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.25.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.25.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.26.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.26.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.27.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.27.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.28.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.28.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.29.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.29.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.3.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.3.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.30.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.30.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.30.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.input_layernorm.weight": "model-00004-of-00004.safetensors", + "model.layers.31.mlp.down_proj.weight": "model-00004-of-00004.safetensors", + "model.layers.31.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.post_attention_layernorm.weight": "model-00004-of-00004.safetensors", + "model.layers.31.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.4.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.4.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.5.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.6.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.7.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.8.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.9.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.9.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.9.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.norm.weight": "model-00004-of-00004.safetensors" + } +} diff --git a/checkpoint-1000/rng_state_0.pth b/checkpoint-1000/rng_state_0.pth new file mode 100644 index 0000000..37c843b --- /dev/null +++ b/checkpoint-1000/rng_state_0.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ddb2e15baa025bd61fe183f8e343e7ff470b9a01aecd8defcf155a1cf00393e3 +size 14917 diff --git a/checkpoint-1000/rng_state_1.pth b/checkpoint-1000/rng_state_1.pth new file mode 100644 index 0000000..d8eccdd --- /dev/null +++ b/checkpoint-1000/rng_state_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:07ee73a4738a457f3198cccec25cf12377bb1eba6c29e95c9fecf83c1487d401 +size 14917 diff --git a/checkpoint-1000/scheduler.pt b/checkpoint-1000/scheduler.pt new file mode 100644 index 0000000..7c34b16 --- /dev/null +++ b/checkpoint-1000/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2f1b2374548a5ec63cdcc0490db6ed738cd23c8550fb53eb0592574609549746 +size 1465 diff --git a/checkpoint-1000/special_tokens_map.json b/checkpoint-1000/special_tokens_map.json new file mode 100644 index 0000000..e5b39b6 --- /dev/null +++ b/checkpoint-1000/special_tokens_map.json @@ -0,0 +1,23 @@ +{ + "bos_token": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "eos_token": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "pad_token": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + } +} diff --git a/checkpoint-1000/tokenizer.json b/checkpoint-1000/tokenizer.json new file mode 100644 index 0000000..03aa64f --- /dev/null +++ b/checkpoint-1000/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0968dcc0ee8e56c7dccd34a7f51f8065ea0cb9e2cc529e3243d1e5c0a4bdaa0c +size 17208754 diff --git a/checkpoint-1000/tokenizer_config.json b/checkpoint-1000/tokenizer_config.json new file mode 100644 index 0000000..877a9a9 --- /dev/null +++ b/checkpoint-1000/tokenizer_config.json @@ -0,0 +1,2063 @@ +{ + "added_tokens_decoder": { + "128000": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128001": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128002": { + "content": "<|reserved_special_token_0|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128003": { + "content": "<|reserved_special_token_1|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128004": { + "content": "<|reserved_special_token_2|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128005": { + "content": "<|reserved_special_token_3|>", + "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": "<|reserved_special_token_4|>", + "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": "<|reserved_special_token_5|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128011": { + "content": "<|reserved_special_token_6|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128012": { + "content": "<|reserved_special_token_7|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128013": { + "content": "<|reserved_special_token_8|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128014": { + "content": "<|reserved_special_token_9|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128015": { + "content": "<|reserved_special_token_10|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128016": { + "content": "<|reserved_special_token_11|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128017": { + "content": "<|reserved_special_token_12|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128018": { + "content": "<|reserved_special_token_13|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128019": { + "content": "<|reserved_special_token_14|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128020": { + "content": "<|reserved_special_token_15|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128021": { + "content": "<|reserved_special_token_16|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128022": { + "content": "<|reserved_special_token_17|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128023": { + "content": "<|reserved_special_token_18|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128024": { + "content": "<|reserved_special_token_19|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128025": { + "content": "<|reserved_special_token_20|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128026": { + "content": "<|reserved_special_token_21|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128027": { + "content": "<|reserved_special_token_22|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128028": { + "content": "<|reserved_special_token_23|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128029": { + "content": "<|reserved_special_token_24|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128030": { + "content": "<|reserved_special_token_25|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128031": { + "content": "<|reserved_special_token_26|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128032": { + "content": "<|reserved_special_token_27|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128033": { + "content": "<|reserved_special_token_28|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128034": { + "content": "<|reserved_special_token_29|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128035": { + "content": "<|reserved_special_token_30|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128036": { + "content": "<|reserved_special_token_31|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128037": { + "content": "<|reserved_special_token_32|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128038": { + "content": "<|reserved_special_token_33|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128039": { + "content": "<|reserved_special_token_34|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128040": { + "content": "<|reserved_special_token_35|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128041": { + "content": "<|reserved_special_token_36|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128042": { + "content": "<|reserved_special_token_37|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128043": { + "content": "<|reserved_special_token_38|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128044": { + "content": "<|reserved_special_token_39|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128045": { + "content": "<|reserved_special_token_40|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128046": { + "content": "<|reserved_special_token_41|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128047": { + "content": "<|reserved_special_token_42|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128048": { + "content": "<|reserved_special_token_43|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128049": { + "content": "<|reserved_special_token_44|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128050": { + "content": "<|reserved_special_token_45|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128051": { + "content": "<|reserved_special_token_46|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128052": { + "content": "<|reserved_special_token_47|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128053": { + "content": "<|reserved_special_token_48|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128054": { + "content": "<|reserved_special_token_49|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128055": { + "content": "<|reserved_special_token_50|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128056": { + "content": "<|reserved_special_token_51|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128057": { + "content": "<|reserved_special_token_52|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128058": { + "content": "<|reserved_special_token_53|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128059": { + "content": "<|reserved_special_token_54|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128060": { + "content": "<|reserved_special_token_55|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128061": { + "content": "<|reserved_special_token_56|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128062": { + "content": "<|reserved_special_token_57|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128063": { + "content": "<|reserved_special_token_58|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128064": { + "content": "<|reserved_special_token_59|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128065": { + "content": "<|reserved_special_token_60|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128066": { + "content": "<|reserved_special_token_61|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128067": { + "content": "<|reserved_special_token_62|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128068": { + "content": "<|reserved_special_token_63|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128069": { + "content": "<|reserved_special_token_64|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128070": { + "content": "<|reserved_special_token_65|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128071": { + "content": "<|reserved_special_token_66|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128072": { + "content": "<|reserved_special_token_67|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128073": { + "content": "<|reserved_special_token_68|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128074": { + "content": "<|reserved_special_token_69|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128075": { + "content": "<|reserved_special_token_70|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128076": { + "content": "<|reserved_special_token_71|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128077": { + "content": "<|reserved_special_token_72|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128078": { + "content": "<|reserved_special_token_73|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128079": { + "content": "<|reserved_special_token_74|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128080": { + "content": "<|reserved_special_token_75|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128081": { + "content": "<|reserved_special_token_76|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128082": { + "content": "<|reserved_special_token_77|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128083": { + "content": "<|reserved_special_token_78|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128084": { + "content": "<|reserved_special_token_79|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128085": { + "content": "<|reserved_special_token_80|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128086": { + "content": "<|reserved_special_token_81|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128087": { + "content": "<|reserved_special_token_82|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128088": { + "content": "<|reserved_special_token_83|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128089": { + "content": "<|reserved_special_token_84|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128090": { + "content": "<|reserved_special_token_85|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128091": { + "content": "<|reserved_special_token_86|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128092": { + "content": "<|reserved_special_token_87|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128093": { + "content": "<|reserved_special_token_88|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128094": { + "content": "<|reserved_special_token_89|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128095": { + "content": "<|reserved_special_token_90|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128096": { + "content": "<|reserved_special_token_91|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128097": { + "content": "<|reserved_special_token_92|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128098": { + "content": "<|reserved_special_token_93|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128099": { + "content": "<|reserved_special_token_94|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128100": { + "content": "<|reserved_special_token_95|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128101": { + "content": "<|reserved_special_token_96|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128102": { + "content": "<|reserved_special_token_97|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128103": { + "content": "<|reserved_special_token_98|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128104": { + "content": "<|reserved_special_token_99|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128105": { + "content": "<|reserved_special_token_100|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128106": { + "content": "<|reserved_special_token_101|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128107": { + "content": "<|reserved_special_token_102|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128108": { + "content": "<|reserved_special_token_103|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128109": { + "content": "<|reserved_special_token_104|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128110": { + "content": "<|reserved_special_token_105|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128111": { + "content": "<|reserved_special_token_106|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128112": { + "content": "<|reserved_special_token_107|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128113": { + "content": "<|reserved_special_token_108|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128114": { + "content": "<|reserved_special_token_109|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128115": { + "content": "<|reserved_special_token_110|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128116": { + "content": "<|reserved_special_token_111|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128117": { + "content": "<|reserved_special_token_112|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128118": { + "content": "<|reserved_special_token_113|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128119": { + "content": "<|reserved_special_token_114|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128120": { + "content": "<|reserved_special_token_115|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128121": { + "content": "<|reserved_special_token_116|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128122": { + "content": "<|reserved_special_token_117|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128123": { + "content": "<|reserved_special_token_118|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128124": { + "content": "<|reserved_special_token_119|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128125": { + "content": "<|reserved_special_token_120|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128126": { + "content": "<|reserved_special_token_121|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128127": { + "content": "<|reserved_special_token_122|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128128": { + "content": "<|reserved_special_token_123|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128129": { + "content": "<|reserved_special_token_124|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128130": { + "content": "<|reserved_special_token_125|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128131": { + "content": "<|reserved_special_token_126|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128132": { + "content": "<|reserved_special_token_127|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128133": { + "content": "<|reserved_special_token_128|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128134": { + "content": "<|reserved_special_token_129|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128135": { + "content": "<|reserved_special_token_130|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128136": { + "content": "<|reserved_special_token_131|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128137": { + "content": "<|reserved_special_token_132|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128138": { + "content": "<|reserved_special_token_133|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128139": { + "content": "<|reserved_special_token_134|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128140": { + "content": "<|reserved_special_token_135|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128141": { + "content": "<|reserved_special_token_136|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128142": { + "content": "<|reserved_special_token_137|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128143": { + "content": "<|reserved_special_token_138|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128144": { + "content": "<|reserved_special_token_139|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128145": { + "content": "<|reserved_special_token_140|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128146": { + "content": "<|reserved_special_token_141|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128147": { + "content": "<|reserved_special_token_142|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128148": { + "content": "<|reserved_special_token_143|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128149": { + "content": "<|reserved_special_token_144|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128150": { + "content": "<|reserved_special_token_145|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128151": { + "content": "<|reserved_special_token_146|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128152": { + "content": "<|reserved_special_token_147|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128153": { + "content": "<|reserved_special_token_148|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128154": { + "content": "<|reserved_special_token_149|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128155": { + "content": "<|reserved_special_token_150|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128156": { + "content": "<|reserved_special_token_151|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128157": { + "content": "<|reserved_special_token_152|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128158": { + "content": "<|reserved_special_token_153|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128159": { + "content": "<|reserved_special_token_154|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128160": { + "content": "<|reserved_special_token_155|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128161": { + "content": "<|reserved_special_token_156|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128162": { + "content": "<|reserved_special_token_157|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128163": { + "content": "<|reserved_special_token_158|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128164": { + "content": "<|reserved_special_token_159|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128165": { + "content": "<|reserved_special_token_160|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128166": { + "content": "<|reserved_special_token_161|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128167": { + "content": "<|reserved_special_token_162|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128168": { + "content": "<|reserved_special_token_163|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128169": { + "content": "<|reserved_special_token_164|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128170": { + "content": "<|reserved_special_token_165|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128171": { + "content": "<|reserved_special_token_166|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128172": { + "content": "<|reserved_special_token_167|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128173": { + "content": "<|reserved_special_token_168|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128174": { + "content": "<|reserved_special_token_169|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128175": { + "content": "<|reserved_special_token_170|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128176": { + "content": "<|reserved_special_token_171|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128177": { + "content": "<|reserved_special_token_172|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128178": { + "content": "<|reserved_special_token_173|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128179": { + "content": "<|reserved_special_token_174|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128180": { + "content": "<|reserved_special_token_175|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128181": { + "content": "<|reserved_special_token_176|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128182": { + "content": "<|reserved_special_token_177|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128183": { + "content": "<|reserved_special_token_178|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128184": { + "content": "<|reserved_special_token_179|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128185": { + "content": "<|reserved_special_token_180|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128186": { + "content": "<|reserved_special_token_181|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128187": { + "content": "<|reserved_special_token_182|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128188": { + "content": "<|reserved_special_token_183|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128189": { + "content": "<|reserved_special_token_184|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128190": { + "content": "<|reserved_special_token_185|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128191": { + "content": "<|reserved_special_token_186|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128192": { + "content": "<|reserved_special_token_187|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128193": { + "content": "<|reserved_special_token_188|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128194": { + "content": "<|reserved_special_token_189|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128195": { + "content": "<|reserved_special_token_190|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128196": { + "content": "<|reserved_special_token_191|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128197": { + "content": "<|reserved_special_token_192|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128198": { + "content": "<|reserved_special_token_193|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128199": { + "content": "<|reserved_special_token_194|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128200": { + "content": "<|reserved_special_token_195|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128201": { + "content": "<|reserved_special_token_196|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128202": { + "content": "<|reserved_special_token_197|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128203": { + "content": "<|reserved_special_token_198|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128204": { + "content": "<|reserved_special_token_199|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128205": { + "content": "<|reserved_special_token_200|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128206": { + "content": "<|reserved_special_token_201|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128207": { + "content": "<|reserved_special_token_202|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128208": { + "content": "<|reserved_special_token_203|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128209": { + "content": "<|reserved_special_token_204|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128210": { + "content": "<|reserved_special_token_205|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128211": { + "content": "<|reserved_special_token_206|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128212": { + "content": "<|reserved_special_token_207|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128213": { + "content": "<|reserved_special_token_208|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128214": { + "content": "<|reserved_special_token_209|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128215": { + "content": "<|reserved_special_token_210|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128216": { + "content": "<|reserved_special_token_211|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128217": { + "content": "<|reserved_special_token_212|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128218": { + "content": "<|reserved_special_token_213|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128219": { + "content": "<|reserved_special_token_214|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128220": { + "content": "<|reserved_special_token_215|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128221": { + "content": "<|reserved_special_token_216|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128222": { + "content": "<|reserved_special_token_217|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128223": { + "content": "<|reserved_special_token_218|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128224": { + "content": "<|reserved_special_token_219|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128225": { + "content": "<|reserved_special_token_220|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128226": { + "content": "<|reserved_special_token_221|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128227": { + "content": "<|reserved_special_token_222|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128228": { + "content": "<|reserved_special_token_223|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128229": { + "content": "<|reserved_special_token_224|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128230": { + "content": "<|reserved_special_token_225|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128231": { + "content": "<|reserved_special_token_226|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128232": { + "content": "<|reserved_special_token_227|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128233": { + "content": "<|reserved_special_token_228|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128234": { + "content": "<|reserved_special_token_229|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128235": { + "content": "<|reserved_special_token_230|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128236": { + "content": "<|reserved_special_token_231|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128237": { + "content": "<|reserved_special_token_232|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128238": { + "content": "<|reserved_special_token_233|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128239": { + "content": "<|reserved_special_token_234|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128240": { + "content": "<|reserved_special_token_235|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128241": { + "content": "<|reserved_special_token_236|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128242": { + "content": "<|reserved_special_token_237|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128243": { + "content": "<|reserved_special_token_238|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128244": { + "content": "<|reserved_special_token_239|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128245": { + "content": "<|reserved_special_token_240|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128246": { + "content": "<|reserved_special_token_241|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128247": { + "content": "<|reserved_special_token_242|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128248": { + "content": "<|reserved_special_token_243|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128249": { + "content": "<|reserved_special_token_244|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128250": { + "content": "<|reserved_special_token_245|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128251": { + "content": "<|reserved_special_token_246|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128252": { + "content": "<|reserved_special_token_247|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128253": { + "content": "<|reserved_special_token_248|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128254": { + "content": "<|reserved_special_token_249|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128255": { + "content": "<|reserved_special_token_250|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + } + }, + "bos_token": "<|begin_of_text|>", + "clean_up_tokenization_spaces": true, + "eos_token": "<|end_of_text|>", + "extra_special_tokens": {}, + "model_input_names": [ + "input_ids", + "attention_mask" + ], + "model_max_length": 32768, + "pad_token": "<|end_of_text|>", + "tokenizer_class": "PreTrainedTokenizerFast" +} diff --git a/checkpoint-1000/trainer_state.json b/checkpoint-1000/trainer_state.json new file mode 100644 index 0000000..66c2089 --- /dev/null +++ b/checkpoint-1000/trainer_state.json @@ -0,0 +1,2199 @@ +{ + "best_global_step": null, + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 0.5234231876472127, + "eval_steps": 500, + "global_step": 1000, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.0005234231876472127, + "grad_norm": 136.4956817626953, + "learning_rate": 0.0, + "logits/chosen": -0.40118408203125, + "logits/rejected": -0.41802978515625, + "logps/chosen": -297.609375, + "logps/rejected": -247.84375, + "logps/weighted_chosen": -4.50634765625, + "logps/weighted_rejected": -3.43408203125, + "loss": 0.6914, + "rewards/accuracies": 0.0, + "rewards/chosen": 0.0, + "rewards/margins": 0.0, + "rewards/rejected": 0.0, + "rewards/weighted_accuracies": 0.0, + "rewards/weighted_chosen": 0.0, + "rewards/weighted_margins": 0.0, + "rewards/weighted_rejected": 0.0, + "step": 1 + }, + { + "epoch": 0.005234231876472127, + "grad_norm": 269.8279113769531, + "learning_rate": 4.6875e-08, + "logits/chosen": -0.3176456093788147, + "logits/rejected": -0.3530849814414978, + "logps/chosen": -275.5694580078125, + "logps/rejected": -255.875, + "logps/weighted_chosen": -2.478325843811035, + "logps/weighted_rejected": -2.635009765625, + "loss": 0.6911, + "rewards/accuracies": 0.2604166567325592, + "rewards/chosen": -0.0588107630610466, + "rewards/margins": -0.0334201380610466, + "rewards/rejected": -0.025390625, + "rewards/weighted_accuracies": 0.3229166567325592, + "rewards/weighted_chosen": 0.0010899438057094812, + "rewards/weighted_margins": 0.0014837053604424, + "rewards/weighted_rejected": -0.00039333768654614687, + "step": 10 + }, + { + "epoch": 0.010468463752944255, + "grad_norm": 12.950738906860352, + "learning_rate": 9.895833333333332e-08, + "logits/chosen": -0.2977302670478821, + "logits/rejected": -0.3139175474643707, + "logps/chosen": -294.37579345703125, + "logps/rejected": -272.7203063964844, + "logps/weighted_chosen": -2.30804443359375, + "logps/weighted_rejected": -2.247570753097534, + "loss": 0.692, + "rewards/accuracies": 0.296875, + "rewards/chosen": -0.05908203125, + "rewards/margins": -0.01513671875, + "rewards/rejected": -0.0439453125, + "rewards/weighted_accuracies": 0.34687501192092896, + "rewards/weighted_chosen": 7.171630568336695e-05, + "rewards/weighted_margins": 3.662109520519152e-05, + "rewards/weighted_rejected": 3.509521411615424e-05, + "step": 20 + }, + { + "epoch": 0.015702695629416383, + "grad_norm": 65.14871215820312, + "learning_rate": 1.5104166666666664e-07, + "logits/chosen": -0.2917991578578949, + "logits/rejected": -0.3373458981513977, + "logps/chosen": -297.9375, + "logps/rejected": -268.0062561035156, + "logps/weighted_chosen": -2.3156495094299316, + "logps/weighted_rejected": -2.529247999191284, + "loss": 0.693, + "rewards/accuracies": 0.28125, + "rewards/chosen": 0.02177734300494194, + "rewards/margins": -0.06005859375, + "rewards/rejected": 0.08183594048023224, + "rewards/weighted_accuracies": 0.34062498807907104, + "rewards/weighted_chosen": 0.0012060165172442794, + "rewards/weighted_margins": -0.0021120072342455387, + "rewards/weighted_rejected": 0.0033180236350744963, + "step": 30 + }, + { + "epoch": 0.02093692750588851, + "grad_norm": 30.699859619140625, + "learning_rate": 2.03125e-07, + "logits/chosen": -0.301095575094223, + "logits/rejected": -0.3433547914028168, + "logps/chosen": -278.7124938964844, + "logps/rejected": -253.84530639648438, + "logps/weighted_chosen": -2.3508667945861816, + "logps/weighted_rejected": -2.55126953125, + "loss": 0.6909, + "rewards/accuracies": 0.2718749940395355, + "rewards/chosen": -0.014892578125, + "rewards/margins": -0.105224609375, + "rewards/rejected": 0.09033203125, + "rewards/weighted_accuracies": 0.359375, + "rewards/weighted_chosen": 0.0031726837623864412, + "rewards/weighted_margins": 0.0010789871448650956, + "rewards/weighted_rejected": 0.002093696501106024, + "step": 40 + }, + { + "epoch": 0.02617115938236064, + "grad_norm": 15.275498390197754, + "learning_rate": 2.552083333333333e-07, + "logits/chosen": -0.28184205293655396, + "logits/rejected": -0.32042425870895386, + "logps/chosen": -280.19140625, + "logps/rejected": -267.4398498535156, + "logps/weighted_chosen": -2.281604051589966, + "logps/weighted_rejected": -2.278125047683716, + "loss": 0.691, + "rewards/accuracies": 0.3125, + "rewards/chosen": 0.009082031436264515, + "rewards/margins": -0.11552734673023224, + "rewards/rejected": 0.12460937350988388, + "rewards/weighted_accuracies": 0.375, + "rewards/weighted_chosen": 0.0036834715865552425, + "rewards/weighted_margins": 0.0022117614280432463, + "rewards/weighted_rejected": 0.0014717101585119963, + "step": 50 + }, + { + "epoch": 0.031405391258832765, + "grad_norm": 42.36075973510742, + "learning_rate": 3.0729166666666665e-07, + "logits/chosen": -0.3147949278354645, + "logits/rejected": -0.3081672787666321, + "logps/chosen": -277.5492248535156, + "logps/rejected": -261.63201904296875, + "logps/weighted_chosen": -2.6925292015075684, + "logps/weighted_rejected": -2.5440917015075684, + "loss": 0.6888, + "rewards/accuracies": 0.39375001192092896, + "rewards/chosen": 0.15625, + "rewards/margins": 0.06132812425494194, + "rewards/rejected": 0.09492187201976776, + "rewards/weighted_accuracies": 0.43437498807907104, + "rewards/weighted_chosen": 0.0014663696056231856, + "rewards/weighted_margins": 0.0070289610885083675, + "rewards/weighted_rejected": -0.005612182430922985, + "step": 60 + }, + { + "epoch": 0.036639623135304895, + "grad_norm": 70.38953399658203, + "learning_rate": 3.59375e-07, + "logits/chosen": -0.31798094511032104, + "logits/rejected": -0.32500457763671875, + "logps/chosen": -289.6695251464844, + "logps/rejected": -244.7351531982422, + "logps/weighted_chosen": -2.19110107421875, + "logps/weighted_rejected": -2.427661180496216, + "loss": 0.6869, + "rewards/accuracies": 0.4375, + "rewards/chosen": 0.36865234375, + "rewards/margins": 0.34619140625, + "rewards/rejected": 0.0224609375, + "rewards/weighted_accuracies": 0.5375000238418579, + "rewards/weighted_chosen": 0.02108154259622097, + "rewards/weighted_margins": 0.011628913693130016, + "rewards/weighted_rejected": 0.009454727172851562, + "step": 70 + }, + { + "epoch": 0.04187385501177702, + "grad_norm": 54.81600570678711, + "learning_rate": 4.114583333333333e-07, + "logits/chosen": -0.2848251461982727, + "logits/rejected": -0.3061889708042145, + "logps/chosen": -289.08203125, + "logps/rejected": -269.95782470703125, + "logps/weighted_chosen": -2.3393311500549316, + "logps/weighted_rejected": -2.5093994140625, + "loss": 0.6827, + "rewards/accuracies": 0.44999998807907104, + "rewards/chosen": 0.679980456829071, + "rewards/margins": 0.3672851622104645, + "rewards/rejected": 0.31269532442092896, + "rewards/weighted_accuracies": 0.5062500238418579, + "rewards/weighted_chosen": 0.048927437514066696, + "rewards/weighted_margins": 0.025808846578001976, + "rewards/weighted_rejected": 0.02311401441693306, + "step": 80 + }, + { + "epoch": 0.04710808688824915, + "grad_norm": 30.603567123413086, + "learning_rate": 4.6354166666666664e-07, + "logits/chosen": -0.312631219625473, + "logits/rejected": -0.30586013197898865, + "logps/chosen": -279.92578125, + "logps/rejected": -257.5234375, + "logps/weighted_chosen": -2.5583739280700684, + "logps/weighted_rejected": -2.6037840843200684, + "loss": 0.6752, + "rewards/accuracies": 0.49687498807907104, + "rewards/chosen": 0.757617175579071, + "rewards/margins": 0.5347656011581421, + "rewards/rejected": 0.22285155951976776, + "rewards/weighted_accuracies": 0.546875, + "rewards/weighted_chosen": 0.08651771396398544, + "rewards/weighted_margins": 0.05735473707318306, + "rewards/weighted_rejected": 0.02917785570025444, + "step": 90 + }, + { + "epoch": 0.05234231876472128, + "grad_norm": 99.52488708496094, + "learning_rate": 5.156249999999999e-07, + "logits/chosen": -0.2793869078159332, + "logits/rejected": -0.3509994447231293, + "logps/chosen": -290.19061279296875, + "logps/rejected": -285.94842529296875, + "logps/weighted_chosen": -2.013537645339966, + "logps/weighted_rejected": -2.5231690406799316, + "loss": 0.6984, + "rewards/accuracies": 0.512499988079071, + "rewards/chosen": 0.72607421875, + "rewards/margins": 0.80712890625, + "rewards/rejected": -0.0810546875, + "rewards/weighted_accuracies": 0.6187499761581421, + "rewards/weighted_chosen": 0.07289886474609375, + "rewards/weighted_margins": 0.019171524792909622, + "rewards/weighted_rejected": 0.05372200161218643, + "step": 100 + }, + { + "epoch": 0.05757655064119341, + "grad_norm": 55.06239318847656, + "learning_rate": 5.677083333333333e-07, + "logits/chosen": -0.3277145326137543, + "logits/rejected": -0.3189544677734375, + "logps/chosen": -296.17889404296875, + "logps/rejected": -261.51953125, + "logps/weighted_chosen": -2.647631883621216, + "logps/weighted_rejected": -2.6458740234375, + "loss": 0.6844, + "rewards/accuracies": 0.518750011920929, + "rewards/chosen": 0.1376953125, + "rewards/margins": 0.789746105670929, + "rewards/rejected": -0.6522461175918579, + "rewards/weighted_accuracies": 0.565625011920929, + "rewards/weighted_chosen": 0.05432138592004776, + "rewards/weighted_margins": 0.03696594387292862, + "rewards/weighted_rejected": 0.01735992357134819, + "step": 110 + }, + { + "epoch": 0.06281078251766553, + "grad_norm": 42.45036315917969, + "learning_rate": 6.197916666666666e-07, + "logits/chosen": -0.33232802152633667, + "logits/rejected": -0.35865744948387146, + "logps/chosen": -294.6078186035156, + "logps/rejected": -255.01327514648438, + "logps/weighted_chosen": -1.970422387123108, + "logps/weighted_rejected": -2.1345458030700684, + "loss": 0.6837, + "rewards/accuracies": 0.609375, + "rewards/chosen": 0.0013671874767169356, + "rewards/margins": 1.2595703601837158, + "rewards/rejected": -1.258203148841858, + "rewards/weighted_accuracies": 0.59375, + "rewards/weighted_chosen": 0.048947714269161224, + "rewards/weighted_margins": 0.03674011304974556, + "rewards/weighted_rejected": 0.012198830023407936, + "step": 120 + }, + { + "epoch": 0.06804501439413765, + "grad_norm": 28.637327194213867, + "learning_rate": 6.718749999999999e-07, + "logits/chosen": -0.2933105528354645, + "logits/rejected": -0.29020920395851135, + "logps/chosen": -305.86602783203125, + "logps/rejected": -279.09844970703125, + "logps/weighted_chosen": -1.9946777820587158, + "logps/weighted_rejected": -2.2933349609375, + "loss": 0.6754, + "rewards/accuracies": 0.625, + "rewards/chosen": 0.285400390625, + "rewards/margins": 1.7852051258087158, + "rewards/rejected": -1.4998047351837158, + "rewards/weighted_accuracies": 0.6187499761581421, + "rewards/weighted_chosen": 0.06223602220416069, + "rewards/weighted_margins": 0.05499267578125, + "rewards/weighted_rejected": 0.0072036744095385075, + "step": 130 + }, + { + "epoch": 0.07327924627060979, + "grad_norm": 20.866477966308594, + "learning_rate": 7.239583333333333e-07, + "logits/chosen": -0.33005982637405396, + "logits/rejected": -0.34638214111328125, + "logps/chosen": -300.7281188964844, + "logps/rejected": -276.9156188964844, + "logps/weighted_chosen": -2.294506788253784, + "logps/weighted_rejected": -2.561084032058716, + "loss": 0.6606, + "rewards/accuracies": 0.628125011920929, + "rewards/chosen": -1.998437523841858, + "rewards/margins": 2.3173828125, + "rewards/rejected": -4.315820217132568, + "rewards/weighted_accuracies": 0.671875, + "rewards/weighted_chosen": 0.04616241529583931, + "rewards/weighted_margins": 0.08864898979663849, + "rewards/weighted_rejected": -0.04253540188074112, + "step": 140 + }, + { + "epoch": 0.07851347814708191, + "grad_norm": 43.497623443603516, + "learning_rate": 7.760416666666666e-07, + "logits/chosen": -0.2890525758266449, + "logits/rejected": -0.30130767822265625, + "logps/chosen": -287.3109436035156, + "logps/rejected": -256.72967529296875, + "logps/weighted_chosen": -2.883862257003784, + "logps/weighted_rejected": -2.796630859375, + "loss": 0.664, + "rewards/accuracies": 0.628125011920929, + "rewards/chosen": -3.108203172683716, + "rewards/margins": 2.7275390625, + "rewards/rejected": -5.835741996765137, + "rewards/weighted_accuracies": 0.6343749761581421, + "rewards/weighted_chosen": 0.04738159105181694, + "rewards/weighted_margins": 0.09773864597082138, + "rewards/weighted_rejected": -0.05034027248620987, + "step": 150 + }, + { + "epoch": 0.08374771002355404, + "grad_norm": 12.989988327026367, + "learning_rate": 8.28125e-07, + "logits/chosen": -0.3334007263183594, + "logits/rejected": -0.3831237852573395, + "logps/chosen": -308.2320251464844, + "logps/rejected": -281.5335998535156, + "logps/weighted_chosen": -2.378857374191284, + "logps/weighted_rejected": -2.3824095726013184, + "loss": 0.6464, + "rewards/accuracies": 0.6499999761581421, + "rewards/chosen": -4.347460746765137, + "rewards/margins": 3.98388671875, + "rewards/rejected": -8.331347465515137, + "rewards/weighted_accuracies": 0.65625, + "rewards/weighted_chosen": 0.04017028957605362, + "rewards/weighted_margins": 0.15716858208179474, + "rewards/weighted_rejected": -0.11710510402917862, + "step": 160 + }, + { + "epoch": 0.08898194190002617, + "grad_norm": 691.15625, + "learning_rate": 8.802083333333333e-07, + "logits/chosen": -0.37783128023147583, + "logits/rejected": -0.37789613008499146, + "logps/chosen": -310.76953125, + "logps/rejected": -270.8374938964844, + "logps/weighted_chosen": -2.6180419921875, + "logps/weighted_rejected": -2.8051514625549316, + "loss": 0.683, + "rewards/accuracies": 0.6156250238418579, + "rewards/chosen": -7.780859470367432, + "rewards/margins": 4.9013671875, + "rewards/rejected": -12.683398246765137, + "rewards/weighted_accuracies": 0.643750011920929, + "rewards/weighted_chosen": 0.06108245998620987, + "rewards/weighted_margins": 0.17230224609375, + "rewards/weighted_rejected": -0.11125946044921875, + "step": 170 + }, + { + "epoch": 0.0942161737764983, + "grad_norm": 65.56312561035156, + "learning_rate": 9.322916666666666e-07, + "logits/chosen": -0.3433074951171875, + "logits/rejected": -0.363290399312973, + "logps/chosen": -285.8812561035156, + "logps/rejected": -274.94842529296875, + "logps/weighted_chosen": -2.50494384765625, + "logps/weighted_rejected": -2.9246826171875, + "loss": 0.6821, + "rewards/accuracies": 0.6343749761581421, + "rewards/chosen": -12.179101943969727, + "rewards/margins": 6.432421684265137, + "rewards/rejected": -18.608592987060547, + "rewards/weighted_accuracies": 0.59375, + "rewards/weighted_chosen": -0.0216827392578125, + "rewards/weighted_margins": 0.19813232123851776, + "rewards/weighted_rejected": -0.2197723388671875, + "step": 180 + }, + { + "epoch": 0.09945040565297043, + "grad_norm": 26.828750610351562, + "learning_rate": 9.84375e-07, + "logits/chosen": -0.36560744047164917, + "logits/rejected": -0.40380173921585083, + "logps/chosen": -321.97967529296875, + "logps/rejected": -287.5570373535156, + "logps/weighted_chosen": -2.3037109375, + "logps/weighted_rejected": -2.64697265625, + "loss": 0.6489, + "rewards/accuracies": 0.596875011920929, + "rewards/chosen": -16.246679306030273, + "rewards/margins": 7.440039157867432, + "rewards/rejected": -23.685937881469727, + "rewards/weighted_accuracies": 0.640625, + "rewards/weighted_chosen": -0.11536254733800888, + "rewards/weighted_margins": 0.19368895888328552, + "rewards/weighted_rejected": -0.309121698141098, + "step": 190 + }, + { + "epoch": 0.10468463752944256, + "grad_norm": 117.84883880615234, + "learning_rate": 9.99959085414323e-07, + "logits/chosen": -0.4036270081996918, + "logits/rejected": -0.43666380643844604, + "logps/chosen": -327.8265686035156, + "logps/rejected": -284.4234313964844, + "logps/weighted_chosen": -2.6368165016174316, + "logps/weighted_rejected": -3.0156006813049316, + "loss": 0.6383, + "rewards/accuracies": 0.6156250238418579, + "rewards/chosen": -18.538379669189453, + "rewards/margins": 8.212695121765137, + "rewards/rejected": -26.748046875, + "rewards/weighted_accuracies": 0.6468750238418579, + "rewards/weighted_chosen": -0.12391586601734161, + "rewards/weighted_margins": 0.28039854764938354, + "rewards/weighted_rejected": -0.404205322265625, + "step": 200 + }, + { + "epoch": 0.10991886940591468, + "grad_norm": 15.525572776794434, + "learning_rate": 9.997587035630105e-07, + "logits/chosen": -0.41632765531539917, + "logits/rejected": -0.4597106873989105, + "logps/chosen": -308.75469970703125, + "logps/rejected": -316.69842529296875, + "logps/weighted_chosen": -2.439282178878784, + "logps/weighted_rejected": -3.152050733566284, + "loss": 0.6437, + "rewards/accuracies": 0.653124988079071, + "rewards/chosen": -21.632617950439453, + "rewards/margins": 10.560155868530273, + "rewards/rejected": -32.18867111206055, + "rewards/weighted_accuracies": 0.671875, + "rewards/weighted_chosen": -0.2110244780778885, + "rewards/weighted_margins": 0.23976440727710724, + "rewards/weighted_rejected": -0.4508819580078125, + "step": 210 + }, + { + "epoch": 0.11515310128238682, + "grad_norm": 31.583986282348633, + "learning_rate": 9.99391406364405e-07, + "logits/chosen": -0.37994080781936646, + "logits/rejected": -0.384744256734848, + "logps/chosen": -318.63592529296875, + "logps/rejected": -305.0, + "logps/weighted_chosen": -2.9573731422424316, + "logps/weighted_rejected": -3.207348585128784, + "loss": 0.7259, + "rewards/accuracies": 0.606249988079071, + "rewards/chosen": -26.522266387939453, + "rewards/margins": 9.657129287719727, + "rewards/rejected": -36.18437576293945, + "rewards/weighted_accuracies": 0.625, + "rewards/weighted_chosen": -0.3494918942451477, + "rewards/weighted_margins": 0.17980042099952698, + "rewards/weighted_rejected": -0.5295776128768921, + "step": 220 + }, + { + "epoch": 0.12038733315885894, + "grad_norm": 611.2006225585938, + "learning_rate": 9.988573164927884e-07, + "logits/chosen": -0.3254447877407074, + "logits/rejected": -0.36442261934280396, + "logps/chosen": -297.16717529296875, + "logps/rejected": -296.48907470703125, + "logps/weighted_chosen": -2.35888671875, + "logps/weighted_rejected": -2.804126024246216, + "loss": 0.6637, + "rewards/accuracies": 0.6656249761581421, + "rewards/chosen": -28.293750762939453, + "rewards/margins": 16.522851943969727, + "rewards/rejected": -44.817970275878906, + "rewards/weighted_accuracies": 0.6812499761581421, + "rewards/weighted_chosen": -0.1478603333234787, + "rewards/weighted_margins": 0.3882904052734375, + "rewards/weighted_rejected": -0.5360336303710938, + "step": 230 + }, + { + "epoch": 0.12562156503533106, + "grad_norm": 34.738059997558594, + "learning_rate": 9.98156612329838e-07, + "logits/chosen": -0.3876676559448242, + "logits/rejected": -0.4364303648471832, + "logps/chosen": -312.5921936035156, + "logps/rejected": -348.60626220703125, + "logps/weighted_chosen": -2.594775438308716, + "logps/weighted_rejected": -3.1068115234375, + "loss": 0.7482, + "rewards/accuracies": 0.637499988079071, + "rewards/chosen": -47.75507736206055, + "rewards/margins": 18.762109756469727, + "rewards/rejected": -66.5289077758789, + "rewards/weighted_accuracies": 0.6499999761581421, + "rewards/weighted_chosen": -0.319406121969223, + "rewards/weighted_margins": 0.204498291015625, + "rewards/weighted_rejected": -0.5241454839706421, + "step": 240 + }, + { + "epoch": 0.13085579691180318, + "grad_norm": 141.2784423828125, + "learning_rate": 9.97289527905053e-07, + "logits/chosen": -0.4159797728061676, + "logits/rejected": -0.4382568299770355, + "logps/chosen": -308.44842529296875, + "logps/rejected": -313.49530029296875, + "logps/weighted_chosen": -2.802807569503784, + "logps/weighted_rejected": -2.9874024391174316, + "loss": 0.7583, + "rewards/accuracies": 0.609375, + "rewards/chosen": -44.16621017456055, + "rewards/margins": 13.382226943969727, + "rewards/rejected": -57.54296875, + "rewards/weighted_accuracies": 0.6187499761581421, + "rewards/weighted_chosen": -0.20551452040672302, + "rewards/weighted_margins": 0.22150878608226776, + "rewards/weighted_rejected": -0.42730408906936646, + "step": 250 + }, + { + "epoch": 0.1360900287882753, + "grad_norm": 75.63247680664062, + "learning_rate": 9.962563528175875e-07, + "logits/chosen": -0.4156021177768707, + "logits/rejected": -0.4493362307548523, + "logps/chosen": -342.1312561035156, + "logps/rejected": -317.84063720703125, + "logps/weighted_chosen": -2.56585693359375, + "logps/weighted_rejected": -3.4276366233825684, + "loss": 0.6673, + "rewards/accuracies": 0.5625, + "rewards/chosen": -43.44648361206055, + "rewards/margins": 13.459375381469727, + "rewards/rejected": -56.90898513793945, + "rewards/weighted_accuracies": 0.6156250238418579, + "rewards/weighted_chosen": -0.310464471578598, + "rewards/weighted_margins": 0.32973939180374146, + "rewards/weighted_rejected": -0.6404876708984375, + "step": 260 + }, + { + "epoch": 0.14132426066474746, + "grad_norm": 32.991607666015625, + "learning_rate": 9.950574321395277e-07, + "logits/chosen": -0.507153332233429, + "logits/rejected": -0.5264068841934204, + "logps/chosen": -317.0335998535156, + "logps/rejected": -300.5718688964844, + "logps/weighted_chosen": -2.669018507003784, + "logps/weighted_rejected": -2.8846678733825684, + "loss": 0.6774, + "rewards/accuracies": 0.609375, + "rewards/chosen": -27.629491806030273, + "rewards/margins": 9.762109756469727, + "rewards/rejected": -37.39765548706055, + "rewards/weighted_accuracies": 0.6000000238418579, + "rewards/weighted_chosen": -0.23316803574562073, + "rewards/weighted_margins": 0.2321929931640625, + "rewards/weighted_rejected": -0.46527099609375, + "step": 270 + }, + { + "epoch": 0.14655849254121958, + "grad_norm": 41.87471389770508, + "learning_rate": 9.936931663006413e-07, + "logits/chosen": -0.5015045404434204, + "logits/rejected": -0.49525564908981323, + "logps/chosen": -319.9296875, + "logps/rejected": -312.28826904296875, + "logps/weighted_chosen": -2.667797803878784, + "logps/weighted_rejected": -3.121386766433716, + "loss": 0.6458, + "rewards/accuracies": 0.6625000238418579, + "rewards/chosen": -17.077342987060547, + "rewards/margins": 15.751367568969727, + "rewards/rejected": -32.82636642456055, + "rewards/weighted_accuracies": 0.703125, + "rewards/weighted_chosen": -0.10239715874195099, + "rewards/weighted_margins": 0.3067169189453125, + "rewards/weighted_rejected": -0.409027099609375, + "step": 280 + }, + { + "epoch": 0.1517927244176917, + "grad_norm": 22.655261993408203, + "learning_rate": 9.921640109546357e-07, + "logits/chosen": -0.4936012327671051, + "logits/rejected": -0.5640270113945007, + "logps/chosen": -286.4117126464844, + "logps/rejected": -286.15313720703125, + "logps/weighted_chosen": -2.640380859375, + "logps/weighted_rejected": -3.774829149246216, + "loss": 0.6369, + "rewards/accuracies": 0.6343749761581421, + "rewards/chosen": -19.629491806030273, + "rewards/margins": 14.98291015625, + "rewards/rejected": -34.61503982543945, + "rewards/weighted_accuracies": 0.668749988079071, + "rewards/weighted_chosen": -0.21005859971046448, + "rewards/weighted_margins": 0.3854110836982727, + "rewards/weighted_rejected": -0.5955780148506165, + "step": 290 + }, + { + "epoch": 0.15702695629416383, + "grad_norm": 33.83458709716797, + "learning_rate": 9.90470476826975e-07, + "logits/chosen": -0.5687958002090454, + "logits/rejected": -0.5700439214706421, + "logps/chosen": -298.06561279296875, + "logps/rejected": -311.8359375, + "logps/weighted_chosen": -2.555248975753784, + "logps/weighted_rejected": -3.0333251953125, + "loss": 0.6326, + "rewards/accuracies": 0.6625000238418579, + "rewards/chosen": -28.754297256469727, + "rewards/margins": 15.813085556030273, + "rewards/rejected": -44.567771911621094, + "rewards/weighted_accuracies": 0.6468750238418579, + "rewards/weighted_chosen": -0.25576478242874146, + "rewards/weighted_margins": 0.3327178955078125, + "rewards/weighted_rejected": -0.5886474847793579, + "step": 300 + }, + { + "epoch": 0.16226118817063595, + "grad_norm": 29.578954696655273, + "learning_rate": 9.886131295443002e-07, + "logits/chosen": -0.6218292117118835, + "logits/rejected": -0.6808746457099915, + "logps/chosen": -313.44061279296875, + "logps/rejected": -297.9156188964844, + "logps/weighted_chosen": -2.719403028488159, + "logps/weighted_rejected": -3.1142334938049316, + "loss": 0.6221, + "rewards/accuracies": 0.653124988079071, + "rewards/chosen": -32.1875, + "rewards/margins": 15.571874618530273, + "rewards/rejected": -47.7666015625, + "rewards/weighted_accuracies": 0.6499999761581421, + "rewards/weighted_chosen": -0.19769592583179474, + "rewards/weighted_margins": 0.521374523639679, + "rewards/weighted_rejected": -0.7188171148300171, + "step": 310 + }, + { + "epoch": 0.16749542004710807, + "grad_norm": 15.948070526123047, + "learning_rate": 9.865925894455166e-07, + "logits/chosen": -0.6636382937431335, + "logits/rejected": -0.6817939877510071, + "logps/chosen": -336.1507873535156, + "logps/rejected": -306.62890625, + "logps/weighted_chosen": -2.83941650390625, + "logps/weighted_rejected": -3.273486375808716, + "loss": 0.7046, + "rewards/accuracies": 0.637499988079071, + "rewards/chosen": -36.21074295043945, + "rewards/margins": 12.9765625, + "rewards/rejected": -49.17656326293945, + "rewards/weighted_accuracies": 0.6343749761581421, + "rewards/weighted_chosen": -0.3679946959018707, + "rewards/weighted_margins": 0.278738409280777, + "rewards/weighted_rejected": -0.6464904546737671, + "step": 320 + }, + { + "epoch": 0.17272965192358022, + "grad_norm": 53.10931396484375, + "learning_rate": 9.84409531374603e-07, + "logits/chosen": -0.6222923398017883, + "logits/rejected": -0.607867419719696, + "logps/chosen": -344.19061279296875, + "logps/rejected": -315.84686279296875, + "logps/weighted_chosen": -2.915234327316284, + "logps/weighted_rejected": -3.346630811691284, + "loss": 0.6334, + "rewards/accuracies": 0.668749988079071, + "rewards/chosen": -39.39531326293945, + "rewards/margins": 13.606640815734863, + "rewards/rejected": -53.00273513793945, + "rewards/weighted_accuracies": 0.65625, + "rewards/weighted_chosen": -0.33662718534469604, + "rewards/weighted_margins": 0.3510375916957855, + "rewards/weighted_rejected": -0.6879855990409851, + "step": 330 + }, + { + "epoch": 0.17796388380005235, + "grad_norm": 13.994344711303711, + "learning_rate": 9.820646844552219e-07, + "logits/chosen": -0.6615959405899048, + "logits/rejected": -0.714038074016571, + "logps/chosen": -312.7328186035156, + "logps/rejected": -323.3359375, + "logps/weighted_chosen": -2.910736083984375, + "logps/weighted_rejected": -3.2298340797424316, + "loss": 0.6376, + "rewards/accuracies": 0.675000011920929, + "rewards/chosen": -36.327537536621094, + "rewards/margins": 21.465625762939453, + "rewards/rejected": -57.79179763793945, + "rewards/weighted_accuracies": 0.675000011920929, + "rewards/weighted_chosen": -0.34371185302734375, + "rewards/weighted_margins": 0.390716552734375, + "rewards/weighted_rejected": -0.734301745891571, + "step": 340 + }, + { + "epoch": 0.18319811567652447, + "grad_norm": 13.682384490966797, + "learning_rate": 9.795588318471964e-07, + "logits/chosen": -0.774444580078125, + "logits/rejected": -0.7887184023857117, + "logps/chosen": -302.76251220703125, + "logps/rejected": -336.90313720703125, + "logps/weighted_chosen": -2.718432664871216, + "logps/weighted_rejected": -3.0622315406799316, + "loss": 0.6451, + "rewards/accuracies": 0.6343749761581421, + "rewards/chosen": -43.29413986206055, + "rewards/margins": 17.229297637939453, + "rewards/rejected": -60.52812576293945, + "rewards/weighted_accuracies": 0.637499988079071, + "rewards/weighted_chosen": -0.33645325899124146, + "rewards/weighted_margins": 0.33460694551467896, + "rewards/weighted_rejected": -0.6709426641464233, + "step": 350 + }, + { + "epoch": 0.1884323475529966, + "grad_norm": 19.35496711730957, + "learning_rate": 9.768928104849415e-07, + "logits/chosen": -0.815447986125946, + "logits/rejected": -0.8216186761856079, + "logps/chosen": -330.66015625, + "logps/rejected": -313.8687438964844, + "logps/weighted_chosen": -2.970752000808716, + "logps/weighted_rejected": -3.1889405250549316, + "loss": 0.6763, + "rewards/accuracies": 0.621874988079071, + "rewards/chosen": -47.90937423706055, + "rewards/margins": 17.064844131469727, + "rewards/rejected": -64.98515319824219, + "rewards/weighted_accuracies": 0.6499999761581421, + "rewards/weighted_chosen": -0.3897460997104645, + "rewards/weighted_margins": 0.384918212890625, + "rewards/weighted_rejected": -0.7745300531387329, + "step": 360 + }, + { + "epoch": 0.19366657942946872, + "grad_norm": 21.70383071899414, + "learning_rate": 9.740675107979355e-07, + "logits/chosen": -0.7939224243164062, + "logits/rejected": -0.827624499797821, + "logps/chosen": -373.0406188964844, + "logps/rejected": -349.3343811035156, + "logps/weighted_chosen": -2.506054639816284, + "logps/weighted_rejected": -3.43896484375, + "loss": 0.6215, + "rewards/accuracies": 0.596875011920929, + "rewards/chosen": -58.082420349121094, + "rewards/margins": 16.594335556030273, + "rewards/rejected": -74.671875, + "rewards/weighted_accuracies": 0.671875, + "rewards/weighted_chosen": -0.48696595430374146, + "rewards/weighted_margins": 0.412750244140625, + "rewards/weighted_rejected": -0.90020751953125, + "step": 370 + }, + { + "epoch": 0.19890081130594087, + "grad_norm": 21.682790756225586, + "learning_rate": 9.71083876413323e-07, + "logits/chosen": -0.77825927734375, + "logits/rejected": -0.7921401858329773, + "logps/chosen": -367.0062561035156, + "logps/rejected": -355.78594970703125, + "logps/weighted_chosen": -2.644775390625, + "logps/weighted_rejected": -3.5374755859375, + "loss": 0.6609, + "rewards/accuracies": 0.628125011920929, + "rewards/chosen": -64.59492492675781, + "rewards/margins": 21.614843368530273, + "rewards/rejected": -86.1988296508789, + "rewards/weighted_accuracies": 0.6625000238418579, + "rewards/weighted_chosen": -0.5685027837753296, + "rewards/weighted_margins": 0.3586792051792145, + "rewards/weighted_rejected": -0.9269439578056335, + "step": 380 + }, + { + "epoch": 0.204135043182413, + "grad_norm": 30.26679229736328, + "learning_rate": 9.67942903840751e-07, + "logits/chosen": -0.8401764035224915, + "logits/rejected": -0.898754894733429, + "logps/chosen": -362.0328063964844, + "logps/rejected": -362.3687438964844, + "logps/weighted_chosen": -2.8179688453674316, + "logps/weighted_rejected": -3.4515624046325684, + "loss": 0.5992, + "rewards/accuracies": 0.668749988079071, + "rewards/chosen": -57.87226486206055, + "rewards/margins": 30.196680068969727, + "rewards/rejected": -88.0804672241211, + "rewards/weighted_accuracies": 0.715624988079071, + "rewards/weighted_chosen": -0.456369012594223, + "rewards/weighted_margins": 0.495849609375, + "rewards/weighted_rejected": -0.951983630657196, + "step": 390 + }, + { + "epoch": 0.2093692750588851, + "grad_norm": 37.81236267089844, + "learning_rate": 9.646456421395447e-07, + "logits/chosen": -0.8902359008789062, + "logits/rejected": -0.9107666015625, + "logps/chosen": -380.6343688964844, + "logps/rejected": -399.7281188964844, + "logps/weighted_chosen": -2.709228515625, + "logps/weighted_rejected": -3.455004930496216, + "loss": 0.6557, + "rewards/accuracies": 0.6812499761581421, + "rewards/chosen": -62.43437576293945, + "rewards/margins": 33.634765625, + "rewards/rejected": -96.0941390991211, + "rewards/weighted_accuracies": 0.653124988079071, + "rewards/weighted_chosen": -0.4688262939453125, + "rewards/weighted_margins": 0.3546081483364105, + "rewards/weighted_rejected": -0.8232513666152954, + "step": 400 + }, + { + "epoch": 0.21460350693535724, + "grad_norm": 21.761716842651367, + "learning_rate": 9.611931925683266e-07, + "logits/chosen": -0.849993109703064, + "logits/rejected": -0.8886367678642273, + "logps/chosen": -368.8374938964844, + "logps/rejected": -353.73126220703125, + "logps/weighted_chosen": -2.533374071121216, + "logps/weighted_rejected": -3.233935594558716, + "loss": 0.591, + "rewards/accuracies": 0.6656249761581421, + "rewards/chosen": -62.575782775878906, + "rewards/margins": 27.423242568969727, + "rewards/rejected": -90.0132827758789, + "rewards/weighted_accuracies": 0.6937500238418579, + "rewards/weighted_chosen": -0.4288879334926605, + "rewards/weighted_margins": 0.507708728313446, + "rewards/weighted_rejected": -0.9364013671875, + "step": 410 + }, + { + "epoch": 0.21983773881182936, + "grad_norm": 28.842378616333008, + "learning_rate": 9.575867082172085e-07, + "logits/chosen": -0.836225152015686, + "logits/rejected": -0.8854034543037415, + "logps/chosen": -371.90625, + "logps/rejected": -369.4125061035156, + "logps/weighted_chosen": -2.9190917015075684, + "logps/weighted_rejected": -3.118579149246216, + "loss": 0.6198, + "rewards/accuracies": 0.668749988079071, + "rewards/chosen": -66.98905944824219, + "rewards/margins": 32.337890625, + "rewards/rejected": -99.3402328491211, + "rewards/weighted_accuracies": 0.6625000238418579, + "rewards/weighted_chosen": -0.4911743104457855, + "rewards/weighted_margins": 0.5335143804550171, + "rewards/weighted_rejected": -1.0242736339569092, + "step": 420 + }, + { + "epoch": 0.22507197068830148, + "grad_norm": 16.805343627929688, + "learning_rate": 9.538273936226673e-07, + "logits/chosen": -0.8646026849746704, + "logits/rejected": -0.898516833782196, + "logps/chosen": -328.94921875, + "logps/rejected": -353.66717529296875, + "logps/weighted_chosen": -3.1983399391174316, + "logps/weighted_rejected": -3.6509766578674316, + "loss": 0.6173, + "rewards/accuracies": 0.625, + "rewards/chosen": -63.52910232543945, + "rewards/margins": 25.6201171875, + "rewards/rejected": -89.1246109008789, + "rewards/weighted_accuracies": 0.6499999761581421, + "rewards/weighted_chosen": -0.436279296875, + "rewards/weighted_margins": 0.438516229391098, + "rewards/weighted_rejected": -0.8747711181640625, + "step": 430 + }, + { + "epoch": 0.23030620256477363, + "grad_norm": 23.78999137878418, + "learning_rate": 9.499165043652391e-07, + "logits/chosen": -0.8759521245956421, + "logits/rejected": -0.8865814208984375, + "logps/chosen": -358.69140625, + "logps/rejected": -358.76251220703125, + "logps/weighted_chosen": -3.2625732421875, + "logps/weighted_rejected": -3.359301805496216, + "loss": 0.6366, + "rewards/accuracies": 0.6187499761581421, + "rewards/chosen": -67.4326171875, + "rewards/margins": 24.610157012939453, + "rewards/rejected": -92.03788757324219, + "rewards/weighted_accuracies": 0.668749988079071, + "rewards/weighted_chosen": -0.5896576046943665, + "rewards/weighted_margins": 0.34770506620407104, + "rewards/weighted_rejected": -0.9376128911972046, + "step": 440 + }, + { + "epoch": 0.23554043444124576, + "grad_norm": 49.646263122558594, + "learning_rate": 9.458553466501665e-07, + "logits/chosen": -0.9158996343612671, + "logits/rejected": -0.9590820074081421, + "logps/chosen": -355.3109436035156, + "logps/rejected": -341.9765625, + "logps/weighted_chosen": -3.231823682785034, + "logps/weighted_rejected": -3.4568848609924316, + "loss": 0.657, + "rewards/accuracies": 0.6312500238418579, + "rewards/chosen": -69.14453125, + "rewards/margins": 28.884180068969727, + "rewards/rejected": -98.0492172241211, + "rewards/weighted_accuracies": 0.6875, + "rewards/weighted_chosen": -0.6470184326171875, + "rewards/weighted_margins": 0.4533935487270355, + "rewards/weighted_rejected": -1.100683569908142, + "step": 450 + }, + { + "epoch": 0.24077466631771788, + "grad_norm": 38.45325469970703, + "learning_rate": 9.416452768711366e-07, + "logits/chosen": -0.9281005859375, + "logits/rejected": -0.9545837640762329, + "logps/chosen": -375.6390686035156, + "logps/rejected": -370.203125, + "logps/weighted_chosen": -2.9860596656799316, + "logps/weighted_rejected": -3.635498046875, + "loss": 0.6457, + "rewards/accuracies": 0.612500011920929, + "rewards/chosen": -83.3871078491211, + "rewards/margins": 26.872264862060547, + "rewards/rejected": -110.26094055175781, + "rewards/weighted_accuracies": 0.6625000238418579, + "rewards/weighted_chosen": -0.6367782354354858, + "rewards/weighted_margins": 0.515301525592804, + "rewards/weighted_rejected": -1.152197241783142, + "step": 460 + }, + { + "epoch": 0.24600889819419, + "grad_norm": 59.67642593383789, + "learning_rate": 9.372877011572557e-07, + "logits/chosen": -0.8797394037246704, + "logits/rejected": -0.904949963092804, + "logps/chosen": -398.8531188964844, + "logps/rejected": -388.22186279296875, + "logps/weighted_chosen": -3.0103516578674316, + "logps/weighted_rejected": -3.409374952316284, + "loss": 0.6181, + "rewards/accuracies": 0.6031249761581421, + "rewards/chosen": -88.2437515258789, + "rewards/margins": 25.542577743530273, + "rewards/rejected": -113.7890625, + "rewards/weighted_accuracies": 0.6781250238418579, + "rewards/weighted_chosen": -0.6517273187637329, + "rewards/weighted_margins": 0.482818603515625, + "rewards/weighted_rejected": -1.1340011358261108, + "step": 470 + }, + { + "epoch": 0.2512431300706621, + "grad_norm": 27.465656280517578, + "learning_rate": 9.327840749034141e-07, + "logits/chosen": -0.910723865032196, + "logits/rejected": -0.9439147710800171, + "logps/chosen": -364.4750061035156, + "logps/rejected": -392.03436279296875, + "logps/weighted_chosen": -2.8880372047424316, + "logps/weighted_rejected": -4.069311618804932, + "loss": 0.6276, + "rewards/accuracies": 0.6812499761581421, + "rewards/chosen": -77.44355773925781, + "rewards/margins": 38.147850036621094, + "rewards/rejected": -115.59648132324219, + "rewards/weighted_accuracies": 0.684374988079071, + "rewards/weighted_chosen": -0.5768798589706421, + "rewards/weighted_margins": 0.541027843952179, + "rewards/weighted_rejected": -1.1178162097930908, + "step": 480 + }, + { + "epoch": 0.2564773619471343, + "grad_norm": 34.35914993286133, + "learning_rate": 9.281359022841965e-07, + "logits/chosen": -0.80584716796875, + "logits/rejected": -0.8291991949081421, + "logps/chosen": -349.86798095703125, + "logps/rejected": -354.75701904296875, + "logps/weighted_chosen": -2.914257764816284, + "logps/weighted_rejected": -4.178680419921875, + "loss": 0.63, + "rewards/accuracies": 0.625, + "rewards/chosen": -74.94511413574219, + "rewards/margins": 34.957618713378906, + "rewards/rejected": -109.90644836425781, + "rewards/weighted_accuracies": 0.690625011920929, + "rewards/weighted_chosen": -0.564099133014679, + "rewards/weighted_margins": 0.658795177936554, + "rewards/weighted_rejected": -1.2229950428009033, + "step": 490 + }, + { + "epoch": 0.26171159382360637, + "grad_norm": 41.01088333129883, + "learning_rate": 9.233447357514989e-07, + "logits/chosen": -0.775347888469696, + "logits/rejected": -0.8226684331893921, + "logps/chosen": -369.1148376464844, + "logps/rejected": -373.8374938964844, + "logps/weighted_chosen": -3.288012742996216, + "logps/weighted_rejected": -3.7696776390075684, + "loss": 0.6225, + "rewards/accuracies": 0.643750011920929, + "rewards/chosen": -71.1283187866211, + "rewards/margins": 32.867576599121094, + "rewards/rejected": -104.0218734741211, + "rewards/weighted_accuracies": 0.6937500238418579, + "rewards/weighted_chosen": -0.5540069341659546, + "rewards/weighted_margins": 0.598828136920929, + "rewards/weighted_rejected": -1.1535155773162842, + "step": 500 + }, + { + "epoch": 0.26171159382360637, + "eval_logits/chosen": -0.8767929673194885, + "eval_logits/rejected": -0.8968105316162109, + "eval_logps/chosen": -369.21099853515625, + "eval_logps/rejected": -379.3609924316406, + "eval_logps/weighted_chosen": -2.957455635070801, + "eval_logps/weighted_rejected": -3.6658689975738525, + "eval_loss": 0.6115986108779907, + "eval_rewards/accuracies": 0.6294999718666077, + "eval_rewards/chosen": -80.43468475341797, + "eval_rewards/margins": 31.297624588012695, + "eval_rewards/rejected": -111.73987579345703, + "eval_rewards/weighted_accuracies": 0.6759999990463257, + "eval_rewards/weighted_chosen": -0.5706117153167725, + "eval_rewards/weighted_margins": 0.5400464534759521, + "eval_rewards/weighted_rejected": -1.1106581687927246, + "eval_runtime": 1361.4222, + "eval_samples_per_second": 1.469, + "eval_steps_per_second": 0.367, + "step": 500 + }, + { + "epoch": 0.2669458257000785, + "grad_norm": 37.983856201171875, + "learning_rate": 9.184121755160232e-07, + "logits/chosen": -0.8435028195381165, + "logits/rejected": -0.8772613406181335, + "logps/chosen": -378.20782470703125, + "logps/rejected": -406.3187561035156, + "logps/weighted_chosen": -3.217089891433716, + "logps/weighted_rejected": -3.7809815406799316, + "loss": 0.6028, + "rewards/accuracies": 0.6656249761581421, + "rewards/chosen": -77.0746078491211, + "rewards/margins": 40.68242263793945, + "rewards/rejected": -117.759765625, + "rewards/weighted_accuracies": 0.6875, + "rewards/weighted_chosen": -0.5556915402412415, + "rewards/weighted_margins": 0.5749145746231079, + "rewards/weighted_rejected": -1.130804419517517, + "step": 510 + }, + { + "epoch": 0.2721800575765506, + "grad_norm": 46.08055877685547, + "learning_rate": 9.133398690128193e-07, + "logits/chosen": -0.894885241985321, + "logits/rejected": -0.9157882928848267, + "logps/chosen": -405.5679626464844, + "logps/rejected": -415.10076904296875, + "logps/weighted_chosen": -3.0770506858825684, + "logps/weighted_rejected": -3.949780225753784, + "loss": 0.5762, + "rewards/accuracies": 0.65625, + "rewards/chosen": -84.00956726074219, + "rewards/margins": 50.470703125, + "rewards/rejected": -134.46737670898438, + "rewards/weighted_accuracies": 0.7250000238418579, + "rewards/weighted_chosen": -0.49931639432907104, + "rewards/weighted_margins": 0.658154308795929, + "rewards/weighted_rejected": -1.1574280261993408, + "step": 520 + }, + { + "epoch": 0.27741428945302277, + "grad_norm": 40.89746856689453, + "learning_rate": 9.081295103510554e-07, + "logits/chosen": -0.8876678347587585, + "logits/rejected": -0.9430938959121704, + "logps/chosen": -383.52734375, + "logps/rejected": -413.60626220703125, + "logps/weighted_chosen": -2.752880811691284, + "logps/weighted_rejected": -4.01953125, + "loss": 0.5471, + "rewards/accuracies": 0.6468750238418579, + "rewards/chosen": -99.4878921508789, + "rewards/margins": 44.412109375, + "rewards/rejected": -143.859375, + "rewards/weighted_accuracies": 0.706250011920929, + "rewards/weighted_chosen": -0.6086364984512329, + "rewards/weighted_margins": 0.8298507928848267, + "rewards/weighted_rejected": -1.438207983970642, + "step": 530 + }, + { + "epoch": 0.2826485213294949, + "grad_norm": 27.849748611450195, + "learning_rate": 9.027828397481989e-07, + "logits/chosen": -0.8531036376953125, + "logits/rejected": -0.885467529296875, + "logps/chosen": -354.4359436035156, + "logps/rejected": -387.6328125, + "logps/weighted_chosen": -3.355029344558716, + "logps/weighted_rejected": -4.168432712554932, + "loss": 0.5956, + "rewards/accuracies": 0.625, + "rewards/chosen": -93.89375305175781, + "rewards/margins": 38.216407775878906, + "rewards/rejected": -132.11776733398438, + "rewards/weighted_accuracies": 0.675000011920929, + "rewards/weighted_chosen": -0.6640685796737671, + "rewards/weighted_margins": 0.6357482671737671, + "rewards/weighted_rejected": -1.2999756336212158, + "step": 540 + }, + { + "epoch": 0.287882753205967, + "grad_norm": 22.531660079956055, + "learning_rate": 8.973016429487988e-07, + "logits/chosen": -0.8864593505859375, + "logits/rejected": -0.9035919308662415, + "logps/chosen": -375.8890686035156, + "logps/rejected": -396.171875, + "logps/weighted_chosen": -3.2917237281799316, + "logps/weighted_rejected": -3.580859422683716, + "loss": 0.6356, + "rewards/accuracies": 0.6499999761581421, + "rewards/chosen": -94.04902648925781, + "rewards/margins": 47.923240661621094, + "rewards/rejected": -141.9646453857422, + "rewards/weighted_accuracies": 0.71875, + "rewards/weighted_chosen": -0.7129974365234375, + "rewards/weighted_margins": 0.53741455078125, + "rewards/weighted_rejected": -1.2503478527069092, + "step": 550 + }, + { + "epoch": 0.29311698508243916, + "grad_norm": 36.68463897705078, + "learning_rate": 8.916877506280601e-07, + "logits/chosen": -0.9174438714981079, + "logits/rejected": -0.9228302240371704, + "logps/chosen": -387.2203063964844, + "logps/rejected": -405.38592529296875, + "logps/weighted_chosen": -3.4023194313049316, + "logps/weighted_rejected": -3.6703858375549316, + "loss": 0.5995, + "rewards/accuracies": 0.6343749761581421, + "rewards/chosen": -106.43359375, + "rewards/margins": 44.87871170043945, + "rewards/rejected": -151.3136749267578, + "rewards/weighted_accuracies": 0.690625011920929, + "rewards/weighted_chosen": -0.7533203363418579, + "rewards/weighted_margins": 0.627819836139679, + "rewards/weighted_rejected": -1.3810546398162842, + "step": 560 + }, + { + "epoch": 0.29835121695891126, + "grad_norm": 14.758890151977539, + "learning_rate": 8.85943037780415e-07, + "logits/chosen": -0.9987869262695312, + "logits/rejected": -1.0033996105194092, + "logps/chosen": -392.72344970703125, + "logps/rejected": -380.15625, + "logps/weighted_chosen": -3.2074217796325684, + "logps/weighted_rejected": -3.7137207984924316, + "loss": 0.6343, + "rewards/accuracies": 0.609375, + "rewards/chosen": -106.4300765991211, + "rewards/margins": 32.053321838378906, + "rewards/rejected": -138.525390625, + "rewards/weighted_accuracies": 0.7093750238418579, + "rewards/weighted_chosen": -0.744158923625946, + "rewards/weighted_margins": 0.547467052936554, + "rewards/weighted_rejected": -1.291876196861267, + "step": 570 + }, + { + "epoch": 0.3035854488353834, + "grad_norm": 18.140382766723633, + "learning_rate": 8.800694230932884e-07, + "logits/chosen": -0.958880603313446, + "logits/rejected": -0.9720214605331421, + "logps/chosen": -390.57342529296875, + "logps/rejected": -398.5562438964844, + "logps/weighted_chosen": -2.86993408203125, + "logps/weighted_rejected": -3.4458250999450684, + "loss": 0.624, + "rewards/accuracies": 0.6031249761581421, + "rewards/chosen": -100.91777038574219, + "rewards/margins": 32.43359375, + "rewards/rejected": -133.35546875, + "rewards/weighted_accuracies": 0.659375011920929, + "rewards/weighted_chosen": -0.558367908000946, + "rewards/weighted_margins": 0.4340454041957855, + "rewards/weighted_rejected": -0.992688000202179, + "step": 580 + }, + { + "epoch": 0.30881968071185556, + "grad_norm": 16.685937881469727, + "learning_rate": 8.740688683062723e-07, + "logits/chosen": -1.0655395984649658, + "logits/rejected": -1.0949890613555908, + "logps/chosen": -449.0078125, + "logps/rejected": -423.9593811035156, + "logps/weighted_chosen": -2.912792921066284, + "logps/weighted_rejected": -3.656201124191284, + "loss": 0.6344, + "rewards/accuracies": 0.6031249761581421, + "rewards/chosen": -119.9111328125, + "rewards/margins": 28.247461318969727, + "rewards/rejected": -148.1476593017578, + "rewards/weighted_accuracies": 0.706250011920929, + "rewards/weighted_chosen": -0.5903121829032898, + "rewards/weighted_margins": 0.525067150592804, + "rewards/weighted_rejected": -1.1153564453125, + "step": 590 + }, + { + "epoch": 0.31405391258832765, + "grad_norm": 20.500003814697266, + "learning_rate": 8.679433775559215e-07, + "logits/chosen": -1.0546143054962158, + "logits/rejected": -1.1019103527069092, + "logps/chosen": -473.6171875, + "logps/rejected": -477.95001220703125, + "logps/weighted_chosen": -2.896728515625, + "logps/weighted_rejected": -3.985156297683716, + "loss": 0.5741, + "rewards/accuracies": 0.606249988079071, + "rewards/chosen": -148.9021453857422, + "rewards/margins": 43.50390625, + "rewards/rejected": -192.4656219482422, + "rewards/weighted_accuracies": 0.71875, + "rewards/weighted_chosen": -0.7574096918106079, + "rewards/weighted_margins": 0.587969958782196, + "rewards/weighted_rejected": -1.345495581626892, + "step": 600 + }, + { + "epoch": 0.3192881444647998, + "grad_norm": 50.760562896728516, + "learning_rate": 8.616949967063871e-07, + "logits/chosen": -1.007971167564392, + "logits/rejected": -1.0581481456756592, + "logps/chosen": -419.1000061035156, + "logps/rejected": -460.9624938964844, + "logps/weighted_chosen": -3.3387207984924316, + "logps/weighted_rejected": -3.981640577316284, + "loss": 0.5971, + "rewards/accuracies": 0.6468750238418579, + "rewards/chosen": -150.87344360351562, + "rewards/margins": 49.451171875, + "rewards/rejected": -200.21444702148438, + "rewards/weighted_accuracies": 0.671875, + "rewards/weighted_chosen": -0.853253185749054, + "rewards/weighted_margins": 0.644561767578125, + "rewards/weighted_rejected": -1.497961401939392, + "step": 610 + }, + { + "epoch": 0.3245223763412719, + "grad_norm": 16.369630813598633, + "learning_rate": 8.553258126661154e-07, + "logits/chosen": -1.0560119152069092, + "logits/rejected": -1.0674712657928467, + "logps/chosen": -442.5859375, + "logps/rejected": -463.5718688964844, + "logps/weighted_chosen": -3.429705858230591, + "logps/weighted_rejected": -4.259472846984863, + "loss": 0.6405, + "rewards/accuracies": 0.621874988079071, + "rewards/chosen": -162.7121124267578, + "rewards/margins": 44.754295349121094, + "rewards/rejected": -207.43008422851562, + "rewards/weighted_accuracies": 0.684374988079071, + "rewards/weighted_chosen": -0.9004287719726562, + "rewards/weighted_margins": 0.6113220453262329, + "rewards/weighted_rejected": -1.511694312095642, + "step": 620 + }, + { + "epoch": 0.32975660821774405, + "grad_norm": 31.255151748657227, + "learning_rate": 8.488379526908368e-07, + "logits/chosen": -1.0301024913787842, + "logits/rejected": -1.0498230457305908, + "logps/chosen": -465.5859375, + "logps/rejected": -505.28125, + "logps/weighted_chosen": -3.3856444358825684, + "logps/weighted_rejected": -4.028759956359863, + "loss": 0.6384, + "rewards/accuracies": 0.621874988079071, + "rewards/chosen": -174.98046875, + "rewards/margins": 61.357032775878906, + "rewards/rejected": -236.38906860351562, + "rewards/weighted_accuracies": 0.6968749761581421, + "rewards/weighted_chosen": -0.969250500202179, + "rewards/weighted_margins": 0.5812011957168579, + "rewards/weighted_rejected": -1.5499999523162842, + "step": 630 + }, + { + "epoch": 0.33499084009421615, + "grad_norm": 23.673486709594727, + "learning_rate": 8.422335836730802e-07, + "logits/chosen": -1.0135101079940796, + "logits/rejected": -1.0169677734375, + "logps/chosen": -426.2875061035156, + "logps/rejected": -497.64373779296875, + "logps/weighted_chosen": -3.098522901535034, + "logps/weighted_rejected": -3.898510694503784, + "loss": 0.6321, + "rewards/accuracies": 0.637499988079071, + "rewards/chosen": -152.58749389648438, + "rewards/margins": 67.427734375, + "rewards/rejected": -220.0578155517578, + "rewards/weighted_accuracies": 0.699999988079071, + "rewards/weighted_chosen": -0.878277599811554, + "rewards/weighted_margins": 0.590301513671875, + "rewards/weighted_rejected": -1.468664526939392, + "step": 640 + }, + { + "epoch": 0.3402250719706883, + "grad_norm": 20.44074821472168, + "learning_rate": 8.355149114184485e-07, + "logits/chosen": -1.105963110923767, + "logits/rejected": -1.095733642578125, + "logps/chosen": -474.7554626464844, + "logps/rejected": -523.0843505859375, + "logps/weighted_chosen": -3.1939697265625, + "logps/weighted_rejected": -3.786669969558716, + "loss": 0.5634, + "rewards/accuracies": 0.628125011920929, + "rewards/chosen": -167.33828735351562, + "rewards/margins": 73.52656555175781, + "rewards/rejected": -240.85702514648438, + "rewards/weighted_accuracies": 0.668749988079071, + "rewards/weighted_chosen": -0.8172271847724915, + "rewards/weighted_margins": 0.6973236203193665, + "rewards/weighted_rejected": -1.514801025390625, + "step": 650 + }, + { + "epoch": 0.34545930384716045, + "grad_norm": 25.958847045898438, + "learning_rate": 8.286841799088963e-07, + "logits/chosen": -1.1746642589569092, + "logits/rejected": -1.179473876953125, + "logps/chosen": -480.34063720703125, + "logps/rejected": -518.9437255859375, + "logps/weighted_chosen": -2.8995361328125, + "logps/weighted_rejected": -3.656787157058716, + "loss": 0.6073, + "rewards/accuracies": 0.5843750238418579, + "rewards/chosen": -196.3136749267578, + "rewards/margins": 59.513671875, + "rewards/rejected": -255.85897827148438, + "rewards/weighted_accuracies": 0.684374988079071, + "rewards/weighted_chosen": -0.8131744265556335, + "rewards/weighted_margins": 0.576770007610321, + "rewards/weighted_rejected": -1.3900268077850342, + "step": 660 + }, + { + "epoch": 0.35069353572363254, + "grad_norm": 15.336409568786621, + "learning_rate": 8.217436705532599e-07, + "logits/chosen": -1.1510436534881592, + "logits/rejected": -1.173730492591858, + "logps/chosen": -508.78594970703125, + "logps/rejected": -500.79217529296875, + "logps/weighted_chosen": -3.0011963844299316, + "logps/weighted_rejected": -3.7521729469299316, + "loss": 0.578, + "rewards/accuracies": 0.6000000238418579, + "rewards/chosen": -205.71914672851562, + "rewards/margins": 31.244531631469727, + "rewards/rejected": -237.14453125, + "rewards/weighted_accuracies": 0.721875011920929, + "rewards/weighted_chosen": -0.8121887445449829, + "rewards/weighted_margins": 0.688189685344696, + "rewards/weighted_rejected": -1.500115990638733, + "step": 670 + }, + { + "epoch": 0.3559277676001047, + "grad_norm": 19.029943466186523, + "learning_rate": 8.14695701425284e-07, + "logits/chosen": -1.079345703125, + "logits/rejected": -1.125024437904358, + "logps/chosen": -466.38360595703125, + "logps/rejected": -470.4234313964844, + "logps/weighted_chosen": -2.9954833984375, + "logps/weighted_rejected": -3.8232421875, + "loss": 0.5882, + "rewards/accuracies": 0.606249988079071, + "rewards/chosen": -160.8718719482422, + "rewards/margins": 47.43242263793945, + "rewards/rejected": -208.275390625, + "rewards/weighted_accuracies": 0.690625011920929, + "rewards/weighted_chosen": -0.7394653558731079, + "rewards/weighted_margins": 0.627307116985321, + "rewards/weighted_rejected": -1.3663330078125, + "step": 680 + }, + { + "epoch": 0.3611619994765768, + "grad_norm": 13.842121124267578, + "learning_rate": 8.075426264894046e-07, + "logits/chosen": -1.013006567955017, + "logits/rejected": -1.0405089855194092, + "logps/chosen": -472.03436279296875, + "logps/rejected": -501.09375, + "logps/weighted_chosen": -3.1174073219299316, + "logps/weighted_rejected": -4.429980278015137, + "loss": 0.5163, + "rewards/accuracies": 0.6656249761581421, + "rewards/chosen": -166.5642547607422, + "rewards/margins": 56.965431213378906, + "rewards/rejected": -223.5593719482422, + "rewards/weighted_accuracies": 0.778124988079071, + "rewards/weighted_chosen": -0.7782195806503296, + "rewards/weighted_margins": 0.887194812297821, + "rewards/weighted_rejected": -1.665307641029358, + "step": 690 + }, + { + "epoch": 0.36639623135304894, + "grad_norm": 31.751785278320312, + "learning_rate": 8.002868348145435e-07, + "logits/chosen": -1.0329437255859375, + "logits/rejected": -1.0401580333709717, + "logps/chosen": -535.6437377929688, + "logps/rejected": -564.8609619140625, + "logps/weighted_chosen": -3.4560546875, + "logps/weighted_rejected": -3.8817505836486816, + "loss": 0.5956, + "rewards/accuracies": 0.625, + "rewards/chosen": -235.3767547607422, + "rewards/margins": 61.516014099121094, + "rewards/rejected": -296.88360595703125, + "rewards/weighted_accuracies": 0.71875, + "rewards/weighted_chosen": -1.1499541997909546, + "rewards/weighted_margins": 0.6718994379043579, + "rewards/weighted_rejected": -1.8215301036834717, + "step": 700 + }, + { + "epoch": 0.3716304632295211, + "grad_norm": 32.05786895751953, + "learning_rate": 7.92930749776179e-07, + "logits/chosen": -1.0492126941680908, + "logits/rejected": -1.092248558998108, + "logps/chosen": -564.2515869140625, + "logps/rejected": -602.4500122070312, + "logps/weighted_chosen": -3.8183350563049316, + "logps/weighted_rejected": -4.569482326507568, + "loss": 0.652, + "rewards/accuracies": 0.581250011920929, + "rewards/chosen": -292.5419921875, + "rewards/margins": 49.53203201293945, + "rewards/rejected": -342.0179748535156, + "rewards/weighted_accuracies": 0.6656249761581421, + "rewards/weighted_chosen": -1.244970679283142, + "rewards/weighted_margins": 0.630566418170929, + "rewards/weighted_rejected": -1.8747985363006592, + "step": 710 + }, + { + "epoch": 0.3768646951059932, + "grad_norm": 23.8446044921875, + "learning_rate": 7.854768282469582e-07, + "logits/chosen": -1.0707213878631592, + "logits/rejected": -1.126983642578125, + "logps/chosen": -462.90155029296875, + "logps/rejected": -542.5218505859375, + "logps/weighted_chosen": -3.208447217941284, + "logps/weighted_rejected": -4.043383598327637, + "loss": 0.6115, + "rewards/accuracies": 0.6499999761581421, + "rewards/chosen": -185.4597625732422, + "rewards/margins": 79.3746109008789, + "rewards/rejected": -264.8812561035156, + "rewards/weighted_accuracies": 0.6781250238418579, + "rewards/weighted_chosen": -0.8762931823730469, + "rewards/weighted_margins": 0.711090087890625, + "rewards/weighted_rejected": -1.5873870849609375, + "step": 720 + }, + { + "epoch": 0.38209892698246534, + "grad_norm": 17.816680908203125, + "learning_rate": 7.779275597761215e-07, + "logits/chosen": -1.036431908607483, + "logits/rejected": -1.105950951576233, + "logps/chosen": -471.43280029296875, + "logps/rejected": -532.6843872070312, + "logps/weighted_chosen": -3.378674268722534, + "logps/weighted_rejected": -4.096581935882568, + "loss": 0.5442, + "rewards/accuracies": 0.6499999761581421, + "rewards/chosen": -195.43984985351562, + "rewards/margins": 68.92460632324219, + "rewards/rejected": -264.296875, + "rewards/weighted_accuracies": 0.706250011920929, + "rewards/weighted_chosen": -0.847393810749054, + "rewards/weighted_margins": 0.777783215045929, + "rewards/weighted_rejected": -1.625451683998108, + "step": 730 + }, + { + "epoch": 0.38733315885893743, + "grad_norm": 29.750192642211914, + "learning_rate": 7.702854657580126e-07, + "logits/chosen": -1.0928223133087158, + "logits/rejected": -1.1139404773712158, + "logps/chosen": -507.796875, + "logps/rejected": -523.7952880859375, + "logps/weighted_chosen": -3.3760986328125, + "logps/weighted_rejected": -4.447509765625, + "loss": 0.5597, + "rewards/accuracies": 0.6343749761581421, + "rewards/chosen": -213.2277374267578, + "rewards/margins": 56.09648513793945, + "rewards/rejected": -269.4203186035156, + "rewards/weighted_accuracies": 0.7406250238418579, + "rewards/weighted_chosen": -0.927258312702179, + "rewards/weighted_margins": 0.8479980230331421, + "rewards/weighted_rejected": -1.7749145030975342, + "step": 740 + }, + { + "epoch": 0.3925673907354096, + "grad_norm": 38.97298812866211, + "learning_rate": 7.625530985899547e-07, + "logits/chosen": -1.0656921863555908, + "logits/rejected": -1.0953140258789062, + "logps/chosen": -491.4609375, + "logps/rejected": -545.0203247070312, + "logps/weighted_chosen": -3.2592406272888184, + "logps/weighted_rejected": -4.605615139007568, + "loss": 0.5776, + "rewards/accuracies": 0.5874999761581421, + "rewards/chosen": -224.6374969482422, + "rewards/margins": 75.23710632324219, + "rewards/rejected": -300.0316467285156, + "rewards/weighted_accuracies": 0.699999988079071, + "rewards/weighted_chosen": -1.093621850013733, + "rewards/weighted_margins": 0.852429211139679, + "rewards/weighted_rejected": -1.9456908702850342, + "step": 750 + }, + { + "epoch": 0.39780162261188173, + "grad_norm": 22.909940719604492, + "learning_rate": 7.547330408197694e-07, + "logits/chosen": -1.0663268566131592, + "logits/rejected": -1.1003234386444092, + "logps/chosen": -572.404296875, + "logps/rejected": -556.4031372070312, + "logps/weighted_chosen": -3.372363328933716, + "logps/weighted_rejected": -4.400781154632568, + "loss": 0.5943, + "rewards/accuracies": 0.6031249761581421, + "rewards/chosen": -276.7529296875, + "rewards/margins": 23.257421493530273, + "rewards/rejected": -300.1558532714844, + "rewards/weighted_accuracies": 0.703125, + "rewards/weighted_chosen": -1.1045868396759033, + "rewards/weighted_margins": 0.7262054681777954, + "rewards/weighted_rejected": -1.8311798572540283, + "step": 760 + }, + { + "epoch": 0.40303585448835383, + "grad_norm": 107.9198989868164, + "learning_rate": 7.468279042832271e-07, + "logits/chosen": -1.067718505859375, + "logits/rejected": -1.095727562904358, + "logps/chosen": -467.5, + "logps/rejected": -559.7703247070312, + "logps/weighted_chosen": -3.0786986351013184, + "logps/weighted_rejected": -3.8573241233825684, + "loss": 0.5995, + "rewards/accuracies": 0.6499999761581421, + "rewards/chosen": -187.9324188232422, + "rewards/margins": 82.28340148925781, + "rewards/rejected": -270.2611389160156, + "rewards/weighted_accuracies": 0.6968749761581421, + "rewards/weighted_chosen": -0.867828369140625, + "rewards/weighted_margins": 0.625274658203125, + "rewards/weighted_rejected": -1.493688941001892, + "step": 770 + }, + { + "epoch": 0.408270086364826, + "grad_norm": 18.453107833862305, + "learning_rate": 7.388403292317154e-07, + "logits/chosen": -1.0655944347381592, + "logits/rejected": -1.121740698814392, + "logps/chosen": -503.8828125, + "logps/rejected": -519.8375244140625, + "logps/weighted_chosen": -3.039843797683716, + "logps/weighted_rejected": -3.848388671875, + "loss": 0.5822, + "rewards/accuracies": 0.606249988079071, + "rewards/chosen": -202.9210968017578, + "rewards/margins": 50.95390701293945, + "rewards/rejected": -253.869140625, + "rewards/weighted_accuracies": 0.731249988079071, + "rewards/weighted_chosen": -0.873791515827179, + "rewards/weighted_margins": 0.6851806640625, + "rewards/weighted_rejected": -1.5595581531524658, + "step": 780 + }, + { + "epoch": 0.4135043182412981, + "grad_norm": 24.38869285583496, + "learning_rate": 7.307729834504154e-07, + "logits/chosen": -1.090570092201233, + "logits/rejected": -1.1371276378631592, + "logps/chosen": -490.86407470703125, + "logps/rejected": -537.7296752929688, + "logps/weighted_chosen": -3.041918992996216, + "logps/weighted_rejected": -4.070898532867432, + "loss": 0.6125, + "rewards/accuracies": 0.6000000238418579, + "rewards/chosen": -209.13241577148438, + "rewards/margins": 62.44609451293945, + "rewards/rejected": -271.58203125, + "rewards/weighted_accuracies": 0.690625011920929, + "rewards/weighted_chosen": -0.947405993938446, + "rewards/weighted_margins": 0.609375, + "rewards/weighted_rejected": -1.556726098060608, + "step": 790 + }, + { + "epoch": 0.4187385501177702, + "grad_norm": 26.962316513061523, + "learning_rate": 7.226285613672847e-07, + "logits/chosen": -1.03680419921875, + "logits/rejected": -1.0742919445037842, + "logps/chosen": -500.84295654296875, + "logps/rejected": -603.6422119140625, + "logps/weighted_chosen": -3.154052734375, + "logps/weighted_rejected": -4.342724800109863, + "loss": 0.5564, + "rewards/accuracies": 0.671875, + "rewards/chosen": -223.51171875, + "rewards/margins": 101.2464828491211, + "rewards/rejected": -324.720703125, + "rewards/weighted_accuracies": 0.7406250238418579, + "rewards/weighted_chosen": -0.996997058391571, + "rewards/weighted_margins": 0.755480945110321, + "rewards/weighted_rejected": -1.752661108970642, + "step": 800 + }, + { + "epoch": 0.4239727819942423, + "grad_norm": 15.780332565307617, + "learning_rate": 7.144097831531398e-07, + "logits/chosen": -1.0250060558319092, + "logits/rejected": -1.052709937095642, + "logps/chosen": -517.7921752929688, + "logps/rejected": -600.5421752929688, + "logps/weighted_chosen": -3.318652391433716, + "logps/weighted_rejected": -4.39453125, + "loss": 0.5695, + "rewards/accuracies": 0.596875011920929, + "rewards/chosen": -247.40664672851562, + "rewards/margins": 88.87187194824219, + "rewards/rejected": -336.234375, + "rewards/weighted_accuracies": 0.703125, + "rewards/weighted_chosen": -1.082605004310608, + "rewards/weighted_margins": 0.89263916015625, + "rewards/weighted_rejected": -1.974450707435608, + "step": 810 + }, + { + "epoch": 0.42920701387071447, + "grad_norm": 24.66672706604004, + "learning_rate": 7.061193938131396e-07, + "logits/chosen": -0.9916015863418579, + "logits/rejected": -1.042016625404358, + "logps/chosen": -607.3375244140625, + "logps/rejected": -642.5671997070312, + "logps/weighted_chosen": -3.689379930496216, + "logps/weighted_rejected": -4.398633003234863, + "loss": 0.5654, + "rewards/accuracies": 0.5874999761581421, + "rewards/chosen": -307.7925720214844, + "rewards/margins": 69.5777359008789, + "rewards/rejected": -377.3050842285156, + "rewards/weighted_accuracies": 0.7124999761581421, + "rewards/weighted_chosen": -1.1716430187225342, + "rewards/weighted_margins": 0.9023987054824829, + "rewards/weighted_rejected": -2.074475049972534, + "step": 820 + }, + { + "epoch": 0.4344412457471866, + "grad_norm": 20.459196090698242, + "learning_rate": 6.977601622699789e-07, + "logits/chosen": -0.9984985589981079, + "logits/rejected": -1.079248070716858, + "logps/chosen": -471.5625, + "logps/rejected": -574.8695068359375, + "logps/weighted_chosen": -3.233837842941284, + "logps/weighted_rejected": -4.211133003234863, + "loss": 0.5097, + "rewards/accuracies": 0.6656249761581421, + "rewards/chosen": -186.7394561767578, + "rewards/margins": 109.28633117675781, + "rewards/rejected": -296.13555908203125, + "rewards/weighted_accuracies": 0.746874988079071, + "rewards/weighted_chosen": -0.760040283203125, + "rewards/weighted_margins": 0.9420166015625, + "rewards/weighted_rejected": -1.7019164562225342, + "step": 830 + }, + { + "epoch": 0.4396754776236587, + "grad_norm": 89.301025390625, + "learning_rate": 6.893348804390882e-07, + "logits/chosen": -1.126031517982483, + "logits/rejected": -1.1464111804962158, + "logps/chosen": -541.5546875, + "logps/rejected": -580.4109497070312, + "logps/weighted_chosen": -3.6358885765075684, + "logps/weighted_rejected": -4.262304782867432, + "loss": 0.566, + "rewards/accuracies": 0.6000000238418579, + "rewards/chosen": -246.6085968017578, + "rewards/margins": 74.0562515258789, + "rewards/rejected": -320.61328125, + "rewards/weighted_accuracies": 0.684374988079071, + "rewards/weighted_chosen": -1.0077393054962158, + "rewards/weighted_margins": 0.7990967035293579, + "rewards/weighted_rejected": -1.806396484375, + "step": 840 + }, + { + "epoch": 0.44490970950013087, + "grad_norm": 20.6274471282959, + "learning_rate": 6.808463622961578e-07, + "logits/chosen": -1.151123046875, + "logits/rejected": -1.2046630382537842, + "logps/chosen": -621.8390502929688, + "logps/rejected": -716.7249755859375, + "logps/weighted_chosen": -3.6482911109924316, + "logps/weighted_rejected": -4.6015625, + "loss": 0.5759, + "rewards/accuracies": 0.6156250238418579, + "rewards/chosen": -328.931640625, + "rewards/margins": 106.0474624633789, + "rewards/rejected": -435.07403564453125, + "rewards/weighted_accuracies": 0.7124999761581421, + "rewards/weighted_chosen": -1.2104400396347046, + "rewards/weighted_margins": 0.9383910894393921, + "rewards/weighted_rejected": -2.1482911109924316, + "step": 850 + }, + { + "epoch": 0.45014394137660296, + "grad_norm": 20.91545295715332, + "learning_rate": 6.722974429372925e-07, + "logits/chosen": -1.1297729015350342, + "logits/rejected": -1.170617699623108, + "logps/chosen": -697.5484619140625, + "logps/rejected": -730.65625, + "logps/weighted_chosen": -3.5313477516174316, + "logps/weighted_rejected": -5.184765815734863, + "loss": 0.5217, + "rewards/accuracies": 0.5718749761581421, + "rewards/chosen": -393.822265625, + "rewards/margins": 74.92500305175781, + "rewards/rejected": -468.8492126464844, + "rewards/weighted_accuracies": 0.7718750238418579, + "rewards/weighted_chosen": -1.393365502357483, + "rewards/weighted_margins": 1.1208374500274658, + "rewards/weighted_rejected": -2.514172315597534, + "step": 860 + }, + { + "epoch": 0.4553781732530751, + "grad_norm": 42.88610076904297, + "learning_rate": 6.636909776321128e-07, + "logits/chosen": -1.1412475109100342, + "logits/rejected": -1.1436645984649658, + "logps/chosen": -549.9671630859375, + "logps/rejected": -677.8781127929688, + "logps/weighted_chosen": -3.683642625808716, + "logps/weighted_rejected": -4.7724609375, + "loss": 0.5195, + "rewards/accuracies": 0.606249988079071, + "rewards/chosen": -288.21484375, + "rewards/margins": 124.931640625, + "rewards/rejected": -413.17266845703125, + "rewards/weighted_accuracies": 0.7250000238418579, + "rewards/weighted_chosen": -1.2497345209121704, + "rewards/weighted_margins": 0.971630871295929, + "rewards/weighted_rejected": -2.2208008766174316, + "step": 870 + }, + { + "epoch": 0.46061240512954726, + "grad_norm": 29.938108444213867, + "learning_rate": 6.550298408701174e-07, + "logits/chosen": -1.0770995616912842, + "logits/rejected": -1.134680151939392, + "logps/chosen": -574.3234252929688, + "logps/rejected": -675.2101440429688, + "logps/weighted_chosen": -3.807568311691284, + "logps/weighted_rejected": -5.117138862609863, + "loss": 0.5413, + "rewards/accuracies": 0.6312500238418579, + "rewards/chosen": -287.63494873046875, + "rewards/margins": 104.92851257324219, + "rewards/rejected": -392.5396423339844, + "rewards/weighted_accuracies": 0.734375, + "rewards/weighted_chosen": -1.1316711902618408, + "rewards/weighted_margins": 0.954119861125946, + "rewards/weighted_rejected": -2.0862364768981934, + "step": 880 + }, + { + "epoch": 0.46584663700601936, + "grad_norm": 27.375259399414062, + "learning_rate": 6.463169254006276e-07, + "logits/chosen": -1.10406494140625, + "logits/rejected": -1.156030297279358, + "logps/chosen": -541.8601684570312, + "logps/rejected": -578.5452880859375, + "logps/weighted_chosen": -3.4522461891174316, + "logps/weighted_rejected": -4.527783393859863, + "loss": 0.5321, + "rewards/accuracies": 0.612500011920929, + "rewards/chosen": -260.65704345703125, + "rewards/margins": 71.1494140625, + "rewards/rejected": -331.875, + "rewards/weighted_accuracies": 0.7281249761581421, + "rewards/weighted_chosen": -0.977038562297821, + "rewards/weighted_margins": 1.022253394126892, + "rewards/weighted_rejected": -1.999292016029358, + "step": 890 + }, + { + "epoch": 0.4710808688824915, + "grad_norm": 32.590675354003906, + "learning_rate": 6.375551412666326e-07, + "logits/chosen": -1.0722472667694092, + "logits/rejected": -1.1129913330078125, + "logps/chosen": -499.8031311035156, + "logps/rejected": -543.9796752929688, + "logps/weighted_chosen": -3.171948194503784, + "logps/weighted_rejected": -4.273193359375, + "loss": 0.6035, + "rewards/accuracies": 0.590624988079071, + "rewards/chosen": -225.7996063232422, + "rewards/margins": 61.669532775878906, + "rewards/rejected": -287.5015563964844, + "rewards/weighted_accuracies": 0.6812499761581421, + "rewards/weighted_chosen": -1.0042235851287842, + "rewards/weighted_margins": 0.7809082269668579, + "rewards/weighted_rejected": -1.784277319908142, + "step": 900 + }, + { + "epoch": 0.4763151007589636, + "grad_norm": 37.88996887207031, + "learning_rate": 6.287474148328583e-07, + "logits/chosen": -1.0318695306777954, + "logits/rejected": -1.0187164545059204, + "logps/chosen": -491.97186279296875, + "logps/rejected": -540.4202880859375, + "logps/weighted_chosen": -3.5292601585388184, + "logps/weighted_rejected": -5.02734375, + "loss": 0.559, + "rewards/accuracies": 0.606249988079071, + "rewards/chosen": -221.07968139648438, + "rewards/margins": 72.57441711425781, + "rewards/rejected": -293.56561279296875, + "rewards/weighted_accuracies": 0.690625011920929, + "rewards/weighted_chosen": -1.0068145990371704, + "rewards/weighted_margins": 0.84521484375, + "rewards/weighted_rejected": -1.851770043373108, + "step": 910 + }, + { + "epoch": 0.48154933263543576, + "grad_norm": 25.340351104736328, + "learning_rate": 6.198966878083857e-07, + "logits/chosen": -1.0930664539337158, + "logits/rejected": -1.12109375, + "logps/chosen": -540.0, + "logps/rejected": -623.390625, + "logps/weighted_chosen": -3.68231201171875, + "logps/weighted_rejected": -4.464990139007568, + "loss": 0.5911, + "rewards/accuracies": 0.606249988079071, + "rewards/chosen": -268.6884765625, + "rewards/margins": 87.7593765258789, + "rewards/rejected": -356.4453125, + "rewards/weighted_accuracies": 0.7124999761581421, + "rewards/weighted_chosen": -1.085455298423767, + "rewards/weighted_margins": 0.79486083984375, + "rewards/weighted_rejected": -1.880859375, + "step": 920 + }, + { + "epoch": 0.48678356451190785, + "grad_norm": 47.73952865600586, + "learning_rate": 6.110059162641439e-07, + "logits/chosen": -1.1020386219024658, + "logits/rejected": -1.1357238292694092, + "logps/chosen": -533.7390747070312, + "logps/rejected": -599.3781127929688, + "logps/weighted_chosen": -2.9868531227111816, + "logps/weighted_rejected": -4.054858207702637, + "loss": 0.5458, + "rewards/accuracies": 0.625, + "rewards/chosen": -251.13125610351562, + "rewards/margins": 80.9544906616211, + "rewards/rejected": -332.2699279785156, + "rewards/weighted_accuracies": 0.734375, + "rewards/weighted_chosen": -0.8615478277206421, + "rewards/weighted_margins": 0.8154662847518921, + "rewards/weighted_rejected": -1.6761963367462158, + "step": 930 + }, + { + "epoch": 0.49201779638838, + "grad_norm": 36.5290412902832, + "learning_rate": 6.020780696456059e-07, + "logits/chosen": -1.133764624595642, + "logits/rejected": -1.177819848060608, + "logps/chosen": -515.2952880859375, + "logps/rejected": -611.4953002929688, + "logps/weighted_chosen": -3.0075440406799316, + "logps/weighted_rejected": -4.250219821929932, + "loss": 0.5361, + "rewards/accuracies": 0.659375011920929, + "rewards/chosen": -244.7570343017578, + "rewards/margins": 99.73710632324219, + "rewards/rejected": -344.3531188964844, + "rewards/weighted_accuracies": 0.731249988079071, + "rewards/weighted_chosen": -1.0392029285430908, + "rewards/weighted_margins": 0.8944336175918579, + "rewards/weighted_rejected": -1.93341064453125, + "step": 940 + }, + { + "epoch": 0.49725202826485215, + "grad_norm": 36.72229766845703, + "learning_rate": 5.931161297810185e-07, + "logits/chosen": -1.2070038318634033, + "logits/rejected": -1.23931884765625, + "logps/chosen": -550.4031372070312, + "logps/rejected": -615.3375244140625, + "logps/weighted_chosen": -3.748730421066284, + "logps/weighted_rejected": -4.64697265625, + "loss": 0.5726, + "rewards/accuracies": 0.5843750238418579, + "rewards/chosen": -277.07733154296875, + "rewards/margins": 77.06562805175781, + "rewards/rejected": -354.22381591796875, + "rewards/weighted_accuracies": 0.6968749761581421, + "rewards/weighted_chosen": -1.09320068359375, + "rewards/weighted_margins": 0.812207043170929, + "rewards/weighted_rejected": -1.906347632408142, + "step": 950 + }, + { + "epoch": 0.5024862601413242, + "grad_norm": 39.58312225341797, + "learning_rate": 5.841230898854959e-07, + "logits/chosen": -1.1735107898712158, + "logits/rejected": -1.2161743640899658, + "logps/chosen": -684.3687744140625, + "logps/rejected": -794.3922119140625, + "logps/weighted_chosen": -3.8369140625, + "logps/weighted_rejected": -4.986083984375, + "loss": 0.6144, + "rewards/accuracies": 0.612500011920929, + "rewards/chosen": -384.17852783203125, + "rewards/margins": 148.6707000732422, + "rewards/rejected": -532.8191528320312, + "rewards/weighted_accuracies": 0.6937500238418579, + "rewards/weighted_chosen": -1.5656402111053467, + "rewards/weighted_margins": 0.9439452886581421, + "rewards/weighted_rejected": -2.5091795921325684, + "step": 960 + }, + { + "epoch": 0.5077204920177963, + "grad_norm": 24.286752700805664, + "learning_rate": 5.751019535613102e-07, + "logits/chosen": -1.0753357410430908, + "logits/rejected": -1.104547142982483, + "logps/chosen": -550.9773559570312, + "logps/rejected": -653.6687622070312, + "logps/weighted_chosen": -3.5946044921875, + "logps/weighted_rejected": -5.038866996765137, + "loss": 0.5266, + "rewards/accuracies": 0.675000011920929, + "rewards/chosen": -291.25390625, + "rewards/margins": 115.4117202758789, + "rewards/rejected": -406.62811279296875, + "rewards/weighted_accuracies": 0.7093750238418579, + "rewards/weighted_chosen": -1.2581603527069092, + "rewards/weighted_margins": 1.127111792564392, + "rewards/weighted_rejected": -2.3849730491638184, + "step": 970 + }, + { + "epoch": 0.5129547238942685, + "grad_norm": 22.55328941345215, + "learning_rate": 5.660557337947117e-07, + "logits/chosen": -1.0788085460662842, + "logits/rejected": -1.109989881515503, + "logps/chosen": -594.8406372070312, + "logps/rejected": -653.6624755859375, + "logps/weighted_chosen": -3.10791015625, + "logps/weighted_rejected": -4.297143459320068, + "loss": 0.5361, + "rewards/accuracies": 0.628125011920929, + "rewards/chosen": -300.37872314453125, + "rewards/margins": 95.828125, + "rewards/rejected": -396.203125, + "rewards/weighted_accuracies": 0.6968749761581421, + "rewards/weighted_chosen": -1.0415496826171875, + "rewards/weighted_margins": 0.88812255859375, + "rewards/weighted_rejected": -1.929406762123108, + "step": 980 + }, + { + "epoch": 0.5181889557707406, + "grad_norm": 53.25763702392578, + "learning_rate": 5.569874519496174e-07, + "logits/chosen": -1.055578589439392, + "logits/rejected": -1.122521996498108, + "logps/chosen": -497.3999938964844, + "logps/rejected": -579.3656005859375, + "logps/weighted_chosen": -3.393261671066284, + "logps/weighted_rejected": -4.673925876617432, + "loss": 0.5302, + "rewards/accuracies": 0.596875011920929, + "rewards/chosen": -222.11483764648438, + "rewards/margins": 90.296875, + "rewards/rejected": -312.42108154296875, + "rewards/weighted_accuracies": 0.699999988079071, + "rewards/weighted_chosen": -0.8590606451034546, + "rewards/weighted_margins": 0.879437267780304, + "rewards/weighted_rejected": -1.7384154796600342, + "step": 990 + }, + { + "epoch": 0.5234231876472127, + "grad_norm": 39.35697555541992, + "learning_rate": 5.47900136758499e-07, + "logits/chosen": -1.0391693115234375, + "logits/rejected": -1.1183960437774658, + "logps/chosen": -589.5906372070312, + "logps/rejected": -653.09375, + "logps/weighted_chosen": -3.634228467941284, + "logps/weighted_rejected": -4.7723388671875, + "loss": 0.5255, + "rewards/accuracies": 0.59375, + "rewards/chosen": -321.95623779296875, + "rewards/margins": 89.08320617675781, + "rewards/rejected": -410.8277282714844, + "rewards/weighted_accuracies": 0.731249988079071, + "rewards/weighted_chosen": -1.187219262123108, + "rewards/weighted_margins": 0.9879394769668579, + "rewards/weighted_rejected": -2.1756300926208496, + "step": 1000 + }, + { + "epoch": 0.5234231876472127, + "eval_logits/chosen": -1.194671869277954, + "eval_logits/rejected": -1.2200820446014404, + "eval_logps/chosen": -645.4445190429688, + "eval_logps/rejected": -726.552001953125, + "eval_logps/weighted_chosen": -3.7153730392456055, + "eval_logps/weighted_rejected": -4.821405410766602, + "eval_loss": 0.5563206672668457, + "eval_rewards/accuracies": 0.5899999737739563, + "eval_rewards/chosen": -356.71826171875, + "eval_rewards/margins": 102.17925262451172, + "eval_rewards/rejected": -458.8554992675781, + "eval_rewards/weighted_accuracies": 0.7135000228881836, + "eval_rewards/weighted_chosen": -1.3285291194915771, + "eval_rewards/weighted_margins": 0.9376659989356995, + "eval_rewards/weighted_rejected": -2.266195297241211, + "eval_runtime": 1365.9873, + "eval_samples_per_second": 1.464, + "eval_steps_per_second": 0.366, + "step": 1000 + } + ], + "logging_steps": 10, + "max_steps": 1911, + "num_input_tokens_seen": 0, + "num_train_epochs": 1, + "save_steps": 500, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 0.0, + "train_batch_size": 1, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoint-1000/training_args.bin b/checkpoint-1000/training_args.bin new file mode 100644 index 0000000..eb22667 --- /dev/null +++ b/checkpoint-1000/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ded8eebbc35e0c697502403c9c9d2a0acb4a3f38810fc064a38066719f316edb +size 8785 diff --git a/checkpoint-1000/zero_to_fp32.py b/checkpoint-1000/zero_to_fp32.py new file mode 100644 index 0000000..0e75914 --- /dev/null +++ b/checkpoint-1000/zero_to_fp32.py @@ -0,0 +1,760 @@ +#!/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 . output_dir/ +# or +# python zero_to_fp32.py . output_dir/ --safe_serialization + +import argparse +import torch +import glob +import math +import os +import re +import gc +import json +import numpy as np +from tqdm import tqdm +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, weights_only=False) + + 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 tqdm(files, desc='Loading checkpoint shards'): + state_dict = torch.load(f, map_location=device, mmap=True, weights_only=False) + # 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}") + + fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] 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") + + +class GatheredTensor: + """ + A pseudo tensor that collects partitioned weights. + It is more memory efficient when there are multiple groups. + """ + + def __init__(self, flat_groups, flat_groups_offset, offset, partitioned_numel, shape): + self.flat_groups = flat_groups + self.flat_groups_offset = flat_groups_offset + self.offset = offset + self.partitioned_numel = partitioned_numel + self.shape = shape + self.dtype = self.flat_groups[0][0].dtype + + def contiguous(self): + """ + Merge partitioned weights from flat_groups into a single tensor. + """ + end_idx = self.offset + self.partitioned_numel + world_size = len(self.flat_groups) + pad_flat_param_chunks = [] + + for rank_i in range(world_size): + # for each rank, we need to collect weights from related group/groups + flat_groups_at_rank_i = self.flat_groups[rank_i] + start_group_id = None + end_group_id = None + for group_id in range(len(self.flat_groups_offset)): + if self.flat_groups_offset[group_id] <= self.offset < self.flat_groups_offset[group_id + 1]: + start_group_id = group_id + if self.flat_groups_offset[group_id] < end_idx <= self.flat_groups_offset[group_id + 1]: + end_group_id = group_id + break + # collect weights from related group/groups + for group_id in range(start_group_id, end_group_id + 1): + flat_tensor = flat_groups_at_rank_i[group_id] + start_offset = self.offset - self.flat_groups_offset[group_id] + end_offset = min(end_idx, self.flat_groups_offset[group_id + 1]) - self.flat_groups_offset[group_id] + pad_flat_param_chunks.append(flat_tensor[start_offset:end_offset]) + + # collect weights from all ranks + pad_flat_param = torch.cat(pad_flat_param_chunks, dim=0) + param = pad_flat_param[:self.shape.numel()].view(self.shape).contiguous() + return param + + +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 = sum([flat_group.numel() for flat_group in fp32_flat_groups[0]]) * 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 + flat_groups_offset = [0] + list(np.cumsum([flat_tensor.numel() for flat_tensor in fp32_flat_groups[0]])) + for name, shape in tqdm(param_shapes.items(), desc='Gathering sharded weights'): + 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}" + ) + + # memory efficient tensor + tensor = GatheredTensor(fp32_flat_groups, flat_groups_offset, offset, partitioned_numel, shape) + state_dict[name] = tensor + 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 to_torch_tensor(state_dict, return_empty_tensor=False): + """ + Convert state_dict of GatheredTensor to torch tensor + """ + torch_state_dict = {} + converted_tensors = {} + for name, tensor in state_dict.items(): + tensor_id = id(tensor) + if tensor_id in converted_tensors: # shared tensors + shared_tensor = torch_state_dict[converted_tensors[tensor_id]] + torch_state_dict[name] = shared_tensor + else: + converted_tensors[tensor_id] = name + if return_empty_tensor: + torch_state_dict[name] = torch.empty(tensor.shape, dtype=tensor.dtype) + else: + torch_state_dict[name] = tensor.contiguous() + return torch_state_dict + + +def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, + tag=None, + exclude_frozen_parameters=False, + lazy_mode=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 + - ``lazy_mode``: get state_dict in lazy mode. It returns a dict of pesduo tensor instead of torch tensor, which is more memory efficient. + Convert the pesduo tensor to torch tensor by ``.contiguous()`` + + Returns: + - pytorch ``state_dict`` + + 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. + + Note: the above usage may not work if your application doesn't have sufficient free CPU memory. + You may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with + the checkpoint. Or you can load state_dict in lazy mode :: + + from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, lazy_mode=True) # not on cpu + for name, lazy_tensor in state_dict.item(): + tensor = lazy_tensor.contiguous() # to cpu + print(name, tensor) + # del tensor to release memory if it no longer in use + """ + 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") + + state_dict = _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters) + if lazy_mode: + return state_dict + else: + return to_torch_tensor(state_dict) + + +def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, + output_dir, + max_shard_size="5GB", + safe_serialization=False, + 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_dir``: directory to the pytorch fp32 state_dict output files + - ``max_shard_size``: the maximum size for a checkpoint before being sharded, default value is 5GB + - ``safe_serialization``: whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`). + - ``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 + """ + + # Dependency pre-check + if safe_serialization: + try: + from safetensors.torch import save_file + except ImportError: + print('If you want to use `safe_serialization`, please `pip install safetensors`') + raise + if max_shard_size is not None: + try: + from huggingface_hub import split_torch_state_dict_into_shards + except ImportError: + print('If you want to use `max_shard_size`, please `pip install huggingface_hub`') + raise + + # Convert zero checkpoint to state_dict + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, + tag, + exclude_frozen_parameters, + lazy_mode=True) + + # Shard the model if it is too big. + weights_name = "model.safetensors" if safe_serialization else "pytorch_model.bin" + if max_shard_size is not None: + filename_pattern = weights_name.replace(".bin", "{suffix}.bin").replace(".safetensors", "{suffix}.safetensors") + # an memory-efficient approach for sharding + empty_state_dict = to_torch_tensor(state_dict, return_empty_tensor=True) + state_dict_split = split_torch_state_dict_into_shards(empty_state_dict, + filename_pattern=filename_pattern, + max_shard_size=max_shard_size) + else: + from collections import namedtuple + StateDictSplit = namedtuple("StateDictSplit", ["is_sharded", "filename_to_tensors"]) + state_dict_split = StateDictSplit(is_sharded=False, + filename_to_tensors={weights_name: list(state_dict.keys())}) + + # Save the model by shard + os.makedirs(output_dir, exist_ok=True) + filename_to_tensors = state_dict_split.filename_to_tensors.items() + for shard_file, tensors in tqdm(filename_to_tensors, desc="Saving checkpoint shards"): + shard_state_dict = {tensor_name: state_dict[tensor_name] for tensor_name in tensors} + shard_state_dict = to_torch_tensor(shard_state_dict) + output_path = os.path.join(output_dir, shard_file) + if safe_serialization: + save_file(shard_state_dict, output_path, metadata={"format": "pt"}) + else: + torch.save(shard_state_dict, output_path) + # release the memory of current shard + for tensor_name in list(shard_state_dict.keys()): + del state_dict[tensor_name] + del shard_state_dict[tensor_name] + del shard_state_dict + gc.collect() + + # Save index if sharded + if state_dict_split.is_sharded: + index = { + "metadata": state_dict_split.metadata, + "weight_map": state_dict_split.tensor_to_filename, + } + save_index_file = "model.safetensors.index.json" if safe_serialization else "pytorch_model.bin.index.json" + save_index_file = os.path.join(output_dir, save_index_file) + with open(save_index_file, "w", encoding="utf-8") as f: + content = json.dumps(index, indent=2, sort_keys=True) + "\n" + f.write(content) + + +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_dir", + type=str, + help="directory to the pytorch fp32 state_dict output files" + "(e.g. path/checkpoint-12-output/)") + parser.add_argument( + "--max_shard_size", + type=str, + default="5GB", + help="The maximum size for a checkpoint before being sharded. Checkpoints shard will then be each of size" + "lower than this size. If expressed as a string, needs to be digits followed by a unit (like `5MB`" + "We default it to 5GB in order for models to be able to run easily on free-tier google colab instances" + "without CPU OOM issues.") + parser.add_argument( + "--safe_serialization", + default=False, + action='store_true', + help="Whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).") + 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_dir, + max_shard_size=args.max_shard_size, + safe_serialization=args.safe_serialization, + tag=args.tag, + exclude_frozen_parameters=args.exclude_frozen_parameters) diff --git a/checkpoint-1500/chat_template.jinja b/checkpoint-1500/chat_template.jinja new file mode 100644 index 0000000..39bd0c9 --- /dev/null +++ b/checkpoint-1500/chat_template.jinja @@ -0,0 +1,5 @@ +{% set loop_messages = messages %}{% for message in loop_messages %}{% set content = '<|start_header_id|>' + message['role'] + '<|end_header_id|> + +'+ message['content'] | trim + '<|eot_id|>' %}{% if loop.index0 == 0 %}{% set content = bos_token + content %}{% endif %}{{ content }}{% endfor %}{% if add_generation_prompt %}{{ '<|start_header_id|>assistant<|end_header_id|> + +' }}{% endif %} \ No newline at end of file diff --git a/checkpoint-1500/config.json b/checkpoint-1500/config.json new file mode 100644 index 0000000..3f8f5c0 --- /dev/null +++ b/checkpoint-1500/config.json @@ -0,0 +1,29 @@ +{ + "architectures": [ + "LlamaForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "bos_token_id": 128000, + "eos_token_id": 128001, + "head_dim": 128, + "hidden_act": "silu", + "hidden_size": 4096, + "initializer_range": 0.02, + "intermediate_size": 14336, + "max_position_embeddings": 8192, + "mlp_bias": false, + "model_type": "llama", + "num_attention_heads": 32, + "num_hidden_layers": 32, + "num_key_value_heads": 8, + "pretraining_tp": 1, + "rms_norm_eps": 1e-05, + "rope_scaling": null, + "rope_theta": 500000.0, + "tie_word_embeddings": false, + "torch_dtype": "bfloat16", + "transformers_version": "4.54.1", + "use_cache": true, + "vocab_size": 128256 +} diff --git a/checkpoint-1500/generation_config.json b/checkpoint-1500/generation_config.json new file mode 100644 index 0000000..fc3c54a --- /dev/null +++ b/checkpoint-1500/generation_config.json @@ -0,0 +1,6 @@ +{ + "_from_model_config": true, + "bos_token_id": 128000, + "eos_token_id": 128001, + "transformers_version": "4.54.1" +} diff --git a/checkpoint-1500/latest b/checkpoint-1500/latest new file mode 100644 index 0000000..c56ff77 --- /dev/null +++ b/checkpoint-1500/latest @@ -0,0 +1 @@ +global_step1500 \ No newline at end of file diff --git a/checkpoint-1500/model-00001-of-00004.safetensors b/checkpoint-1500/model-00001-of-00004.safetensors new file mode 100644 index 0000000..de4eae2 --- /dev/null +++ b/checkpoint-1500/model-00001-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a68f8cbf09ff722196857350483041d4a0f107b7574c50164a4e6c805b2313a2 +size 4976698672 diff --git a/checkpoint-1500/model-00002-of-00004.safetensors b/checkpoint-1500/model-00002-of-00004.safetensors new file mode 100644 index 0000000..9454574 --- /dev/null +++ b/checkpoint-1500/model-00002-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0a5b21870dab59d064ab4776664a034a06f074ca178fd2e775015307171f95ba +size 4999802720 diff --git a/checkpoint-1500/model-00003-of-00004.safetensors b/checkpoint-1500/model-00003-of-00004.safetensors new file mode 100644 index 0000000..fca4f43 --- /dev/null +++ b/checkpoint-1500/model-00003-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cae407564c5181e88262b0b52f35fc38e31dc08f950c09b712da48a7aa256021 +size 4915916176 diff --git a/checkpoint-1500/model-00004-of-00004.safetensors b/checkpoint-1500/model-00004-of-00004.safetensors new file mode 100644 index 0000000..bd29bf9 --- /dev/null +++ b/checkpoint-1500/model-00004-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bec51b93372ad223c14be596d9f8a28f65c413ac8af2ce315dda1b23d06851ca +size 1168138808 diff --git a/checkpoint-1500/model.safetensors.index.json b/checkpoint-1500/model.safetensors.index.json new file mode 100644 index 0000000..58a7ef4 --- /dev/null +++ b/checkpoint-1500/model.safetensors.index.json @@ -0,0 +1,299 @@ +{ + "metadata": { + "total_parameters": 266240, + "total_size": 16060522496 + }, + "weight_map": { + "lm_head.weight": "model-00004-of-00004.safetensors", + "model.embed_tokens.weight": "model-00001-of-00004.safetensors", + "model.layers.0.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.0.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.1.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.10.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.10.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.10.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.11.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.11.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.12.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.12.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.13.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.13.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.14.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.14.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.15.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.15.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.16.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.16.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.17.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.17.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.18.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.18.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.19.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.19.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.2.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.2.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.20.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.20.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.20.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.20.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.20.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.20.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.20.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.20.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.20.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.21.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.21.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.21.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.22.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.22.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.23.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.23.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.24.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.24.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.25.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.25.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.26.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.26.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.27.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.27.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.28.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.28.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.29.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.29.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.3.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.3.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.30.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.30.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.30.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.input_layernorm.weight": "model-00004-of-00004.safetensors", + "model.layers.31.mlp.down_proj.weight": "model-00004-of-00004.safetensors", + "model.layers.31.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.post_attention_layernorm.weight": "model-00004-of-00004.safetensors", + "model.layers.31.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.4.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.4.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.5.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.6.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.7.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.8.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.9.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.9.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.9.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.norm.weight": "model-00004-of-00004.safetensors" + } +} diff --git a/checkpoint-1500/rng_state_0.pth b/checkpoint-1500/rng_state_0.pth new file mode 100644 index 0000000..8d84687 --- /dev/null +++ b/checkpoint-1500/rng_state_0.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:02ea5dcfd1b4a49b41b4fa01a8b24bba6186957162c3fd555ebff28620c7268b +size 14917 diff --git a/checkpoint-1500/rng_state_1.pth b/checkpoint-1500/rng_state_1.pth new file mode 100644 index 0000000..54f0119 --- /dev/null +++ b/checkpoint-1500/rng_state_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c2a5af18bb5eae8b7fd6bdef66259014d98ba87ffb16d614bba38f2c32030798 +size 14917 diff --git a/checkpoint-1500/scheduler.pt b/checkpoint-1500/scheduler.pt new file mode 100644 index 0000000..fabfbf2 --- /dev/null +++ b/checkpoint-1500/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5360a9ca3afd27044b0e3751f896f0dd514aa42a145ec88e8857a2bb4c8588f4 +size 1465 diff --git a/checkpoint-1500/special_tokens_map.json b/checkpoint-1500/special_tokens_map.json new file mode 100644 index 0000000..e5b39b6 --- /dev/null +++ b/checkpoint-1500/special_tokens_map.json @@ -0,0 +1,23 @@ +{ + "bos_token": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "eos_token": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "pad_token": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + } +} diff --git a/checkpoint-1500/tokenizer.json b/checkpoint-1500/tokenizer.json new file mode 100644 index 0000000..03aa64f --- /dev/null +++ b/checkpoint-1500/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0968dcc0ee8e56c7dccd34a7f51f8065ea0cb9e2cc529e3243d1e5c0a4bdaa0c +size 17208754 diff --git a/checkpoint-1500/tokenizer_config.json b/checkpoint-1500/tokenizer_config.json new file mode 100644 index 0000000..877a9a9 --- /dev/null +++ b/checkpoint-1500/tokenizer_config.json @@ -0,0 +1,2063 @@ +{ + "added_tokens_decoder": { + "128000": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128001": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128002": { + "content": "<|reserved_special_token_0|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128003": { + "content": "<|reserved_special_token_1|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128004": { + "content": "<|reserved_special_token_2|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128005": { + "content": "<|reserved_special_token_3|>", + "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": "<|reserved_special_token_4|>", + "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": "<|reserved_special_token_5|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128011": { + "content": "<|reserved_special_token_6|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128012": { + "content": "<|reserved_special_token_7|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128013": { + "content": "<|reserved_special_token_8|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128014": { + "content": "<|reserved_special_token_9|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128015": { + "content": "<|reserved_special_token_10|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128016": { + "content": "<|reserved_special_token_11|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128017": { + "content": "<|reserved_special_token_12|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128018": { + "content": "<|reserved_special_token_13|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128019": { + "content": "<|reserved_special_token_14|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128020": { + "content": "<|reserved_special_token_15|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128021": { + "content": "<|reserved_special_token_16|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128022": { + "content": "<|reserved_special_token_17|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128023": { + "content": "<|reserved_special_token_18|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128024": { + "content": "<|reserved_special_token_19|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128025": { + "content": "<|reserved_special_token_20|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128026": { + "content": "<|reserved_special_token_21|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128027": { + "content": "<|reserved_special_token_22|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128028": { + "content": "<|reserved_special_token_23|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128029": { + "content": "<|reserved_special_token_24|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128030": { + "content": "<|reserved_special_token_25|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128031": { + "content": "<|reserved_special_token_26|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128032": { + "content": "<|reserved_special_token_27|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128033": { + "content": "<|reserved_special_token_28|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128034": { + "content": "<|reserved_special_token_29|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128035": { + "content": "<|reserved_special_token_30|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128036": { + "content": "<|reserved_special_token_31|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128037": { + "content": "<|reserved_special_token_32|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128038": { + "content": "<|reserved_special_token_33|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128039": { + "content": "<|reserved_special_token_34|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128040": { + "content": "<|reserved_special_token_35|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128041": { + "content": "<|reserved_special_token_36|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128042": { + "content": "<|reserved_special_token_37|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128043": { + "content": "<|reserved_special_token_38|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128044": { + "content": "<|reserved_special_token_39|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128045": { + "content": "<|reserved_special_token_40|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128046": { + "content": "<|reserved_special_token_41|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128047": { + "content": "<|reserved_special_token_42|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128048": { + "content": "<|reserved_special_token_43|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128049": { + "content": "<|reserved_special_token_44|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128050": { + "content": "<|reserved_special_token_45|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128051": { + "content": "<|reserved_special_token_46|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128052": { + "content": "<|reserved_special_token_47|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128053": { + "content": "<|reserved_special_token_48|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128054": { + "content": "<|reserved_special_token_49|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128055": { + "content": "<|reserved_special_token_50|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128056": { + "content": "<|reserved_special_token_51|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128057": { + "content": "<|reserved_special_token_52|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128058": { + "content": "<|reserved_special_token_53|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128059": { + "content": "<|reserved_special_token_54|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128060": { + "content": "<|reserved_special_token_55|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128061": { + "content": "<|reserved_special_token_56|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128062": { + "content": "<|reserved_special_token_57|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128063": { + "content": "<|reserved_special_token_58|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128064": { + "content": "<|reserved_special_token_59|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128065": { + "content": "<|reserved_special_token_60|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128066": { + "content": "<|reserved_special_token_61|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128067": { + "content": "<|reserved_special_token_62|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128068": { + "content": "<|reserved_special_token_63|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128069": { + "content": "<|reserved_special_token_64|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128070": { + "content": "<|reserved_special_token_65|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128071": { + "content": "<|reserved_special_token_66|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128072": { + "content": "<|reserved_special_token_67|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128073": { + "content": "<|reserved_special_token_68|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128074": { + "content": "<|reserved_special_token_69|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128075": { + "content": "<|reserved_special_token_70|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128076": { + "content": "<|reserved_special_token_71|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128077": { + "content": "<|reserved_special_token_72|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128078": { + "content": "<|reserved_special_token_73|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128079": { + "content": "<|reserved_special_token_74|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128080": { + "content": "<|reserved_special_token_75|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128081": { + "content": "<|reserved_special_token_76|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128082": { + "content": "<|reserved_special_token_77|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128083": { + "content": "<|reserved_special_token_78|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128084": { + "content": "<|reserved_special_token_79|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128085": { + "content": "<|reserved_special_token_80|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128086": { + "content": "<|reserved_special_token_81|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128087": { + "content": "<|reserved_special_token_82|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128088": { + "content": "<|reserved_special_token_83|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128089": { + "content": "<|reserved_special_token_84|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128090": { + "content": "<|reserved_special_token_85|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128091": { + "content": "<|reserved_special_token_86|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128092": { + "content": "<|reserved_special_token_87|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128093": { + "content": "<|reserved_special_token_88|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128094": { + "content": "<|reserved_special_token_89|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128095": { + "content": "<|reserved_special_token_90|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128096": { + "content": "<|reserved_special_token_91|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128097": { + "content": "<|reserved_special_token_92|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128098": { + "content": "<|reserved_special_token_93|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128099": { + "content": "<|reserved_special_token_94|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128100": { + "content": "<|reserved_special_token_95|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128101": { + "content": "<|reserved_special_token_96|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128102": { + "content": "<|reserved_special_token_97|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128103": { + "content": "<|reserved_special_token_98|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128104": { + "content": "<|reserved_special_token_99|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128105": { + "content": "<|reserved_special_token_100|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128106": { + "content": "<|reserved_special_token_101|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128107": { + "content": "<|reserved_special_token_102|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128108": { + "content": "<|reserved_special_token_103|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128109": { + "content": "<|reserved_special_token_104|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128110": { + "content": "<|reserved_special_token_105|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128111": { + "content": "<|reserved_special_token_106|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128112": { + "content": "<|reserved_special_token_107|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128113": { + "content": "<|reserved_special_token_108|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128114": { + "content": "<|reserved_special_token_109|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128115": { + "content": "<|reserved_special_token_110|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128116": { + "content": "<|reserved_special_token_111|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128117": { + "content": "<|reserved_special_token_112|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128118": { + "content": "<|reserved_special_token_113|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128119": { + "content": "<|reserved_special_token_114|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128120": { + "content": "<|reserved_special_token_115|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128121": { + "content": "<|reserved_special_token_116|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128122": { + "content": "<|reserved_special_token_117|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128123": { + "content": "<|reserved_special_token_118|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128124": { + "content": "<|reserved_special_token_119|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128125": { + "content": "<|reserved_special_token_120|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128126": { + "content": "<|reserved_special_token_121|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128127": { + "content": "<|reserved_special_token_122|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128128": { + "content": "<|reserved_special_token_123|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128129": { + "content": "<|reserved_special_token_124|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128130": { + "content": "<|reserved_special_token_125|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128131": { + "content": "<|reserved_special_token_126|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128132": { + "content": "<|reserved_special_token_127|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128133": { + "content": "<|reserved_special_token_128|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128134": { + "content": "<|reserved_special_token_129|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128135": { + "content": "<|reserved_special_token_130|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128136": { + "content": "<|reserved_special_token_131|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128137": { + "content": "<|reserved_special_token_132|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128138": { + "content": "<|reserved_special_token_133|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128139": { + "content": "<|reserved_special_token_134|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128140": { + "content": "<|reserved_special_token_135|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128141": { + "content": "<|reserved_special_token_136|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128142": { + "content": "<|reserved_special_token_137|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128143": { + "content": "<|reserved_special_token_138|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128144": { + "content": "<|reserved_special_token_139|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128145": { + "content": "<|reserved_special_token_140|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128146": { + "content": "<|reserved_special_token_141|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128147": { + "content": "<|reserved_special_token_142|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128148": { + "content": "<|reserved_special_token_143|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128149": { + "content": "<|reserved_special_token_144|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128150": { + "content": "<|reserved_special_token_145|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128151": { + "content": "<|reserved_special_token_146|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128152": { + "content": "<|reserved_special_token_147|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128153": { + "content": "<|reserved_special_token_148|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128154": { + "content": "<|reserved_special_token_149|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128155": { + "content": "<|reserved_special_token_150|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128156": { + "content": "<|reserved_special_token_151|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128157": { + "content": "<|reserved_special_token_152|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128158": { + "content": "<|reserved_special_token_153|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128159": { + "content": "<|reserved_special_token_154|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128160": { + "content": "<|reserved_special_token_155|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128161": { + "content": "<|reserved_special_token_156|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128162": { + "content": "<|reserved_special_token_157|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128163": { + "content": "<|reserved_special_token_158|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128164": { + "content": "<|reserved_special_token_159|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128165": { + "content": "<|reserved_special_token_160|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128166": { + "content": "<|reserved_special_token_161|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128167": { + "content": "<|reserved_special_token_162|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128168": { + "content": "<|reserved_special_token_163|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128169": { + "content": "<|reserved_special_token_164|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128170": { + "content": "<|reserved_special_token_165|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128171": { + "content": "<|reserved_special_token_166|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128172": { + "content": "<|reserved_special_token_167|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128173": { + "content": "<|reserved_special_token_168|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128174": { + "content": "<|reserved_special_token_169|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128175": { + "content": "<|reserved_special_token_170|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128176": { + "content": "<|reserved_special_token_171|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128177": { + "content": "<|reserved_special_token_172|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128178": { + "content": "<|reserved_special_token_173|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128179": { + "content": "<|reserved_special_token_174|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128180": { + "content": "<|reserved_special_token_175|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128181": { + "content": "<|reserved_special_token_176|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128182": { + "content": "<|reserved_special_token_177|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128183": { + "content": "<|reserved_special_token_178|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128184": { + "content": "<|reserved_special_token_179|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128185": { + "content": "<|reserved_special_token_180|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128186": { + "content": "<|reserved_special_token_181|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128187": { + "content": "<|reserved_special_token_182|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128188": { + "content": "<|reserved_special_token_183|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128189": { + "content": "<|reserved_special_token_184|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128190": { + "content": "<|reserved_special_token_185|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128191": { + "content": "<|reserved_special_token_186|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128192": { + "content": "<|reserved_special_token_187|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128193": { + "content": "<|reserved_special_token_188|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128194": { + "content": "<|reserved_special_token_189|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128195": { + "content": "<|reserved_special_token_190|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128196": { + "content": "<|reserved_special_token_191|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128197": { + "content": "<|reserved_special_token_192|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128198": { + "content": "<|reserved_special_token_193|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128199": { + "content": "<|reserved_special_token_194|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128200": { + "content": "<|reserved_special_token_195|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128201": { + "content": "<|reserved_special_token_196|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128202": { + "content": "<|reserved_special_token_197|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128203": { + "content": "<|reserved_special_token_198|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128204": { + "content": "<|reserved_special_token_199|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128205": { + "content": "<|reserved_special_token_200|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128206": { + "content": "<|reserved_special_token_201|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128207": { + "content": "<|reserved_special_token_202|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128208": { + "content": "<|reserved_special_token_203|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128209": { + "content": "<|reserved_special_token_204|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128210": { + "content": "<|reserved_special_token_205|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128211": { + "content": "<|reserved_special_token_206|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128212": { + "content": "<|reserved_special_token_207|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128213": { + "content": "<|reserved_special_token_208|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128214": { + "content": "<|reserved_special_token_209|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128215": { + "content": "<|reserved_special_token_210|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128216": { + "content": "<|reserved_special_token_211|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128217": { + "content": "<|reserved_special_token_212|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128218": { + "content": "<|reserved_special_token_213|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128219": { + "content": "<|reserved_special_token_214|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128220": { + "content": "<|reserved_special_token_215|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128221": { + "content": "<|reserved_special_token_216|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128222": { + "content": "<|reserved_special_token_217|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128223": { + "content": "<|reserved_special_token_218|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128224": { + "content": "<|reserved_special_token_219|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128225": { + "content": "<|reserved_special_token_220|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128226": { + "content": "<|reserved_special_token_221|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128227": { + "content": "<|reserved_special_token_222|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128228": { + "content": "<|reserved_special_token_223|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128229": { + "content": "<|reserved_special_token_224|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128230": { + "content": "<|reserved_special_token_225|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128231": { + "content": "<|reserved_special_token_226|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128232": { + "content": "<|reserved_special_token_227|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128233": { + "content": "<|reserved_special_token_228|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128234": { + "content": "<|reserved_special_token_229|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128235": { + "content": "<|reserved_special_token_230|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128236": { + "content": "<|reserved_special_token_231|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128237": { + "content": "<|reserved_special_token_232|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128238": { + "content": "<|reserved_special_token_233|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128239": { + "content": "<|reserved_special_token_234|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128240": { + "content": "<|reserved_special_token_235|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128241": { + "content": "<|reserved_special_token_236|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128242": { + "content": "<|reserved_special_token_237|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128243": { + "content": "<|reserved_special_token_238|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128244": { + "content": "<|reserved_special_token_239|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128245": { + "content": "<|reserved_special_token_240|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128246": { + "content": "<|reserved_special_token_241|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128247": { + "content": "<|reserved_special_token_242|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128248": { + "content": "<|reserved_special_token_243|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128249": { + "content": "<|reserved_special_token_244|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128250": { + "content": "<|reserved_special_token_245|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128251": { + "content": "<|reserved_special_token_246|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128252": { + "content": "<|reserved_special_token_247|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128253": { + "content": "<|reserved_special_token_248|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128254": { + "content": "<|reserved_special_token_249|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128255": { + "content": "<|reserved_special_token_250|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + } + }, + "bos_token": "<|begin_of_text|>", + "clean_up_tokenization_spaces": true, + "eos_token": "<|end_of_text|>", + "extra_special_tokens": {}, + "model_input_names": [ + "input_ids", + "attention_mask" + ], + "model_max_length": 32768, + "pad_token": "<|end_of_text|>", + "tokenizer_class": "PreTrainedTokenizerFast" +} diff --git a/checkpoint-1500/trainer_state.json b/checkpoint-1500/trainer_state.json new file mode 100644 index 0000000..349a839 --- /dev/null +++ b/checkpoint-1500/trainer_state.json @@ -0,0 +1,3271 @@ +{ + "best_global_step": null, + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 0.7851347814708192, + "eval_steps": 500, + "global_step": 1500, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.0005234231876472127, + "grad_norm": 136.4956817626953, + "learning_rate": 0.0, + "logits/chosen": -0.40118408203125, + "logits/rejected": -0.41802978515625, + "logps/chosen": -297.609375, + "logps/rejected": -247.84375, + "logps/weighted_chosen": -4.50634765625, + "logps/weighted_rejected": -3.43408203125, + "loss": 0.6914, + "rewards/accuracies": 0.0, + "rewards/chosen": 0.0, + "rewards/margins": 0.0, + "rewards/rejected": 0.0, + "rewards/weighted_accuracies": 0.0, + "rewards/weighted_chosen": 0.0, + "rewards/weighted_margins": 0.0, + "rewards/weighted_rejected": 0.0, + "step": 1 + }, + { + "epoch": 0.005234231876472127, + "grad_norm": 269.8279113769531, + "learning_rate": 4.6875e-08, + "logits/chosen": -0.3176456093788147, + "logits/rejected": -0.3530849814414978, + "logps/chosen": -275.5694580078125, + "logps/rejected": -255.875, + "logps/weighted_chosen": -2.478325843811035, + "logps/weighted_rejected": -2.635009765625, + "loss": 0.6911, + "rewards/accuracies": 0.2604166567325592, + "rewards/chosen": -0.0588107630610466, + "rewards/margins": -0.0334201380610466, + "rewards/rejected": -0.025390625, + "rewards/weighted_accuracies": 0.3229166567325592, + "rewards/weighted_chosen": 0.0010899438057094812, + "rewards/weighted_margins": 0.0014837053604424, + "rewards/weighted_rejected": -0.00039333768654614687, + "step": 10 + }, + { + "epoch": 0.010468463752944255, + "grad_norm": 12.950738906860352, + "learning_rate": 9.895833333333332e-08, + "logits/chosen": -0.2977302670478821, + "logits/rejected": -0.3139175474643707, + "logps/chosen": -294.37579345703125, + "logps/rejected": -272.7203063964844, + "logps/weighted_chosen": -2.30804443359375, + "logps/weighted_rejected": -2.247570753097534, + "loss": 0.692, + "rewards/accuracies": 0.296875, + "rewards/chosen": -0.05908203125, + "rewards/margins": -0.01513671875, + "rewards/rejected": -0.0439453125, + "rewards/weighted_accuracies": 0.34687501192092896, + "rewards/weighted_chosen": 7.171630568336695e-05, + "rewards/weighted_margins": 3.662109520519152e-05, + "rewards/weighted_rejected": 3.509521411615424e-05, + "step": 20 + }, + { + "epoch": 0.015702695629416383, + "grad_norm": 65.14871215820312, + "learning_rate": 1.5104166666666664e-07, + "logits/chosen": -0.2917991578578949, + "logits/rejected": -0.3373458981513977, + "logps/chosen": -297.9375, + "logps/rejected": -268.0062561035156, + "logps/weighted_chosen": -2.3156495094299316, + "logps/weighted_rejected": -2.529247999191284, + "loss": 0.693, + "rewards/accuracies": 0.28125, + "rewards/chosen": 0.02177734300494194, + "rewards/margins": -0.06005859375, + "rewards/rejected": 0.08183594048023224, + "rewards/weighted_accuracies": 0.34062498807907104, + "rewards/weighted_chosen": 0.0012060165172442794, + "rewards/weighted_margins": -0.0021120072342455387, + "rewards/weighted_rejected": 0.0033180236350744963, + "step": 30 + }, + { + "epoch": 0.02093692750588851, + "grad_norm": 30.699859619140625, + "learning_rate": 2.03125e-07, + "logits/chosen": -0.301095575094223, + "logits/rejected": -0.3433547914028168, + "logps/chosen": -278.7124938964844, + "logps/rejected": -253.84530639648438, + "logps/weighted_chosen": -2.3508667945861816, + "logps/weighted_rejected": -2.55126953125, + "loss": 0.6909, + "rewards/accuracies": 0.2718749940395355, + "rewards/chosen": -0.014892578125, + "rewards/margins": -0.105224609375, + "rewards/rejected": 0.09033203125, + "rewards/weighted_accuracies": 0.359375, + "rewards/weighted_chosen": 0.0031726837623864412, + "rewards/weighted_margins": 0.0010789871448650956, + "rewards/weighted_rejected": 0.002093696501106024, + "step": 40 + }, + { + "epoch": 0.02617115938236064, + "grad_norm": 15.275498390197754, + "learning_rate": 2.552083333333333e-07, + "logits/chosen": -0.28184205293655396, + "logits/rejected": -0.32042425870895386, + "logps/chosen": -280.19140625, + "logps/rejected": -267.4398498535156, + "logps/weighted_chosen": -2.281604051589966, + "logps/weighted_rejected": -2.278125047683716, + "loss": 0.691, + "rewards/accuracies": 0.3125, + "rewards/chosen": 0.009082031436264515, + "rewards/margins": -0.11552734673023224, + "rewards/rejected": 0.12460937350988388, + "rewards/weighted_accuracies": 0.375, + "rewards/weighted_chosen": 0.0036834715865552425, + "rewards/weighted_margins": 0.0022117614280432463, + "rewards/weighted_rejected": 0.0014717101585119963, + "step": 50 + }, + { + "epoch": 0.031405391258832765, + "grad_norm": 42.36075973510742, + "learning_rate": 3.0729166666666665e-07, + "logits/chosen": -0.3147949278354645, + "logits/rejected": -0.3081672787666321, + "logps/chosen": -277.5492248535156, + "logps/rejected": -261.63201904296875, + "logps/weighted_chosen": -2.6925292015075684, + "logps/weighted_rejected": -2.5440917015075684, + "loss": 0.6888, + "rewards/accuracies": 0.39375001192092896, + "rewards/chosen": 0.15625, + "rewards/margins": 0.06132812425494194, + "rewards/rejected": 0.09492187201976776, + "rewards/weighted_accuracies": 0.43437498807907104, + "rewards/weighted_chosen": 0.0014663696056231856, + "rewards/weighted_margins": 0.0070289610885083675, + "rewards/weighted_rejected": -0.005612182430922985, + "step": 60 + }, + { + "epoch": 0.036639623135304895, + "grad_norm": 70.38953399658203, + "learning_rate": 3.59375e-07, + "logits/chosen": -0.31798094511032104, + "logits/rejected": -0.32500457763671875, + "logps/chosen": -289.6695251464844, + "logps/rejected": -244.7351531982422, + "logps/weighted_chosen": -2.19110107421875, + "logps/weighted_rejected": -2.427661180496216, + "loss": 0.6869, + "rewards/accuracies": 0.4375, + "rewards/chosen": 0.36865234375, + "rewards/margins": 0.34619140625, + "rewards/rejected": 0.0224609375, + "rewards/weighted_accuracies": 0.5375000238418579, + "rewards/weighted_chosen": 0.02108154259622097, + "rewards/weighted_margins": 0.011628913693130016, + "rewards/weighted_rejected": 0.009454727172851562, + "step": 70 + }, + { + "epoch": 0.04187385501177702, + "grad_norm": 54.81600570678711, + "learning_rate": 4.114583333333333e-07, + "logits/chosen": -0.2848251461982727, + "logits/rejected": -0.3061889708042145, + "logps/chosen": -289.08203125, + "logps/rejected": -269.95782470703125, + "logps/weighted_chosen": -2.3393311500549316, + "logps/weighted_rejected": -2.5093994140625, + "loss": 0.6827, + "rewards/accuracies": 0.44999998807907104, + "rewards/chosen": 0.679980456829071, + "rewards/margins": 0.3672851622104645, + "rewards/rejected": 0.31269532442092896, + "rewards/weighted_accuracies": 0.5062500238418579, + "rewards/weighted_chosen": 0.048927437514066696, + "rewards/weighted_margins": 0.025808846578001976, + "rewards/weighted_rejected": 0.02311401441693306, + "step": 80 + }, + { + "epoch": 0.04710808688824915, + "grad_norm": 30.603567123413086, + "learning_rate": 4.6354166666666664e-07, + "logits/chosen": -0.312631219625473, + "logits/rejected": -0.30586013197898865, + "logps/chosen": -279.92578125, + "logps/rejected": -257.5234375, + "logps/weighted_chosen": -2.5583739280700684, + "logps/weighted_rejected": -2.6037840843200684, + "loss": 0.6752, + "rewards/accuracies": 0.49687498807907104, + "rewards/chosen": 0.757617175579071, + "rewards/margins": 0.5347656011581421, + "rewards/rejected": 0.22285155951976776, + "rewards/weighted_accuracies": 0.546875, + "rewards/weighted_chosen": 0.08651771396398544, + "rewards/weighted_margins": 0.05735473707318306, + "rewards/weighted_rejected": 0.02917785570025444, + "step": 90 + }, + { + "epoch": 0.05234231876472128, + "grad_norm": 99.52488708496094, + "learning_rate": 5.156249999999999e-07, + "logits/chosen": -0.2793869078159332, + "logits/rejected": -0.3509994447231293, + "logps/chosen": -290.19061279296875, + "logps/rejected": -285.94842529296875, + "logps/weighted_chosen": -2.013537645339966, + "logps/weighted_rejected": -2.5231690406799316, + "loss": 0.6984, + "rewards/accuracies": 0.512499988079071, + "rewards/chosen": 0.72607421875, + "rewards/margins": 0.80712890625, + "rewards/rejected": -0.0810546875, + "rewards/weighted_accuracies": 0.6187499761581421, + "rewards/weighted_chosen": 0.07289886474609375, + "rewards/weighted_margins": 0.019171524792909622, + "rewards/weighted_rejected": 0.05372200161218643, + "step": 100 + }, + { + "epoch": 0.05757655064119341, + "grad_norm": 55.06239318847656, + "learning_rate": 5.677083333333333e-07, + "logits/chosen": -0.3277145326137543, + "logits/rejected": -0.3189544677734375, + "logps/chosen": -296.17889404296875, + "logps/rejected": -261.51953125, + "logps/weighted_chosen": -2.647631883621216, + "logps/weighted_rejected": -2.6458740234375, + "loss": 0.6844, + "rewards/accuracies": 0.518750011920929, + "rewards/chosen": 0.1376953125, + "rewards/margins": 0.789746105670929, + "rewards/rejected": -0.6522461175918579, + "rewards/weighted_accuracies": 0.565625011920929, + "rewards/weighted_chosen": 0.05432138592004776, + "rewards/weighted_margins": 0.03696594387292862, + "rewards/weighted_rejected": 0.01735992357134819, + "step": 110 + }, + { + "epoch": 0.06281078251766553, + "grad_norm": 42.45036315917969, + "learning_rate": 6.197916666666666e-07, + "logits/chosen": -0.33232802152633667, + "logits/rejected": -0.35865744948387146, + "logps/chosen": -294.6078186035156, + "logps/rejected": -255.01327514648438, + "logps/weighted_chosen": -1.970422387123108, + "logps/weighted_rejected": -2.1345458030700684, + "loss": 0.6837, + "rewards/accuracies": 0.609375, + "rewards/chosen": 0.0013671874767169356, + "rewards/margins": 1.2595703601837158, + "rewards/rejected": -1.258203148841858, + "rewards/weighted_accuracies": 0.59375, + "rewards/weighted_chosen": 0.048947714269161224, + "rewards/weighted_margins": 0.03674011304974556, + "rewards/weighted_rejected": 0.012198830023407936, + "step": 120 + }, + { + "epoch": 0.06804501439413765, + "grad_norm": 28.637327194213867, + "learning_rate": 6.718749999999999e-07, + "logits/chosen": -0.2933105528354645, + "logits/rejected": -0.29020920395851135, + "logps/chosen": -305.86602783203125, + "logps/rejected": -279.09844970703125, + "logps/weighted_chosen": -1.9946777820587158, + "logps/weighted_rejected": -2.2933349609375, + "loss": 0.6754, + "rewards/accuracies": 0.625, + "rewards/chosen": 0.285400390625, + "rewards/margins": 1.7852051258087158, + "rewards/rejected": -1.4998047351837158, + "rewards/weighted_accuracies": 0.6187499761581421, + "rewards/weighted_chosen": 0.06223602220416069, + "rewards/weighted_margins": 0.05499267578125, + "rewards/weighted_rejected": 0.0072036744095385075, + "step": 130 + }, + { + "epoch": 0.07327924627060979, + "grad_norm": 20.866477966308594, + "learning_rate": 7.239583333333333e-07, + "logits/chosen": -0.33005982637405396, + "logits/rejected": -0.34638214111328125, + "logps/chosen": -300.7281188964844, + "logps/rejected": -276.9156188964844, + "logps/weighted_chosen": -2.294506788253784, + "logps/weighted_rejected": -2.561084032058716, + "loss": 0.6606, + "rewards/accuracies": 0.628125011920929, + "rewards/chosen": -1.998437523841858, + "rewards/margins": 2.3173828125, + "rewards/rejected": -4.315820217132568, + "rewards/weighted_accuracies": 0.671875, + "rewards/weighted_chosen": 0.04616241529583931, + "rewards/weighted_margins": 0.08864898979663849, + "rewards/weighted_rejected": -0.04253540188074112, + "step": 140 + }, + { + "epoch": 0.07851347814708191, + "grad_norm": 43.497623443603516, + "learning_rate": 7.760416666666666e-07, + "logits/chosen": -0.2890525758266449, + "logits/rejected": -0.30130767822265625, + "logps/chosen": -287.3109436035156, + "logps/rejected": -256.72967529296875, + "logps/weighted_chosen": -2.883862257003784, + "logps/weighted_rejected": -2.796630859375, + "loss": 0.664, + "rewards/accuracies": 0.628125011920929, + "rewards/chosen": -3.108203172683716, + "rewards/margins": 2.7275390625, + "rewards/rejected": -5.835741996765137, + "rewards/weighted_accuracies": 0.6343749761581421, + "rewards/weighted_chosen": 0.04738159105181694, + "rewards/weighted_margins": 0.09773864597082138, + "rewards/weighted_rejected": -0.05034027248620987, + "step": 150 + }, + { + "epoch": 0.08374771002355404, + "grad_norm": 12.989988327026367, + "learning_rate": 8.28125e-07, + "logits/chosen": -0.3334007263183594, + "logits/rejected": -0.3831237852573395, + "logps/chosen": -308.2320251464844, + "logps/rejected": -281.5335998535156, + "logps/weighted_chosen": -2.378857374191284, + "logps/weighted_rejected": -2.3824095726013184, + "loss": 0.6464, + "rewards/accuracies": 0.6499999761581421, + "rewards/chosen": -4.347460746765137, + "rewards/margins": 3.98388671875, + "rewards/rejected": -8.331347465515137, + "rewards/weighted_accuracies": 0.65625, + "rewards/weighted_chosen": 0.04017028957605362, + "rewards/weighted_margins": 0.15716858208179474, + "rewards/weighted_rejected": -0.11710510402917862, + "step": 160 + }, + { + "epoch": 0.08898194190002617, + "grad_norm": 691.15625, + "learning_rate": 8.802083333333333e-07, + "logits/chosen": -0.37783128023147583, + "logits/rejected": -0.37789613008499146, + "logps/chosen": -310.76953125, + "logps/rejected": -270.8374938964844, + "logps/weighted_chosen": -2.6180419921875, + "logps/weighted_rejected": -2.8051514625549316, + "loss": 0.683, + "rewards/accuracies": 0.6156250238418579, + "rewards/chosen": -7.780859470367432, + "rewards/margins": 4.9013671875, + "rewards/rejected": -12.683398246765137, + "rewards/weighted_accuracies": 0.643750011920929, + "rewards/weighted_chosen": 0.06108245998620987, + "rewards/weighted_margins": 0.17230224609375, + "rewards/weighted_rejected": -0.11125946044921875, + "step": 170 + }, + { + "epoch": 0.0942161737764983, + "grad_norm": 65.56312561035156, + "learning_rate": 9.322916666666666e-07, + "logits/chosen": -0.3433074951171875, + "logits/rejected": -0.363290399312973, + "logps/chosen": -285.8812561035156, + "logps/rejected": -274.94842529296875, + "logps/weighted_chosen": -2.50494384765625, + "logps/weighted_rejected": -2.9246826171875, + "loss": 0.6821, + "rewards/accuracies": 0.6343749761581421, + "rewards/chosen": -12.179101943969727, + "rewards/margins": 6.432421684265137, + "rewards/rejected": -18.608592987060547, + "rewards/weighted_accuracies": 0.59375, + "rewards/weighted_chosen": -0.0216827392578125, + "rewards/weighted_margins": 0.19813232123851776, + "rewards/weighted_rejected": -0.2197723388671875, + "step": 180 + }, + { + "epoch": 0.09945040565297043, + "grad_norm": 26.828750610351562, + "learning_rate": 9.84375e-07, + "logits/chosen": -0.36560744047164917, + "logits/rejected": -0.40380173921585083, + "logps/chosen": -321.97967529296875, + "logps/rejected": -287.5570373535156, + "logps/weighted_chosen": -2.3037109375, + "logps/weighted_rejected": -2.64697265625, + "loss": 0.6489, + "rewards/accuracies": 0.596875011920929, + "rewards/chosen": -16.246679306030273, + "rewards/margins": 7.440039157867432, + "rewards/rejected": -23.685937881469727, + "rewards/weighted_accuracies": 0.640625, + "rewards/weighted_chosen": -0.11536254733800888, + "rewards/weighted_margins": 0.19368895888328552, + "rewards/weighted_rejected": -0.309121698141098, + "step": 190 + }, + { + "epoch": 0.10468463752944256, + "grad_norm": 117.84883880615234, + "learning_rate": 9.99959085414323e-07, + "logits/chosen": -0.4036270081996918, + "logits/rejected": -0.43666380643844604, + "logps/chosen": -327.8265686035156, + "logps/rejected": -284.4234313964844, + "logps/weighted_chosen": -2.6368165016174316, + "logps/weighted_rejected": -3.0156006813049316, + "loss": 0.6383, + "rewards/accuracies": 0.6156250238418579, + "rewards/chosen": -18.538379669189453, + "rewards/margins": 8.212695121765137, + "rewards/rejected": -26.748046875, + "rewards/weighted_accuracies": 0.6468750238418579, + "rewards/weighted_chosen": -0.12391586601734161, + "rewards/weighted_margins": 0.28039854764938354, + "rewards/weighted_rejected": -0.404205322265625, + "step": 200 + }, + { + "epoch": 0.10991886940591468, + "grad_norm": 15.525572776794434, + "learning_rate": 9.997587035630105e-07, + "logits/chosen": -0.41632765531539917, + "logits/rejected": -0.4597106873989105, + "logps/chosen": -308.75469970703125, + "logps/rejected": -316.69842529296875, + "logps/weighted_chosen": -2.439282178878784, + "logps/weighted_rejected": -3.152050733566284, + "loss": 0.6437, + "rewards/accuracies": 0.653124988079071, + "rewards/chosen": -21.632617950439453, + "rewards/margins": 10.560155868530273, + "rewards/rejected": -32.18867111206055, + "rewards/weighted_accuracies": 0.671875, + "rewards/weighted_chosen": -0.2110244780778885, + "rewards/weighted_margins": 0.23976440727710724, + "rewards/weighted_rejected": -0.4508819580078125, + "step": 210 + }, + { + "epoch": 0.11515310128238682, + "grad_norm": 31.583986282348633, + "learning_rate": 9.99391406364405e-07, + "logits/chosen": -0.37994080781936646, + "logits/rejected": -0.384744256734848, + "logps/chosen": -318.63592529296875, + "logps/rejected": -305.0, + "logps/weighted_chosen": -2.9573731422424316, + "logps/weighted_rejected": -3.207348585128784, + "loss": 0.7259, + "rewards/accuracies": 0.606249988079071, + "rewards/chosen": -26.522266387939453, + "rewards/margins": 9.657129287719727, + "rewards/rejected": -36.18437576293945, + "rewards/weighted_accuracies": 0.625, + "rewards/weighted_chosen": -0.3494918942451477, + "rewards/weighted_margins": 0.17980042099952698, + "rewards/weighted_rejected": -0.5295776128768921, + "step": 220 + }, + { + "epoch": 0.12038733315885894, + "grad_norm": 611.2006225585938, + "learning_rate": 9.988573164927884e-07, + "logits/chosen": -0.3254447877407074, + "logits/rejected": -0.36442261934280396, + "logps/chosen": -297.16717529296875, + "logps/rejected": -296.48907470703125, + "logps/weighted_chosen": -2.35888671875, + "logps/weighted_rejected": -2.804126024246216, + "loss": 0.6637, + "rewards/accuracies": 0.6656249761581421, + "rewards/chosen": -28.293750762939453, + "rewards/margins": 16.522851943969727, + "rewards/rejected": -44.817970275878906, + "rewards/weighted_accuracies": 0.6812499761581421, + "rewards/weighted_chosen": -0.1478603333234787, + "rewards/weighted_margins": 0.3882904052734375, + "rewards/weighted_rejected": -0.5360336303710938, + "step": 230 + }, + { + "epoch": 0.12562156503533106, + "grad_norm": 34.738059997558594, + "learning_rate": 9.98156612329838e-07, + "logits/chosen": -0.3876676559448242, + "logits/rejected": -0.4364303648471832, + "logps/chosen": -312.5921936035156, + "logps/rejected": -348.60626220703125, + "logps/weighted_chosen": -2.594775438308716, + "logps/weighted_rejected": -3.1068115234375, + "loss": 0.7482, + "rewards/accuracies": 0.637499988079071, + "rewards/chosen": -47.75507736206055, + "rewards/margins": 18.762109756469727, + "rewards/rejected": -66.5289077758789, + "rewards/weighted_accuracies": 0.6499999761581421, + "rewards/weighted_chosen": -0.319406121969223, + "rewards/weighted_margins": 0.204498291015625, + "rewards/weighted_rejected": -0.5241454839706421, + "step": 240 + }, + { + "epoch": 0.13085579691180318, + "grad_norm": 141.2784423828125, + "learning_rate": 9.97289527905053e-07, + "logits/chosen": -0.4159797728061676, + "logits/rejected": -0.4382568299770355, + "logps/chosen": -308.44842529296875, + "logps/rejected": -313.49530029296875, + "logps/weighted_chosen": -2.802807569503784, + "logps/weighted_rejected": -2.9874024391174316, + "loss": 0.7583, + "rewards/accuracies": 0.609375, + "rewards/chosen": -44.16621017456055, + "rewards/margins": 13.382226943969727, + "rewards/rejected": -57.54296875, + "rewards/weighted_accuracies": 0.6187499761581421, + "rewards/weighted_chosen": -0.20551452040672302, + "rewards/weighted_margins": 0.22150878608226776, + "rewards/weighted_rejected": -0.42730408906936646, + "step": 250 + }, + { + "epoch": 0.1360900287882753, + "grad_norm": 75.63247680664062, + "learning_rate": 9.962563528175875e-07, + "logits/chosen": -0.4156021177768707, + "logits/rejected": -0.4493362307548523, + "logps/chosen": -342.1312561035156, + "logps/rejected": -317.84063720703125, + "logps/weighted_chosen": -2.56585693359375, + "logps/weighted_rejected": -3.4276366233825684, + "loss": 0.6673, + "rewards/accuracies": 0.5625, + "rewards/chosen": -43.44648361206055, + "rewards/margins": 13.459375381469727, + "rewards/rejected": -56.90898513793945, + "rewards/weighted_accuracies": 0.6156250238418579, + "rewards/weighted_chosen": -0.310464471578598, + "rewards/weighted_margins": 0.32973939180374146, + "rewards/weighted_rejected": -0.6404876708984375, + "step": 260 + }, + { + "epoch": 0.14132426066474746, + "grad_norm": 32.991607666015625, + "learning_rate": 9.950574321395277e-07, + "logits/chosen": -0.507153332233429, + "logits/rejected": -0.5264068841934204, + "logps/chosen": -317.0335998535156, + "logps/rejected": -300.5718688964844, + "logps/weighted_chosen": -2.669018507003784, + "logps/weighted_rejected": -2.8846678733825684, + "loss": 0.6774, + "rewards/accuracies": 0.609375, + "rewards/chosen": -27.629491806030273, + "rewards/margins": 9.762109756469727, + "rewards/rejected": -37.39765548706055, + "rewards/weighted_accuracies": 0.6000000238418579, + "rewards/weighted_chosen": -0.23316803574562073, + "rewards/weighted_margins": 0.2321929931640625, + "rewards/weighted_rejected": -0.46527099609375, + "step": 270 + }, + { + "epoch": 0.14655849254121958, + "grad_norm": 41.87471389770508, + "learning_rate": 9.936931663006413e-07, + "logits/chosen": -0.5015045404434204, + "logits/rejected": -0.49525564908981323, + "logps/chosen": -319.9296875, + "logps/rejected": -312.28826904296875, + "logps/weighted_chosen": -2.667797803878784, + "logps/weighted_rejected": -3.121386766433716, + "loss": 0.6458, + "rewards/accuracies": 0.6625000238418579, + "rewards/chosen": -17.077342987060547, + "rewards/margins": 15.751367568969727, + "rewards/rejected": -32.82636642456055, + "rewards/weighted_accuracies": 0.703125, + "rewards/weighted_chosen": -0.10239715874195099, + "rewards/weighted_margins": 0.3067169189453125, + "rewards/weighted_rejected": -0.409027099609375, + "step": 280 + }, + { + "epoch": 0.1517927244176917, + "grad_norm": 22.655261993408203, + "learning_rate": 9.921640109546357e-07, + "logits/chosen": -0.4936012327671051, + "logits/rejected": -0.5640270113945007, + "logps/chosen": -286.4117126464844, + "logps/rejected": -286.15313720703125, + "logps/weighted_chosen": -2.640380859375, + "logps/weighted_rejected": -3.774829149246216, + "loss": 0.6369, + "rewards/accuracies": 0.6343749761581421, + "rewards/chosen": -19.629491806030273, + "rewards/margins": 14.98291015625, + "rewards/rejected": -34.61503982543945, + "rewards/weighted_accuracies": 0.668749988079071, + "rewards/weighted_chosen": -0.21005859971046448, + "rewards/weighted_margins": 0.3854110836982727, + "rewards/weighted_rejected": -0.5955780148506165, + "step": 290 + }, + { + "epoch": 0.15702695629416383, + "grad_norm": 33.83458709716797, + "learning_rate": 9.90470476826975e-07, + "logits/chosen": -0.5687958002090454, + "logits/rejected": -0.5700439214706421, + "logps/chosen": -298.06561279296875, + "logps/rejected": -311.8359375, + "logps/weighted_chosen": -2.555248975753784, + "logps/weighted_rejected": -3.0333251953125, + "loss": 0.6326, + "rewards/accuracies": 0.6625000238418579, + "rewards/chosen": -28.754297256469727, + "rewards/margins": 15.813085556030273, + "rewards/rejected": -44.567771911621094, + "rewards/weighted_accuracies": 0.6468750238418579, + "rewards/weighted_chosen": -0.25576478242874146, + "rewards/weighted_margins": 0.3327178955078125, + "rewards/weighted_rejected": -0.5886474847793579, + "step": 300 + }, + { + "epoch": 0.16226118817063595, + "grad_norm": 29.578954696655273, + "learning_rate": 9.886131295443002e-07, + "logits/chosen": -0.6218292117118835, + "logits/rejected": -0.6808746457099915, + "logps/chosen": -313.44061279296875, + "logps/rejected": -297.9156188964844, + "logps/weighted_chosen": -2.719403028488159, + "logps/weighted_rejected": -3.1142334938049316, + "loss": 0.6221, + "rewards/accuracies": 0.653124988079071, + "rewards/chosen": -32.1875, + "rewards/margins": 15.571874618530273, + "rewards/rejected": -47.7666015625, + "rewards/weighted_accuracies": 0.6499999761581421, + "rewards/weighted_chosen": -0.19769592583179474, + "rewards/weighted_margins": 0.521374523639679, + "rewards/weighted_rejected": -0.7188171148300171, + "step": 310 + }, + { + "epoch": 0.16749542004710807, + "grad_norm": 15.948070526123047, + "learning_rate": 9.865925894455166e-07, + "logits/chosen": -0.6636382937431335, + "logits/rejected": -0.6817939877510071, + "logps/chosen": -336.1507873535156, + "logps/rejected": -306.62890625, + "logps/weighted_chosen": -2.83941650390625, + "logps/weighted_rejected": -3.273486375808716, + "loss": 0.7046, + "rewards/accuracies": 0.637499988079071, + "rewards/chosen": -36.21074295043945, + "rewards/margins": 12.9765625, + "rewards/rejected": -49.17656326293945, + "rewards/weighted_accuracies": 0.6343749761581421, + "rewards/weighted_chosen": -0.3679946959018707, + "rewards/weighted_margins": 0.278738409280777, + "rewards/weighted_rejected": -0.6464904546737671, + "step": 320 + }, + { + "epoch": 0.17272965192358022, + "grad_norm": 53.10931396484375, + "learning_rate": 9.84409531374603e-07, + "logits/chosen": -0.6222923398017883, + "logits/rejected": -0.607867419719696, + "logps/chosen": -344.19061279296875, + "logps/rejected": -315.84686279296875, + "logps/weighted_chosen": -2.915234327316284, + "logps/weighted_rejected": -3.346630811691284, + "loss": 0.6334, + "rewards/accuracies": 0.668749988079071, + "rewards/chosen": -39.39531326293945, + "rewards/margins": 13.606640815734863, + "rewards/rejected": -53.00273513793945, + "rewards/weighted_accuracies": 0.65625, + "rewards/weighted_chosen": -0.33662718534469604, + "rewards/weighted_margins": 0.3510375916957855, + "rewards/weighted_rejected": -0.6879855990409851, + "step": 330 + }, + { + "epoch": 0.17796388380005235, + "grad_norm": 13.994344711303711, + "learning_rate": 9.820646844552219e-07, + "logits/chosen": -0.6615959405899048, + "logits/rejected": -0.714038074016571, + "logps/chosen": -312.7328186035156, + "logps/rejected": -323.3359375, + "logps/weighted_chosen": -2.910736083984375, + "logps/weighted_rejected": -3.2298340797424316, + "loss": 0.6376, + "rewards/accuracies": 0.675000011920929, + "rewards/chosen": -36.327537536621094, + "rewards/margins": 21.465625762939453, + "rewards/rejected": -57.79179763793945, + "rewards/weighted_accuracies": 0.675000011920929, + "rewards/weighted_chosen": -0.34371185302734375, + "rewards/weighted_margins": 0.390716552734375, + "rewards/weighted_rejected": -0.734301745891571, + "step": 340 + }, + { + "epoch": 0.18319811567652447, + "grad_norm": 13.682384490966797, + "learning_rate": 9.795588318471964e-07, + "logits/chosen": -0.774444580078125, + "logits/rejected": -0.7887184023857117, + "logps/chosen": -302.76251220703125, + "logps/rejected": -336.90313720703125, + "logps/weighted_chosen": -2.718432664871216, + "logps/weighted_rejected": -3.0622315406799316, + "loss": 0.6451, + "rewards/accuracies": 0.6343749761581421, + "rewards/chosen": -43.29413986206055, + "rewards/margins": 17.229297637939453, + "rewards/rejected": -60.52812576293945, + "rewards/weighted_accuracies": 0.637499988079071, + "rewards/weighted_chosen": -0.33645325899124146, + "rewards/weighted_margins": 0.33460694551467896, + "rewards/weighted_rejected": -0.6709426641464233, + "step": 350 + }, + { + "epoch": 0.1884323475529966, + "grad_norm": 19.35496711730957, + "learning_rate": 9.768928104849415e-07, + "logits/chosen": -0.815447986125946, + "logits/rejected": -0.8216186761856079, + "logps/chosen": -330.66015625, + "logps/rejected": -313.8687438964844, + "logps/weighted_chosen": -2.970752000808716, + "logps/weighted_rejected": -3.1889405250549316, + "loss": 0.6763, + "rewards/accuracies": 0.621874988079071, + "rewards/chosen": -47.90937423706055, + "rewards/margins": 17.064844131469727, + "rewards/rejected": -64.98515319824219, + "rewards/weighted_accuracies": 0.6499999761581421, + "rewards/weighted_chosen": -0.3897460997104645, + "rewards/weighted_margins": 0.384918212890625, + "rewards/weighted_rejected": -0.7745300531387329, + "step": 360 + }, + { + "epoch": 0.19366657942946872, + "grad_norm": 21.70383071899414, + "learning_rate": 9.740675107979355e-07, + "logits/chosen": -0.7939224243164062, + "logits/rejected": -0.827624499797821, + "logps/chosen": -373.0406188964844, + "logps/rejected": -349.3343811035156, + "logps/weighted_chosen": -2.506054639816284, + "logps/weighted_rejected": -3.43896484375, + "loss": 0.6215, + "rewards/accuracies": 0.596875011920929, + "rewards/chosen": -58.082420349121094, + "rewards/margins": 16.594335556030273, + "rewards/rejected": -74.671875, + "rewards/weighted_accuracies": 0.671875, + "rewards/weighted_chosen": -0.48696595430374146, + "rewards/weighted_margins": 0.412750244140625, + "rewards/weighted_rejected": -0.90020751953125, + "step": 370 + }, + { + "epoch": 0.19890081130594087, + "grad_norm": 21.682790756225586, + "learning_rate": 9.71083876413323e-07, + "logits/chosen": -0.77825927734375, + "logits/rejected": -0.7921401858329773, + "logps/chosen": -367.0062561035156, + "logps/rejected": -355.78594970703125, + "logps/weighted_chosen": -2.644775390625, + "logps/weighted_rejected": -3.5374755859375, + "loss": 0.6609, + "rewards/accuracies": 0.628125011920929, + "rewards/chosen": -64.59492492675781, + "rewards/margins": 21.614843368530273, + "rewards/rejected": -86.1988296508789, + "rewards/weighted_accuracies": 0.6625000238418579, + "rewards/weighted_chosen": -0.5685027837753296, + "rewards/weighted_margins": 0.3586792051792145, + "rewards/weighted_rejected": -0.9269439578056335, + "step": 380 + }, + { + "epoch": 0.204135043182413, + "grad_norm": 30.26679229736328, + "learning_rate": 9.67942903840751e-07, + "logits/chosen": -0.8401764035224915, + "logits/rejected": -0.898754894733429, + "logps/chosen": -362.0328063964844, + "logps/rejected": -362.3687438964844, + "logps/weighted_chosen": -2.8179688453674316, + "logps/weighted_rejected": -3.4515624046325684, + "loss": 0.5992, + "rewards/accuracies": 0.668749988079071, + "rewards/chosen": -57.87226486206055, + "rewards/margins": 30.196680068969727, + "rewards/rejected": -88.0804672241211, + "rewards/weighted_accuracies": 0.715624988079071, + "rewards/weighted_chosen": -0.456369012594223, + "rewards/weighted_margins": 0.495849609375, + "rewards/weighted_rejected": -0.951983630657196, + "step": 390 + }, + { + "epoch": 0.2093692750588851, + "grad_norm": 37.81236267089844, + "learning_rate": 9.646456421395447e-07, + "logits/chosen": -0.8902359008789062, + "logits/rejected": -0.9107666015625, + "logps/chosen": -380.6343688964844, + "logps/rejected": -399.7281188964844, + "logps/weighted_chosen": -2.709228515625, + "logps/weighted_rejected": -3.455004930496216, + "loss": 0.6557, + "rewards/accuracies": 0.6812499761581421, + "rewards/chosen": -62.43437576293945, + "rewards/margins": 33.634765625, + "rewards/rejected": -96.0941390991211, + "rewards/weighted_accuracies": 0.653124988079071, + "rewards/weighted_chosen": -0.4688262939453125, + "rewards/weighted_margins": 0.3546081483364105, + "rewards/weighted_rejected": -0.8232513666152954, + "step": 400 + }, + { + "epoch": 0.21460350693535724, + "grad_norm": 21.761716842651367, + "learning_rate": 9.611931925683266e-07, + "logits/chosen": -0.849993109703064, + "logits/rejected": -0.8886367678642273, + "logps/chosen": -368.8374938964844, + "logps/rejected": -353.73126220703125, + "logps/weighted_chosen": -2.533374071121216, + "logps/weighted_rejected": -3.233935594558716, + "loss": 0.591, + "rewards/accuracies": 0.6656249761581421, + "rewards/chosen": -62.575782775878906, + "rewards/margins": 27.423242568969727, + "rewards/rejected": -90.0132827758789, + "rewards/weighted_accuracies": 0.6937500238418579, + "rewards/weighted_chosen": -0.4288879334926605, + "rewards/weighted_margins": 0.507708728313446, + "rewards/weighted_rejected": -0.9364013671875, + "step": 410 + }, + { + "epoch": 0.21983773881182936, + "grad_norm": 28.842378616333008, + "learning_rate": 9.575867082172085e-07, + "logits/chosen": -0.836225152015686, + "logits/rejected": -0.8854034543037415, + "logps/chosen": -371.90625, + "logps/rejected": -369.4125061035156, + "logps/weighted_chosen": -2.9190917015075684, + "logps/weighted_rejected": -3.118579149246216, + "loss": 0.6198, + "rewards/accuracies": 0.668749988079071, + "rewards/chosen": -66.98905944824219, + "rewards/margins": 32.337890625, + "rewards/rejected": -99.3402328491211, + "rewards/weighted_accuracies": 0.6625000238418579, + "rewards/weighted_chosen": -0.4911743104457855, + "rewards/weighted_margins": 0.5335143804550171, + "rewards/weighted_rejected": -1.0242736339569092, + "step": 420 + }, + { + "epoch": 0.22507197068830148, + "grad_norm": 16.805343627929688, + "learning_rate": 9.538273936226673e-07, + "logits/chosen": -0.8646026849746704, + "logits/rejected": -0.898516833782196, + "logps/chosen": -328.94921875, + "logps/rejected": -353.66717529296875, + "logps/weighted_chosen": -3.1983399391174316, + "logps/weighted_rejected": -3.6509766578674316, + "loss": 0.6173, + "rewards/accuracies": 0.625, + "rewards/chosen": -63.52910232543945, + "rewards/margins": 25.6201171875, + "rewards/rejected": -89.1246109008789, + "rewards/weighted_accuracies": 0.6499999761581421, + "rewards/weighted_chosen": -0.436279296875, + "rewards/weighted_margins": 0.438516229391098, + "rewards/weighted_rejected": -0.8747711181640625, + "step": 430 + }, + { + "epoch": 0.23030620256477363, + "grad_norm": 23.78999137878418, + "learning_rate": 9.499165043652391e-07, + "logits/chosen": -0.8759521245956421, + "logits/rejected": -0.8865814208984375, + "logps/chosen": -358.69140625, + "logps/rejected": -358.76251220703125, + "logps/weighted_chosen": -3.2625732421875, + "logps/weighted_rejected": -3.359301805496216, + "loss": 0.6366, + "rewards/accuracies": 0.6187499761581421, + "rewards/chosen": -67.4326171875, + "rewards/margins": 24.610157012939453, + "rewards/rejected": -92.03788757324219, + "rewards/weighted_accuracies": 0.668749988079071, + "rewards/weighted_chosen": -0.5896576046943665, + "rewards/weighted_margins": 0.34770506620407104, + "rewards/weighted_rejected": -0.9376128911972046, + "step": 440 + }, + { + "epoch": 0.23554043444124576, + "grad_norm": 49.646263122558594, + "learning_rate": 9.458553466501665e-07, + "logits/chosen": -0.9158996343612671, + "logits/rejected": -0.9590820074081421, + "logps/chosen": -355.3109436035156, + "logps/rejected": -341.9765625, + "logps/weighted_chosen": -3.231823682785034, + "logps/weighted_rejected": -3.4568848609924316, + "loss": 0.657, + "rewards/accuracies": 0.6312500238418579, + "rewards/chosen": -69.14453125, + "rewards/margins": 28.884180068969727, + "rewards/rejected": -98.0492172241211, + "rewards/weighted_accuracies": 0.6875, + "rewards/weighted_chosen": -0.6470184326171875, + "rewards/weighted_margins": 0.4533935487270355, + "rewards/weighted_rejected": -1.100683569908142, + "step": 450 + }, + { + "epoch": 0.24077466631771788, + "grad_norm": 38.45325469970703, + "learning_rate": 9.416452768711366e-07, + "logits/chosen": -0.9281005859375, + "logits/rejected": -0.9545837640762329, + "logps/chosen": -375.6390686035156, + "logps/rejected": -370.203125, + "logps/weighted_chosen": -2.9860596656799316, + "logps/weighted_rejected": -3.635498046875, + "loss": 0.6457, + "rewards/accuracies": 0.612500011920929, + "rewards/chosen": -83.3871078491211, + "rewards/margins": 26.872264862060547, + "rewards/rejected": -110.26094055175781, + "rewards/weighted_accuracies": 0.6625000238418579, + "rewards/weighted_chosen": -0.6367782354354858, + "rewards/weighted_margins": 0.515301525592804, + "rewards/weighted_rejected": -1.152197241783142, + "step": 460 + }, + { + "epoch": 0.24600889819419, + "grad_norm": 59.67642593383789, + "learning_rate": 9.372877011572557e-07, + "logits/chosen": -0.8797394037246704, + "logits/rejected": -0.904949963092804, + "logps/chosen": -398.8531188964844, + "logps/rejected": -388.22186279296875, + "logps/weighted_chosen": -3.0103516578674316, + "logps/weighted_rejected": -3.409374952316284, + "loss": 0.6181, + "rewards/accuracies": 0.6031249761581421, + "rewards/chosen": -88.2437515258789, + "rewards/margins": 25.542577743530273, + "rewards/rejected": -113.7890625, + "rewards/weighted_accuracies": 0.6781250238418579, + "rewards/weighted_chosen": -0.6517273187637329, + "rewards/weighted_margins": 0.482818603515625, + "rewards/weighted_rejected": -1.1340011358261108, + "step": 470 + }, + { + "epoch": 0.2512431300706621, + "grad_norm": 27.465656280517578, + "learning_rate": 9.327840749034141e-07, + "logits/chosen": -0.910723865032196, + "logits/rejected": -0.9439147710800171, + "logps/chosen": -364.4750061035156, + "logps/rejected": -392.03436279296875, + "logps/weighted_chosen": -2.8880372047424316, + "logps/weighted_rejected": -4.069311618804932, + "loss": 0.6276, + "rewards/accuracies": 0.6812499761581421, + "rewards/chosen": -77.44355773925781, + "rewards/margins": 38.147850036621094, + "rewards/rejected": -115.59648132324219, + "rewards/weighted_accuracies": 0.684374988079071, + "rewards/weighted_chosen": -0.5768798589706421, + "rewards/weighted_margins": 0.541027843952179, + "rewards/weighted_rejected": -1.1178162097930908, + "step": 480 + }, + { + "epoch": 0.2564773619471343, + "grad_norm": 34.35914993286133, + "learning_rate": 9.281359022841965e-07, + "logits/chosen": -0.80584716796875, + "logits/rejected": -0.8291991949081421, + "logps/chosen": -349.86798095703125, + "logps/rejected": -354.75701904296875, + "logps/weighted_chosen": -2.914257764816284, + "logps/weighted_rejected": -4.178680419921875, + "loss": 0.63, + "rewards/accuracies": 0.625, + "rewards/chosen": -74.94511413574219, + "rewards/margins": 34.957618713378906, + "rewards/rejected": -109.90644836425781, + "rewards/weighted_accuracies": 0.690625011920929, + "rewards/weighted_chosen": -0.564099133014679, + "rewards/weighted_margins": 0.658795177936554, + "rewards/weighted_rejected": -1.2229950428009033, + "step": 490 + }, + { + "epoch": 0.26171159382360637, + "grad_norm": 41.01088333129883, + "learning_rate": 9.233447357514989e-07, + "logits/chosen": -0.775347888469696, + "logits/rejected": -0.8226684331893921, + "logps/chosen": -369.1148376464844, + "logps/rejected": -373.8374938964844, + "logps/weighted_chosen": -3.288012742996216, + "logps/weighted_rejected": -3.7696776390075684, + "loss": 0.6225, + "rewards/accuracies": 0.643750011920929, + "rewards/chosen": -71.1283187866211, + "rewards/margins": 32.867576599121094, + "rewards/rejected": -104.0218734741211, + "rewards/weighted_accuracies": 0.6937500238418579, + "rewards/weighted_chosen": -0.5540069341659546, + "rewards/weighted_margins": 0.598828136920929, + "rewards/weighted_rejected": -1.1535155773162842, + "step": 500 + }, + { + "epoch": 0.26171159382360637, + "eval_logits/chosen": -0.8767929673194885, + "eval_logits/rejected": -0.8968105316162109, + "eval_logps/chosen": -369.21099853515625, + "eval_logps/rejected": -379.3609924316406, + "eval_logps/weighted_chosen": -2.957455635070801, + "eval_logps/weighted_rejected": -3.6658689975738525, + "eval_loss": 0.6115986108779907, + "eval_rewards/accuracies": 0.6294999718666077, + "eval_rewards/chosen": -80.43468475341797, + "eval_rewards/margins": 31.297624588012695, + "eval_rewards/rejected": -111.73987579345703, + "eval_rewards/weighted_accuracies": 0.6759999990463257, + "eval_rewards/weighted_chosen": -0.5706117153167725, + "eval_rewards/weighted_margins": 0.5400464534759521, + "eval_rewards/weighted_rejected": -1.1106581687927246, + "eval_runtime": 1361.4222, + "eval_samples_per_second": 1.469, + "eval_steps_per_second": 0.367, + "step": 500 + }, + { + "epoch": 0.2669458257000785, + "grad_norm": 37.983856201171875, + "learning_rate": 9.184121755160232e-07, + "logits/chosen": -0.8435028195381165, + "logits/rejected": -0.8772613406181335, + "logps/chosen": -378.20782470703125, + "logps/rejected": -406.3187561035156, + "logps/weighted_chosen": -3.217089891433716, + "logps/weighted_rejected": -3.7809815406799316, + "loss": 0.6028, + "rewards/accuracies": 0.6656249761581421, + "rewards/chosen": -77.0746078491211, + "rewards/margins": 40.68242263793945, + "rewards/rejected": -117.759765625, + "rewards/weighted_accuracies": 0.6875, + "rewards/weighted_chosen": -0.5556915402412415, + "rewards/weighted_margins": 0.5749145746231079, + "rewards/weighted_rejected": -1.130804419517517, + "step": 510 + }, + { + "epoch": 0.2721800575765506, + "grad_norm": 46.08055877685547, + "learning_rate": 9.133398690128193e-07, + "logits/chosen": -0.894885241985321, + "logits/rejected": -0.9157882928848267, + "logps/chosen": -405.5679626464844, + "logps/rejected": -415.10076904296875, + "logps/weighted_chosen": -3.0770506858825684, + "logps/weighted_rejected": -3.949780225753784, + "loss": 0.5762, + "rewards/accuracies": 0.65625, + "rewards/chosen": -84.00956726074219, + "rewards/margins": 50.470703125, + "rewards/rejected": -134.46737670898438, + "rewards/weighted_accuracies": 0.7250000238418579, + "rewards/weighted_chosen": -0.49931639432907104, + "rewards/weighted_margins": 0.658154308795929, + "rewards/weighted_rejected": -1.1574280261993408, + "step": 520 + }, + { + "epoch": 0.27741428945302277, + "grad_norm": 40.89746856689453, + "learning_rate": 9.081295103510554e-07, + "logits/chosen": -0.8876678347587585, + "logits/rejected": -0.9430938959121704, + "logps/chosen": -383.52734375, + "logps/rejected": -413.60626220703125, + "logps/weighted_chosen": -2.752880811691284, + "logps/weighted_rejected": -4.01953125, + "loss": 0.5471, + "rewards/accuracies": 0.6468750238418579, + "rewards/chosen": -99.4878921508789, + "rewards/margins": 44.412109375, + "rewards/rejected": -143.859375, + "rewards/weighted_accuracies": 0.706250011920929, + "rewards/weighted_chosen": -0.6086364984512329, + "rewards/weighted_margins": 0.8298507928848267, + "rewards/weighted_rejected": -1.438207983970642, + "step": 530 + }, + { + "epoch": 0.2826485213294949, + "grad_norm": 27.849748611450195, + "learning_rate": 9.027828397481989e-07, + "logits/chosen": -0.8531036376953125, + "logits/rejected": -0.885467529296875, + "logps/chosen": -354.4359436035156, + "logps/rejected": -387.6328125, + "logps/weighted_chosen": -3.355029344558716, + "logps/weighted_rejected": -4.168432712554932, + "loss": 0.5956, + "rewards/accuracies": 0.625, + "rewards/chosen": -93.89375305175781, + "rewards/margins": 38.216407775878906, + "rewards/rejected": -132.11776733398438, + "rewards/weighted_accuracies": 0.675000011920929, + "rewards/weighted_chosen": -0.6640685796737671, + "rewards/weighted_margins": 0.6357482671737671, + "rewards/weighted_rejected": -1.2999756336212158, + "step": 540 + }, + { + "epoch": 0.287882753205967, + "grad_norm": 22.531660079956055, + "learning_rate": 8.973016429487988e-07, + "logits/chosen": -0.8864593505859375, + "logits/rejected": -0.9035919308662415, + "logps/chosen": -375.8890686035156, + "logps/rejected": -396.171875, + "logps/weighted_chosen": -3.2917237281799316, + "logps/weighted_rejected": -3.580859422683716, + "loss": 0.6356, + "rewards/accuracies": 0.6499999761581421, + "rewards/chosen": -94.04902648925781, + "rewards/margins": 47.923240661621094, + "rewards/rejected": -141.9646453857422, + "rewards/weighted_accuracies": 0.71875, + "rewards/weighted_chosen": -0.7129974365234375, + "rewards/weighted_margins": 0.53741455078125, + "rewards/weighted_rejected": -1.2503478527069092, + "step": 550 + }, + { + "epoch": 0.29311698508243916, + "grad_norm": 36.68463897705078, + "learning_rate": 8.916877506280601e-07, + "logits/chosen": -0.9174438714981079, + "logits/rejected": -0.9228302240371704, + "logps/chosen": -387.2203063964844, + "logps/rejected": -405.38592529296875, + "logps/weighted_chosen": -3.4023194313049316, + "logps/weighted_rejected": -3.6703858375549316, + "loss": 0.5995, + "rewards/accuracies": 0.6343749761581421, + "rewards/chosen": -106.43359375, + "rewards/margins": 44.87871170043945, + "rewards/rejected": -151.3136749267578, + "rewards/weighted_accuracies": 0.690625011920929, + "rewards/weighted_chosen": -0.7533203363418579, + "rewards/weighted_margins": 0.627819836139679, + "rewards/weighted_rejected": -1.3810546398162842, + "step": 560 + }, + { + "epoch": 0.29835121695891126, + "grad_norm": 14.758890151977539, + "learning_rate": 8.85943037780415e-07, + "logits/chosen": -0.9987869262695312, + "logits/rejected": -1.0033996105194092, + "logps/chosen": -392.72344970703125, + "logps/rejected": -380.15625, + "logps/weighted_chosen": -3.2074217796325684, + "logps/weighted_rejected": -3.7137207984924316, + "loss": 0.6343, + "rewards/accuracies": 0.609375, + "rewards/chosen": -106.4300765991211, + "rewards/margins": 32.053321838378906, + "rewards/rejected": -138.525390625, + "rewards/weighted_accuracies": 0.7093750238418579, + "rewards/weighted_chosen": -0.744158923625946, + "rewards/weighted_margins": 0.547467052936554, + "rewards/weighted_rejected": -1.291876196861267, + "step": 570 + }, + { + "epoch": 0.3035854488353834, + "grad_norm": 18.140382766723633, + "learning_rate": 8.800694230932884e-07, + "logits/chosen": -0.958880603313446, + "logits/rejected": -0.9720214605331421, + "logps/chosen": -390.57342529296875, + "logps/rejected": -398.5562438964844, + "logps/weighted_chosen": -2.86993408203125, + "logps/weighted_rejected": -3.4458250999450684, + "loss": 0.624, + "rewards/accuracies": 0.6031249761581421, + "rewards/chosen": -100.91777038574219, + "rewards/margins": 32.43359375, + "rewards/rejected": -133.35546875, + "rewards/weighted_accuracies": 0.659375011920929, + "rewards/weighted_chosen": -0.558367908000946, + "rewards/weighted_margins": 0.4340454041957855, + "rewards/weighted_rejected": -0.992688000202179, + "step": 580 + }, + { + "epoch": 0.30881968071185556, + "grad_norm": 16.685937881469727, + "learning_rate": 8.740688683062723e-07, + "logits/chosen": -1.0655395984649658, + "logits/rejected": -1.0949890613555908, + "logps/chosen": -449.0078125, + "logps/rejected": -423.9593811035156, + "logps/weighted_chosen": -2.912792921066284, + "logps/weighted_rejected": -3.656201124191284, + "loss": 0.6344, + "rewards/accuracies": 0.6031249761581421, + "rewards/chosen": -119.9111328125, + "rewards/margins": 28.247461318969727, + "rewards/rejected": -148.1476593017578, + "rewards/weighted_accuracies": 0.706250011920929, + "rewards/weighted_chosen": -0.5903121829032898, + "rewards/weighted_margins": 0.525067150592804, + "rewards/weighted_rejected": -1.1153564453125, + "step": 590 + }, + { + "epoch": 0.31405391258832765, + "grad_norm": 20.500003814697266, + "learning_rate": 8.679433775559215e-07, + "logits/chosen": -1.0546143054962158, + "logits/rejected": -1.1019103527069092, + "logps/chosen": -473.6171875, + "logps/rejected": -477.95001220703125, + "logps/weighted_chosen": -2.896728515625, + "logps/weighted_rejected": -3.985156297683716, + "loss": 0.5741, + "rewards/accuracies": 0.606249988079071, + "rewards/chosen": -148.9021453857422, + "rewards/margins": 43.50390625, + "rewards/rejected": -192.4656219482422, + "rewards/weighted_accuracies": 0.71875, + "rewards/weighted_chosen": -0.7574096918106079, + "rewards/weighted_margins": 0.587969958782196, + "rewards/weighted_rejected": -1.345495581626892, + "step": 600 + }, + { + "epoch": 0.3192881444647998, + "grad_norm": 50.760562896728516, + "learning_rate": 8.616949967063871e-07, + "logits/chosen": -1.007971167564392, + "logits/rejected": -1.0581481456756592, + "logps/chosen": -419.1000061035156, + "logps/rejected": -460.9624938964844, + "logps/weighted_chosen": -3.3387207984924316, + "logps/weighted_rejected": -3.981640577316284, + "loss": 0.5971, + "rewards/accuracies": 0.6468750238418579, + "rewards/chosen": -150.87344360351562, + "rewards/margins": 49.451171875, + "rewards/rejected": -200.21444702148438, + "rewards/weighted_accuracies": 0.671875, + "rewards/weighted_chosen": -0.853253185749054, + "rewards/weighted_margins": 0.644561767578125, + "rewards/weighted_rejected": -1.497961401939392, + "step": 610 + }, + { + "epoch": 0.3245223763412719, + "grad_norm": 16.369630813598633, + "learning_rate": 8.553258126661154e-07, + "logits/chosen": -1.0560119152069092, + "logits/rejected": -1.0674712657928467, + "logps/chosen": -442.5859375, + "logps/rejected": -463.5718688964844, + "logps/weighted_chosen": -3.429705858230591, + "logps/weighted_rejected": -4.259472846984863, + "loss": 0.6405, + "rewards/accuracies": 0.621874988079071, + "rewards/chosen": -162.7121124267578, + "rewards/margins": 44.754295349121094, + "rewards/rejected": -207.43008422851562, + "rewards/weighted_accuracies": 0.684374988079071, + "rewards/weighted_chosen": -0.9004287719726562, + "rewards/weighted_margins": 0.6113220453262329, + "rewards/weighted_rejected": -1.511694312095642, + "step": 620 + }, + { + "epoch": 0.32975660821774405, + "grad_norm": 31.255151748657227, + "learning_rate": 8.488379526908368e-07, + "logits/chosen": -1.0301024913787842, + "logits/rejected": -1.0498230457305908, + "logps/chosen": -465.5859375, + "logps/rejected": -505.28125, + "logps/weighted_chosen": -3.3856444358825684, + "logps/weighted_rejected": -4.028759956359863, + "loss": 0.6384, + "rewards/accuracies": 0.621874988079071, + "rewards/chosen": -174.98046875, + "rewards/margins": 61.357032775878906, + "rewards/rejected": -236.38906860351562, + "rewards/weighted_accuracies": 0.6968749761581421, + "rewards/weighted_chosen": -0.969250500202179, + "rewards/weighted_margins": 0.5812011957168579, + "rewards/weighted_rejected": -1.5499999523162842, + "step": 630 + }, + { + "epoch": 0.33499084009421615, + "grad_norm": 23.673486709594727, + "learning_rate": 8.422335836730802e-07, + "logits/chosen": -1.0135101079940796, + "logits/rejected": -1.0169677734375, + "logps/chosen": -426.2875061035156, + "logps/rejected": -497.64373779296875, + "logps/weighted_chosen": -3.098522901535034, + "logps/weighted_rejected": -3.898510694503784, + "loss": 0.6321, + "rewards/accuracies": 0.637499988079071, + "rewards/chosen": -152.58749389648438, + "rewards/margins": 67.427734375, + "rewards/rejected": -220.0578155517578, + "rewards/weighted_accuracies": 0.699999988079071, + "rewards/weighted_chosen": -0.878277599811554, + "rewards/weighted_margins": 0.590301513671875, + "rewards/weighted_rejected": -1.468664526939392, + "step": 640 + }, + { + "epoch": 0.3402250719706883, + "grad_norm": 20.44074821472168, + "learning_rate": 8.355149114184485e-07, + "logits/chosen": -1.105963110923767, + "logits/rejected": -1.095733642578125, + "logps/chosen": -474.7554626464844, + "logps/rejected": -523.0843505859375, + "logps/weighted_chosen": -3.1939697265625, + "logps/weighted_rejected": -3.786669969558716, + "loss": 0.5634, + "rewards/accuracies": 0.628125011920929, + "rewards/chosen": -167.33828735351562, + "rewards/margins": 73.52656555175781, + "rewards/rejected": -240.85702514648438, + "rewards/weighted_accuracies": 0.668749988079071, + "rewards/weighted_chosen": -0.8172271847724915, + "rewards/weighted_margins": 0.6973236203193665, + "rewards/weighted_rejected": -1.514801025390625, + "step": 650 + }, + { + "epoch": 0.34545930384716045, + "grad_norm": 25.958847045898438, + "learning_rate": 8.286841799088963e-07, + "logits/chosen": -1.1746642589569092, + "logits/rejected": -1.179473876953125, + "logps/chosen": -480.34063720703125, + "logps/rejected": -518.9437255859375, + "logps/weighted_chosen": -2.8995361328125, + "logps/weighted_rejected": -3.656787157058716, + "loss": 0.6073, + "rewards/accuracies": 0.5843750238418579, + "rewards/chosen": -196.3136749267578, + "rewards/margins": 59.513671875, + "rewards/rejected": -255.85897827148438, + "rewards/weighted_accuracies": 0.684374988079071, + "rewards/weighted_chosen": -0.8131744265556335, + "rewards/weighted_margins": 0.576770007610321, + "rewards/weighted_rejected": -1.3900268077850342, + "step": 660 + }, + { + "epoch": 0.35069353572363254, + "grad_norm": 15.336409568786621, + "learning_rate": 8.217436705532599e-07, + "logits/chosen": -1.1510436534881592, + "logits/rejected": -1.173730492591858, + "logps/chosen": -508.78594970703125, + "logps/rejected": -500.79217529296875, + "logps/weighted_chosen": -3.0011963844299316, + "logps/weighted_rejected": -3.7521729469299316, + "loss": 0.578, + "rewards/accuracies": 0.6000000238418579, + "rewards/chosen": -205.71914672851562, + "rewards/margins": 31.244531631469727, + "rewards/rejected": -237.14453125, + "rewards/weighted_accuracies": 0.721875011920929, + "rewards/weighted_chosen": -0.8121887445449829, + "rewards/weighted_margins": 0.688189685344696, + "rewards/weighted_rejected": -1.500115990638733, + "step": 670 + }, + { + "epoch": 0.3559277676001047, + "grad_norm": 19.029943466186523, + "learning_rate": 8.14695701425284e-07, + "logits/chosen": -1.079345703125, + "logits/rejected": -1.125024437904358, + "logps/chosen": -466.38360595703125, + "logps/rejected": -470.4234313964844, + "logps/weighted_chosen": -2.9954833984375, + "logps/weighted_rejected": -3.8232421875, + "loss": 0.5882, + "rewards/accuracies": 0.606249988079071, + "rewards/chosen": -160.8718719482422, + "rewards/margins": 47.43242263793945, + "rewards/rejected": -208.275390625, + "rewards/weighted_accuracies": 0.690625011920929, + "rewards/weighted_chosen": -0.7394653558731079, + "rewards/weighted_margins": 0.627307116985321, + "rewards/weighted_rejected": -1.3663330078125, + "step": 680 + }, + { + "epoch": 0.3611619994765768, + "grad_norm": 13.842121124267578, + "learning_rate": 8.075426264894046e-07, + "logits/chosen": -1.013006567955017, + "logits/rejected": -1.0405089855194092, + "logps/chosen": -472.03436279296875, + "logps/rejected": -501.09375, + "logps/weighted_chosen": -3.1174073219299316, + "logps/weighted_rejected": -4.429980278015137, + "loss": 0.5163, + "rewards/accuracies": 0.6656249761581421, + "rewards/chosen": -166.5642547607422, + "rewards/margins": 56.965431213378906, + "rewards/rejected": -223.5593719482422, + "rewards/weighted_accuracies": 0.778124988079071, + "rewards/weighted_chosen": -0.7782195806503296, + "rewards/weighted_margins": 0.887194812297821, + "rewards/weighted_rejected": -1.665307641029358, + "step": 690 + }, + { + "epoch": 0.36639623135304894, + "grad_norm": 31.751785278320312, + "learning_rate": 8.002868348145435e-07, + "logits/chosen": -1.0329437255859375, + "logits/rejected": -1.0401580333709717, + "logps/chosen": -535.6437377929688, + "logps/rejected": -564.8609619140625, + "logps/weighted_chosen": -3.4560546875, + "logps/weighted_rejected": -3.8817505836486816, + "loss": 0.5956, + "rewards/accuracies": 0.625, + "rewards/chosen": -235.3767547607422, + "rewards/margins": 61.516014099121094, + "rewards/rejected": -296.88360595703125, + "rewards/weighted_accuracies": 0.71875, + "rewards/weighted_chosen": -1.1499541997909546, + "rewards/weighted_margins": 0.6718994379043579, + "rewards/weighted_rejected": -1.8215301036834717, + "step": 700 + }, + { + "epoch": 0.3716304632295211, + "grad_norm": 32.05786895751953, + "learning_rate": 7.92930749776179e-07, + "logits/chosen": -1.0492126941680908, + "logits/rejected": -1.092248558998108, + "logps/chosen": -564.2515869140625, + "logps/rejected": -602.4500122070312, + "logps/weighted_chosen": -3.8183350563049316, + "logps/weighted_rejected": -4.569482326507568, + "loss": 0.652, + "rewards/accuracies": 0.581250011920929, + "rewards/chosen": -292.5419921875, + "rewards/margins": 49.53203201293945, + "rewards/rejected": -342.0179748535156, + "rewards/weighted_accuracies": 0.6656249761581421, + "rewards/weighted_chosen": -1.244970679283142, + "rewards/weighted_margins": 0.630566418170929, + "rewards/weighted_rejected": -1.8747985363006592, + "step": 710 + }, + { + "epoch": 0.3768646951059932, + "grad_norm": 23.8446044921875, + "learning_rate": 7.854768282469582e-07, + "logits/chosen": -1.0707213878631592, + "logits/rejected": -1.126983642578125, + "logps/chosen": -462.90155029296875, + "logps/rejected": -542.5218505859375, + "logps/weighted_chosen": -3.208447217941284, + "logps/weighted_rejected": -4.043383598327637, + "loss": 0.6115, + "rewards/accuracies": 0.6499999761581421, + "rewards/chosen": -185.4597625732422, + "rewards/margins": 79.3746109008789, + "rewards/rejected": -264.8812561035156, + "rewards/weighted_accuracies": 0.6781250238418579, + "rewards/weighted_chosen": -0.8762931823730469, + "rewards/weighted_margins": 0.711090087890625, + "rewards/weighted_rejected": -1.5873870849609375, + "step": 720 + }, + { + "epoch": 0.38209892698246534, + "grad_norm": 17.816680908203125, + "learning_rate": 7.779275597761215e-07, + "logits/chosen": -1.036431908607483, + "logits/rejected": -1.105950951576233, + "logps/chosen": -471.43280029296875, + "logps/rejected": -532.6843872070312, + "logps/weighted_chosen": -3.378674268722534, + "logps/weighted_rejected": -4.096581935882568, + "loss": 0.5442, + "rewards/accuracies": 0.6499999761581421, + "rewards/chosen": -195.43984985351562, + "rewards/margins": 68.92460632324219, + "rewards/rejected": -264.296875, + "rewards/weighted_accuracies": 0.706250011920929, + "rewards/weighted_chosen": -0.847393810749054, + "rewards/weighted_margins": 0.777783215045929, + "rewards/weighted_rejected": -1.625451683998108, + "step": 730 + }, + { + "epoch": 0.38733315885893743, + "grad_norm": 29.750192642211914, + "learning_rate": 7.702854657580126e-07, + "logits/chosen": -1.0928223133087158, + "logits/rejected": -1.1139404773712158, + "logps/chosen": -507.796875, + "logps/rejected": -523.7952880859375, + "logps/weighted_chosen": -3.3760986328125, + "logps/weighted_rejected": -4.447509765625, + "loss": 0.5597, + "rewards/accuracies": 0.6343749761581421, + "rewards/chosen": -213.2277374267578, + "rewards/margins": 56.09648513793945, + "rewards/rejected": -269.4203186035156, + "rewards/weighted_accuracies": 0.7406250238418579, + "rewards/weighted_chosen": -0.927258312702179, + "rewards/weighted_margins": 0.8479980230331421, + "rewards/weighted_rejected": -1.7749145030975342, + "step": 740 + }, + { + "epoch": 0.3925673907354096, + "grad_norm": 38.97298812866211, + "learning_rate": 7.625530985899547e-07, + "logits/chosen": -1.0656921863555908, + "logits/rejected": -1.0953140258789062, + "logps/chosen": -491.4609375, + "logps/rejected": -545.0203247070312, + "logps/weighted_chosen": -3.2592406272888184, + "logps/weighted_rejected": -4.605615139007568, + "loss": 0.5776, + "rewards/accuracies": 0.5874999761581421, + "rewards/chosen": -224.6374969482422, + "rewards/margins": 75.23710632324219, + "rewards/rejected": -300.0316467285156, + "rewards/weighted_accuracies": 0.699999988079071, + "rewards/weighted_chosen": -1.093621850013733, + "rewards/weighted_margins": 0.852429211139679, + "rewards/weighted_rejected": -1.9456908702850342, + "step": 750 + }, + { + "epoch": 0.39780162261188173, + "grad_norm": 22.909940719604492, + "learning_rate": 7.547330408197694e-07, + "logits/chosen": -1.0663268566131592, + "logits/rejected": -1.1003234386444092, + "logps/chosen": -572.404296875, + "logps/rejected": -556.4031372070312, + "logps/weighted_chosen": -3.372363328933716, + "logps/weighted_rejected": -4.400781154632568, + "loss": 0.5943, + "rewards/accuracies": 0.6031249761581421, + "rewards/chosen": -276.7529296875, + "rewards/margins": 23.257421493530273, + "rewards/rejected": -300.1558532714844, + "rewards/weighted_accuracies": 0.703125, + "rewards/weighted_chosen": -1.1045868396759033, + "rewards/weighted_margins": 0.7262054681777954, + "rewards/weighted_rejected": -1.8311798572540283, + "step": 760 + }, + { + "epoch": 0.40303585448835383, + "grad_norm": 107.9198989868164, + "learning_rate": 7.468279042832271e-07, + "logits/chosen": -1.067718505859375, + "logits/rejected": -1.095727562904358, + "logps/chosen": -467.5, + "logps/rejected": -559.7703247070312, + "logps/weighted_chosen": -3.0786986351013184, + "logps/weighted_rejected": -3.8573241233825684, + "loss": 0.5995, + "rewards/accuracies": 0.6499999761581421, + "rewards/chosen": -187.9324188232422, + "rewards/margins": 82.28340148925781, + "rewards/rejected": -270.2611389160156, + "rewards/weighted_accuracies": 0.6968749761581421, + "rewards/weighted_chosen": -0.867828369140625, + "rewards/weighted_margins": 0.625274658203125, + "rewards/weighted_rejected": -1.493688941001892, + "step": 770 + }, + { + "epoch": 0.408270086364826, + "grad_norm": 18.453107833862305, + "learning_rate": 7.388403292317154e-07, + "logits/chosen": -1.0655944347381592, + "logits/rejected": -1.121740698814392, + "logps/chosen": -503.8828125, + "logps/rejected": -519.8375244140625, + "logps/weighted_chosen": -3.039843797683716, + "logps/weighted_rejected": -3.848388671875, + "loss": 0.5822, + "rewards/accuracies": 0.606249988079071, + "rewards/chosen": -202.9210968017578, + "rewards/margins": 50.95390701293945, + "rewards/rejected": -253.869140625, + "rewards/weighted_accuracies": 0.731249988079071, + "rewards/weighted_chosen": -0.873791515827179, + "rewards/weighted_margins": 0.6851806640625, + "rewards/weighted_rejected": -1.5595581531524658, + "step": 780 + }, + { + "epoch": 0.4135043182412981, + "grad_norm": 24.38869285583496, + "learning_rate": 7.307729834504154e-07, + "logits/chosen": -1.090570092201233, + "logits/rejected": -1.1371276378631592, + "logps/chosen": -490.86407470703125, + "logps/rejected": -537.7296752929688, + "logps/weighted_chosen": -3.041918992996216, + "logps/weighted_rejected": -4.070898532867432, + "loss": 0.6125, + "rewards/accuracies": 0.6000000238418579, + "rewards/chosen": -209.13241577148438, + "rewards/margins": 62.44609451293945, + "rewards/rejected": -271.58203125, + "rewards/weighted_accuracies": 0.690625011920929, + "rewards/weighted_chosen": -0.947405993938446, + "rewards/weighted_margins": 0.609375, + "rewards/weighted_rejected": -1.556726098060608, + "step": 790 + }, + { + "epoch": 0.4187385501177702, + "grad_norm": 26.962316513061523, + "learning_rate": 7.226285613672847e-07, + "logits/chosen": -1.03680419921875, + "logits/rejected": -1.0742919445037842, + "logps/chosen": -500.84295654296875, + "logps/rejected": -603.6422119140625, + "logps/weighted_chosen": -3.154052734375, + "logps/weighted_rejected": -4.342724800109863, + "loss": 0.5564, + "rewards/accuracies": 0.671875, + "rewards/chosen": -223.51171875, + "rewards/margins": 101.2464828491211, + "rewards/rejected": -324.720703125, + "rewards/weighted_accuracies": 0.7406250238418579, + "rewards/weighted_chosen": -0.996997058391571, + "rewards/weighted_margins": 0.755480945110321, + "rewards/weighted_rejected": -1.752661108970642, + "step": 800 + }, + { + "epoch": 0.4239727819942423, + "grad_norm": 15.780332565307617, + "learning_rate": 7.144097831531398e-07, + "logits/chosen": -1.0250060558319092, + "logits/rejected": -1.052709937095642, + "logps/chosen": -517.7921752929688, + "logps/rejected": -600.5421752929688, + "logps/weighted_chosen": -3.318652391433716, + "logps/weighted_rejected": -4.39453125, + "loss": 0.5695, + "rewards/accuracies": 0.596875011920929, + "rewards/chosen": -247.40664672851562, + "rewards/margins": 88.87187194824219, + "rewards/rejected": -336.234375, + "rewards/weighted_accuracies": 0.703125, + "rewards/weighted_chosen": -1.082605004310608, + "rewards/weighted_margins": 0.89263916015625, + "rewards/weighted_rejected": -1.974450707435608, + "step": 810 + }, + { + "epoch": 0.42920701387071447, + "grad_norm": 24.66672706604004, + "learning_rate": 7.061193938131396e-07, + "logits/chosen": -0.9916015863418579, + "logits/rejected": -1.042016625404358, + "logps/chosen": -607.3375244140625, + "logps/rejected": -642.5671997070312, + "logps/weighted_chosen": -3.689379930496216, + "logps/weighted_rejected": -4.398633003234863, + "loss": 0.5654, + "rewards/accuracies": 0.5874999761581421, + "rewards/chosen": -307.7925720214844, + "rewards/margins": 69.5777359008789, + "rewards/rejected": -377.3050842285156, + "rewards/weighted_accuracies": 0.7124999761581421, + "rewards/weighted_chosen": -1.1716430187225342, + "rewards/weighted_margins": 0.9023987054824829, + "rewards/weighted_rejected": -2.074475049972534, + "step": 820 + }, + { + "epoch": 0.4344412457471866, + "grad_norm": 20.459196090698242, + "learning_rate": 6.977601622699789e-07, + "logits/chosen": -0.9984985589981079, + "logits/rejected": -1.079248070716858, + "logps/chosen": -471.5625, + "logps/rejected": -574.8695068359375, + "logps/weighted_chosen": -3.233837842941284, + "logps/weighted_rejected": -4.211133003234863, + "loss": 0.5097, + "rewards/accuracies": 0.6656249761581421, + "rewards/chosen": -186.7394561767578, + "rewards/margins": 109.28633117675781, + "rewards/rejected": -296.13555908203125, + "rewards/weighted_accuracies": 0.746874988079071, + "rewards/weighted_chosen": -0.760040283203125, + "rewards/weighted_margins": 0.9420166015625, + "rewards/weighted_rejected": -1.7019164562225342, + "step": 830 + }, + { + "epoch": 0.4396754776236587, + "grad_norm": 89.301025390625, + "learning_rate": 6.893348804390882e-07, + "logits/chosen": -1.126031517982483, + "logits/rejected": -1.1464111804962158, + "logps/chosen": -541.5546875, + "logps/rejected": -580.4109497070312, + "logps/weighted_chosen": -3.6358885765075684, + "logps/weighted_rejected": -4.262304782867432, + "loss": 0.566, + "rewards/accuracies": 0.6000000238418579, + "rewards/chosen": -246.6085968017578, + "rewards/margins": 74.0562515258789, + "rewards/rejected": -320.61328125, + "rewards/weighted_accuracies": 0.684374988079071, + "rewards/weighted_chosen": -1.0077393054962158, + "rewards/weighted_margins": 0.7990967035293579, + "rewards/weighted_rejected": -1.806396484375, + "step": 840 + }, + { + "epoch": 0.44490970950013087, + "grad_norm": 20.6274471282959, + "learning_rate": 6.808463622961578e-07, + "logits/chosen": -1.151123046875, + "logits/rejected": -1.2046630382537842, + "logps/chosen": -621.8390502929688, + "logps/rejected": -716.7249755859375, + "logps/weighted_chosen": -3.6482911109924316, + "logps/weighted_rejected": -4.6015625, + "loss": 0.5759, + "rewards/accuracies": 0.6156250238418579, + "rewards/chosen": -328.931640625, + "rewards/margins": 106.0474624633789, + "rewards/rejected": -435.07403564453125, + "rewards/weighted_accuracies": 0.7124999761581421, + "rewards/weighted_chosen": -1.2104400396347046, + "rewards/weighted_margins": 0.9383910894393921, + "rewards/weighted_rejected": -2.1482911109924316, + "step": 850 + }, + { + "epoch": 0.45014394137660296, + "grad_norm": 20.91545295715332, + "learning_rate": 6.722974429372925e-07, + "logits/chosen": -1.1297729015350342, + "logits/rejected": -1.170617699623108, + "logps/chosen": -697.5484619140625, + "logps/rejected": -730.65625, + "logps/weighted_chosen": -3.5313477516174316, + "logps/weighted_rejected": -5.184765815734863, + "loss": 0.5217, + "rewards/accuracies": 0.5718749761581421, + "rewards/chosen": -393.822265625, + "rewards/margins": 74.92500305175781, + "rewards/rejected": -468.8492126464844, + "rewards/weighted_accuracies": 0.7718750238418579, + "rewards/weighted_chosen": -1.393365502357483, + "rewards/weighted_margins": 1.1208374500274658, + "rewards/weighted_rejected": -2.514172315597534, + "step": 860 + }, + { + "epoch": 0.4553781732530751, + "grad_norm": 42.88610076904297, + "learning_rate": 6.636909776321128e-07, + "logits/chosen": -1.1412475109100342, + "logits/rejected": -1.1436645984649658, + "logps/chosen": -549.9671630859375, + "logps/rejected": -677.8781127929688, + "logps/weighted_chosen": -3.683642625808716, + "logps/weighted_rejected": -4.7724609375, + "loss": 0.5195, + "rewards/accuracies": 0.606249988079071, + "rewards/chosen": -288.21484375, + "rewards/margins": 124.931640625, + "rewards/rejected": -413.17266845703125, + "rewards/weighted_accuracies": 0.7250000238418579, + "rewards/weighted_chosen": -1.2497345209121704, + "rewards/weighted_margins": 0.971630871295929, + "rewards/weighted_rejected": -2.2208008766174316, + "step": 870 + }, + { + "epoch": 0.46061240512954726, + "grad_norm": 29.938108444213867, + "learning_rate": 6.550298408701174e-07, + "logits/chosen": -1.0770995616912842, + "logits/rejected": -1.134680151939392, + "logps/chosen": -574.3234252929688, + "logps/rejected": -675.2101440429688, + "logps/weighted_chosen": -3.807568311691284, + "logps/weighted_rejected": -5.117138862609863, + "loss": 0.5413, + "rewards/accuracies": 0.6312500238418579, + "rewards/chosen": -287.63494873046875, + "rewards/margins": 104.92851257324219, + "rewards/rejected": -392.5396423339844, + "rewards/weighted_accuracies": 0.734375, + "rewards/weighted_chosen": -1.1316711902618408, + "rewards/weighted_margins": 0.954119861125946, + "rewards/weighted_rejected": -2.0862364768981934, + "step": 880 + }, + { + "epoch": 0.46584663700601936, + "grad_norm": 27.375259399414062, + "learning_rate": 6.463169254006276e-07, + "logits/chosen": -1.10406494140625, + "logits/rejected": -1.156030297279358, + "logps/chosen": -541.8601684570312, + "logps/rejected": -578.5452880859375, + "logps/weighted_chosen": -3.4522461891174316, + "logps/weighted_rejected": -4.527783393859863, + "loss": 0.5321, + "rewards/accuracies": 0.612500011920929, + "rewards/chosen": -260.65704345703125, + "rewards/margins": 71.1494140625, + "rewards/rejected": -331.875, + "rewards/weighted_accuracies": 0.7281249761581421, + "rewards/weighted_chosen": -0.977038562297821, + "rewards/weighted_margins": 1.022253394126892, + "rewards/weighted_rejected": -1.999292016029358, + "step": 890 + }, + { + "epoch": 0.4710808688824915, + "grad_norm": 32.590675354003906, + "learning_rate": 6.375551412666326e-07, + "logits/chosen": -1.0722472667694092, + "logits/rejected": -1.1129913330078125, + "logps/chosen": -499.8031311035156, + "logps/rejected": -543.9796752929688, + "logps/weighted_chosen": -3.171948194503784, + "logps/weighted_rejected": -4.273193359375, + "loss": 0.6035, + "rewards/accuracies": 0.590624988079071, + "rewards/chosen": -225.7996063232422, + "rewards/margins": 61.669532775878906, + "rewards/rejected": -287.5015563964844, + "rewards/weighted_accuracies": 0.6812499761581421, + "rewards/weighted_chosen": -1.0042235851287842, + "rewards/weighted_margins": 0.7809082269668579, + "rewards/weighted_rejected": -1.784277319908142, + "step": 900 + }, + { + "epoch": 0.4763151007589636, + "grad_norm": 37.88996887207031, + "learning_rate": 6.287474148328583e-07, + "logits/chosen": -1.0318695306777954, + "logits/rejected": -1.0187164545059204, + "logps/chosen": -491.97186279296875, + "logps/rejected": -540.4202880859375, + "logps/weighted_chosen": -3.5292601585388184, + "logps/weighted_rejected": -5.02734375, + "loss": 0.559, + "rewards/accuracies": 0.606249988079071, + "rewards/chosen": -221.07968139648438, + "rewards/margins": 72.57441711425781, + "rewards/rejected": -293.56561279296875, + "rewards/weighted_accuracies": 0.690625011920929, + "rewards/weighted_chosen": -1.0068145990371704, + "rewards/weighted_margins": 0.84521484375, + "rewards/weighted_rejected": -1.851770043373108, + "step": 910 + }, + { + "epoch": 0.48154933263543576, + "grad_norm": 25.340351104736328, + "learning_rate": 6.198966878083857e-07, + "logits/chosen": -1.0930664539337158, + "logits/rejected": -1.12109375, + "logps/chosen": -540.0, + "logps/rejected": -623.390625, + "logps/weighted_chosen": -3.68231201171875, + "logps/weighted_rejected": -4.464990139007568, + "loss": 0.5911, + "rewards/accuracies": 0.606249988079071, + "rewards/chosen": -268.6884765625, + "rewards/margins": 87.7593765258789, + "rewards/rejected": -356.4453125, + "rewards/weighted_accuracies": 0.7124999761581421, + "rewards/weighted_chosen": -1.085455298423767, + "rewards/weighted_margins": 0.79486083984375, + "rewards/weighted_rejected": -1.880859375, + "step": 920 + }, + { + "epoch": 0.48678356451190785, + "grad_norm": 47.73952865600586, + "learning_rate": 6.110059162641439e-07, + "logits/chosen": -1.1020386219024658, + "logits/rejected": -1.1357238292694092, + "logps/chosen": -533.7390747070312, + "logps/rejected": -599.3781127929688, + "logps/weighted_chosen": -2.9868531227111816, + "logps/weighted_rejected": -4.054858207702637, + "loss": 0.5458, + "rewards/accuracies": 0.625, + "rewards/chosen": -251.13125610351562, + "rewards/margins": 80.9544906616211, + "rewards/rejected": -332.2699279785156, + "rewards/weighted_accuracies": 0.734375, + "rewards/weighted_chosen": -0.8615478277206421, + "rewards/weighted_margins": 0.8154662847518921, + "rewards/weighted_rejected": -1.6761963367462158, + "step": 930 + }, + { + "epoch": 0.49201779638838, + "grad_norm": 36.5290412902832, + "learning_rate": 6.020780696456059e-07, + "logits/chosen": -1.133764624595642, + "logits/rejected": -1.177819848060608, + "logps/chosen": -515.2952880859375, + "logps/rejected": -611.4953002929688, + "logps/weighted_chosen": -3.0075440406799316, + "logps/weighted_rejected": -4.250219821929932, + "loss": 0.5361, + "rewards/accuracies": 0.659375011920929, + "rewards/chosen": -244.7570343017578, + "rewards/margins": 99.73710632324219, + "rewards/rejected": -344.3531188964844, + "rewards/weighted_accuracies": 0.731249988079071, + "rewards/weighted_chosen": -1.0392029285430908, + "rewards/weighted_margins": 0.8944336175918579, + "rewards/weighted_rejected": -1.93341064453125, + "step": 940 + }, + { + "epoch": 0.49725202826485215, + "grad_norm": 36.72229766845703, + "learning_rate": 5.931161297810185e-07, + "logits/chosen": -1.2070038318634033, + "logits/rejected": -1.23931884765625, + "logps/chosen": -550.4031372070312, + "logps/rejected": -615.3375244140625, + "logps/weighted_chosen": -3.748730421066284, + "logps/weighted_rejected": -4.64697265625, + "loss": 0.5726, + "rewards/accuracies": 0.5843750238418579, + "rewards/chosen": -277.07733154296875, + "rewards/margins": 77.06562805175781, + "rewards/rejected": -354.22381591796875, + "rewards/weighted_accuracies": 0.6968749761581421, + "rewards/weighted_chosen": -1.09320068359375, + "rewards/weighted_margins": 0.812207043170929, + "rewards/weighted_rejected": -1.906347632408142, + "step": 950 + }, + { + "epoch": 0.5024862601413242, + "grad_norm": 39.58312225341797, + "learning_rate": 5.841230898854959e-07, + "logits/chosen": -1.1735107898712158, + "logits/rejected": -1.2161743640899658, + "logps/chosen": -684.3687744140625, + "logps/rejected": -794.3922119140625, + "logps/weighted_chosen": -3.8369140625, + "logps/weighted_rejected": -4.986083984375, + "loss": 0.6144, + "rewards/accuracies": 0.612500011920929, + "rewards/chosen": -384.17852783203125, + "rewards/margins": 148.6707000732422, + "rewards/rejected": -532.8191528320312, + "rewards/weighted_accuracies": 0.6937500238418579, + "rewards/weighted_chosen": -1.5656402111053467, + "rewards/weighted_margins": 0.9439452886581421, + "rewards/weighted_rejected": -2.5091795921325684, + "step": 960 + }, + { + "epoch": 0.5077204920177963, + "grad_norm": 24.286752700805664, + "learning_rate": 5.751019535613102e-07, + "logits/chosen": -1.0753357410430908, + "logits/rejected": -1.104547142982483, + "logps/chosen": -550.9773559570312, + "logps/rejected": -653.6687622070312, + "logps/weighted_chosen": -3.5946044921875, + "logps/weighted_rejected": -5.038866996765137, + "loss": 0.5266, + "rewards/accuracies": 0.675000011920929, + "rewards/chosen": -291.25390625, + "rewards/margins": 115.4117202758789, + "rewards/rejected": -406.62811279296875, + "rewards/weighted_accuracies": 0.7093750238418579, + "rewards/weighted_chosen": -1.2581603527069092, + "rewards/weighted_margins": 1.127111792564392, + "rewards/weighted_rejected": -2.3849730491638184, + "step": 970 + }, + { + "epoch": 0.5129547238942685, + "grad_norm": 22.55328941345215, + "learning_rate": 5.660557337947117e-07, + "logits/chosen": -1.0788085460662842, + "logits/rejected": -1.109989881515503, + "logps/chosen": -594.8406372070312, + "logps/rejected": -653.6624755859375, + "logps/weighted_chosen": -3.10791015625, + "logps/weighted_rejected": -4.297143459320068, + "loss": 0.5361, + "rewards/accuracies": 0.628125011920929, + "rewards/chosen": -300.37872314453125, + "rewards/margins": 95.828125, + "rewards/rejected": -396.203125, + "rewards/weighted_accuracies": 0.6968749761581421, + "rewards/weighted_chosen": -1.0415496826171875, + "rewards/weighted_margins": 0.88812255859375, + "rewards/weighted_rejected": -1.929406762123108, + "step": 980 + }, + { + "epoch": 0.5181889557707406, + "grad_norm": 53.25763702392578, + "learning_rate": 5.569874519496174e-07, + "logits/chosen": -1.055578589439392, + "logits/rejected": -1.122521996498108, + "logps/chosen": -497.3999938964844, + "logps/rejected": -579.3656005859375, + "logps/weighted_chosen": -3.393261671066284, + "logps/weighted_rejected": -4.673925876617432, + "loss": 0.5302, + "rewards/accuracies": 0.596875011920929, + "rewards/chosen": -222.11483764648438, + "rewards/margins": 90.296875, + "rewards/rejected": -312.42108154296875, + "rewards/weighted_accuracies": 0.699999988079071, + "rewards/weighted_chosen": -0.8590606451034546, + "rewards/weighted_margins": 0.879437267780304, + "rewards/weighted_rejected": -1.7384154796600342, + "step": 990 + }, + { + "epoch": 0.5234231876472127, + "grad_norm": 39.35697555541992, + "learning_rate": 5.47900136758499e-07, + "logits/chosen": -1.0391693115234375, + "logits/rejected": -1.1183960437774658, + "logps/chosen": -589.5906372070312, + "logps/rejected": -653.09375, + "logps/weighted_chosen": -3.634228467941284, + "logps/weighted_rejected": -4.7723388671875, + "loss": 0.5255, + "rewards/accuracies": 0.59375, + "rewards/chosen": -321.95623779296875, + "rewards/margins": 89.08320617675781, + "rewards/rejected": -410.8277282714844, + "rewards/weighted_accuracies": 0.731249988079071, + "rewards/weighted_chosen": -1.187219262123108, + "rewards/weighted_margins": 0.9879394769668579, + "rewards/weighted_rejected": -2.1756300926208496, + "step": 1000 + }, + { + "epoch": 0.5234231876472127, + "eval_logits/chosen": -1.194671869277954, + "eval_logits/rejected": -1.2200820446014404, + "eval_logps/chosen": -645.4445190429688, + "eval_logps/rejected": -726.552001953125, + "eval_logps/weighted_chosen": -3.7153730392456055, + "eval_logps/weighted_rejected": -4.821405410766602, + "eval_loss": 0.5563206672668457, + "eval_rewards/accuracies": 0.5899999737739563, + "eval_rewards/chosen": -356.71826171875, + "eval_rewards/margins": 102.17925262451172, + "eval_rewards/rejected": -458.8554992675781, + "eval_rewards/weighted_accuracies": 0.7135000228881836, + "eval_rewards/weighted_chosen": -1.3285291194915771, + "eval_rewards/weighted_margins": 0.9376659989356995, + "eval_rewards/weighted_rejected": -2.266195297241211, + "eval_runtime": 1365.9873, + "eval_samples_per_second": 1.464, + "eval_steps_per_second": 0.366, + "step": 1000 + }, + { + "epoch": 0.528657419523685, + "grad_norm": 45.573585510253906, + "learning_rate": 5.387968233108113e-07, + "logits/chosen": -1.087103247642517, + "logits/rejected": -1.085015892982483, + "logps/chosen": -669.9202880859375, + "logps/rejected": -781.1453247070312, + "logps/weighted_chosen": -4.16748046875, + "logps/weighted_rejected": -5.289355278015137, + "loss": 0.5551, + "rewards/accuracies": 0.5843750238418579, + "rewards/chosen": -385.9546813964844, + "rewards/margins": 140.77304077148438, + "rewards/rejected": -526.48046875, + "rewards/weighted_accuracies": 0.746874988079071, + "rewards/weighted_chosen": -1.512841820716858, + "rewards/weighted_margins": 1.02880859375, + "rewards/weighted_rejected": -2.542529344558716, + "step": 1010 + }, + { + "epoch": 0.533891651400157, + "grad_norm": 304.9451904296875, + "learning_rate": 5.296805520392962e-07, + "logits/chosen": -1.160436987876892, + "logits/rejected": -1.2073547840118408, + "logps/chosen": -662.7718505859375, + "logps/rejected": -731.2015380859375, + "logps/weighted_chosen": -3.366748094558716, + "logps/weighted_rejected": -4.887963771820068, + "loss": 0.6053, + "rewards/accuracies": 0.5874999761581421, + "rewards/chosen": -350.32733154296875, + "rewards/margins": 98.02656555175781, + "rewards/rejected": -448.2242126464844, + "rewards/weighted_accuracies": 0.703125, + "rewards/weighted_chosen": -1.51910400390625, + "rewards/weighted_margins": 0.8489990234375, + "rewards/weighted_rejected": -2.3676390647888184, + "step": 1020 + }, + { + "epoch": 0.5391258832766291, + "grad_norm": 186.6497344970703, + "learning_rate": 5.205543677045049e-07, + "logits/chosen": -1.0708801746368408, + "logits/rejected": -1.114501953125, + "logps/chosen": -556.8343505859375, + "logps/rejected": -603.2718505859375, + "logps/weighted_chosen": -3.673095703125, + "logps/weighted_rejected": -4.930810451507568, + "loss": 0.5157, + "rewards/accuracies": 0.606249988079071, + "rewards/chosen": -282.62188720703125, + "rewards/margins": 75.29374694824219, + "rewards/rejected": -357.86053466796875, + "rewards/weighted_accuracies": 0.746874988079071, + "rewards/weighted_chosen": -1.1851806640625, + "rewards/weighted_margins": 1.0668213367462158, + "rewards/weighted_rejected": -2.2524254322052, + "step": 1030 + }, + { + "epoch": 0.5443601151531012, + "grad_norm": 88.95674896240234, + "learning_rate": 5.114213183778697e-07, + "logits/chosen": -1.123449683189392, + "logits/rejected": -1.1691772937774658, + "logps/chosen": -555.5703125, + "logps/rejected": -638.265625, + "logps/weighted_chosen": -4.111181735992432, + "logps/weighted_rejected": -4.94482421875, + "loss": 0.5669, + "rewards/accuracies": 0.625, + "rewards/chosen": -275.11407470703125, + "rewards/margins": 106.10546875, + "rewards/rejected": -381.244140625, + "rewards/weighted_accuracies": 0.7124999761581421, + "rewards/weighted_chosen": -1.225732445716858, + "rewards/weighted_margins": 0.956982433795929, + "rewards/weighted_rejected": -2.182055711746216, + "step": 1040 + }, + { + "epoch": 0.5495943470295734, + "grad_norm": 227.4050750732422, + "learning_rate": 5.022844544236754e-07, + "logits/chosen": -1.054589867591858, + "logits/rejected": -1.088189721107483, + "logps/chosen": -567.5890502929688, + "logps/rejected": -656.4656372070312, + "logps/weighted_chosen": -3.9198241233825684, + "logps/weighted_rejected": -4.956591606140137, + "loss": 0.572, + "rewards/accuracies": 0.5843750238418579, + "rewards/chosen": -287.54962158203125, + "rewards/margins": 113.6656265258789, + "rewards/rejected": -401.166015625, + "rewards/weighted_accuracies": 0.715624988079071, + "rewards/weighted_chosen": -1.267511010169983, + "rewards/weighted_margins": 0.9030395746231079, + "rewards/weighted_rejected": -2.170910596847534, + "step": 1050 + }, + { + "epoch": 0.5548285789060455, + "grad_norm": 1079.8702392578125, + "learning_rate": 4.931468274802608e-07, + "logits/chosen": -1.070989966392517, + "logits/rejected": -1.1126830577850342, + "logps/chosen": -568.2453002929688, + "logps/rejected": -670.5187377929688, + "logps/weighted_chosen": -3.2323241233825684, + "logps/weighted_rejected": -4.560888767242432, + "loss": 0.5383, + "rewards/accuracies": 0.621874988079071, + "rewards/chosen": -285.74139404296875, + "rewards/margins": 123.85664367675781, + "rewards/rejected": -409.6675720214844, + "rewards/weighted_accuracies": 0.7250000238418579, + "rewards/weighted_chosen": -1.198522925376892, + "rewards/weighted_margins": 0.938366711139679, + "rewards/weighted_rejected": -2.136554002761841, + "step": 1060 + }, + { + "epoch": 0.5600628107825176, + "grad_norm": 56.90155792236328, + "learning_rate": 4.840114894407974e-07, + "logits/chosen": -1.10162353515625, + "logits/rejected": -1.1522216796875, + "logps/chosen": -528.5078125, + "logps/rejected": -583.03125, + "logps/weighted_chosen": -3.598315477371216, + "logps/weighted_rejected": -4.284228324890137, + "loss": 0.5674, + "rewards/accuracies": 0.6343749761581421, + "rewards/chosen": -252.3361358642578, + "rewards/margins": 79.1089859008789, + "rewards/rejected": -331.48748779296875, + "rewards/weighted_accuracies": 0.731249988079071, + "rewards/weighted_chosen": -1.137750267982483, + "rewards/weighted_margins": 0.870159924030304, + "rewards/weighted_rejected": -2.008807420730591, + "step": 1070 + }, + { + "epoch": 0.5652970426589898, + "grad_norm": 46.219932556152344, + "learning_rate": 4.748814914339811e-07, + "logits/chosen": -1.0657470226287842, + "logits/rejected": -1.1109619140625, + "logps/chosen": -569.3312377929688, + "logps/rejected": -623.2874755859375, + "logps/weighted_chosen": -3.5493407249450684, + "logps/weighted_rejected": -4.359619140625, + "loss": 0.5449, + "rewards/accuracies": 0.621874988079071, + "rewards/chosen": -284.59649658203125, + "rewards/margins": 83.34257507324219, + "rewards/rejected": -368.00311279296875, + "rewards/weighted_accuracies": 0.715624988079071, + "rewards/weighted_chosen": -1.1415588855743408, + "rewards/weighted_margins": 1.012670874595642, + "rewards/weighted_rejected": -2.1540770530700684, + "step": 1080 + }, + { + "epoch": 0.5705312745354619, + "grad_norm": 86.65474700927734, + "learning_rate": 4.657598828049801e-07, + "logits/chosen": -1.1116211414337158, + "logits/rejected": -1.1939513683319092, + "logps/chosen": -578.01953125, + "logps/rejected": -681.3382568359375, + "logps/weighted_chosen": -3.48370361328125, + "logps/weighted_rejected": -4.3099365234375, + "loss": 0.5293, + "rewards/accuracies": 0.612500011920929, + "rewards/chosen": -287.9331970214844, + "rewards/margins": 112.53828430175781, + "rewards/rejected": -400.390625, + "rewards/weighted_accuracies": 0.7281249761581421, + "rewards/weighted_chosen": -1.1633422374725342, + "rewards/weighted_margins": 0.9815673828125, + "rewards/weighted_rejected": -2.144360303878784, + "step": 1090 + }, + { + "epoch": 0.575765506411934, + "grad_norm": 419.50701904296875, + "learning_rate": 4.566497100969792e-07, + "logits/chosen": -1.1083495616912842, + "logits/rejected": -1.1515319347381592, + "logps/chosen": -681.3523559570312, + "logps/rejected": -761.9328002929688, + "logps/weighted_chosen": -3.8998045921325684, + "logps/weighted_rejected": -4.931884765625, + "loss": 0.5762, + "rewards/accuracies": 0.6625000238418579, + "rewards/chosen": -358.6919860839844, + "rewards/margins": 104.8218765258789, + "rewards/rejected": -463.4956970214844, + "rewards/weighted_accuracies": 0.734375, + "rewards/weighted_chosen": -1.351904273033142, + "rewards/weighted_margins": 1.048761010169983, + "rewards/weighted_rejected": -2.40142822265625, + "step": 1100 + }, + { + "epoch": 0.5809997382884062, + "grad_norm": 56.882720947265625, + "learning_rate": 4.475540160336576e-07, + "logits/chosen": -1.1287415027618408, + "logits/rejected": -1.183984398841858, + "logps/chosen": -593.2249755859375, + "logps/rejected": -657.9328002929688, + "logps/weighted_chosen": -3.99072265625, + "logps/weighted_rejected": -5.168993949890137, + "loss": 0.5249, + "rewards/accuracies": 0.628125011920929, + "rewards/chosen": -294.9878845214844, + "rewards/margins": 104.7953109741211, + "rewards/rejected": -399.80743408203125, + "rewards/weighted_accuracies": 0.7250000238418579, + "rewards/weighted_chosen": -1.250451683998108, + "rewards/weighted_margins": 1.0572631359100342, + "rewards/weighted_rejected": -2.306506395339966, + "step": 1110 + }, + { + "epoch": 0.5862339701648783, + "grad_norm": 633.436279296875, + "learning_rate": 4.3847583850294565e-07, + "logits/chosen": -1.109594702720642, + "logits/rejected": -1.14373779296875, + "logps/chosen": -646.6163940429688, + "logps/rejected": -697.4781494140625, + "logps/weighted_chosen": -4.158276557922363, + "logps/weighted_rejected": -4.803124904632568, + "loss": 0.5622, + "rewards/accuracies": 0.6656249761581421, + "rewards/chosen": -355.50311279296875, + "rewards/margins": 82.48612976074219, + "rewards/rejected": -437.83984375, + "rewards/weighted_accuracies": 0.7406250238418579, + "rewards/weighted_chosen": -1.350854516029358, + "rewards/weighted_margins": 1.000707983970642, + "rewards/weighted_rejected": -2.352795362472534, + "step": 1120 + }, + { + "epoch": 0.5914682020413504, + "grad_norm": 512.849853515625, + "learning_rate": 4.294182095423934e-07, + "logits/chosen": -1.0823333263397217, + "logits/rejected": -1.1519775390625, + "logps/chosen": -601.3250122070312, + "logps/rejected": -686.8515625, + "logps/weighted_chosen": -3.522167921066284, + "logps/weighted_rejected": -4.769873142242432, + "loss": 0.5542, + "rewards/accuracies": 0.590624988079071, + "rewards/chosen": -315.3882751464844, + "rewards/margins": 110.39375305175781, + "rewards/rejected": -425.638671875, + "rewards/weighted_accuracies": 0.703125, + "rewards/weighted_chosen": -1.290283203125, + "rewards/weighted_margins": 1.020105004310608, + "rewards/weighted_rejected": -2.3109498023986816, + "step": 1130 + }, + { + "epoch": 0.5967024339178225, + "grad_norm": 65.05388641357422, + "learning_rate": 4.20384154326496e-07, + "logits/chosen": -1.0805175304412842, + "logits/rejected": -1.14312744140625, + "logps/chosen": -516.7843627929688, + "logps/rejected": -553.4617309570312, + "logps/weighted_chosen": -3.43231201171875, + "logps/weighted_rejected": -4.573815822601318, + "loss": 0.5682, + "rewards/accuracies": 0.6187499761581421, + "rewards/chosen": -256.2328186035156, + "rewards/margins": 61.6025390625, + "rewards/rejected": -317.79412841796875, + "rewards/weighted_accuracies": 0.7124999761581421, + "rewards/weighted_chosen": -1.086877465248108, + "rewards/weighted_margins": 0.8528991937637329, + "rewards/weighted_rejected": -1.9411132335662842, + "step": 1140 + }, + { + "epoch": 0.6019366657942947, + "grad_norm": 75.07366943359375, + "learning_rate": 4.1137669015630863e-07, + "logits/chosen": -1.077880859375, + "logits/rejected": -1.1405029296875, + "logps/chosen": -537.2109375, + "logps/rejected": -624.4749755859375, + "logps/weighted_chosen": -3.0843262672424316, + "logps/weighted_rejected": -4.2431640625, + "loss": 0.5219, + "rewards/accuracies": 0.675000011920929, + "rewards/chosen": -241.1707000732422, + "rewards/margins": 99.6703109741211, + "rewards/rejected": -340.8062438964844, + "rewards/weighted_accuracies": 0.7124999761581421, + "rewards/weighted_chosen": -0.9316955804824829, + "rewards/weighted_margins": 0.957165539264679, + "rewards/weighted_rejected": -1.889074683189392, + "step": 1150 + }, + { + "epoch": 0.6071708976707668, + "grad_norm": 53.81990432739258, + "learning_rate": 4.023988254516943e-07, + "logits/chosen": -1.059973120689392, + "logits/rejected": -1.120214819908142, + "logps/chosen": -530.0851440429688, + "logps/rejected": -566.734375, + "logps/weighted_chosen": -3.7030029296875, + "logps/weighted_rejected": -4.251660346984863, + "loss": 0.5091, + "rewards/accuracies": 0.653124988079071, + "rewards/chosen": -229.01113891601562, + "rewards/margins": 70.63789367675781, + "rewards/rejected": -299.630859375, + "rewards/weighted_accuracies": 0.75, + "rewards/weighted_chosen": -0.8327713012695312, + "rewards/weighted_margins": 0.8830932378768921, + "rewards/weighted_rejected": -1.715429663658142, + "step": 1160 + }, + { + "epoch": 0.6124051295472389, + "grad_norm": 50.65428161621094, + "learning_rate": 3.9345355874653366e-07, + "logits/chosen": -1.058905005455017, + "logits/rejected": -1.0953247547149658, + "logps/chosen": -535.8616943359375, + "logps/rejected": -544.9773559570312, + "logps/weighted_chosen": -3.3579344749450684, + "logps/weighted_rejected": -4.205273628234863, + "loss": 0.5906, + "rewards/accuracies": 0.609375, + "rewards/chosen": -238.26016235351562, + "rewards/margins": 56.61601638793945, + "rewards/rejected": -294.7398376464844, + "rewards/weighted_accuracies": 0.6968749761581421, + "rewards/weighted_chosen": -0.988323986530304, + "rewards/weighted_margins": 0.763989269733429, + "rewards/weighted_rejected": -1.7522430419921875, + "step": 1170 + }, + { + "epoch": 0.6176393614237111, + "grad_norm": 44.16736602783203, + "learning_rate": 3.8454387768724157e-07, + "logits/chosen": -1.097143530845642, + "logits/rejected": -1.1111571788787842, + "logps/chosen": -475.8023376464844, + "logps/rejected": -494.05078125, + "logps/weighted_chosen": -3.5390625, + "logps/weighted_rejected": -4.489306449890137, + "loss": 0.5337, + "rewards/accuracies": 0.578125, + "rewards/chosen": -202.70858764648438, + "rewards/margins": 60.640235900878906, + "rewards/rejected": -263.4339904785156, + "rewards/weighted_accuracies": 0.7124999761581421, + "rewards/weighted_chosen": -0.8571746945381165, + "rewards/weighted_margins": 0.9136596918106079, + "rewards/weighted_rejected": -1.7705810070037842, + "step": 1180 + }, + { + "epoch": 0.6228735933001832, + "grad_norm": 402.7128601074219, + "learning_rate": 3.7567275803491525e-07, + "logits/chosen": -1.1082000732421875, + "logits/rejected": -1.155554175376892, + "logps/chosen": -559.3796997070312, + "logps/rejected": -586.2781372070312, + "logps/weighted_chosen": -3.2266845703125, + "logps/weighted_rejected": -4.550146579742432, + "loss": 0.5018, + "rewards/accuracies": 0.6312500238418579, + "rewards/chosen": -257.08673095703125, + "rewards/margins": 75.61640930175781, + "rewards/rejected": -332.78204345703125, + "rewards/weighted_accuracies": 0.746874988079071, + "rewards/weighted_chosen": -0.8839172124862671, + "rewards/weighted_margins": 1.0030639171600342, + "rewards/weighted_rejected": -1.886468529701233, + "step": 1190 + }, + { + "epoch": 0.6281078251766553, + "grad_norm": 393.51788330078125, + "learning_rate": 3.66843162671456e-07, + "logits/chosen": -1.1154296398162842, + "logits/rejected": -1.1315186023712158, + "logps/chosen": -552.2093505859375, + "logps/rejected": -680.3843994140625, + "logps/weighted_chosen": -3.9300293922424316, + "logps/weighted_rejected": -4.456152439117432, + "loss": 0.5959, + "rewards/accuracies": 0.6031249761581421, + "rewards/chosen": -278.14764404296875, + "rewards/margins": 130.2062530517578, + "rewards/rejected": -408.24530029296875, + "rewards/weighted_accuracies": 0.671875, + "rewards/weighted_chosen": -1.1631286144256592, + "rewards/weighted_margins": 0.8199828863143921, + "rewards/weighted_rejected": -1.9827392101287842, + "step": 1200 + }, + { + "epoch": 0.6333420570531274, + "grad_norm": 915.16943359375, + "learning_rate": 3.5805804060998924e-07, + "logits/chosen": -1.1075561046600342, + "logits/rejected": -1.1484863758087158, + "logps/chosen": -600.2609252929688, + "logps/rejected": -717.1453247070312, + "logps/weighted_chosen": -3.2332520484924316, + "logps/weighted_rejected": -4.710791110992432, + "loss": 0.499, + "rewards/accuracies": 0.668749988079071, + "rewards/chosen": -299.421875, + "rewards/margins": 147.1335906982422, + "rewards/rejected": -446.451171875, + "rewards/weighted_accuracies": 0.75, + "rewards/weighted_chosen": -1.162207007408142, + "rewards/weighted_margins": 1.1646606922149658, + "rewards/weighted_rejected": -2.326342821121216, + "step": 1210 + }, + { + "epoch": 0.6385762889295996, + "grad_norm": 685.1804809570312, + "learning_rate": 3.493203260099197e-07, + "logits/chosen": -1.1248352527618408, + "logits/rejected": -1.195556640625, + "logps/chosen": -622.3953247070312, + "logps/rejected": -694.5250244140625, + "logps/weighted_chosen": -3.321972608566284, + "logps/weighted_rejected": -4.685571193695068, + "loss": 0.5497, + "rewards/accuracies": 0.581250011920929, + "rewards/chosen": -334.587890625, + "rewards/margins": 87.31562805175781, + "rewards/rejected": -421.9072265625, + "rewards/weighted_accuracies": 0.706250011920929, + "rewards/weighted_chosen": -1.299523949623108, + "rewards/weighted_margins": 0.9338439702987671, + "rewards/weighted_rejected": -2.2326292991638184, + "step": 1220 + }, + { + "epoch": 0.6438105208060717, + "grad_norm": 740.0750732421875, + "learning_rate": 3.4063293719694407e-07, + "logits/chosen": -1.126245141029358, + "logits/rejected": -1.180517554283142, + "logps/chosen": -567.4812622070312, + "logps/rejected": -648.1539306640625, + "logps/weighted_chosen": -3.639111280441284, + "logps/weighted_rejected": -4.597058296203613, + "loss": 0.6, + "rewards/accuracies": 0.6312500238418579, + "rewards/chosen": -290.95684814453125, + "rewards/margins": 99.4990234375, + "rewards/rejected": -390.62872314453125, + "rewards/weighted_accuracies": 0.6812499761581421, + "rewards/weighted_chosen": -1.1769592761993408, + "rewards/weighted_margins": 0.8424926996231079, + "rewards/weighted_rejected": -2.0197691917419434, + "step": 1230 + }, + { + "epoch": 0.6490447526825438, + "grad_norm": 342.7829895019531, + "learning_rate": 3.319987756883559e-07, + "logits/chosen": -1.156945824623108, + "logits/rejected": -1.198333740234375, + "logps/chosen": -572.3328247070312, + "logps/rejected": -671.7633056640625, + "logps/weighted_chosen": -3.338793992996216, + "logps/weighted_rejected": -4.542822360992432, + "loss": 0.5046, + "rewards/accuracies": 0.65625, + "rewards/chosen": -294.57342529296875, + "rewards/margins": 108.07109069824219, + "rewards/rejected": -402.66680908203125, + "rewards/weighted_accuracies": 0.746874988079071, + "rewards/weighted_chosen": -1.004339575767517, + "rewards/weighted_margins": 1.094323754310608, + "rewards/weighted_rejected": -2.0980224609375, + "step": 1240 + }, + { + "epoch": 0.654278984559016, + "grad_norm": 406.2123107910156, + "learning_rate": 3.234207252239607e-07, + "logits/chosen": -1.1353332996368408, + "logits/rejected": -1.184594750404358, + "logps/chosen": -643.1171875, + "logps/rejected": -680.65625, + "logps/weighted_chosen": -3.9437499046325684, + "logps/weighted_rejected": -4.764257907867432, + "loss": 0.5738, + "rewards/accuracies": 0.6000000238418579, + "rewards/chosen": -350.52264404296875, + "rewards/margins": 77.30488586425781, + "rewards/rejected": -428.01641845703125, + "rewards/weighted_accuracies": 0.675000011920929, + "rewards/weighted_chosen": -1.3293838500976562, + "rewards/weighted_margins": 0.863757312297821, + "rewards/weighted_rejected": -2.193603515625, + "step": 1250 + }, + { + "epoch": 0.6595132164354881, + "grad_norm": 518.19775390625, + "learning_rate": 3.1490165080293175e-07, + "logits/chosen": -1.1296905279159546, + "logits/rejected": -1.1927978992462158, + "logps/chosen": -573.6226806640625, + "logps/rejected": -675.9562377929688, + "logps/weighted_chosen": -3.474902391433716, + "logps/weighted_rejected": -4.43798828125, + "loss": 0.5317, + "rewards/accuracies": 0.590624988079071, + "rewards/chosen": -306.15155029296875, + "rewards/margins": 106.2289047241211, + "rewards/rejected": -412.40234375, + "rewards/weighted_accuracies": 0.7124999761581421, + "rewards/weighted_chosen": -1.2220337390899658, + "rewards/weighted_margins": 0.9764038324356079, + "rewards/weighted_rejected": -2.197509765625, + "step": 1260 + }, + { + "epoch": 0.6647474483119602, + "grad_norm": 152.93478393554688, + "learning_rate": 3.06444397726922e-07, + "logits/chosen": -1.082611083984375, + "logits/rejected": -1.1652100086212158, + "logps/chosen": -609.8375244140625, + "logps/rejected": -677.9749755859375, + "logps/weighted_chosen": -3.4163575172424316, + "logps/weighted_rejected": -5.1240234375, + "loss": 0.5052, + "rewards/accuracies": 0.6187499761581421, + "rewards/chosen": -314.2923889160156, + "rewards/margins": 101.6371078491211, + "rewards/rejected": -415.8968811035156, + "rewards/weighted_accuracies": 0.71875, + "rewards/weighted_chosen": -1.135400414466858, + "rewards/weighted_margins": 1.23980712890625, + "rewards/weighted_rejected": -2.375744581222534, + "step": 1270 + }, + { + "epoch": 0.6699816801884323, + "grad_norm": 285.5711669921875, + "learning_rate": 2.980517906497586e-07, + "logits/chosen": -1.1207764148712158, + "logits/rejected": -1.1819336414337158, + "logps/chosen": -597.5718994140625, + "logps/rejected": -703.9953002929688, + "logps/weighted_chosen": -3.632373094558716, + "logps/weighted_rejected": -4.974999904632568, + "loss": 0.5807, + "rewards/accuracies": 0.6312500238418579, + "rewards/chosen": -304.43048095703125, + "rewards/margins": 127.4535140991211, + "rewards/rejected": -431.8851623535156, + "rewards/weighted_accuracies": 0.721875011920929, + "rewards/weighted_chosen": -1.202392578125, + "rewards/weighted_margins": 0.9914306402206421, + "rewards/weighted_rejected": -2.193042039871216, + "step": 1280 + }, + { + "epoch": 0.6752159120649045, + "grad_norm": 82.95504760742188, + "learning_rate": 2.89726632634029e-07, + "logits/chosen": -1.113470435142517, + "logits/rejected": -1.1529052257537842, + "logps/chosen": -599.0562744140625, + "logps/rejected": -677.3453369140625, + "logps/weighted_chosen": -3.367846727371216, + "logps/weighted_rejected": -4.500195503234863, + "loss": 0.574, + "rewards/accuracies": 0.590624988079071, + "rewards/chosen": -324.748046875, + "rewards/margins": 97.44921875, + "rewards/rejected": -422.11798095703125, + "rewards/weighted_accuracies": 0.6937500238418579, + "rewards/weighted_chosen": -1.2255462408065796, + "rewards/weighted_margins": 0.875323474407196, + "rewards/weighted_rejected": -2.1007933616638184, + "step": 1290 + }, + { + "epoch": 0.6804501439413766, + "grad_norm": 2586.47119140625, + "learning_rate": 2.814717042148827e-07, + "logits/chosen": -1.131799340248108, + "logits/rejected": -1.1629149913787842, + "logps/chosen": -566.9031372070312, + "logps/rejected": -659.9281005859375, + "logps/weighted_chosen": -4.212353706359863, + "logps/weighted_rejected": -4.885107517242432, + "loss": 0.606, + "rewards/accuracies": 0.5874999761581421, + "rewards/chosen": -292.85858154296875, + "rewards/margins": 106.29667663574219, + "rewards/rejected": -399.13201904296875, + "rewards/weighted_accuracies": 0.684374988079071, + "rewards/weighted_chosen": -1.2742125988006592, + "rewards/weighted_margins": 0.8754211664199829, + "rewards/weighted_rejected": -2.149639844894409, + "step": 1300 + }, + { + "epoch": 0.6856843758178487, + "grad_norm": 20.005109786987305, + "learning_rate": 2.7328976247135416e-07, + "logits/chosen": -1.156762719154358, + "logits/rejected": -1.1973998546600342, + "logps/chosen": -577.4781494140625, + "logps/rejected": -642.1609497070312, + "logps/weighted_chosen": -3.5555663108825684, + "logps/weighted_rejected": -4.547216892242432, + "loss": 0.6037, + "rewards/accuracies": 0.6187499761581421, + "rewards/chosen": -287.97772216796875, + "rewards/margins": 101.68867492675781, + "rewards/rejected": -389.75079345703125, + "rewards/weighted_accuracies": 0.690625011920929, + "rewards/weighted_chosen": -1.2163574695587158, + "rewards/weighted_margins": 0.806256115436554, + "rewards/weighted_rejected": -2.022289991378784, + "step": 1310 + }, + { + "epoch": 0.6909186076943209, + "grad_norm": 151.74427795410156, + "learning_rate": 2.651835401055217e-07, + "logits/chosen": -1.149621605873108, + "logits/rejected": -1.183782935142517, + "logps/chosen": -597.9265747070312, + "logps/rejected": -678.625, + "logps/weighted_chosen": -3.490771532058716, + "logps/weighted_rejected": -4.412158012390137, + "loss": 0.5855, + "rewards/accuracies": 0.609375, + "rewards/chosen": -320.44451904296875, + "rewards/margins": 92.2328109741211, + "rewards/rejected": -412.76953125, + "rewards/weighted_accuracies": 0.7124999761581421, + "rewards/weighted_chosen": -1.213354468345642, + "rewards/weighted_margins": 0.8388611078262329, + "rewards/weighted_rejected": -2.0516357421875, + "step": 1320 + }, + { + "epoch": 0.696152839570793, + "grad_norm": 47.340484619140625, + "learning_rate": 2.571557445298055e-07, + "logits/chosen": -1.135412573814392, + "logits/rejected": -1.1872680187225342, + "logps/chosen": -558.5109252929688, + "logps/rejected": -635.0374755859375, + "logps/weighted_chosen": -3.464648485183716, + "logps/weighted_rejected": -4.4168701171875, + "loss": 0.5666, + "rewards/accuracies": 0.625, + "rewards/chosen": -280.0796813964844, + "rewards/margins": 90.732421875, + "rewards/rejected": -370.8531188964844, + "rewards/weighted_accuracies": 0.675000011920929, + "rewards/weighted_chosen": -1.1060791015625, + "rewards/weighted_margins": 0.821246325969696, + "rewards/weighted_rejected": -1.9268798828125, + "step": 1330 + }, + { + "epoch": 0.7013870714472651, + "grad_norm": 49.26417922973633, + "learning_rate": 2.49209056962716e-07, + "logits/chosen": -1.146185278892517, + "logits/rejected": -1.186279296875, + "logps/chosen": -612.96875, + "logps/rejected": -650.0765380859375, + "logps/weighted_chosen": -3.656054735183716, + "logps/weighted_rejected": -4.784960746765137, + "loss": 0.5627, + "rewards/accuracies": 0.6031249761581421, + "rewards/chosen": -298.53436279296875, + "rewards/margins": 84.52030944824219, + "rewards/rejected": -383.0699157714844, + "rewards/weighted_accuracies": 0.706250011920929, + "rewards/weighted_chosen": -1.108679175376892, + "rewards/weighted_margins": 0.8104248046875, + "rewards/weighted_rejected": -1.9180908203125, + "step": 1340 + }, + { + "epoch": 0.7066213033237373, + "grad_norm": 43.23517990112305, + "learning_rate": 2.41346131533347e-07, + "logits/chosen": -1.1938354969024658, + "logits/rejected": -1.217199683189392, + "logps/chosen": -628.8203125, + "logps/rejected": -686.6749877929688, + "logps/weighted_chosen": -3.1535401344299316, + "logps/weighted_rejected": -4.464306831359863, + "loss": 0.546, + "rewards/accuracies": 0.625, + "rewards/chosen": -322.18829345703125, + "rewards/margins": 89.7154312133789, + "rewards/rejected": -411.8656311035156, + "rewards/weighted_accuracies": 0.737500011920929, + "rewards/weighted_chosen": -1.133204698562622, + "rewards/weighted_margins": 0.804534912109375, + "rewards/weighted_rejected": -1.9378173351287842, + "step": 1350 + }, + { + "epoch": 0.7118555352002094, + "grad_norm": 34.37676239013672, + "learning_rate": 2.3356959439491898e-07, + "logits/chosen": -1.1357300281524658, + "logits/rejected": -1.2062256336212158, + "logps/chosen": -595.59375, + "logps/rejected": -672.6422119140625, + "logps/weighted_chosen": -3.9969725608825684, + "logps/weighted_rejected": -4.689795017242432, + "loss": 0.5808, + "rewards/accuracies": 0.625, + "rewards/chosen": -325.7347717285156, + "rewards/margins": 104.75312805175781, + "rewards/rejected": -430.67108154296875, + "rewards/weighted_accuracies": 0.699999988079071, + "rewards/weighted_chosen": -1.2155777215957642, + "rewards/weighted_margins": 0.9368530511856079, + "rewards/weighted_rejected": -2.152294874191284, + "step": 1360 + }, + { + "epoch": 0.7170897670766815, + "grad_norm": 51.87490463256836, + "learning_rate": 2.258820428476645e-07, + "logits/chosen": -1.1657593250274658, + "logits/rejected": -1.2217528820037842, + "logps/chosen": -631.8890380859375, + "logps/rejected": -746.7828369140625, + "logps/weighted_chosen": -3.470263719558716, + "logps/weighted_rejected": -4.346484184265137, + "loss": 0.5501, + "rewards/accuracies": 0.612500011920929, + "rewards/chosen": -352.91015625, + "rewards/margins": 124.40898132324219, + "rewards/rejected": -477.31951904296875, + "rewards/weighted_accuracies": 0.734375, + "rewards/weighted_chosen": -1.186987280845642, + "rewards/weighted_margins": 0.836499035358429, + "rewards/weighted_rejected": -2.023852586746216, + "step": 1370 + }, + { + "epoch": 0.7223239989531536, + "grad_norm": 416.21136474609375, + "learning_rate": 2.1828604447135245e-07, + "logits/chosen": -1.11541748046875, + "logits/rejected": -1.1730468273162842, + "logps/chosen": -694.0234375, + "logps/rejected": -737.5452880859375, + "logps/weighted_chosen": -3.9507813453674316, + "logps/weighted_rejected": -5.173535346984863, + "loss": 0.5413, + "rewards/accuracies": 0.5874999761581421, + "rewards/chosen": -423.78436279296875, + "rewards/margins": 56.84648513793945, + "rewards/rejected": -480.90936279296875, + "rewards/weighted_accuracies": 0.6875, + "rewards/weighted_chosen": -1.29058837890625, + "rewards/weighted_margins": 0.98492431640625, + "rewards/weighted_rejected": -2.275256395339966, + "step": 1380 + }, + { + "epoch": 0.7275582308296258, + "grad_norm": 68.63409423828125, + "learning_rate": 2.1078413626773545e-07, + "logits/chosen": -1.174198865890503, + "logits/rejected": -1.2014648914337158, + "logps/chosen": -640.5343627929688, + "logps/rejected": -758.4734497070312, + "logps/weighted_chosen": -3.5299744606018066, + "logps/weighted_rejected": -5.219091892242432, + "loss": 0.5755, + "rewards/accuracies": 0.6031249761581421, + "rewards/chosen": -360.80780029296875, + "rewards/margins": 128.4132843017578, + "rewards/rejected": -489.3277282714844, + "rewards/weighted_accuracies": 0.715624988079071, + "rewards/weighted_chosen": -1.248020887374878, + "rewards/weighted_margins": 0.9331451654434204, + "rewards/weighted_rejected": -2.181262254714966, + "step": 1390 + }, + { + "epoch": 0.7327924627060979, + "grad_norm": 323.93890380859375, + "learning_rate": 2.0337882381321347e-07, + "logits/chosen": -1.1679198741912842, + "logits/rejected": -1.1945312023162842, + "logps/chosen": -653.7203369140625, + "logps/rejected": -707.4640502929688, + "logps/weighted_chosen": -3.5541749000549316, + "logps/weighted_rejected": -4.433495998382568, + "loss": 0.5551, + "rewards/accuracies": 0.612500011920929, + "rewards/chosen": -362.28399658203125, + "rewards/margins": 91.1234359741211, + "rewards/rejected": -453.5562438964844, + "rewards/weighted_accuracies": 0.7093750238418579, + "rewards/weighted_chosen": -1.336692452430725, + "rewards/weighted_margins": 0.8907715082168579, + "rewards/weighted_rejected": -2.227795362472534, + "step": 1400 + }, + { + "epoch": 0.73802669458257, + "grad_norm": 74.13733673095703, + "learning_rate": 1.960725804219905e-07, + "logits/chosen": -1.149194359779358, + "logits/rejected": -1.2033812999725342, + "logps/chosen": -661.9656372070312, + "logps/rejected": -768.3421630859375, + "logps/weighted_chosen": -4.091479301452637, + "logps/weighted_rejected": -4.242919921875, + "loss": 0.5754, + "rewards/accuracies": 0.5718749761581421, + "rewards/chosen": -362.45428466796875, + "rewards/margins": 122.5601577758789, + "rewards/rejected": -484.8101501464844, + "rewards/weighted_accuracies": 0.6937500238418579, + "rewards/weighted_chosen": -1.2423064708709717, + "rewards/weighted_margins": 0.8276306390762329, + "rewards/weighted_rejected": -2.069873094558716, + "step": 1410 + }, + { + "epoch": 0.7432609264590422, + "grad_norm": 237.26011657714844, + "learning_rate": 1.8886784632000824e-07, + "logits/chosen": -1.1620361804962158, + "logits/rejected": -1.206933617591858, + "logps/chosen": -640.3624877929688, + "logps/rejected": -783.7437744140625, + "logps/weighted_chosen": -3.4447264671325684, + "logps/weighted_rejected": -4.929491996765137, + "loss": 0.5398, + "rewards/accuracies": 0.65625, + "rewards/chosen": -354.3374938964844, + "rewards/margins": 162.05233764648438, + "rewards/rejected": -516.314453125, + "rewards/weighted_accuracies": 0.7250000238418579, + "rewards/weighted_chosen": -1.1836121082305908, + "rewards/weighted_margins": 1.072717308998108, + "rewards/weighted_rejected": -2.256579637527466, + "step": 1420 + }, + { + "epoch": 0.7484951583355143, + "grad_norm": 482.7161560058594, + "learning_rate": 1.8176702782993025e-07, + "logits/chosen": -1.166711449623108, + "logits/rejected": -1.185546875, + "logps/chosen": -614.01953125, + "logps/rejected": -700.3484497070312, + "logps/weighted_chosen": -3.3925538063049316, + "logps/weighted_rejected": -4.749658107757568, + "loss": 0.5436, + "rewards/accuracies": 0.6031249761581421, + "rewards/chosen": -352.78125, + "rewards/margins": 93.58476257324219, + "rewards/rejected": -446.29296875, + "rewards/weighted_accuracies": 0.706250011920929, + "rewards/weighted_chosen": -1.2850220203399658, + "rewards/weighted_margins": 0.9663940668106079, + "rewards/weighted_rejected": -2.2514891624450684, + "step": 1430 + }, + { + "epoch": 0.7537293902119864, + "grad_norm": 48.64420700073242, + "learning_rate": 1.7477249656745034e-07, + "logits/chosen": -1.1055176258087158, + "logits/rejected": -1.157617211341858, + "logps/chosen": -559.9140625, + "logps/rejected": -613.4656372070312, + "logps/weighted_chosen": -3.6603026390075684, + "logps/weighted_rejected": -4.916943550109863, + "loss": 0.5473, + "rewards/accuracies": 0.5874999761581421, + "rewards/chosen": -309.0234375, + "rewards/margins": 77.943359375, + "rewards/rejected": -387.1015625, + "rewards/weighted_accuracies": 0.731249988079071, + "rewards/weighted_chosen": -1.15887451171875, + "rewards/weighted_margins": 1.0418579578399658, + "rewards/weighted_rejected": -2.2011961936950684, + "step": 1440 + }, + { + "epoch": 0.7589636220884585, + "grad_norm": 61.97262954711914, + "learning_rate": 1.6788658864919118e-07, + "logits/chosen": -1.111364722251892, + "logits/rejected": -1.2080078125, + "logps/chosen": -681.9781494140625, + "logps/rejected": -766.1875, + "logps/weighted_chosen": -3.5382080078125, + "logps/weighted_rejected": -4.320166110992432, + "loss": 0.5212, + "rewards/accuracies": 0.625, + "rewards/chosen": -367.3251953125, + "rewards/margins": 115.8160171508789, + "rewards/rejected": -482.9632873535156, + "rewards/weighted_accuracies": 0.715624988079071, + "rewards/weighted_chosen": -1.1493957042694092, + "rewards/weighted_margins": 1.0002562999725342, + "rewards/weighted_rejected": -2.149792432785034, + "step": 1450 + }, + { + "epoch": 0.7641978539649307, + "grad_norm": 33.67172622680664, + "learning_rate": 1.611116039124613e-07, + "logits/chosen": -1.0988738536834717, + "logits/rejected": -1.160314917564392, + "logps/chosen": -592.9827880859375, + "logps/rejected": -634.5718994140625, + "logps/weighted_chosen": -3.8994140625, + "logps/weighted_rejected": -4.845922946929932, + "loss": 0.5548, + "rewards/accuracies": 0.609375, + "rewards/chosen": -330.28790283203125, + "rewards/margins": 74.87812805175781, + "rewards/rejected": -405.23126220703125, + "rewards/weighted_accuracies": 0.6812499761581421, + "rewards/weighted_chosen": -1.151556372642517, + "rewards/weighted_margins": 0.903613269329071, + "rewards/weighted_rejected": -2.055248975753784, + "step": 1460 + }, + { + "epoch": 0.7694320858414028, + "grad_norm": 144.79293823242188, + "learning_rate": 1.5444980514712723e-07, + "logits/chosen": -1.196630835533142, + "logits/rejected": -1.222875952720642, + "logps/chosen": -668.71875, + "logps/rejected": -797.6375122070312, + "logps/weighted_chosen": -3.661572217941284, + "logps/weighted_rejected": -4.466870307922363, + "loss": 0.6131, + "rewards/accuracies": 0.621874988079071, + "rewards/chosen": -357.33673095703125, + "rewards/margins": 134.50155639648438, + "rewards/rejected": -491.7562561035156, + "rewards/weighted_accuracies": 0.6781250238418579, + "rewards/weighted_chosen": -1.314672827720642, + "rewards/weighted_margins": 0.78485107421875, + "rewards/weighted_rejected": -2.099261522293091, + "step": 1470 + }, + { + "epoch": 0.7746663177178749, + "grad_norm": 40.60256576538086, + "learning_rate": 1.4790341733986083e-07, + "logits/chosen": -1.150146484375, + "logits/rejected": -1.1887328624725342, + "logps/chosen": -620.7406005859375, + "logps/rejected": -698.5499877929688, + "logps/weighted_chosen": -3.9625487327575684, + "logps/weighted_rejected": -4.390429496765137, + "loss": 0.5719, + "rewards/accuracies": 0.6000000238418579, + "rewards/chosen": -333.4839782714844, + "rewards/margins": 102.4716796875, + "rewards/rejected": -436.12420654296875, + "rewards/weighted_accuracies": 0.7093750238418579, + "rewards/weighted_chosen": -1.234228491783142, + "rewards/weighted_margins": 0.851452648639679, + "rewards/weighted_rejected": -2.085034132003784, + "step": 1480 + }, + { + "epoch": 0.7799005495943471, + "grad_norm": 217.96365356445312, + "learning_rate": 1.4147462693101108e-07, + "logits/chosen": -1.131872534751892, + "logits/rejected": -1.179223656654358, + "logps/chosen": -645.7999877929688, + "logps/rejected": -766.2390747070312, + "logps/weighted_chosen": -3.494384765625, + "logps/weighted_rejected": -4.757568359375, + "loss": 0.5302, + "rewards/accuracies": 0.640625, + "rewards/chosen": -358.8677673339844, + "rewards/margins": 136.17617797851562, + "rewards/rejected": -495.09765625, + "rewards/weighted_accuracies": 0.734375, + "rewards/weighted_chosen": -1.171356201171875, + "rewards/weighted_margins": 1.06304931640625, + "rewards/weighted_rejected": -2.235211133956909, + "step": 1490 + }, + { + "epoch": 0.7851347814708192, + "grad_norm": 655.3441162109375, + "learning_rate": 1.3516558108435177e-07, + "logits/chosen": -1.133544921875, + "logits/rejected": -1.154302954673767, + "logps/chosen": -594.9124755859375, + "logps/rejected": -729.9781494140625, + "logps/weighted_chosen": -3.306835889816284, + "logps/weighted_rejected": -4.992211818695068, + "loss": 0.508, + "rewards/accuracies": 0.606249988079071, + "rewards/chosen": -315.57147216796875, + "rewards/margins": 149.4695281982422, + "rewards/rejected": -465.1382751464844, + "rewards/weighted_accuracies": 0.737500011920929, + "rewards/weighted_chosen": -1.0454895496368408, + "rewards/weighted_margins": 1.0130126476287842, + "rewards/weighted_rejected": -2.058490037918091, + "step": 1500 + }, + { + "epoch": 0.7851347814708192, + "eval_logits/chosen": -1.2306523323059082, + "eval_logits/rejected": -1.26123046875, + "eval_logps/chosen": -633.6630249023438, + "eval_logps/rejected": -713.4450073242188, + "eval_logps/weighted_chosen": -3.545902967453003, + "eval_logps/weighted_rejected": -4.605466842651367, + "eval_loss": 0.5521173477172852, + "eval_rewards/accuracies": 0.593999981880188, + "eval_rewards/chosen": -344.9159851074219, + "eval_rewards/margins": 100.7760009765625, + "eval_rewards/rejected": -445.6925048828125, + "eval_rewards/weighted_accuracies": 0.7149999737739563, + "eval_rewards/weighted_chosen": -1.159058928489685, + "eval_rewards/weighted_margins": 0.8911977410316467, + "eval_rewards/weighted_rejected": -2.0502567291259766, + "eval_runtime": 1316.3607, + "eval_samples_per_second": 1.519, + "eval_steps_per_second": 0.38, + "step": 1500 + } + ], + "logging_steps": 10, + "max_steps": 1911, + "num_input_tokens_seen": 0, + "num_train_epochs": 1, + "save_steps": 500, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 0.0, + "train_batch_size": 1, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoint-1500/training_args.bin b/checkpoint-1500/training_args.bin new file mode 100644 index 0000000..eb22667 --- /dev/null +++ b/checkpoint-1500/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ded8eebbc35e0c697502403c9c9d2a0acb4a3f38810fc064a38066719f316edb +size 8785 diff --git a/checkpoint-1500/zero_to_fp32.py b/checkpoint-1500/zero_to_fp32.py new file mode 100644 index 0000000..0e75914 --- /dev/null +++ b/checkpoint-1500/zero_to_fp32.py @@ -0,0 +1,760 @@ +#!/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 . output_dir/ +# or +# python zero_to_fp32.py . output_dir/ --safe_serialization + +import argparse +import torch +import glob +import math +import os +import re +import gc +import json +import numpy as np +from tqdm import tqdm +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, weights_only=False) + + 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 tqdm(files, desc='Loading checkpoint shards'): + state_dict = torch.load(f, map_location=device, mmap=True, weights_only=False) + # 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}") + + fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] 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") + + +class GatheredTensor: + """ + A pseudo tensor that collects partitioned weights. + It is more memory efficient when there are multiple groups. + """ + + def __init__(self, flat_groups, flat_groups_offset, offset, partitioned_numel, shape): + self.flat_groups = flat_groups + self.flat_groups_offset = flat_groups_offset + self.offset = offset + self.partitioned_numel = partitioned_numel + self.shape = shape + self.dtype = self.flat_groups[0][0].dtype + + def contiguous(self): + """ + Merge partitioned weights from flat_groups into a single tensor. + """ + end_idx = self.offset + self.partitioned_numel + world_size = len(self.flat_groups) + pad_flat_param_chunks = [] + + for rank_i in range(world_size): + # for each rank, we need to collect weights from related group/groups + flat_groups_at_rank_i = self.flat_groups[rank_i] + start_group_id = None + end_group_id = None + for group_id in range(len(self.flat_groups_offset)): + if self.flat_groups_offset[group_id] <= self.offset < self.flat_groups_offset[group_id + 1]: + start_group_id = group_id + if self.flat_groups_offset[group_id] < end_idx <= self.flat_groups_offset[group_id + 1]: + end_group_id = group_id + break + # collect weights from related group/groups + for group_id in range(start_group_id, end_group_id + 1): + flat_tensor = flat_groups_at_rank_i[group_id] + start_offset = self.offset - self.flat_groups_offset[group_id] + end_offset = min(end_idx, self.flat_groups_offset[group_id + 1]) - self.flat_groups_offset[group_id] + pad_flat_param_chunks.append(flat_tensor[start_offset:end_offset]) + + # collect weights from all ranks + pad_flat_param = torch.cat(pad_flat_param_chunks, dim=0) + param = pad_flat_param[:self.shape.numel()].view(self.shape).contiguous() + return param + + +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 = sum([flat_group.numel() for flat_group in fp32_flat_groups[0]]) * 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 + flat_groups_offset = [0] + list(np.cumsum([flat_tensor.numel() for flat_tensor in fp32_flat_groups[0]])) + for name, shape in tqdm(param_shapes.items(), desc='Gathering sharded weights'): + 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}" + ) + + # memory efficient tensor + tensor = GatheredTensor(fp32_flat_groups, flat_groups_offset, offset, partitioned_numel, shape) + state_dict[name] = tensor + 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 to_torch_tensor(state_dict, return_empty_tensor=False): + """ + Convert state_dict of GatheredTensor to torch tensor + """ + torch_state_dict = {} + converted_tensors = {} + for name, tensor in state_dict.items(): + tensor_id = id(tensor) + if tensor_id in converted_tensors: # shared tensors + shared_tensor = torch_state_dict[converted_tensors[tensor_id]] + torch_state_dict[name] = shared_tensor + else: + converted_tensors[tensor_id] = name + if return_empty_tensor: + torch_state_dict[name] = torch.empty(tensor.shape, dtype=tensor.dtype) + else: + torch_state_dict[name] = tensor.contiguous() + return torch_state_dict + + +def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, + tag=None, + exclude_frozen_parameters=False, + lazy_mode=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 + - ``lazy_mode``: get state_dict in lazy mode. It returns a dict of pesduo tensor instead of torch tensor, which is more memory efficient. + Convert the pesduo tensor to torch tensor by ``.contiguous()`` + + Returns: + - pytorch ``state_dict`` + + 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. + + Note: the above usage may not work if your application doesn't have sufficient free CPU memory. + You may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with + the checkpoint. Or you can load state_dict in lazy mode :: + + from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, lazy_mode=True) # not on cpu + for name, lazy_tensor in state_dict.item(): + tensor = lazy_tensor.contiguous() # to cpu + print(name, tensor) + # del tensor to release memory if it no longer in use + """ + 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") + + state_dict = _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters) + if lazy_mode: + return state_dict + else: + return to_torch_tensor(state_dict) + + +def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, + output_dir, + max_shard_size="5GB", + safe_serialization=False, + 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_dir``: directory to the pytorch fp32 state_dict output files + - ``max_shard_size``: the maximum size for a checkpoint before being sharded, default value is 5GB + - ``safe_serialization``: whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`). + - ``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 + """ + + # Dependency pre-check + if safe_serialization: + try: + from safetensors.torch import save_file + except ImportError: + print('If you want to use `safe_serialization`, please `pip install safetensors`') + raise + if max_shard_size is not None: + try: + from huggingface_hub import split_torch_state_dict_into_shards + except ImportError: + print('If you want to use `max_shard_size`, please `pip install huggingface_hub`') + raise + + # Convert zero checkpoint to state_dict + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, + tag, + exclude_frozen_parameters, + lazy_mode=True) + + # Shard the model if it is too big. + weights_name = "model.safetensors" if safe_serialization else "pytorch_model.bin" + if max_shard_size is not None: + filename_pattern = weights_name.replace(".bin", "{suffix}.bin").replace(".safetensors", "{suffix}.safetensors") + # an memory-efficient approach for sharding + empty_state_dict = to_torch_tensor(state_dict, return_empty_tensor=True) + state_dict_split = split_torch_state_dict_into_shards(empty_state_dict, + filename_pattern=filename_pattern, + max_shard_size=max_shard_size) + else: + from collections import namedtuple + StateDictSplit = namedtuple("StateDictSplit", ["is_sharded", "filename_to_tensors"]) + state_dict_split = StateDictSplit(is_sharded=False, + filename_to_tensors={weights_name: list(state_dict.keys())}) + + # Save the model by shard + os.makedirs(output_dir, exist_ok=True) + filename_to_tensors = state_dict_split.filename_to_tensors.items() + for shard_file, tensors in tqdm(filename_to_tensors, desc="Saving checkpoint shards"): + shard_state_dict = {tensor_name: state_dict[tensor_name] for tensor_name in tensors} + shard_state_dict = to_torch_tensor(shard_state_dict) + output_path = os.path.join(output_dir, shard_file) + if safe_serialization: + save_file(shard_state_dict, output_path, metadata={"format": "pt"}) + else: + torch.save(shard_state_dict, output_path) + # release the memory of current shard + for tensor_name in list(shard_state_dict.keys()): + del state_dict[tensor_name] + del shard_state_dict[tensor_name] + del shard_state_dict + gc.collect() + + # Save index if sharded + if state_dict_split.is_sharded: + index = { + "metadata": state_dict_split.metadata, + "weight_map": state_dict_split.tensor_to_filename, + } + save_index_file = "model.safetensors.index.json" if safe_serialization else "pytorch_model.bin.index.json" + save_index_file = os.path.join(output_dir, save_index_file) + with open(save_index_file, "w", encoding="utf-8") as f: + content = json.dumps(index, indent=2, sort_keys=True) + "\n" + f.write(content) + + +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_dir", + type=str, + help="directory to the pytorch fp32 state_dict output files" + "(e.g. path/checkpoint-12-output/)") + parser.add_argument( + "--max_shard_size", + type=str, + default="5GB", + help="The maximum size for a checkpoint before being sharded. Checkpoints shard will then be each of size" + "lower than this size. If expressed as a string, needs to be digits followed by a unit (like `5MB`" + "We default it to 5GB in order for models to be able to run easily on free-tier google colab instances" + "without CPU OOM issues.") + parser.add_argument( + "--safe_serialization", + default=False, + action='store_true', + help="Whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).") + 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_dir, + max_shard_size=args.max_shard_size, + safe_serialization=args.safe_serialization, + tag=args.tag, + exclude_frozen_parameters=args.exclude_frozen_parameters) diff --git a/checkpoint-1911/chat_template.jinja b/checkpoint-1911/chat_template.jinja new file mode 100644 index 0000000..39bd0c9 --- /dev/null +++ b/checkpoint-1911/chat_template.jinja @@ -0,0 +1,5 @@ +{% set loop_messages = messages %}{% for message in loop_messages %}{% set content = '<|start_header_id|>' + message['role'] + '<|end_header_id|> + +'+ message['content'] | trim + '<|eot_id|>' %}{% if loop.index0 == 0 %}{% set content = bos_token + content %}{% endif %}{{ content }}{% endfor %}{% if add_generation_prompt %}{{ '<|start_header_id|>assistant<|end_header_id|> + +' }}{% endif %} \ No newline at end of file diff --git a/checkpoint-1911/config.json b/checkpoint-1911/config.json new file mode 100644 index 0000000..3f8f5c0 --- /dev/null +++ b/checkpoint-1911/config.json @@ -0,0 +1,29 @@ +{ + "architectures": [ + "LlamaForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "bos_token_id": 128000, + "eos_token_id": 128001, + "head_dim": 128, + "hidden_act": "silu", + "hidden_size": 4096, + "initializer_range": 0.02, + "intermediate_size": 14336, + "max_position_embeddings": 8192, + "mlp_bias": false, + "model_type": "llama", + "num_attention_heads": 32, + "num_hidden_layers": 32, + "num_key_value_heads": 8, + "pretraining_tp": 1, + "rms_norm_eps": 1e-05, + "rope_scaling": null, + "rope_theta": 500000.0, + "tie_word_embeddings": false, + "torch_dtype": "bfloat16", + "transformers_version": "4.54.1", + "use_cache": true, + "vocab_size": 128256 +} diff --git a/checkpoint-1911/generation_config.json b/checkpoint-1911/generation_config.json new file mode 100644 index 0000000..fc3c54a --- /dev/null +++ b/checkpoint-1911/generation_config.json @@ -0,0 +1,6 @@ +{ + "_from_model_config": true, + "bos_token_id": 128000, + "eos_token_id": 128001, + "transformers_version": "4.54.1" +} diff --git a/checkpoint-1911/latest b/checkpoint-1911/latest new file mode 100644 index 0000000..44970e7 --- /dev/null +++ b/checkpoint-1911/latest @@ -0,0 +1 @@ +global_step1910 \ No newline at end of file diff --git a/checkpoint-1911/model-00001-of-00004.safetensors b/checkpoint-1911/model-00001-of-00004.safetensors new file mode 100644 index 0000000..fe3fa95 --- /dev/null +++ b/checkpoint-1911/model-00001-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8209180ff91c21e5237d8af84b424786491a3be181c7fdd56dfbbf618026f5ab +size 4976698672 diff --git a/checkpoint-1911/model-00002-of-00004.safetensors b/checkpoint-1911/model-00002-of-00004.safetensors new file mode 100644 index 0000000..b1ca729 --- /dev/null +++ b/checkpoint-1911/model-00002-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:057704766c2933c8127ef9ff0f2d95990eaad4c77689f3ca38fc64e15b61462a +size 4999802720 diff --git a/checkpoint-1911/model-00003-of-00004.safetensors b/checkpoint-1911/model-00003-of-00004.safetensors new file mode 100644 index 0000000..62a518a --- /dev/null +++ b/checkpoint-1911/model-00003-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dd5511a86272c495069570a333effca3e14edab8427d25d9f9d62034005354ad +size 4915916176 diff --git a/checkpoint-1911/model-00004-of-00004.safetensors b/checkpoint-1911/model-00004-of-00004.safetensors new file mode 100644 index 0000000..0d5a6a1 --- /dev/null +++ b/checkpoint-1911/model-00004-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7107756fe86b7f389d0887d3517180fc28a3d56fc2e46efcc95cb8aadc7dc253 +size 1168138808 diff --git a/checkpoint-1911/model.safetensors.index.json b/checkpoint-1911/model.safetensors.index.json new file mode 100644 index 0000000..58a7ef4 --- /dev/null +++ b/checkpoint-1911/model.safetensors.index.json @@ -0,0 +1,299 @@ +{ + "metadata": { + "total_parameters": 266240, + "total_size": 16060522496 + }, + "weight_map": { + "lm_head.weight": "model-00004-of-00004.safetensors", + "model.embed_tokens.weight": "model-00001-of-00004.safetensors", + "model.layers.0.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.0.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.1.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.10.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.10.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.10.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.11.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.11.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.12.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.12.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.13.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.13.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.14.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.14.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.15.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.15.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.16.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.16.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.17.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.17.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.18.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.18.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.19.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.19.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.2.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.2.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.20.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.20.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.20.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.20.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.20.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.20.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.20.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.20.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.20.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.21.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.21.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.21.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.22.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.22.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.23.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.23.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.24.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.24.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.25.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.25.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.26.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.26.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.27.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.27.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.28.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.28.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.29.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.29.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.3.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.3.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.30.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.30.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.30.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.input_layernorm.weight": "model-00004-of-00004.safetensors", + "model.layers.31.mlp.down_proj.weight": "model-00004-of-00004.safetensors", + "model.layers.31.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.post_attention_layernorm.weight": "model-00004-of-00004.safetensors", + "model.layers.31.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.4.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.4.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.5.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.6.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.7.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.8.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.9.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.9.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.9.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.norm.weight": "model-00004-of-00004.safetensors" + } +} diff --git a/checkpoint-1911/rng_state_0.pth b/checkpoint-1911/rng_state_0.pth new file mode 100644 index 0000000..8d84687 --- /dev/null +++ b/checkpoint-1911/rng_state_0.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:02ea5dcfd1b4a49b41b4fa01a8b24bba6186957162c3fd555ebff28620c7268b +size 14917 diff --git a/checkpoint-1911/rng_state_1.pth b/checkpoint-1911/rng_state_1.pth new file mode 100644 index 0000000..54f0119 --- /dev/null +++ b/checkpoint-1911/rng_state_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c2a5af18bb5eae8b7fd6bdef66259014d98ba87ffb16d614bba38f2c32030798 +size 14917 diff --git a/checkpoint-1911/scheduler.pt b/checkpoint-1911/scheduler.pt new file mode 100644 index 0000000..b593d50 --- /dev/null +++ b/checkpoint-1911/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f5b59ab434923fa38180a582743116cb648692f61b72d4eddecc5e7980e6087d +size 1465 diff --git a/checkpoint-1911/special_tokens_map.json b/checkpoint-1911/special_tokens_map.json new file mode 100644 index 0000000..e5b39b6 --- /dev/null +++ b/checkpoint-1911/special_tokens_map.json @@ -0,0 +1,23 @@ +{ + "bos_token": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "eos_token": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "pad_token": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + } +} diff --git a/checkpoint-1911/tokenizer.json b/checkpoint-1911/tokenizer.json new file mode 100644 index 0000000..03aa64f --- /dev/null +++ b/checkpoint-1911/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0968dcc0ee8e56c7dccd34a7f51f8065ea0cb9e2cc529e3243d1e5c0a4bdaa0c +size 17208754 diff --git a/checkpoint-1911/tokenizer_config.json b/checkpoint-1911/tokenizer_config.json new file mode 100644 index 0000000..877a9a9 --- /dev/null +++ b/checkpoint-1911/tokenizer_config.json @@ -0,0 +1,2063 @@ +{ + "added_tokens_decoder": { + "128000": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128001": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128002": { + "content": "<|reserved_special_token_0|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128003": { + "content": "<|reserved_special_token_1|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128004": { + "content": "<|reserved_special_token_2|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128005": { + "content": "<|reserved_special_token_3|>", + "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": "<|reserved_special_token_4|>", + "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": "<|reserved_special_token_5|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128011": { + "content": "<|reserved_special_token_6|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128012": { + "content": "<|reserved_special_token_7|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128013": { + "content": "<|reserved_special_token_8|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128014": { + "content": "<|reserved_special_token_9|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128015": { + "content": "<|reserved_special_token_10|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128016": { + "content": "<|reserved_special_token_11|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128017": { + "content": "<|reserved_special_token_12|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128018": { + "content": "<|reserved_special_token_13|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128019": { + "content": "<|reserved_special_token_14|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128020": { + "content": "<|reserved_special_token_15|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128021": { + "content": "<|reserved_special_token_16|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128022": { + "content": "<|reserved_special_token_17|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128023": { + "content": "<|reserved_special_token_18|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128024": { + "content": "<|reserved_special_token_19|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128025": { + "content": "<|reserved_special_token_20|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128026": { + "content": "<|reserved_special_token_21|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128027": { + "content": "<|reserved_special_token_22|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128028": { + "content": "<|reserved_special_token_23|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128029": { + "content": "<|reserved_special_token_24|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128030": { + "content": "<|reserved_special_token_25|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128031": { + "content": "<|reserved_special_token_26|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128032": { + "content": "<|reserved_special_token_27|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128033": { + "content": "<|reserved_special_token_28|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128034": { + "content": "<|reserved_special_token_29|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128035": { + "content": "<|reserved_special_token_30|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128036": { + "content": "<|reserved_special_token_31|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128037": { + "content": "<|reserved_special_token_32|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128038": { + "content": "<|reserved_special_token_33|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128039": { + "content": "<|reserved_special_token_34|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128040": { + "content": "<|reserved_special_token_35|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128041": { + "content": "<|reserved_special_token_36|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128042": { + "content": "<|reserved_special_token_37|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128043": { + "content": "<|reserved_special_token_38|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128044": { + "content": "<|reserved_special_token_39|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128045": { + "content": "<|reserved_special_token_40|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128046": { + "content": "<|reserved_special_token_41|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128047": { + "content": "<|reserved_special_token_42|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128048": { + "content": "<|reserved_special_token_43|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128049": { + "content": "<|reserved_special_token_44|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128050": { + "content": "<|reserved_special_token_45|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128051": { + "content": "<|reserved_special_token_46|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128052": { + "content": "<|reserved_special_token_47|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128053": { + "content": "<|reserved_special_token_48|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128054": { + "content": "<|reserved_special_token_49|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128055": { + "content": "<|reserved_special_token_50|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128056": { + "content": "<|reserved_special_token_51|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128057": { + "content": "<|reserved_special_token_52|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128058": { + "content": "<|reserved_special_token_53|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128059": { + "content": "<|reserved_special_token_54|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128060": { + "content": "<|reserved_special_token_55|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128061": { + "content": "<|reserved_special_token_56|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128062": { + "content": "<|reserved_special_token_57|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128063": { + "content": "<|reserved_special_token_58|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128064": { + "content": "<|reserved_special_token_59|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128065": { + "content": "<|reserved_special_token_60|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128066": { + "content": "<|reserved_special_token_61|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128067": { + "content": "<|reserved_special_token_62|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128068": { + "content": "<|reserved_special_token_63|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128069": { + "content": "<|reserved_special_token_64|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128070": { + "content": "<|reserved_special_token_65|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128071": { + "content": "<|reserved_special_token_66|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128072": { + "content": "<|reserved_special_token_67|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128073": { + "content": "<|reserved_special_token_68|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128074": { + "content": "<|reserved_special_token_69|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128075": { + "content": "<|reserved_special_token_70|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128076": { + "content": "<|reserved_special_token_71|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128077": { + "content": "<|reserved_special_token_72|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128078": { + "content": "<|reserved_special_token_73|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128079": { + "content": "<|reserved_special_token_74|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128080": { + "content": "<|reserved_special_token_75|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128081": { + "content": "<|reserved_special_token_76|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128082": { + "content": "<|reserved_special_token_77|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128083": { + "content": "<|reserved_special_token_78|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128084": { + "content": "<|reserved_special_token_79|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128085": { + "content": "<|reserved_special_token_80|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128086": { + "content": "<|reserved_special_token_81|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128087": { + "content": "<|reserved_special_token_82|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128088": { + "content": "<|reserved_special_token_83|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128089": { + "content": "<|reserved_special_token_84|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128090": { + "content": "<|reserved_special_token_85|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128091": { + "content": "<|reserved_special_token_86|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128092": { + "content": "<|reserved_special_token_87|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128093": { + "content": "<|reserved_special_token_88|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128094": { + "content": "<|reserved_special_token_89|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128095": { + "content": "<|reserved_special_token_90|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128096": { + "content": "<|reserved_special_token_91|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128097": { + "content": "<|reserved_special_token_92|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128098": { + "content": "<|reserved_special_token_93|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128099": { + "content": "<|reserved_special_token_94|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128100": { + "content": "<|reserved_special_token_95|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128101": { + "content": "<|reserved_special_token_96|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128102": { + "content": "<|reserved_special_token_97|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128103": { + "content": "<|reserved_special_token_98|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128104": { + "content": "<|reserved_special_token_99|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128105": { + "content": "<|reserved_special_token_100|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128106": { + "content": "<|reserved_special_token_101|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128107": { + "content": "<|reserved_special_token_102|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128108": { + "content": "<|reserved_special_token_103|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128109": { + "content": "<|reserved_special_token_104|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128110": { + "content": "<|reserved_special_token_105|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128111": { + "content": "<|reserved_special_token_106|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128112": { + "content": "<|reserved_special_token_107|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128113": { + "content": "<|reserved_special_token_108|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128114": { + "content": "<|reserved_special_token_109|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128115": { + "content": "<|reserved_special_token_110|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128116": { + "content": "<|reserved_special_token_111|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128117": { + "content": "<|reserved_special_token_112|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128118": { + "content": "<|reserved_special_token_113|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128119": { + "content": "<|reserved_special_token_114|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128120": { + "content": "<|reserved_special_token_115|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128121": { + "content": "<|reserved_special_token_116|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128122": { + "content": "<|reserved_special_token_117|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128123": { + "content": "<|reserved_special_token_118|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128124": { + "content": "<|reserved_special_token_119|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128125": { + "content": "<|reserved_special_token_120|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128126": { + "content": "<|reserved_special_token_121|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128127": { + "content": "<|reserved_special_token_122|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128128": { + "content": "<|reserved_special_token_123|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128129": { + "content": "<|reserved_special_token_124|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128130": { + "content": "<|reserved_special_token_125|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128131": { + "content": "<|reserved_special_token_126|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128132": { + "content": "<|reserved_special_token_127|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128133": { + "content": "<|reserved_special_token_128|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128134": { + "content": "<|reserved_special_token_129|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128135": { + "content": "<|reserved_special_token_130|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128136": { + "content": "<|reserved_special_token_131|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128137": { + "content": "<|reserved_special_token_132|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128138": { + "content": "<|reserved_special_token_133|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128139": { + "content": "<|reserved_special_token_134|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128140": { + "content": "<|reserved_special_token_135|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128141": { + "content": "<|reserved_special_token_136|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128142": { + "content": "<|reserved_special_token_137|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128143": { + "content": "<|reserved_special_token_138|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128144": { + "content": "<|reserved_special_token_139|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128145": { + "content": "<|reserved_special_token_140|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128146": { + "content": "<|reserved_special_token_141|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128147": { + "content": "<|reserved_special_token_142|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128148": { + "content": "<|reserved_special_token_143|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128149": { + "content": "<|reserved_special_token_144|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128150": { + "content": "<|reserved_special_token_145|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128151": { + "content": "<|reserved_special_token_146|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128152": { + "content": "<|reserved_special_token_147|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128153": { + "content": "<|reserved_special_token_148|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128154": { + "content": "<|reserved_special_token_149|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128155": { + "content": "<|reserved_special_token_150|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128156": { + "content": "<|reserved_special_token_151|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128157": { + "content": "<|reserved_special_token_152|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128158": { + "content": "<|reserved_special_token_153|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128159": { + "content": "<|reserved_special_token_154|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128160": { + "content": "<|reserved_special_token_155|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128161": { + "content": "<|reserved_special_token_156|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128162": { + "content": "<|reserved_special_token_157|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128163": { + "content": "<|reserved_special_token_158|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128164": { + "content": "<|reserved_special_token_159|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128165": { + "content": "<|reserved_special_token_160|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128166": { + "content": "<|reserved_special_token_161|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128167": { + "content": "<|reserved_special_token_162|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128168": { + "content": "<|reserved_special_token_163|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128169": { + "content": "<|reserved_special_token_164|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128170": { + "content": "<|reserved_special_token_165|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128171": { + "content": "<|reserved_special_token_166|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128172": { + "content": "<|reserved_special_token_167|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128173": { + "content": "<|reserved_special_token_168|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128174": { + "content": "<|reserved_special_token_169|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128175": { + "content": "<|reserved_special_token_170|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128176": { + "content": "<|reserved_special_token_171|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128177": { + "content": "<|reserved_special_token_172|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128178": { + "content": "<|reserved_special_token_173|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128179": { + "content": "<|reserved_special_token_174|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128180": { + "content": "<|reserved_special_token_175|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128181": { + "content": "<|reserved_special_token_176|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128182": { + "content": "<|reserved_special_token_177|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128183": { + "content": "<|reserved_special_token_178|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128184": { + "content": "<|reserved_special_token_179|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128185": { + "content": "<|reserved_special_token_180|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128186": { + "content": "<|reserved_special_token_181|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128187": { + "content": "<|reserved_special_token_182|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128188": { + "content": "<|reserved_special_token_183|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128189": { + "content": "<|reserved_special_token_184|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128190": { + "content": "<|reserved_special_token_185|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128191": { + "content": "<|reserved_special_token_186|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128192": { + "content": "<|reserved_special_token_187|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128193": { + "content": "<|reserved_special_token_188|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128194": { + "content": "<|reserved_special_token_189|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128195": { + "content": "<|reserved_special_token_190|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128196": { + "content": "<|reserved_special_token_191|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128197": { + "content": "<|reserved_special_token_192|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128198": { + "content": "<|reserved_special_token_193|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128199": { + "content": "<|reserved_special_token_194|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128200": { + "content": "<|reserved_special_token_195|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128201": { + "content": "<|reserved_special_token_196|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128202": { + "content": "<|reserved_special_token_197|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128203": { + "content": "<|reserved_special_token_198|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128204": { + "content": "<|reserved_special_token_199|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128205": { + "content": "<|reserved_special_token_200|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128206": { + "content": "<|reserved_special_token_201|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128207": { + "content": "<|reserved_special_token_202|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128208": { + "content": "<|reserved_special_token_203|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128209": { + "content": "<|reserved_special_token_204|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128210": { + "content": "<|reserved_special_token_205|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128211": { + "content": "<|reserved_special_token_206|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128212": { + "content": "<|reserved_special_token_207|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128213": { + "content": "<|reserved_special_token_208|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128214": { + "content": "<|reserved_special_token_209|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128215": { + "content": "<|reserved_special_token_210|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128216": { + "content": "<|reserved_special_token_211|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128217": { + "content": "<|reserved_special_token_212|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128218": { + "content": "<|reserved_special_token_213|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128219": { + "content": "<|reserved_special_token_214|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128220": { + "content": "<|reserved_special_token_215|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128221": { + "content": "<|reserved_special_token_216|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128222": { + "content": "<|reserved_special_token_217|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128223": { + "content": "<|reserved_special_token_218|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128224": { + "content": "<|reserved_special_token_219|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128225": { + "content": "<|reserved_special_token_220|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128226": { + "content": "<|reserved_special_token_221|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128227": { + "content": "<|reserved_special_token_222|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128228": { + "content": "<|reserved_special_token_223|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128229": { + "content": "<|reserved_special_token_224|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128230": { + "content": "<|reserved_special_token_225|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128231": { + "content": "<|reserved_special_token_226|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128232": { + "content": "<|reserved_special_token_227|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128233": { + "content": "<|reserved_special_token_228|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128234": { + "content": "<|reserved_special_token_229|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128235": { + "content": "<|reserved_special_token_230|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128236": { + "content": "<|reserved_special_token_231|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128237": { + "content": "<|reserved_special_token_232|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128238": { + "content": "<|reserved_special_token_233|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128239": { + "content": "<|reserved_special_token_234|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128240": { + "content": "<|reserved_special_token_235|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128241": { + "content": "<|reserved_special_token_236|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128242": { + "content": "<|reserved_special_token_237|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128243": { + "content": "<|reserved_special_token_238|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128244": { + "content": "<|reserved_special_token_239|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128245": { + "content": "<|reserved_special_token_240|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128246": { + "content": "<|reserved_special_token_241|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128247": { + "content": "<|reserved_special_token_242|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128248": { + "content": "<|reserved_special_token_243|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128249": { + "content": "<|reserved_special_token_244|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128250": { + "content": "<|reserved_special_token_245|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128251": { + "content": "<|reserved_special_token_246|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128252": { + "content": "<|reserved_special_token_247|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128253": { + "content": "<|reserved_special_token_248|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128254": { + "content": "<|reserved_special_token_249|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128255": { + "content": "<|reserved_special_token_250|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + } + }, + "bos_token": "<|begin_of_text|>", + "clean_up_tokenization_spaces": true, + "eos_token": "<|end_of_text|>", + "extra_special_tokens": {}, + "model_input_names": [ + "input_ids", + "attention_mask" + ], + "model_max_length": 32768, + "pad_token": "<|end_of_text|>", + "tokenizer_class": "PreTrainedTokenizerFast" +} diff --git a/checkpoint-1911/trainer_state.json b/checkpoint-1911/trainer_state.json new file mode 100644 index 0000000..d18e848 --- /dev/null +++ b/checkpoint-1911/trainer_state.json @@ -0,0 +1,4132 @@ +{ + "best_global_step": null, + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 1.0, + "eval_steps": 500, + "global_step": 1911, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.0005234231876472127, + "grad_norm": 136.4956817626953, + "learning_rate": 0.0, + "logits/chosen": -0.40118408203125, + "logits/rejected": -0.41802978515625, + "logps/chosen": -297.609375, + "logps/rejected": -247.84375, + "logps/weighted_chosen": -4.50634765625, + "logps/weighted_rejected": -3.43408203125, + "loss": 0.6914, + "rewards/accuracies": 0.0, + "rewards/chosen": 0.0, + "rewards/margins": 0.0, + "rewards/rejected": 0.0, + "rewards/weighted_accuracies": 0.0, + "rewards/weighted_chosen": 0.0, + "rewards/weighted_margins": 0.0, + "rewards/weighted_rejected": 0.0, + "step": 1 + }, + { + "epoch": 0.005234231876472127, + "grad_norm": 269.8279113769531, + "learning_rate": 4.6875e-08, + "logits/chosen": -0.3176456093788147, + "logits/rejected": -0.3530849814414978, + "logps/chosen": -275.5694580078125, + "logps/rejected": -255.875, + "logps/weighted_chosen": -2.478325843811035, + "logps/weighted_rejected": -2.635009765625, + "loss": 0.6911, + "rewards/accuracies": 0.2604166567325592, + "rewards/chosen": -0.0588107630610466, + "rewards/margins": -0.0334201380610466, + "rewards/rejected": -0.025390625, + "rewards/weighted_accuracies": 0.3229166567325592, + "rewards/weighted_chosen": 0.0010899438057094812, + "rewards/weighted_margins": 0.0014837053604424, + "rewards/weighted_rejected": -0.00039333768654614687, + "step": 10 + }, + { + "epoch": 0.010468463752944255, + "grad_norm": 12.950738906860352, + "learning_rate": 9.895833333333332e-08, + "logits/chosen": -0.2977302670478821, + "logits/rejected": -0.3139175474643707, + "logps/chosen": -294.37579345703125, + "logps/rejected": -272.7203063964844, + "logps/weighted_chosen": -2.30804443359375, + "logps/weighted_rejected": -2.247570753097534, + "loss": 0.692, + "rewards/accuracies": 0.296875, + "rewards/chosen": -0.05908203125, + "rewards/margins": -0.01513671875, + "rewards/rejected": -0.0439453125, + "rewards/weighted_accuracies": 0.34687501192092896, + "rewards/weighted_chosen": 7.171630568336695e-05, + "rewards/weighted_margins": 3.662109520519152e-05, + "rewards/weighted_rejected": 3.509521411615424e-05, + "step": 20 + }, + { + "epoch": 0.015702695629416383, + "grad_norm": 65.14871215820312, + "learning_rate": 1.5104166666666664e-07, + "logits/chosen": -0.2917991578578949, + "logits/rejected": -0.3373458981513977, + "logps/chosen": -297.9375, + "logps/rejected": -268.0062561035156, + "logps/weighted_chosen": -2.3156495094299316, + "logps/weighted_rejected": -2.529247999191284, + "loss": 0.693, + "rewards/accuracies": 0.28125, + "rewards/chosen": 0.02177734300494194, + "rewards/margins": -0.06005859375, + "rewards/rejected": 0.08183594048023224, + "rewards/weighted_accuracies": 0.34062498807907104, + "rewards/weighted_chosen": 0.0012060165172442794, + "rewards/weighted_margins": -0.0021120072342455387, + "rewards/weighted_rejected": 0.0033180236350744963, + "step": 30 + }, + { + "epoch": 0.02093692750588851, + "grad_norm": 30.699859619140625, + "learning_rate": 2.03125e-07, + "logits/chosen": -0.301095575094223, + "logits/rejected": -0.3433547914028168, + "logps/chosen": -278.7124938964844, + "logps/rejected": -253.84530639648438, + "logps/weighted_chosen": -2.3508667945861816, + "logps/weighted_rejected": -2.55126953125, + "loss": 0.6909, + "rewards/accuracies": 0.2718749940395355, + "rewards/chosen": -0.014892578125, + "rewards/margins": -0.105224609375, + "rewards/rejected": 0.09033203125, + "rewards/weighted_accuracies": 0.359375, + "rewards/weighted_chosen": 0.0031726837623864412, + "rewards/weighted_margins": 0.0010789871448650956, + "rewards/weighted_rejected": 0.002093696501106024, + "step": 40 + }, + { + "epoch": 0.02617115938236064, + "grad_norm": 15.275498390197754, + "learning_rate": 2.552083333333333e-07, + "logits/chosen": -0.28184205293655396, + "logits/rejected": -0.32042425870895386, + "logps/chosen": -280.19140625, + "logps/rejected": -267.4398498535156, + "logps/weighted_chosen": -2.281604051589966, + "logps/weighted_rejected": -2.278125047683716, + "loss": 0.691, + "rewards/accuracies": 0.3125, + "rewards/chosen": 0.009082031436264515, + "rewards/margins": -0.11552734673023224, + "rewards/rejected": 0.12460937350988388, + "rewards/weighted_accuracies": 0.375, + "rewards/weighted_chosen": 0.0036834715865552425, + "rewards/weighted_margins": 0.0022117614280432463, + "rewards/weighted_rejected": 0.0014717101585119963, + "step": 50 + }, + { + "epoch": 0.031405391258832765, + "grad_norm": 42.36075973510742, + "learning_rate": 3.0729166666666665e-07, + "logits/chosen": -0.3147949278354645, + "logits/rejected": -0.3081672787666321, + "logps/chosen": -277.5492248535156, + "logps/rejected": -261.63201904296875, + "logps/weighted_chosen": -2.6925292015075684, + "logps/weighted_rejected": -2.5440917015075684, + "loss": 0.6888, + "rewards/accuracies": 0.39375001192092896, + "rewards/chosen": 0.15625, + "rewards/margins": 0.06132812425494194, + "rewards/rejected": 0.09492187201976776, + "rewards/weighted_accuracies": 0.43437498807907104, + "rewards/weighted_chosen": 0.0014663696056231856, + "rewards/weighted_margins": 0.0070289610885083675, + "rewards/weighted_rejected": -0.005612182430922985, + "step": 60 + }, + { + "epoch": 0.036639623135304895, + "grad_norm": 70.38953399658203, + "learning_rate": 3.59375e-07, + "logits/chosen": -0.31798094511032104, + "logits/rejected": -0.32500457763671875, + "logps/chosen": -289.6695251464844, + "logps/rejected": -244.7351531982422, + "logps/weighted_chosen": -2.19110107421875, + "logps/weighted_rejected": -2.427661180496216, + "loss": 0.6869, + "rewards/accuracies": 0.4375, + "rewards/chosen": 0.36865234375, + "rewards/margins": 0.34619140625, + "rewards/rejected": 0.0224609375, + "rewards/weighted_accuracies": 0.5375000238418579, + "rewards/weighted_chosen": 0.02108154259622097, + "rewards/weighted_margins": 0.011628913693130016, + "rewards/weighted_rejected": 0.009454727172851562, + "step": 70 + }, + { + "epoch": 0.04187385501177702, + "grad_norm": 54.81600570678711, + "learning_rate": 4.114583333333333e-07, + "logits/chosen": -0.2848251461982727, + "logits/rejected": -0.3061889708042145, + "logps/chosen": -289.08203125, + "logps/rejected": -269.95782470703125, + "logps/weighted_chosen": -2.3393311500549316, + "logps/weighted_rejected": -2.5093994140625, + "loss": 0.6827, + "rewards/accuracies": 0.44999998807907104, + "rewards/chosen": 0.679980456829071, + "rewards/margins": 0.3672851622104645, + "rewards/rejected": 0.31269532442092896, + "rewards/weighted_accuracies": 0.5062500238418579, + "rewards/weighted_chosen": 0.048927437514066696, + "rewards/weighted_margins": 0.025808846578001976, + "rewards/weighted_rejected": 0.02311401441693306, + "step": 80 + }, + { + "epoch": 0.04710808688824915, + "grad_norm": 30.603567123413086, + "learning_rate": 4.6354166666666664e-07, + "logits/chosen": -0.312631219625473, + "logits/rejected": -0.30586013197898865, + "logps/chosen": -279.92578125, + "logps/rejected": -257.5234375, + "logps/weighted_chosen": -2.5583739280700684, + "logps/weighted_rejected": -2.6037840843200684, + "loss": 0.6752, + "rewards/accuracies": 0.49687498807907104, + "rewards/chosen": 0.757617175579071, + "rewards/margins": 0.5347656011581421, + "rewards/rejected": 0.22285155951976776, + "rewards/weighted_accuracies": 0.546875, + "rewards/weighted_chosen": 0.08651771396398544, + "rewards/weighted_margins": 0.05735473707318306, + "rewards/weighted_rejected": 0.02917785570025444, + "step": 90 + }, + { + "epoch": 0.05234231876472128, + "grad_norm": 99.52488708496094, + "learning_rate": 5.156249999999999e-07, + "logits/chosen": -0.2793869078159332, + "logits/rejected": -0.3509994447231293, + "logps/chosen": -290.19061279296875, + "logps/rejected": -285.94842529296875, + "logps/weighted_chosen": -2.013537645339966, + "logps/weighted_rejected": -2.5231690406799316, + "loss": 0.6984, + "rewards/accuracies": 0.512499988079071, + "rewards/chosen": 0.72607421875, + "rewards/margins": 0.80712890625, + "rewards/rejected": -0.0810546875, + "rewards/weighted_accuracies": 0.6187499761581421, + "rewards/weighted_chosen": 0.07289886474609375, + "rewards/weighted_margins": 0.019171524792909622, + "rewards/weighted_rejected": 0.05372200161218643, + "step": 100 + }, + { + "epoch": 0.05757655064119341, + "grad_norm": 55.06239318847656, + "learning_rate": 5.677083333333333e-07, + "logits/chosen": -0.3277145326137543, + "logits/rejected": -0.3189544677734375, + "logps/chosen": -296.17889404296875, + "logps/rejected": -261.51953125, + "logps/weighted_chosen": -2.647631883621216, + "logps/weighted_rejected": -2.6458740234375, + "loss": 0.6844, + "rewards/accuracies": 0.518750011920929, + "rewards/chosen": 0.1376953125, + "rewards/margins": 0.789746105670929, + "rewards/rejected": -0.6522461175918579, + "rewards/weighted_accuracies": 0.565625011920929, + "rewards/weighted_chosen": 0.05432138592004776, + "rewards/weighted_margins": 0.03696594387292862, + "rewards/weighted_rejected": 0.01735992357134819, + "step": 110 + }, + { + "epoch": 0.06281078251766553, + "grad_norm": 42.45036315917969, + "learning_rate": 6.197916666666666e-07, + "logits/chosen": -0.33232802152633667, + "logits/rejected": -0.35865744948387146, + "logps/chosen": -294.6078186035156, + "logps/rejected": -255.01327514648438, + "logps/weighted_chosen": -1.970422387123108, + "logps/weighted_rejected": -2.1345458030700684, + "loss": 0.6837, + "rewards/accuracies": 0.609375, + "rewards/chosen": 0.0013671874767169356, + "rewards/margins": 1.2595703601837158, + "rewards/rejected": -1.258203148841858, + "rewards/weighted_accuracies": 0.59375, + "rewards/weighted_chosen": 0.048947714269161224, + "rewards/weighted_margins": 0.03674011304974556, + "rewards/weighted_rejected": 0.012198830023407936, + "step": 120 + }, + { + "epoch": 0.06804501439413765, + "grad_norm": 28.637327194213867, + "learning_rate": 6.718749999999999e-07, + "logits/chosen": -0.2933105528354645, + "logits/rejected": -0.29020920395851135, + "logps/chosen": -305.86602783203125, + "logps/rejected": -279.09844970703125, + "logps/weighted_chosen": -1.9946777820587158, + "logps/weighted_rejected": -2.2933349609375, + "loss": 0.6754, + "rewards/accuracies": 0.625, + "rewards/chosen": 0.285400390625, + "rewards/margins": 1.7852051258087158, + "rewards/rejected": -1.4998047351837158, + "rewards/weighted_accuracies": 0.6187499761581421, + "rewards/weighted_chosen": 0.06223602220416069, + "rewards/weighted_margins": 0.05499267578125, + "rewards/weighted_rejected": 0.0072036744095385075, + "step": 130 + }, + { + "epoch": 0.07327924627060979, + "grad_norm": 20.866477966308594, + "learning_rate": 7.239583333333333e-07, + "logits/chosen": -0.33005982637405396, + "logits/rejected": -0.34638214111328125, + "logps/chosen": -300.7281188964844, + "logps/rejected": -276.9156188964844, + "logps/weighted_chosen": -2.294506788253784, + "logps/weighted_rejected": -2.561084032058716, + "loss": 0.6606, + "rewards/accuracies": 0.628125011920929, + "rewards/chosen": -1.998437523841858, + "rewards/margins": 2.3173828125, + "rewards/rejected": -4.315820217132568, + "rewards/weighted_accuracies": 0.671875, + "rewards/weighted_chosen": 0.04616241529583931, + "rewards/weighted_margins": 0.08864898979663849, + "rewards/weighted_rejected": -0.04253540188074112, + "step": 140 + }, + { + "epoch": 0.07851347814708191, + "grad_norm": 43.497623443603516, + "learning_rate": 7.760416666666666e-07, + "logits/chosen": -0.2890525758266449, + "logits/rejected": -0.30130767822265625, + "logps/chosen": -287.3109436035156, + "logps/rejected": -256.72967529296875, + "logps/weighted_chosen": -2.883862257003784, + "logps/weighted_rejected": -2.796630859375, + "loss": 0.664, + "rewards/accuracies": 0.628125011920929, + "rewards/chosen": -3.108203172683716, + "rewards/margins": 2.7275390625, + "rewards/rejected": -5.835741996765137, + "rewards/weighted_accuracies": 0.6343749761581421, + "rewards/weighted_chosen": 0.04738159105181694, + "rewards/weighted_margins": 0.09773864597082138, + "rewards/weighted_rejected": -0.05034027248620987, + "step": 150 + }, + { + "epoch": 0.08374771002355404, + "grad_norm": 12.989988327026367, + "learning_rate": 8.28125e-07, + "logits/chosen": -0.3334007263183594, + "logits/rejected": -0.3831237852573395, + "logps/chosen": -308.2320251464844, + "logps/rejected": -281.5335998535156, + "logps/weighted_chosen": -2.378857374191284, + "logps/weighted_rejected": -2.3824095726013184, + "loss": 0.6464, + "rewards/accuracies": 0.6499999761581421, + "rewards/chosen": -4.347460746765137, + "rewards/margins": 3.98388671875, + "rewards/rejected": -8.331347465515137, + "rewards/weighted_accuracies": 0.65625, + "rewards/weighted_chosen": 0.04017028957605362, + "rewards/weighted_margins": 0.15716858208179474, + "rewards/weighted_rejected": -0.11710510402917862, + "step": 160 + }, + { + "epoch": 0.08898194190002617, + "grad_norm": 691.15625, + "learning_rate": 8.802083333333333e-07, + "logits/chosen": -0.37783128023147583, + "logits/rejected": -0.37789613008499146, + "logps/chosen": -310.76953125, + "logps/rejected": -270.8374938964844, + "logps/weighted_chosen": -2.6180419921875, + "logps/weighted_rejected": -2.8051514625549316, + "loss": 0.683, + "rewards/accuracies": 0.6156250238418579, + "rewards/chosen": -7.780859470367432, + "rewards/margins": 4.9013671875, + "rewards/rejected": -12.683398246765137, + "rewards/weighted_accuracies": 0.643750011920929, + "rewards/weighted_chosen": 0.06108245998620987, + "rewards/weighted_margins": 0.17230224609375, + "rewards/weighted_rejected": -0.11125946044921875, + "step": 170 + }, + { + "epoch": 0.0942161737764983, + "grad_norm": 65.56312561035156, + "learning_rate": 9.322916666666666e-07, + "logits/chosen": -0.3433074951171875, + "logits/rejected": -0.363290399312973, + "logps/chosen": -285.8812561035156, + "logps/rejected": -274.94842529296875, + "logps/weighted_chosen": -2.50494384765625, + "logps/weighted_rejected": -2.9246826171875, + "loss": 0.6821, + "rewards/accuracies": 0.6343749761581421, + "rewards/chosen": -12.179101943969727, + "rewards/margins": 6.432421684265137, + "rewards/rejected": -18.608592987060547, + "rewards/weighted_accuracies": 0.59375, + "rewards/weighted_chosen": -0.0216827392578125, + "rewards/weighted_margins": 0.19813232123851776, + "rewards/weighted_rejected": -0.2197723388671875, + "step": 180 + }, + { + "epoch": 0.09945040565297043, + "grad_norm": 26.828750610351562, + "learning_rate": 9.84375e-07, + "logits/chosen": -0.36560744047164917, + "logits/rejected": -0.40380173921585083, + "logps/chosen": -321.97967529296875, + "logps/rejected": -287.5570373535156, + "logps/weighted_chosen": -2.3037109375, + "logps/weighted_rejected": -2.64697265625, + "loss": 0.6489, + "rewards/accuracies": 0.596875011920929, + "rewards/chosen": -16.246679306030273, + "rewards/margins": 7.440039157867432, + "rewards/rejected": -23.685937881469727, + "rewards/weighted_accuracies": 0.640625, + "rewards/weighted_chosen": -0.11536254733800888, + "rewards/weighted_margins": 0.19368895888328552, + "rewards/weighted_rejected": -0.309121698141098, + "step": 190 + }, + { + "epoch": 0.10468463752944256, + "grad_norm": 117.84883880615234, + "learning_rate": 9.99959085414323e-07, + "logits/chosen": -0.4036270081996918, + "logits/rejected": -0.43666380643844604, + "logps/chosen": -327.8265686035156, + "logps/rejected": -284.4234313964844, + "logps/weighted_chosen": -2.6368165016174316, + "logps/weighted_rejected": -3.0156006813049316, + "loss": 0.6383, + "rewards/accuracies": 0.6156250238418579, + "rewards/chosen": -18.538379669189453, + "rewards/margins": 8.212695121765137, + "rewards/rejected": -26.748046875, + "rewards/weighted_accuracies": 0.6468750238418579, + "rewards/weighted_chosen": -0.12391586601734161, + "rewards/weighted_margins": 0.28039854764938354, + "rewards/weighted_rejected": -0.404205322265625, + "step": 200 + }, + { + "epoch": 0.10991886940591468, + "grad_norm": 15.525572776794434, + "learning_rate": 9.997587035630105e-07, + "logits/chosen": -0.41632765531539917, + "logits/rejected": -0.4597106873989105, + "logps/chosen": -308.75469970703125, + "logps/rejected": -316.69842529296875, + "logps/weighted_chosen": -2.439282178878784, + "logps/weighted_rejected": -3.152050733566284, + "loss": 0.6437, + "rewards/accuracies": 0.653124988079071, + "rewards/chosen": -21.632617950439453, + "rewards/margins": 10.560155868530273, + "rewards/rejected": -32.18867111206055, + "rewards/weighted_accuracies": 0.671875, + "rewards/weighted_chosen": -0.2110244780778885, + "rewards/weighted_margins": 0.23976440727710724, + "rewards/weighted_rejected": -0.4508819580078125, + "step": 210 + }, + { + "epoch": 0.11515310128238682, + "grad_norm": 31.583986282348633, + "learning_rate": 9.99391406364405e-07, + "logits/chosen": -0.37994080781936646, + "logits/rejected": -0.384744256734848, + "logps/chosen": -318.63592529296875, + "logps/rejected": -305.0, + "logps/weighted_chosen": -2.9573731422424316, + "logps/weighted_rejected": -3.207348585128784, + "loss": 0.7259, + "rewards/accuracies": 0.606249988079071, + "rewards/chosen": -26.522266387939453, + "rewards/margins": 9.657129287719727, + "rewards/rejected": -36.18437576293945, + "rewards/weighted_accuracies": 0.625, + "rewards/weighted_chosen": -0.3494918942451477, + "rewards/weighted_margins": 0.17980042099952698, + "rewards/weighted_rejected": -0.5295776128768921, + "step": 220 + }, + { + "epoch": 0.12038733315885894, + "grad_norm": 611.2006225585938, + "learning_rate": 9.988573164927884e-07, + "logits/chosen": -0.3254447877407074, + "logits/rejected": -0.36442261934280396, + "logps/chosen": -297.16717529296875, + "logps/rejected": -296.48907470703125, + "logps/weighted_chosen": -2.35888671875, + "logps/weighted_rejected": -2.804126024246216, + "loss": 0.6637, + "rewards/accuracies": 0.6656249761581421, + "rewards/chosen": -28.293750762939453, + "rewards/margins": 16.522851943969727, + "rewards/rejected": -44.817970275878906, + "rewards/weighted_accuracies": 0.6812499761581421, + "rewards/weighted_chosen": -0.1478603333234787, + "rewards/weighted_margins": 0.3882904052734375, + "rewards/weighted_rejected": -0.5360336303710938, + "step": 230 + }, + { + "epoch": 0.12562156503533106, + "grad_norm": 34.738059997558594, + "learning_rate": 9.98156612329838e-07, + "logits/chosen": -0.3876676559448242, + "logits/rejected": -0.4364303648471832, + "logps/chosen": -312.5921936035156, + "logps/rejected": -348.60626220703125, + "logps/weighted_chosen": -2.594775438308716, + "logps/weighted_rejected": -3.1068115234375, + "loss": 0.7482, + "rewards/accuracies": 0.637499988079071, + "rewards/chosen": -47.75507736206055, + "rewards/margins": 18.762109756469727, + "rewards/rejected": -66.5289077758789, + "rewards/weighted_accuracies": 0.6499999761581421, + "rewards/weighted_chosen": -0.319406121969223, + "rewards/weighted_margins": 0.204498291015625, + "rewards/weighted_rejected": -0.5241454839706421, + "step": 240 + }, + { + "epoch": 0.13085579691180318, + "grad_norm": 141.2784423828125, + "learning_rate": 9.97289527905053e-07, + "logits/chosen": -0.4159797728061676, + "logits/rejected": -0.4382568299770355, + "logps/chosen": -308.44842529296875, + "logps/rejected": -313.49530029296875, + "logps/weighted_chosen": -2.802807569503784, + "logps/weighted_rejected": -2.9874024391174316, + "loss": 0.7583, + "rewards/accuracies": 0.609375, + "rewards/chosen": -44.16621017456055, + "rewards/margins": 13.382226943969727, + "rewards/rejected": -57.54296875, + "rewards/weighted_accuracies": 0.6187499761581421, + "rewards/weighted_chosen": -0.20551452040672302, + "rewards/weighted_margins": 0.22150878608226776, + "rewards/weighted_rejected": -0.42730408906936646, + "step": 250 + }, + { + "epoch": 0.1360900287882753, + "grad_norm": 75.63247680664062, + "learning_rate": 9.962563528175875e-07, + "logits/chosen": -0.4156021177768707, + "logits/rejected": -0.4493362307548523, + "logps/chosen": -342.1312561035156, + "logps/rejected": -317.84063720703125, + "logps/weighted_chosen": -2.56585693359375, + "logps/weighted_rejected": -3.4276366233825684, + "loss": 0.6673, + "rewards/accuracies": 0.5625, + "rewards/chosen": -43.44648361206055, + "rewards/margins": 13.459375381469727, + "rewards/rejected": -56.90898513793945, + "rewards/weighted_accuracies": 0.6156250238418579, + "rewards/weighted_chosen": -0.310464471578598, + "rewards/weighted_margins": 0.32973939180374146, + "rewards/weighted_rejected": -0.6404876708984375, + "step": 260 + }, + { + "epoch": 0.14132426066474746, + "grad_norm": 32.991607666015625, + "learning_rate": 9.950574321395277e-07, + "logits/chosen": -0.507153332233429, + "logits/rejected": -0.5264068841934204, + "logps/chosen": -317.0335998535156, + "logps/rejected": -300.5718688964844, + "logps/weighted_chosen": -2.669018507003784, + "logps/weighted_rejected": -2.8846678733825684, + "loss": 0.6774, + "rewards/accuracies": 0.609375, + "rewards/chosen": -27.629491806030273, + "rewards/margins": 9.762109756469727, + "rewards/rejected": -37.39765548706055, + "rewards/weighted_accuracies": 0.6000000238418579, + "rewards/weighted_chosen": -0.23316803574562073, + "rewards/weighted_margins": 0.2321929931640625, + "rewards/weighted_rejected": -0.46527099609375, + "step": 270 + }, + { + "epoch": 0.14655849254121958, + "grad_norm": 41.87471389770508, + "learning_rate": 9.936931663006413e-07, + "logits/chosen": -0.5015045404434204, + "logits/rejected": -0.49525564908981323, + "logps/chosen": -319.9296875, + "logps/rejected": -312.28826904296875, + "logps/weighted_chosen": -2.667797803878784, + "logps/weighted_rejected": -3.121386766433716, + "loss": 0.6458, + "rewards/accuracies": 0.6625000238418579, + "rewards/chosen": -17.077342987060547, + "rewards/margins": 15.751367568969727, + "rewards/rejected": -32.82636642456055, + "rewards/weighted_accuracies": 0.703125, + "rewards/weighted_chosen": -0.10239715874195099, + "rewards/weighted_margins": 0.3067169189453125, + "rewards/weighted_rejected": -0.409027099609375, + "step": 280 + }, + { + "epoch": 0.1517927244176917, + "grad_norm": 22.655261993408203, + "learning_rate": 9.921640109546357e-07, + "logits/chosen": -0.4936012327671051, + "logits/rejected": -0.5640270113945007, + "logps/chosen": -286.4117126464844, + "logps/rejected": -286.15313720703125, + "logps/weighted_chosen": -2.640380859375, + "logps/weighted_rejected": -3.774829149246216, + "loss": 0.6369, + "rewards/accuracies": 0.6343749761581421, + "rewards/chosen": -19.629491806030273, + "rewards/margins": 14.98291015625, + "rewards/rejected": -34.61503982543945, + "rewards/weighted_accuracies": 0.668749988079071, + "rewards/weighted_chosen": -0.21005859971046448, + "rewards/weighted_margins": 0.3854110836982727, + "rewards/weighted_rejected": -0.5955780148506165, + "step": 290 + }, + { + "epoch": 0.15702695629416383, + "grad_norm": 33.83458709716797, + "learning_rate": 9.90470476826975e-07, + "logits/chosen": -0.5687958002090454, + "logits/rejected": -0.5700439214706421, + "logps/chosen": -298.06561279296875, + "logps/rejected": -311.8359375, + "logps/weighted_chosen": -2.555248975753784, + "logps/weighted_rejected": -3.0333251953125, + "loss": 0.6326, + "rewards/accuracies": 0.6625000238418579, + "rewards/chosen": -28.754297256469727, + "rewards/margins": 15.813085556030273, + "rewards/rejected": -44.567771911621094, + "rewards/weighted_accuracies": 0.6468750238418579, + "rewards/weighted_chosen": -0.25576478242874146, + "rewards/weighted_margins": 0.3327178955078125, + "rewards/weighted_rejected": -0.5886474847793579, + "step": 300 + }, + { + "epoch": 0.16226118817063595, + "grad_norm": 29.578954696655273, + "learning_rate": 9.886131295443002e-07, + "logits/chosen": -0.6218292117118835, + "logits/rejected": -0.6808746457099915, + "logps/chosen": -313.44061279296875, + "logps/rejected": -297.9156188964844, + "logps/weighted_chosen": -2.719403028488159, + "logps/weighted_rejected": -3.1142334938049316, + "loss": 0.6221, + "rewards/accuracies": 0.653124988079071, + "rewards/chosen": -32.1875, + "rewards/margins": 15.571874618530273, + "rewards/rejected": -47.7666015625, + "rewards/weighted_accuracies": 0.6499999761581421, + "rewards/weighted_chosen": -0.19769592583179474, + "rewards/weighted_margins": 0.521374523639679, + "rewards/weighted_rejected": -0.7188171148300171, + "step": 310 + }, + { + "epoch": 0.16749542004710807, + "grad_norm": 15.948070526123047, + "learning_rate": 9.865925894455166e-07, + "logits/chosen": -0.6636382937431335, + "logits/rejected": -0.6817939877510071, + "logps/chosen": -336.1507873535156, + "logps/rejected": -306.62890625, + "logps/weighted_chosen": -2.83941650390625, + "logps/weighted_rejected": -3.273486375808716, + "loss": 0.7046, + "rewards/accuracies": 0.637499988079071, + "rewards/chosen": -36.21074295043945, + "rewards/margins": 12.9765625, + "rewards/rejected": -49.17656326293945, + "rewards/weighted_accuracies": 0.6343749761581421, + "rewards/weighted_chosen": -0.3679946959018707, + "rewards/weighted_margins": 0.278738409280777, + "rewards/weighted_rejected": -0.6464904546737671, + "step": 320 + }, + { + "epoch": 0.17272965192358022, + "grad_norm": 53.10931396484375, + "learning_rate": 9.84409531374603e-07, + "logits/chosen": -0.6222923398017883, + "logits/rejected": -0.607867419719696, + "logps/chosen": -344.19061279296875, + "logps/rejected": -315.84686279296875, + "logps/weighted_chosen": -2.915234327316284, + "logps/weighted_rejected": -3.346630811691284, + "loss": 0.6334, + "rewards/accuracies": 0.668749988079071, + "rewards/chosen": -39.39531326293945, + "rewards/margins": 13.606640815734863, + "rewards/rejected": -53.00273513793945, + "rewards/weighted_accuracies": 0.65625, + "rewards/weighted_chosen": -0.33662718534469604, + "rewards/weighted_margins": 0.3510375916957855, + "rewards/weighted_rejected": -0.6879855990409851, + "step": 330 + }, + { + "epoch": 0.17796388380005235, + "grad_norm": 13.994344711303711, + "learning_rate": 9.820646844552219e-07, + "logits/chosen": -0.6615959405899048, + "logits/rejected": -0.714038074016571, + "logps/chosen": -312.7328186035156, + "logps/rejected": -323.3359375, + "logps/weighted_chosen": -2.910736083984375, + "logps/weighted_rejected": -3.2298340797424316, + "loss": 0.6376, + "rewards/accuracies": 0.675000011920929, + "rewards/chosen": -36.327537536621094, + "rewards/margins": 21.465625762939453, + "rewards/rejected": -57.79179763793945, + "rewards/weighted_accuracies": 0.675000011920929, + "rewards/weighted_chosen": -0.34371185302734375, + "rewards/weighted_margins": 0.390716552734375, + "rewards/weighted_rejected": -0.734301745891571, + "step": 340 + }, + { + "epoch": 0.18319811567652447, + "grad_norm": 13.682384490966797, + "learning_rate": 9.795588318471964e-07, + "logits/chosen": -0.774444580078125, + "logits/rejected": -0.7887184023857117, + "logps/chosen": -302.76251220703125, + "logps/rejected": -336.90313720703125, + "logps/weighted_chosen": -2.718432664871216, + "logps/weighted_rejected": -3.0622315406799316, + "loss": 0.6451, + "rewards/accuracies": 0.6343749761581421, + "rewards/chosen": -43.29413986206055, + "rewards/margins": 17.229297637939453, + "rewards/rejected": -60.52812576293945, + "rewards/weighted_accuracies": 0.637499988079071, + "rewards/weighted_chosen": -0.33645325899124146, + "rewards/weighted_margins": 0.33460694551467896, + "rewards/weighted_rejected": -0.6709426641464233, + "step": 350 + }, + { + "epoch": 0.1884323475529966, + "grad_norm": 19.35496711730957, + "learning_rate": 9.768928104849415e-07, + "logits/chosen": -0.815447986125946, + "logits/rejected": -0.8216186761856079, + "logps/chosen": -330.66015625, + "logps/rejected": -313.8687438964844, + "logps/weighted_chosen": -2.970752000808716, + "logps/weighted_rejected": -3.1889405250549316, + "loss": 0.6763, + "rewards/accuracies": 0.621874988079071, + "rewards/chosen": -47.90937423706055, + "rewards/margins": 17.064844131469727, + "rewards/rejected": -64.98515319824219, + "rewards/weighted_accuracies": 0.6499999761581421, + "rewards/weighted_chosen": -0.3897460997104645, + "rewards/weighted_margins": 0.384918212890625, + "rewards/weighted_rejected": -0.7745300531387329, + "step": 360 + }, + { + "epoch": 0.19366657942946872, + "grad_norm": 21.70383071899414, + "learning_rate": 9.740675107979355e-07, + "logits/chosen": -0.7939224243164062, + "logits/rejected": -0.827624499797821, + "logps/chosen": -373.0406188964844, + "logps/rejected": -349.3343811035156, + "logps/weighted_chosen": -2.506054639816284, + "logps/weighted_rejected": -3.43896484375, + "loss": 0.6215, + "rewards/accuracies": 0.596875011920929, + "rewards/chosen": -58.082420349121094, + "rewards/margins": 16.594335556030273, + "rewards/rejected": -74.671875, + "rewards/weighted_accuracies": 0.671875, + "rewards/weighted_chosen": -0.48696595430374146, + "rewards/weighted_margins": 0.412750244140625, + "rewards/weighted_rejected": -0.90020751953125, + "step": 370 + }, + { + "epoch": 0.19890081130594087, + "grad_norm": 21.682790756225586, + "learning_rate": 9.71083876413323e-07, + "logits/chosen": -0.77825927734375, + "logits/rejected": -0.7921401858329773, + "logps/chosen": -367.0062561035156, + "logps/rejected": -355.78594970703125, + "logps/weighted_chosen": -2.644775390625, + "logps/weighted_rejected": -3.5374755859375, + "loss": 0.6609, + "rewards/accuracies": 0.628125011920929, + "rewards/chosen": -64.59492492675781, + "rewards/margins": 21.614843368530273, + "rewards/rejected": -86.1988296508789, + "rewards/weighted_accuracies": 0.6625000238418579, + "rewards/weighted_chosen": -0.5685027837753296, + "rewards/weighted_margins": 0.3586792051792145, + "rewards/weighted_rejected": -0.9269439578056335, + "step": 380 + }, + { + "epoch": 0.204135043182413, + "grad_norm": 30.26679229736328, + "learning_rate": 9.67942903840751e-07, + "logits/chosen": -0.8401764035224915, + "logits/rejected": -0.898754894733429, + "logps/chosen": -362.0328063964844, + "logps/rejected": -362.3687438964844, + "logps/weighted_chosen": -2.8179688453674316, + "logps/weighted_rejected": -3.4515624046325684, + "loss": 0.5992, + "rewards/accuracies": 0.668749988079071, + "rewards/chosen": -57.87226486206055, + "rewards/margins": 30.196680068969727, + "rewards/rejected": -88.0804672241211, + "rewards/weighted_accuracies": 0.715624988079071, + "rewards/weighted_chosen": -0.456369012594223, + "rewards/weighted_margins": 0.495849609375, + "rewards/weighted_rejected": -0.951983630657196, + "step": 390 + }, + { + "epoch": 0.2093692750588851, + "grad_norm": 37.81236267089844, + "learning_rate": 9.646456421395447e-07, + "logits/chosen": -0.8902359008789062, + "logits/rejected": -0.9107666015625, + "logps/chosen": -380.6343688964844, + "logps/rejected": -399.7281188964844, + "logps/weighted_chosen": -2.709228515625, + "logps/weighted_rejected": -3.455004930496216, + "loss": 0.6557, + "rewards/accuracies": 0.6812499761581421, + "rewards/chosen": -62.43437576293945, + "rewards/margins": 33.634765625, + "rewards/rejected": -96.0941390991211, + "rewards/weighted_accuracies": 0.653124988079071, + "rewards/weighted_chosen": -0.4688262939453125, + "rewards/weighted_margins": 0.3546081483364105, + "rewards/weighted_rejected": -0.8232513666152954, + "step": 400 + }, + { + "epoch": 0.21460350693535724, + "grad_norm": 21.761716842651367, + "learning_rate": 9.611931925683266e-07, + "logits/chosen": -0.849993109703064, + "logits/rejected": -0.8886367678642273, + "logps/chosen": -368.8374938964844, + "logps/rejected": -353.73126220703125, + "logps/weighted_chosen": -2.533374071121216, + "logps/weighted_rejected": -3.233935594558716, + "loss": 0.591, + "rewards/accuracies": 0.6656249761581421, + "rewards/chosen": -62.575782775878906, + "rewards/margins": 27.423242568969727, + "rewards/rejected": -90.0132827758789, + "rewards/weighted_accuracies": 0.6937500238418579, + "rewards/weighted_chosen": -0.4288879334926605, + "rewards/weighted_margins": 0.507708728313446, + "rewards/weighted_rejected": -0.9364013671875, + "step": 410 + }, + { + "epoch": 0.21983773881182936, + "grad_norm": 28.842378616333008, + "learning_rate": 9.575867082172085e-07, + "logits/chosen": -0.836225152015686, + "logits/rejected": -0.8854034543037415, + "logps/chosen": -371.90625, + "logps/rejected": -369.4125061035156, + "logps/weighted_chosen": -2.9190917015075684, + "logps/weighted_rejected": -3.118579149246216, + "loss": 0.6198, + "rewards/accuracies": 0.668749988079071, + "rewards/chosen": -66.98905944824219, + "rewards/margins": 32.337890625, + "rewards/rejected": -99.3402328491211, + "rewards/weighted_accuracies": 0.6625000238418579, + "rewards/weighted_chosen": -0.4911743104457855, + "rewards/weighted_margins": 0.5335143804550171, + "rewards/weighted_rejected": -1.0242736339569092, + "step": 420 + }, + { + "epoch": 0.22507197068830148, + "grad_norm": 16.805343627929688, + "learning_rate": 9.538273936226673e-07, + "logits/chosen": -0.8646026849746704, + "logits/rejected": -0.898516833782196, + "logps/chosen": -328.94921875, + "logps/rejected": -353.66717529296875, + "logps/weighted_chosen": -3.1983399391174316, + "logps/weighted_rejected": -3.6509766578674316, + "loss": 0.6173, + "rewards/accuracies": 0.625, + "rewards/chosen": -63.52910232543945, + "rewards/margins": 25.6201171875, + "rewards/rejected": -89.1246109008789, + "rewards/weighted_accuracies": 0.6499999761581421, + "rewards/weighted_chosen": -0.436279296875, + "rewards/weighted_margins": 0.438516229391098, + "rewards/weighted_rejected": -0.8747711181640625, + "step": 430 + }, + { + "epoch": 0.23030620256477363, + "grad_norm": 23.78999137878418, + "learning_rate": 9.499165043652391e-07, + "logits/chosen": -0.8759521245956421, + "logits/rejected": -0.8865814208984375, + "logps/chosen": -358.69140625, + "logps/rejected": -358.76251220703125, + "logps/weighted_chosen": -3.2625732421875, + "logps/weighted_rejected": -3.359301805496216, + "loss": 0.6366, + "rewards/accuracies": 0.6187499761581421, + "rewards/chosen": -67.4326171875, + "rewards/margins": 24.610157012939453, + "rewards/rejected": -92.03788757324219, + "rewards/weighted_accuracies": 0.668749988079071, + "rewards/weighted_chosen": -0.5896576046943665, + "rewards/weighted_margins": 0.34770506620407104, + "rewards/weighted_rejected": -0.9376128911972046, + "step": 440 + }, + { + "epoch": 0.23554043444124576, + "grad_norm": 49.646263122558594, + "learning_rate": 9.458553466501665e-07, + "logits/chosen": -0.9158996343612671, + "logits/rejected": -0.9590820074081421, + "logps/chosen": -355.3109436035156, + "logps/rejected": -341.9765625, + "logps/weighted_chosen": -3.231823682785034, + "logps/weighted_rejected": -3.4568848609924316, + "loss": 0.657, + "rewards/accuracies": 0.6312500238418579, + "rewards/chosen": -69.14453125, + "rewards/margins": 28.884180068969727, + "rewards/rejected": -98.0492172241211, + "rewards/weighted_accuracies": 0.6875, + "rewards/weighted_chosen": -0.6470184326171875, + "rewards/weighted_margins": 0.4533935487270355, + "rewards/weighted_rejected": -1.100683569908142, + "step": 450 + }, + { + "epoch": 0.24077466631771788, + "grad_norm": 38.45325469970703, + "learning_rate": 9.416452768711366e-07, + "logits/chosen": -0.9281005859375, + "logits/rejected": -0.9545837640762329, + "logps/chosen": -375.6390686035156, + "logps/rejected": -370.203125, + "logps/weighted_chosen": -2.9860596656799316, + "logps/weighted_rejected": -3.635498046875, + "loss": 0.6457, + "rewards/accuracies": 0.612500011920929, + "rewards/chosen": -83.3871078491211, + "rewards/margins": 26.872264862060547, + "rewards/rejected": -110.26094055175781, + "rewards/weighted_accuracies": 0.6625000238418579, + "rewards/weighted_chosen": -0.6367782354354858, + "rewards/weighted_margins": 0.515301525592804, + "rewards/weighted_rejected": -1.152197241783142, + "step": 460 + }, + { + "epoch": 0.24600889819419, + "grad_norm": 59.67642593383789, + "learning_rate": 9.372877011572557e-07, + "logits/chosen": -0.8797394037246704, + "logits/rejected": -0.904949963092804, + "logps/chosen": -398.8531188964844, + "logps/rejected": -388.22186279296875, + "logps/weighted_chosen": -3.0103516578674316, + "logps/weighted_rejected": -3.409374952316284, + "loss": 0.6181, + "rewards/accuracies": 0.6031249761581421, + "rewards/chosen": -88.2437515258789, + "rewards/margins": 25.542577743530273, + "rewards/rejected": -113.7890625, + "rewards/weighted_accuracies": 0.6781250238418579, + "rewards/weighted_chosen": -0.6517273187637329, + "rewards/weighted_margins": 0.482818603515625, + "rewards/weighted_rejected": -1.1340011358261108, + "step": 470 + }, + { + "epoch": 0.2512431300706621, + "grad_norm": 27.465656280517578, + "learning_rate": 9.327840749034141e-07, + "logits/chosen": -0.910723865032196, + "logits/rejected": -0.9439147710800171, + "logps/chosen": -364.4750061035156, + "logps/rejected": -392.03436279296875, + "logps/weighted_chosen": -2.8880372047424316, + "logps/weighted_rejected": -4.069311618804932, + "loss": 0.6276, + "rewards/accuracies": 0.6812499761581421, + "rewards/chosen": -77.44355773925781, + "rewards/margins": 38.147850036621094, + "rewards/rejected": -115.59648132324219, + "rewards/weighted_accuracies": 0.684374988079071, + "rewards/weighted_chosen": -0.5768798589706421, + "rewards/weighted_margins": 0.541027843952179, + "rewards/weighted_rejected": -1.1178162097930908, + "step": 480 + }, + { + "epoch": 0.2564773619471343, + "grad_norm": 34.35914993286133, + "learning_rate": 9.281359022841965e-07, + "logits/chosen": -0.80584716796875, + "logits/rejected": -0.8291991949081421, + "logps/chosen": -349.86798095703125, + "logps/rejected": -354.75701904296875, + "logps/weighted_chosen": -2.914257764816284, + "logps/weighted_rejected": -4.178680419921875, + "loss": 0.63, + "rewards/accuracies": 0.625, + "rewards/chosen": -74.94511413574219, + "rewards/margins": 34.957618713378906, + "rewards/rejected": -109.90644836425781, + "rewards/weighted_accuracies": 0.690625011920929, + "rewards/weighted_chosen": -0.564099133014679, + "rewards/weighted_margins": 0.658795177936554, + "rewards/weighted_rejected": -1.2229950428009033, + "step": 490 + }, + { + "epoch": 0.26171159382360637, + "grad_norm": 41.01088333129883, + "learning_rate": 9.233447357514989e-07, + "logits/chosen": -0.775347888469696, + "logits/rejected": -0.8226684331893921, + "logps/chosen": -369.1148376464844, + "logps/rejected": -373.8374938964844, + "logps/weighted_chosen": -3.288012742996216, + "logps/weighted_rejected": -3.7696776390075684, + "loss": 0.6225, + "rewards/accuracies": 0.643750011920929, + "rewards/chosen": -71.1283187866211, + "rewards/margins": 32.867576599121094, + "rewards/rejected": -104.0218734741211, + "rewards/weighted_accuracies": 0.6937500238418579, + "rewards/weighted_chosen": -0.5540069341659546, + "rewards/weighted_margins": 0.598828136920929, + "rewards/weighted_rejected": -1.1535155773162842, + "step": 500 + }, + { + "epoch": 0.26171159382360637, + "eval_logits/chosen": -0.8767929673194885, + "eval_logits/rejected": -0.8968105316162109, + "eval_logps/chosen": -369.21099853515625, + "eval_logps/rejected": -379.3609924316406, + "eval_logps/weighted_chosen": -2.957455635070801, + "eval_logps/weighted_rejected": -3.6658689975738525, + "eval_loss": 0.6115986108779907, + "eval_rewards/accuracies": 0.6294999718666077, + "eval_rewards/chosen": -80.43468475341797, + "eval_rewards/margins": 31.297624588012695, + "eval_rewards/rejected": -111.73987579345703, + "eval_rewards/weighted_accuracies": 0.6759999990463257, + "eval_rewards/weighted_chosen": -0.5706117153167725, + "eval_rewards/weighted_margins": 0.5400464534759521, + "eval_rewards/weighted_rejected": -1.1106581687927246, + "eval_runtime": 1361.4222, + "eval_samples_per_second": 1.469, + "eval_steps_per_second": 0.367, + "step": 500 + }, + { + "epoch": 0.2669458257000785, + "grad_norm": 37.983856201171875, + "learning_rate": 9.184121755160232e-07, + "logits/chosen": -0.8435028195381165, + "logits/rejected": -0.8772613406181335, + "logps/chosen": -378.20782470703125, + "logps/rejected": -406.3187561035156, + "logps/weighted_chosen": -3.217089891433716, + "logps/weighted_rejected": -3.7809815406799316, + "loss": 0.6028, + "rewards/accuracies": 0.6656249761581421, + "rewards/chosen": -77.0746078491211, + "rewards/margins": 40.68242263793945, + "rewards/rejected": -117.759765625, + "rewards/weighted_accuracies": 0.6875, + "rewards/weighted_chosen": -0.5556915402412415, + "rewards/weighted_margins": 0.5749145746231079, + "rewards/weighted_rejected": -1.130804419517517, + "step": 510 + }, + { + "epoch": 0.2721800575765506, + "grad_norm": 46.08055877685547, + "learning_rate": 9.133398690128193e-07, + "logits/chosen": -0.894885241985321, + "logits/rejected": -0.9157882928848267, + "logps/chosen": -405.5679626464844, + "logps/rejected": -415.10076904296875, + "logps/weighted_chosen": -3.0770506858825684, + "logps/weighted_rejected": -3.949780225753784, + "loss": 0.5762, + "rewards/accuracies": 0.65625, + "rewards/chosen": -84.00956726074219, + "rewards/margins": 50.470703125, + "rewards/rejected": -134.46737670898438, + "rewards/weighted_accuracies": 0.7250000238418579, + "rewards/weighted_chosen": -0.49931639432907104, + "rewards/weighted_margins": 0.658154308795929, + "rewards/weighted_rejected": -1.1574280261993408, + "step": 520 + }, + { + "epoch": 0.27741428945302277, + "grad_norm": 40.89746856689453, + "learning_rate": 9.081295103510554e-07, + "logits/chosen": -0.8876678347587585, + "logits/rejected": -0.9430938959121704, + "logps/chosen": -383.52734375, + "logps/rejected": -413.60626220703125, + "logps/weighted_chosen": -2.752880811691284, + "logps/weighted_rejected": -4.01953125, + "loss": 0.5471, + "rewards/accuracies": 0.6468750238418579, + "rewards/chosen": -99.4878921508789, + "rewards/margins": 44.412109375, + "rewards/rejected": -143.859375, + "rewards/weighted_accuracies": 0.706250011920929, + "rewards/weighted_chosen": -0.6086364984512329, + "rewards/weighted_margins": 0.8298507928848267, + "rewards/weighted_rejected": -1.438207983970642, + "step": 530 + }, + { + "epoch": 0.2826485213294949, + "grad_norm": 27.849748611450195, + "learning_rate": 9.027828397481989e-07, + "logits/chosen": -0.8531036376953125, + "logits/rejected": -0.885467529296875, + "logps/chosen": -354.4359436035156, + "logps/rejected": -387.6328125, + "logps/weighted_chosen": -3.355029344558716, + "logps/weighted_rejected": -4.168432712554932, + "loss": 0.5956, + "rewards/accuracies": 0.625, + "rewards/chosen": -93.89375305175781, + "rewards/margins": 38.216407775878906, + "rewards/rejected": -132.11776733398438, + "rewards/weighted_accuracies": 0.675000011920929, + "rewards/weighted_chosen": -0.6640685796737671, + "rewards/weighted_margins": 0.6357482671737671, + "rewards/weighted_rejected": -1.2999756336212158, + "step": 540 + }, + { + "epoch": 0.287882753205967, + "grad_norm": 22.531660079956055, + "learning_rate": 8.973016429487988e-07, + "logits/chosen": -0.8864593505859375, + "logits/rejected": -0.9035919308662415, + "logps/chosen": -375.8890686035156, + "logps/rejected": -396.171875, + "logps/weighted_chosen": -3.2917237281799316, + "logps/weighted_rejected": -3.580859422683716, + "loss": 0.6356, + "rewards/accuracies": 0.6499999761581421, + "rewards/chosen": -94.04902648925781, + "rewards/margins": 47.923240661621094, + "rewards/rejected": -141.9646453857422, + "rewards/weighted_accuracies": 0.71875, + "rewards/weighted_chosen": -0.7129974365234375, + "rewards/weighted_margins": 0.53741455078125, + "rewards/weighted_rejected": -1.2503478527069092, + "step": 550 + }, + { + "epoch": 0.29311698508243916, + "grad_norm": 36.68463897705078, + "learning_rate": 8.916877506280601e-07, + "logits/chosen": -0.9174438714981079, + "logits/rejected": -0.9228302240371704, + "logps/chosen": -387.2203063964844, + "logps/rejected": -405.38592529296875, + "logps/weighted_chosen": -3.4023194313049316, + "logps/weighted_rejected": -3.6703858375549316, + "loss": 0.5995, + "rewards/accuracies": 0.6343749761581421, + "rewards/chosen": -106.43359375, + "rewards/margins": 44.87871170043945, + "rewards/rejected": -151.3136749267578, + "rewards/weighted_accuracies": 0.690625011920929, + "rewards/weighted_chosen": -0.7533203363418579, + "rewards/weighted_margins": 0.627819836139679, + "rewards/weighted_rejected": -1.3810546398162842, + "step": 560 + }, + { + "epoch": 0.29835121695891126, + "grad_norm": 14.758890151977539, + "learning_rate": 8.85943037780415e-07, + "logits/chosen": -0.9987869262695312, + "logits/rejected": -1.0033996105194092, + "logps/chosen": -392.72344970703125, + "logps/rejected": -380.15625, + "logps/weighted_chosen": -3.2074217796325684, + "logps/weighted_rejected": -3.7137207984924316, + "loss": 0.6343, + "rewards/accuracies": 0.609375, + "rewards/chosen": -106.4300765991211, + "rewards/margins": 32.053321838378906, + "rewards/rejected": -138.525390625, + "rewards/weighted_accuracies": 0.7093750238418579, + "rewards/weighted_chosen": -0.744158923625946, + "rewards/weighted_margins": 0.547467052936554, + "rewards/weighted_rejected": -1.291876196861267, + "step": 570 + }, + { + "epoch": 0.3035854488353834, + "grad_norm": 18.140382766723633, + "learning_rate": 8.800694230932884e-07, + "logits/chosen": -0.958880603313446, + "logits/rejected": -0.9720214605331421, + "logps/chosen": -390.57342529296875, + "logps/rejected": -398.5562438964844, + "logps/weighted_chosen": -2.86993408203125, + "logps/weighted_rejected": -3.4458250999450684, + "loss": 0.624, + "rewards/accuracies": 0.6031249761581421, + "rewards/chosen": -100.91777038574219, + "rewards/margins": 32.43359375, + "rewards/rejected": -133.35546875, + "rewards/weighted_accuracies": 0.659375011920929, + "rewards/weighted_chosen": -0.558367908000946, + "rewards/weighted_margins": 0.4340454041957855, + "rewards/weighted_rejected": -0.992688000202179, + "step": 580 + }, + { + "epoch": 0.30881968071185556, + "grad_norm": 16.685937881469727, + "learning_rate": 8.740688683062723e-07, + "logits/chosen": -1.0655395984649658, + "logits/rejected": -1.0949890613555908, + "logps/chosen": -449.0078125, + "logps/rejected": -423.9593811035156, + "logps/weighted_chosen": -2.912792921066284, + "logps/weighted_rejected": -3.656201124191284, + "loss": 0.6344, + "rewards/accuracies": 0.6031249761581421, + "rewards/chosen": -119.9111328125, + "rewards/margins": 28.247461318969727, + "rewards/rejected": -148.1476593017578, + "rewards/weighted_accuracies": 0.706250011920929, + "rewards/weighted_chosen": -0.5903121829032898, + "rewards/weighted_margins": 0.525067150592804, + "rewards/weighted_rejected": -1.1153564453125, + "step": 590 + }, + { + "epoch": 0.31405391258832765, + "grad_norm": 20.500003814697266, + "learning_rate": 8.679433775559215e-07, + "logits/chosen": -1.0546143054962158, + "logits/rejected": -1.1019103527069092, + "logps/chosen": -473.6171875, + "logps/rejected": -477.95001220703125, + "logps/weighted_chosen": -2.896728515625, + "logps/weighted_rejected": -3.985156297683716, + "loss": 0.5741, + "rewards/accuracies": 0.606249988079071, + "rewards/chosen": -148.9021453857422, + "rewards/margins": 43.50390625, + "rewards/rejected": -192.4656219482422, + "rewards/weighted_accuracies": 0.71875, + "rewards/weighted_chosen": -0.7574096918106079, + "rewards/weighted_margins": 0.587969958782196, + "rewards/weighted_rejected": -1.345495581626892, + "step": 600 + }, + { + "epoch": 0.3192881444647998, + "grad_norm": 50.760562896728516, + "learning_rate": 8.616949967063871e-07, + "logits/chosen": -1.007971167564392, + "logits/rejected": -1.0581481456756592, + "logps/chosen": -419.1000061035156, + "logps/rejected": -460.9624938964844, + "logps/weighted_chosen": -3.3387207984924316, + "logps/weighted_rejected": -3.981640577316284, + "loss": 0.5971, + "rewards/accuracies": 0.6468750238418579, + "rewards/chosen": -150.87344360351562, + "rewards/margins": 49.451171875, + "rewards/rejected": -200.21444702148438, + "rewards/weighted_accuracies": 0.671875, + "rewards/weighted_chosen": -0.853253185749054, + "rewards/weighted_margins": 0.644561767578125, + "rewards/weighted_rejected": -1.497961401939392, + "step": 610 + }, + { + "epoch": 0.3245223763412719, + "grad_norm": 16.369630813598633, + "learning_rate": 8.553258126661154e-07, + "logits/chosen": -1.0560119152069092, + "logits/rejected": -1.0674712657928467, + "logps/chosen": -442.5859375, + "logps/rejected": -463.5718688964844, + "logps/weighted_chosen": -3.429705858230591, + "logps/weighted_rejected": -4.259472846984863, + "loss": 0.6405, + "rewards/accuracies": 0.621874988079071, + "rewards/chosen": -162.7121124267578, + "rewards/margins": 44.754295349121094, + "rewards/rejected": -207.43008422851562, + "rewards/weighted_accuracies": 0.684374988079071, + "rewards/weighted_chosen": -0.9004287719726562, + "rewards/weighted_margins": 0.6113220453262329, + "rewards/weighted_rejected": -1.511694312095642, + "step": 620 + }, + { + "epoch": 0.32975660821774405, + "grad_norm": 31.255151748657227, + "learning_rate": 8.488379526908368e-07, + "logits/chosen": -1.0301024913787842, + "logits/rejected": -1.0498230457305908, + "logps/chosen": -465.5859375, + "logps/rejected": -505.28125, + "logps/weighted_chosen": -3.3856444358825684, + "logps/weighted_rejected": -4.028759956359863, + "loss": 0.6384, + "rewards/accuracies": 0.621874988079071, + "rewards/chosen": -174.98046875, + "rewards/margins": 61.357032775878906, + "rewards/rejected": -236.38906860351562, + "rewards/weighted_accuracies": 0.6968749761581421, + "rewards/weighted_chosen": -0.969250500202179, + "rewards/weighted_margins": 0.5812011957168579, + "rewards/weighted_rejected": -1.5499999523162842, + "step": 630 + }, + { + "epoch": 0.33499084009421615, + "grad_norm": 23.673486709594727, + "learning_rate": 8.422335836730802e-07, + "logits/chosen": -1.0135101079940796, + "logits/rejected": -1.0169677734375, + "logps/chosen": -426.2875061035156, + "logps/rejected": -497.64373779296875, + "logps/weighted_chosen": -3.098522901535034, + "logps/weighted_rejected": -3.898510694503784, + "loss": 0.6321, + "rewards/accuracies": 0.637499988079071, + "rewards/chosen": -152.58749389648438, + "rewards/margins": 67.427734375, + "rewards/rejected": -220.0578155517578, + "rewards/weighted_accuracies": 0.699999988079071, + "rewards/weighted_chosen": -0.878277599811554, + "rewards/weighted_margins": 0.590301513671875, + "rewards/weighted_rejected": -1.468664526939392, + "step": 640 + }, + { + "epoch": 0.3402250719706883, + "grad_norm": 20.44074821472168, + "learning_rate": 8.355149114184485e-07, + "logits/chosen": -1.105963110923767, + "logits/rejected": -1.095733642578125, + "logps/chosen": -474.7554626464844, + "logps/rejected": -523.0843505859375, + "logps/weighted_chosen": -3.1939697265625, + "logps/weighted_rejected": -3.786669969558716, + "loss": 0.5634, + "rewards/accuracies": 0.628125011920929, + "rewards/chosen": -167.33828735351562, + "rewards/margins": 73.52656555175781, + "rewards/rejected": -240.85702514648438, + "rewards/weighted_accuracies": 0.668749988079071, + "rewards/weighted_chosen": -0.8172271847724915, + "rewards/weighted_margins": 0.6973236203193665, + "rewards/weighted_rejected": -1.514801025390625, + "step": 650 + }, + { + "epoch": 0.34545930384716045, + "grad_norm": 25.958847045898438, + "learning_rate": 8.286841799088963e-07, + "logits/chosen": -1.1746642589569092, + "logits/rejected": -1.179473876953125, + "logps/chosen": -480.34063720703125, + "logps/rejected": -518.9437255859375, + "logps/weighted_chosen": -2.8995361328125, + "logps/weighted_rejected": -3.656787157058716, + "loss": 0.6073, + "rewards/accuracies": 0.5843750238418579, + "rewards/chosen": -196.3136749267578, + "rewards/margins": 59.513671875, + "rewards/rejected": -255.85897827148438, + "rewards/weighted_accuracies": 0.684374988079071, + "rewards/weighted_chosen": -0.8131744265556335, + "rewards/weighted_margins": 0.576770007610321, + "rewards/weighted_rejected": -1.3900268077850342, + "step": 660 + }, + { + "epoch": 0.35069353572363254, + "grad_norm": 15.336409568786621, + "learning_rate": 8.217436705532599e-07, + "logits/chosen": -1.1510436534881592, + "logits/rejected": -1.173730492591858, + "logps/chosen": -508.78594970703125, + "logps/rejected": -500.79217529296875, + "logps/weighted_chosen": -3.0011963844299316, + "logps/weighted_rejected": -3.7521729469299316, + "loss": 0.578, + "rewards/accuracies": 0.6000000238418579, + "rewards/chosen": -205.71914672851562, + "rewards/margins": 31.244531631469727, + "rewards/rejected": -237.14453125, + "rewards/weighted_accuracies": 0.721875011920929, + "rewards/weighted_chosen": -0.8121887445449829, + "rewards/weighted_margins": 0.688189685344696, + "rewards/weighted_rejected": -1.500115990638733, + "step": 670 + }, + { + "epoch": 0.3559277676001047, + "grad_norm": 19.029943466186523, + "learning_rate": 8.14695701425284e-07, + "logits/chosen": -1.079345703125, + "logits/rejected": -1.125024437904358, + "logps/chosen": -466.38360595703125, + "logps/rejected": -470.4234313964844, + "logps/weighted_chosen": -2.9954833984375, + "logps/weighted_rejected": -3.8232421875, + "loss": 0.5882, + "rewards/accuracies": 0.606249988079071, + "rewards/chosen": -160.8718719482422, + "rewards/margins": 47.43242263793945, + "rewards/rejected": -208.275390625, + "rewards/weighted_accuracies": 0.690625011920929, + "rewards/weighted_chosen": -0.7394653558731079, + "rewards/weighted_margins": 0.627307116985321, + "rewards/weighted_rejected": -1.3663330078125, + "step": 680 + }, + { + "epoch": 0.3611619994765768, + "grad_norm": 13.842121124267578, + "learning_rate": 8.075426264894046e-07, + "logits/chosen": -1.013006567955017, + "logits/rejected": -1.0405089855194092, + "logps/chosen": -472.03436279296875, + "logps/rejected": -501.09375, + "logps/weighted_chosen": -3.1174073219299316, + "logps/weighted_rejected": -4.429980278015137, + "loss": 0.5163, + "rewards/accuracies": 0.6656249761581421, + "rewards/chosen": -166.5642547607422, + "rewards/margins": 56.965431213378906, + "rewards/rejected": -223.5593719482422, + "rewards/weighted_accuracies": 0.778124988079071, + "rewards/weighted_chosen": -0.7782195806503296, + "rewards/weighted_margins": 0.887194812297821, + "rewards/weighted_rejected": -1.665307641029358, + "step": 690 + }, + { + "epoch": 0.36639623135304894, + "grad_norm": 31.751785278320312, + "learning_rate": 8.002868348145435e-07, + "logits/chosen": -1.0329437255859375, + "logits/rejected": -1.0401580333709717, + "logps/chosen": -535.6437377929688, + "logps/rejected": -564.8609619140625, + "logps/weighted_chosen": -3.4560546875, + "logps/weighted_rejected": -3.8817505836486816, + "loss": 0.5956, + "rewards/accuracies": 0.625, + "rewards/chosen": -235.3767547607422, + "rewards/margins": 61.516014099121094, + "rewards/rejected": -296.88360595703125, + "rewards/weighted_accuracies": 0.71875, + "rewards/weighted_chosen": -1.1499541997909546, + "rewards/weighted_margins": 0.6718994379043579, + "rewards/weighted_rejected": -1.8215301036834717, + "step": 700 + }, + { + "epoch": 0.3716304632295211, + "grad_norm": 32.05786895751953, + "learning_rate": 7.92930749776179e-07, + "logits/chosen": -1.0492126941680908, + "logits/rejected": -1.092248558998108, + "logps/chosen": -564.2515869140625, + "logps/rejected": -602.4500122070312, + "logps/weighted_chosen": -3.8183350563049316, + "logps/weighted_rejected": -4.569482326507568, + "loss": 0.652, + "rewards/accuracies": 0.581250011920929, + "rewards/chosen": -292.5419921875, + "rewards/margins": 49.53203201293945, + "rewards/rejected": -342.0179748535156, + "rewards/weighted_accuracies": 0.6656249761581421, + "rewards/weighted_chosen": -1.244970679283142, + "rewards/weighted_margins": 0.630566418170929, + "rewards/weighted_rejected": -1.8747985363006592, + "step": 710 + }, + { + "epoch": 0.3768646951059932, + "grad_norm": 23.8446044921875, + "learning_rate": 7.854768282469582e-07, + "logits/chosen": -1.0707213878631592, + "logits/rejected": -1.126983642578125, + "logps/chosen": -462.90155029296875, + "logps/rejected": -542.5218505859375, + "logps/weighted_chosen": -3.208447217941284, + "logps/weighted_rejected": -4.043383598327637, + "loss": 0.6115, + "rewards/accuracies": 0.6499999761581421, + "rewards/chosen": -185.4597625732422, + "rewards/margins": 79.3746109008789, + "rewards/rejected": -264.8812561035156, + "rewards/weighted_accuracies": 0.6781250238418579, + "rewards/weighted_chosen": -0.8762931823730469, + "rewards/weighted_margins": 0.711090087890625, + "rewards/weighted_rejected": -1.5873870849609375, + "step": 720 + }, + { + "epoch": 0.38209892698246534, + "grad_norm": 17.816680908203125, + "learning_rate": 7.779275597761215e-07, + "logits/chosen": -1.036431908607483, + "logits/rejected": -1.105950951576233, + "logps/chosen": -471.43280029296875, + "logps/rejected": -532.6843872070312, + "logps/weighted_chosen": -3.378674268722534, + "logps/weighted_rejected": -4.096581935882568, + "loss": 0.5442, + "rewards/accuracies": 0.6499999761581421, + "rewards/chosen": -195.43984985351562, + "rewards/margins": 68.92460632324219, + "rewards/rejected": -264.296875, + "rewards/weighted_accuracies": 0.706250011920929, + "rewards/weighted_chosen": -0.847393810749054, + "rewards/weighted_margins": 0.777783215045929, + "rewards/weighted_rejected": -1.625451683998108, + "step": 730 + }, + { + "epoch": 0.38733315885893743, + "grad_norm": 29.750192642211914, + "learning_rate": 7.702854657580126e-07, + "logits/chosen": -1.0928223133087158, + "logits/rejected": -1.1139404773712158, + "logps/chosen": -507.796875, + "logps/rejected": -523.7952880859375, + "logps/weighted_chosen": -3.3760986328125, + "logps/weighted_rejected": -4.447509765625, + "loss": 0.5597, + "rewards/accuracies": 0.6343749761581421, + "rewards/chosen": -213.2277374267578, + "rewards/margins": 56.09648513793945, + "rewards/rejected": -269.4203186035156, + "rewards/weighted_accuracies": 0.7406250238418579, + "rewards/weighted_chosen": -0.927258312702179, + "rewards/weighted_margins": 0.8479980230331421, + "rewards/weighted_rejected": -1.7749145030975342, + "step": 740 + }, + { + "epoch": 0.3925673907354096, + "grad_norm": 38.97298812866211, + "learning_rate": 7.625530985899547e-07, + "logits/chosen": -1.0656921863555908, + "logits/rejected": -1.0953140258789062, + "logps/chosen": -491.4609375, + "logps/rejected": -545.0203247070312, + "logps/weighted_chosen": -3.2592406272888184, + "logps/weighted_rejected": -4.605615139007568, + "loss": 0.5776, + "rewards/accuracies": 0.5874999761581421, + "rewards/chosen": -224.6374969482422, + "rewards/margins": 75.23710632324219, + "rewards/rejected": -300.0316467285156, + "rewards/weighted_accuracies": 0.699999988079071, + "rewards/weighted_chosen": -1.093621850013733, + "rewards/weighted_margins": 0.852429211139679, + "rewards/weighted_rejected": -1.9456908702850342, + "step": 750 + }, + { + "epoch": 0.39780162261188173, + "grad_norm": 22.909940719604492, + "learning_rate": 7.547330408197694e-07, + "logits/chosen": -1.0663268566131592, + "logits/rejected": -1.1003234386444092, + "logps/chosen": -572.404296875, + "logps/rejected": -556.4031372070312, + "logps/weighted_chosen": -3.372363328933716, + "logps/weighted_rejected": -4.400781154632568, + "loss": 0.5943, + "rewards/accuracies": 0.6031249761581421, + "rewards/chosen": -276.7529296875, + "rewards/margins": 23.257421493530273, + "rewards/rejected": -300.1558532714844, + "rewards/weighted_accuracies": 0.703125, + "rewards/weighted_chosen": -1.1045868396759033, + "rewards/weighted_margins": 0.7262054681777954, + "rewards/weighted_rejected": -1.8311798572540283, + "step": 760 + }, + { + "epoch": 0.40303585448835383, + "grad_norm": 107.9198989868164, + "learning_rate": 7.468279042832271e-07, + "logits/chosen": -1.067718505859375, + "logits/rejected": -1.095727562904358, + "logps/chosen": -467.5, + "logps/rejected": -559.7703247070312, + "logps/weighted_chosen": -3.0786986351013184, + "logps/weighted_rejected": -3.8573241233825684, + "loss": 0.5995, + "rewards/accuracies": 0.6499999761581421, + "rewards/chosen": -187.9324188232422, + "rewards/margins": 82.28340148925781, + "rewards/rejected": -270.2611389160156, + "rewards/weighted_accuracies": 0.6968749761581421, + "rewards/weighted_chosen": -0.867828369140625, + "rewards/weighted_margins": 0.625274658203125, + "rewards/weighted_rejected": -1.493688941001892, + "step": 770 + }, + { + "epoch": 0.408270086364826, + "grad_norm": 18.453107833862305, + "learning_rate": 7.388403292317154e-07, + "logits/chosen": -1.0655944347381592, + "logits/rejected": -1.121740698814392, + "logps/chosen": -503.8828125, + "logps/rejected": -519.8375244140625, + "logps/weighted_chosen": -3.039843797683716, + "logps/weighted_rejected": -3.848388671875, + "loss": 0.5822, + "rewards/accuracies": 0.606249988079071, + "rewards/chosen": -202.9210968017578, + "rewards/margins": 50.95390701293945, + "rewards/rejected": -253.869140625, + "rewards/weighted_accuracies": 0.731249988079071, + "rewards/weighted_chosen": -0.873791515827179, + "rewards/weighted_margins": 0.6851806640625, + "rewards/weighted_rejected": -1.5595581531524658, + "step": 780 + }, + { + "epoch": 0.4135043182412981, + "grad_norm": 24.38869285583496, + "learning_rate": 7.307729834504154e-07, + "logits/chosen": -1.090570092201233, + "logits/rejected": -1.1371276378631592, + "logps/chosen": -490.86407470703125, + "logps/rejected": -537.7296752929688, + "logps/weighted_chosen": -3.041918992996216, + "logps/weighted_rejected": -4.070898532867432, + "loss": 0.6125, + "rewards/accuracies": 0.6000000238418579, + "rewards/chosen": -209.13241577148438, + "rewards/margins": 62.44609451293945, + "rewards/rejected": -271.58203125, + "rewards/weighted_accuracies": 0.690625011920929, + "rewards/weighted_chosen": -0.947405993938446, + "rewards/weighted_margins": 0.609375, + "rewards/weighted_rejected": -1.556726098060608, + "step": 790 + }, + { + "epoch": 0.4187385501177702, + "grad_norm": 26.962316513061523, + "learning_rate": 7.226285613672847e-07, + "logits/chosen": -1.03680419921875, + "logits/rejected": -1.0742919445037842, + "logps/chosen": -500.84295654296875, + "logps/rejected": -603.6422119140625, + "logps/weighted_chosen": -3.154052734375, + "logps/weighted_rejected": -4.342724800109863, + "loss": 0.5564, + "rewards/accuracies": 0.671875, + "rewards/chosen": -223.51171875, + "rewards/margins": 101.2464828491211, + "rewards/rejected": -324.720703125, + "rewards/weighted_accuracies": 0.7406250238418579, + "rewards/weighted_chosen": -0.996997058391571, + "rewards/weighted_margins": 0.755480945110321, + "rewards/weighted_rejected": -1.752661108970642, + "step": 800 + }, + { + "epoch": 0.4239727819942423, + "grad_norm": 15.780332565307617, + "learning_rate": 7.144097831531398e-07, + "logits/chosen": -1.0250060558319092, + "logits/rejected": -1.052709937095642, + "logps/chosen": -517.7921752929688, + "logps/rejected": -600.5421752929688, + "logps/weighted_chosen": -3.318652391433716, + "logps/weighted_rejected": -4.39453125, + "loss": 0.5695, + "rewards/accuracies": 0.596875011920929, + "rewards/chosen": -247.40664672851562, + "rewards/margins": 88.87187194824219, + "rewards/rejected": -336.234375, + "rewards/weighted_accuracies": 0.703125, + "rewards/weighted_chosen": -1.082605004310608, + "rewards/weighted_margins": 0.89263916015625, + "rewards/weighted_rejected": -1.974450707435608, + "step": 810 + }, + { + "epoch": 0.42920701387071447, + "grad_norm": 24.66672706604004, + "learning_rate": 7.061193938131396e-07, + "logits/chosen": -0.9916015863418579, + "logits/rejected": -1.042016625404358, + "logps/chosen": -607.3375244140625, + "logps/rejected": -642.5671997070312, + "logps/weighted_chosen": -3.689379930496216, + "logps/weighted_rejected": -4.398633003234863, + "loss": 0.5654, + "rewards/accuracies": 0.5874999761581421, + "rewards/chosen": -307.7925720214844, + "rewards/margins": 69.5777359008789, + "rewards/rejected": -377.3050842285156, + "rewards/weighted_accuracies": 0.7124999761581421, + "rewards/weighted_chosen": -1.1716430187225342, + "rewards/weighted_margins": 0.9023987054824829, + "rewards/weighted_rejected": -2.074475049972534, + "step": 820 + }, + { + "epoch": 0.4344412457471866, + "grad_norm": 20.459196090698242, + "learning_rate": 6.977601622699789e-07, + "logits/chosen": -0.9984985589981079, + "logits/rejected": -1.079248070716858, + "logps/chosen": -471.5625, + "logps/rejected": -574.8695068359375, + "logps/weighted_chosen": -3.233837842941284, + "logps/weighted_rejected": -4.211133003234863, + "loss": 0.5097, + "rewards/accuracies": 0.6656249761581421, + "rewards/chosen": -186.7394561767578, + "rewards/margins": 109.28633117675781, + "rewards/rejected": -296.13555908203125, + "rewards/weighted_accuracies": 0.746874988079071, + "rewards/weighted_chosen": -0.760040283203125, + "rewards/weighted_margins": 0.9420166015625, + "rewards/weighted_rejected": -1.7019164562225342, + "step": 830 + }, + { + "epoch": 0.4396754776236587, + "grad_norm": 89.301025390625, + "learning_rate": 6.893348804390882e-07, + "logits/chosen": -1.126031517982483, + "logits/rejected": -1.1464111804962158, + "logps/chosen": -541.5546875, + "logps/rejected": -580.4109497070312, + "logps/weighted_chosen": -3.6358885765075684, + "logps/weighted_rejected": -4.262304782867432, + "loss": 0.566, + "rewards/accuracies": 0.6000000238418579, + "rewards/chosen": -246.6085968017578, + "rewards/margins": 74.0562515258789, + "rewards/rejected": -320.61328125, + "rewards/weighted_accuracies": 0.684374988079071, + "rewards/weighted_chosen": -1.0077393054962158, + "rewards/weighted_margins": 0.7990967035293579, + "rewards/weighted_rejected": -1.806396484375, + "step": 840 + }, + { + "epoch": 0.44490970950013087, + "grad_norm": 20.6274471282959, + "learning_rate": 6.808463622961578e-07, + "logits/chosen": -1.151123046875, + "logits/rejected": -1.2046630382537842, + "logps/chosen": -621.8390502929688, + "logps/rejected": -716.7249755859375, + "logps/weighted_chosen": -3.6482911109924316, + "logps/weighted_rejected": -4.6015625, + "loss": 0.5759, + "rewards/accuracies": 0.6156250238418579, + "rewards/chosen": -328.931640625, + "rewards/margins": 106.0474624633789, + "rewards/rejected": -435.07403564453125, + "rewards/weighted_accuracies": 0.7124999761581421, + "rewards/weighted_chosen": -1.2104400396347046, + "rewards/weighted_margins": 0.9383910894393921, + "rewards/weighted_rejected": -2.1482911109924316, + "step": 850 + }, + { + "epoch": 0.45014394137660296, + "grad_norm": 20.91545295715332, + "learning_rate": 6.722974429372925e-07, + "logits/chosen": -1.1297729015350342, + "logits/rejected": -1.170617699623108, + "logps/chosen": -697.5484619140625, + "logps/rejected": -730.65625, + "logps/weighted_chosen": -3.5313477516174316, + "logps/weighted_rejected": -5.184765815734863, + "loss": 0.5217, + "rewards/accuracies": 0.5718749761581421, + "rewards/chosen": -393.822265625, + "rewards/margins": 74.92500305175781, + "rewards/rejected": -468.8492126464844, + "rewards/weighted_accuracies": 0.7718750238418579, + "rewards/weighted_chosen": -1.393365502357483, + "rewards/weighted_margins": 1.1208374500274658, + "rewards/weighted_rejected": -2.514172315597534, + "step": 860 + }, + { + "epoch": 0.4553781732530751, + "grad_norm": 42.88610076904297, + "learning_rate": 6.636909776321128e-07, + "logits/chosen": -1.1412475109100342, + "logits/rejected": -1.1436645984649658, + "logps/chosen": -549.9671630859375, + "logps/rejected": -677.8781127929688, + "logps/weighted_chosen": -3.683642625808716, + "logps/weighted_rejected": -4.7724609375, + "loss": 0.5195, + "rewards/accuracies": 0.606249988079071, + "rewards/chosen": -288.21484375, + "rewards/margins": 124.931640625, + "rewards/rejected": -413.17266845703125, + "rewards/weighted_accuracies": 0.7250000238418579, + "rewards/weighted_chosen": -1.2497345209121704, + "rewards/weighted_margins": 0.971630871295929, + "rewards/weighted_rejected": -2.2208008766174316, + "step": 870 + }, + { + "epoch": 0.46061240512954726, + "grad_norm": 29.938108444213867, + "learning_rate": 6.550298408701174e-07, + "logits/chosen": -1.0770995616912842, + "logits/rejected": -1.134680151939392, + "logps/chosen": -574.3234252929688, + "logps/rejected": -675.2101440429688, + "logps/weighted_chosen": -3.807568311691284, + "logps/weighted_rejected": -5.117138862609863, + "loss": 0.5413, + "rewards/accuracies": 0.6312500238418579, + "rewards/chosen": -287.63494873046875, + "rewards/margins": 104.92851257324219, + "rewards/rejected": -392.5396423339844, + "rewards/weighted_accuracies": 0.734375, + "rewards/weighted_chosen": -1.1316711902618408, + "rewards/weighted_margins": 0.954119861125946, + "rewards/weighted_rejected": -2.0862364768981934, + "step": 880 + }, + { + "epoch": 0.46584663700601936, + "grad_norm": 27.375259399414062, + "learning_rate": 6.463169254006276e-07, + "logits/chosen": -1.10406494140625, + "logits/rejected": -1.156030297279358, + "logps/chosen": -541.8601684570312, + "logps/rejected": -578.5452880859375, + "logps/weighted_chosen": -3.4522461891174316, + "logps/weighted_rejected": -4.527783393859863, + "loss": 0.5321, + "rewards/accuracies": 0.612500011920929, + "rewards/chosen": -260.65704345703125, + "rewards/margins": 71.1494140625, + "rewards/rejected": -331.875, + "rewards/weighted_accuracies": 0.7281249761581421, + "rewards/weighted_chosen": -0.977038562297821, + "rewards/weighted_margins": 1.022253394126892, + "rewards/weighted_rejected": -1.999292016029358, + "step": 890 + }, + { + "epoch": 0.4710808688824915, + "grad_norm": 32.590675354003906, + "learning_rate": 6.375551412666326e-07, + "logits/chosen": -1.0722472667694092, + "logits/rejected": -1.1129913330078125, + "logps/chosen": -499.8031311035156, + "logps/rejected": -543.9796752929688, + "logps/weighted_chosen": -3.171948194503784, + "logps/weighted_rejected": -4.273193359375, + "loss": 0.6035, + "rewards/accuracies": 0.590624988079071, + "rewards/chosen": -225.7996063232422, + "rewards/margins": 61.669532775878906, + "rewards/rejected": -287.5015563964844, + "rewards/weighted_accuracies": 0.6812499761581421, + "rewards/weighted_chosen": -1.0042235851287842, + "rewards/weighted_margins": 0.7809082269668579, + "rewards/weighted_rejected": -1.784277319908142, + "step": 900 + }, + { + "epoch": 0.4763151007589636, + "grad_norm": 37.88996887207031, + "learning_rate": 6.287474148328583e-07, + "logits/chosen": -1.0318695306777954, + "logits/rejected": -1.0187164545059204, + "logps/chosen": -491.97186279296875, + "logps/rejected": -540.4202880859375, + "logps/weighted_chosen": -3.5292601585388184, + "logps/weighted_rejected": -5.02734375, + "loss": 0.559, + "rewards/accuracies": 0.606249988079071, + "rewards/chosen": -221.07968139648438, + "rewards/margins": 72.57441711425781, + "rewards/rejected": -293.56561279296875, + "rewards/weighted_accuracies": 0.690625011920929, + "rewards/weighted_chosen": -1.0068145990371704, + "rewards/weighted_margins": 0.84521484375, + "rewards/weighted_rejected": -1.851770043373108, + "step": 910 + }, + { + "epoch": 0.48154933263543576, + "grad_norm": 25.340351104736328, + "learning_rate": 6.198966878083857e-07, + "logits/chosen": -1.0930664539337158, + "logits/rejected": -1.12109375, + "logps/chosen": -540.0, + "logps/rejected": -623.390625, + "logps/weighted_chosen": -3.68231201171875, + "logps/weighted_rejected": -4.464990139007568, + "loss": 0.5911, + "rewards/accuracies": 0.606249988079071, + "rewards/chosen": -268.6884765625, + "rewards/margins": 87.7593765258789, + "rewards/rejected": -356.4453125, + "rewards/weighted_accuracies": 0.7124999761581421, + "rewards/weighted_chosen": -1.085455298423767, + "rewards/weighted_margins": 0.79486083984375, + "rewards/weighted_rejected": -1.880859375, + "step": 920 + }, + { + "epoch": 0.48678356451190785, + "grad_norm": 47.73952865600586, + "learning_rate": 6.110059162641439e-07, + "logits/chosen": -1.1020386219024658, + "logits/rejected": -1.1357238292694092, + "logps/chosen": -533.7390747070312, + "logps/rejected": -599.3781127929688, + "logps/weighted_chosen": -2.9868531227111816, + "logps/weighted_rejected": -4.054858207702637, + "loss": 0.5458, + "rewards/accuracies": 0.625, + "rewards/chosen": -251.13125610351562, + "rewards/margins": 80.9544906616211, + "rewards/rejected": -332.2699279785156, + "rewards/weighted_accuracies": 0.734375, + "rewards/weighted_chosen": -0.8615478277206421, + "rewards/weighted_margins": 0.8154662847518921, + "rewards/weighted_rejected": -1.6761963367462158, + "step": 930 + }, + { + "epoch": 0.49201779638838, + "grad_norm": 36.5290412902832, + "learning_rate": 6.020780696456059e-07, + "logits/chosen": -1.133764624595642, + "logits/rejected": -1.177819848060608, + "logps/chosen": -515.2952880859375, + "logps/rejected": -611.4953002929688, + "logps/weighted_chosen": -3.0075440406799316, + "logps/weighted_rejected": -4.250219821929932, + "loss": 0.5361, + "rewards/accuracies": 0.659375011920929, + "rewards/chosen": -244.7570343017578, + "rewards/margins": 99.73710632324219, + "rewards/rejected": -344.3531188964844, + "rewards/weighted_accuracies": 0.731249988079071, + "rewards/weighted_chosen": -1.0392029285430908, + "rewards/weighted_margins": 0.8944336175918579, + "rewards/weighted_rejected": -1.93341064453125, + "step": 940 + }, + { + "epoch": 0.49725202826485215, + "grad_norm": 36.72229766845703, + "learning_rate": 5.931161297810185e-07, + "logits/chosen": -1.2070038318634033, + "logits/rejected": -1.23931884765625, + "logps/chosen": -550.4031372070312, + "logps/rejected": -615.3375244140625, + "logps/weighted_chosen": -3.748730421066284, + "logps/weighted_rejected": -4.64697265625, + "loss": 0.5726, + "rewards/accuracies": 0.5843750238418579, + "rewards/chosen": -277.07733154296875, + "rewards/margins": 77.06562805175781, + "rewards/rejected": -354.22381591796875, + "rewards/weighted_accuracies": 0.6968749761581421, + "rewards/weighted_chosen": -1.09320068359375, + "rewards/weighted_margins": 0.812207043170929, + "rewards/weighted_rejected": -1.906347632408142, + "step": 950 + }, + { + "epoch": 0.5024862601413242, + "grad_norm": 39.58312225341797, + "learning_rate": 5.841230898854959e-07, + "logits/chosen": -1.1735107898712158, + "logits/rejected": -1.2161743640899658, + "logps/chosen": -684.3687744140625, + "logps/rejected": -794.3922119140625, + "logps/weighted_chosen": -3.8369140625, + "logps/weighted_rejected": -4.986083984375, + "loss": 0.6144, + "rewards/accuracies": 0.612500011920929, + "rewards/chosen": -384.17852783203125, + "rewards/margins": 148.6707000732422, + "rewards/rejected": -532.8191528320312, + "rewards/weighted_accuracies": 0.6937500238418579, + "rewards/weighted_chosen": -1.5656402111053467, + "rewards/weighted_margins": 0.9439452886581421, + "rewards/weighted_rejected": -2.5091795921325684, + "step": 960 + }, + { + "epoch": 0.5077204920177963, + "grad_norm": 24.286752700805664, + "learning_rate": 5.751019535613102e-07, + "logits/chosen": -1.0753357410430908, + "logits/rejected": -1.104547142982483, + "logps/chosen": -550.9773559570312, + "logps/rejected": -653.6687622070312, + "logps/weighted_chosen": -3.5946044921875, + "logps/weighted_rejected": -5.038866996765137, + "loss": 0.5266, + "rewards/accuracies": 0.675000011920929, + "rewards/chosen": -291.25390625, + "rewards/margins": 115.4117202758789, + "rewards/rejected": -406.62811279296875, + "rewards/weighted_accuracies": 0.7093750238418579, + "rewards/weighted_chosen": -1.2581603527069092, + "rewards/weighted_margins": 1.127111792564392, + "rewards/weighted_rejected": -2.3849730491638184, + "step": 970 + }, + { + "epoch": 0.5129547238942685, + "grad_norm": 22.55328941345215, + "learning_rate": 5.660557337947117e-07, + "logits/chosen": -1.0788085460662842, + "logits/rejected": -1.109989881515503, + "logps/chosen": -594.8406372070312, + "logps/rejected": -653.6624755859375, + "logps/weighted_chosen": -3.10791015625, + "logps/weighted_rejected": -4.297143459320068, + "loss": 0.5361, + "rewards/accuracies": 0.628125011920929, + "rewards/chosen": -300.37872314453125, + "rewards/margins": 95.828125, + "rewards/rejected": -396.203125, + "rewards/weighted_accuracies": 0.6968749761581421, + "rewards/weighted_chosen": -1.0415496826171875, + "rewards/weighted_margins": 0.88812255859375, + "rewards/weighted_rejected": -1.929406762123108, + "step": 980 + }, + { + "epoch": 0.5181889557707406, + "grad_norm": 53.25763702392578, + "learning_rate": 5.569874519496174e-07, + "logits/chosen": -1.055578589439392, + "logits/rejected": -1.122521996498108, + "logps/chosen": -497.3999938964844, + "logps/rejected": -579.3656005859375, + "logps/weighted_chosen": -3.393261671066284, + "logps/weighted_rejected": -4.673925876617432, + "loss": 0.5302, + "rewards/accuracies": 0.596875011920929, + "rewards/chosen": -222.11483764648438, + "rewards/margins": 90.296875, + "rewards/rejected": -312.42108154296875, + "rewards/weighted_accuracies": 0.699999988079071, + "rewards/weighted_chosen": -0.8590606451034546, + "rewards/weighted_margins": 0.879437267780304, + "rewards/weighted_rejected": -1.7384154796600342, + "step": 990 + }, + { + "epoch": 0.5234231876472127, + "grad_norm": 39.35697555541992, + "learning_rate": 5.47900136758499e-07, + "logits/chosen": -1.0391693115234375, + "logits/rejected": -1.1183960437774658, + "logps/chosen": -589.5906372070312, + "logps/rejected": -653.09375, + "logps/weighted_chosen": -3.634228467941284, + "logps/weighted_rejected": -4.7723388671875, + "loss": 0.5255, + "rewards/accuracies": 0.59375, + "rewards/chosen": -321.95623779296875, + "rewards/margins": 89.08320617675781, + "rewards/rejected": -410.8277282714844, + "rewards/weighted_accuracies": 0.731249988079071, + "rewards/weighted_chosen": -1.187219262123108, + "rewards/weighted_margins": 0.9879394769668579, + "rewards/weighted_rejected": -2.1756300926208496, + "step": 1000 + }, + { + "epoch": 0.5234231876472127, + "eval_logits/chosen": -1.194671869277954, + "eval_logits/rejected": -1.2200820446014404, + "eval_logps/chosen": -645.4445190429688, + "eval_logps/rejected": -726.552001953125, + "eval_logps/weighted_chosen": -3.7153730392456055, + "eval_logps/weighted_rejected": -4.821405410766602, + "eval_loss": 0.5563206672668457, + "eval_rewards/accuracies": 0.5899999737739563, + "eval_rewards/chosen": -356.71826171875, + "eval_rewards/margins": 102.17925262451172, + "eval_rewards/rejected": -458.8554992675781, + "eval_rewards/weighted_accuracies": 0.7135000228881836, + "eval_rewards/weighted_chosen": -1.3285291194915771, + "eval_rewards/weighted_margins": 0.9376659989356995, + "eval_rewards/weighted_rejected": -2.266195297241211, + "eval_runtime": 1365.9873, + "eval_samples_per_second": 1.464, + "eval_steps_per_second": 0.366, + "step": 1000 + }, + { + "epoch": 0.528657419523685, + "grad_norm": 45.573585510253906, + "learning_rate": 5.387968233108113e-07, + "logits/chosen": -1.087103247642517, + "logits/rejected": -1.085015892982483, + "logps/chosen": -669.9202880859375, + "logps/rejected": -781.1453247070312, + "logps/weighted_chosen": -4.16748046875, + "logps/weighted_rejected": -5.289355278015137, + "loss": 0.5551, + "rewards/accuracies": 0.5843750238418579, + "rewards/chosen": -385.9546813964844, + "rewards/margins": 140.77304077148438, + "rewards/rejected": -526.48046875, + "rewards/weighted_accuracies": 0.746874988079071, + "rewards/weighted_chosen": -1.512841820716858, + "rewards/weighted_margins": 1.02880859375, + "rewards/weighted_rejected": -2.542529344558716, + "step": 1010 + }, + { + "epoch": 0.533891651400157, + "grad_norm": 304.9451904296875, + "learning_rate": 5.296805520392962e-07, + "logits/chosen": -1.160436987876892, + "logits/rejected": -1.2073547840118408, + "logps/chosen": -662.7718505859375, + "logps/rejected": -731.2015380859375, + "logps/weighted_chosen": -3.366748094558716, + "logps/weighted_rejected": -4.887963771820068, + "loss": 0.6053, + "rewards/accuracies": 0.5874999761581421, + "rewards/chosen": -350.32733154296875, + "rewards/margins": 98.02656555175781, + "rewards/rejected": -448.2242126464844, + "rewards/weighted_accuracies": 0.703125, + "rewards/weighted_chosen": -1.51910400390625, + "rewards/weighted_margins": 0.8489990234375, + "rewards/weighted_rejected": -2.3676390647888184, + "step": 1020 + }, + { + "epoch": 0.5391258832766291, + "grad_norm": 186.6497344970703, + "learning_rate": 5.205543677045049e-07, + "logits/chosen": -1.0708801746368408, + "logits/rejected": -1.114501953125, + "logps/chosen": -556.8343505859375, + "logps/rejected": -603.2718505859375, + "logps/weighted_chosen": -3.673095703125, + "logps/weighted_rejected": -4.930810451507568, + "loss": 0.5157, + "rewards/accuracies": 0.606249988079071, + "rewards/chosen": -282.62188720703125, + "rewards/margins": 75.29374694824219, + "rewards/rejected": -357.86053466796875, + "rewards/weighted_accuracies": 0.746874988079071, + "rewards/weighted_chosen": -1.1851806640625, + "rewards/weighted_margins": 1.0668213367462158, + "rewards/weighted_rejected": -2.2524254322052, + "step": 1030 + }, + { + "epoch": 0.5443601151531012, + "grad_norm": 88.95674896240234, + "learning_rate": 5.114213183778697e-07, + "logits/chosen": -1.123449683189392, + "logits/rejected": -1.1691772937774658, + "logps/chosen": -555.5703125, + "logps/rejected": -638.265625, + "logps/weighted_chosen": -4.111181735992432, + "logps/weighted_rejected": -4.94482421875, + "loss": 0.5669, + "rewards/accuracies": 0.625, + "rewards/chosen": -275.11407470703125, + "rewards/margins": 106.10546875, + "rewards/rejected": -381.244140625, + "rewards/weighted_accuracies": 0.7124999761581421, + "rewards/weighted_chosen": -1.225732445716858, + "rewards/weighted_margins": 0.956982433795929, + "rewards/weighted_rejected": -2.182055711746216, + "step": 1040 + }, + { + "epoch": 0.5495943470295734, + "grad_norm": 227.4050750732422, + "learning_rate": 5.022844544236754e-07, + "logits/chosen": -1.054589867591858, + "logits/rejected": -1.088189721107483, + "logps/chosen": -567.5890502929688, + "logps/rejected": -656.4656372070312, + "logps/weighted_chosen": -3.9198241233825684, + "logps/weighted_rejected": -4.956591606140137, + "loss": 0.572, + "rewards/accuracies": 0.5843750238418579, + "rewards/chosen": -287.54962158203125, + "rewards/margins": 113.6656265258789, + "rewards/rejected": -401.166015625, + "rewards/weighted_accuracies": 0.715624988079071, + "rewards/weighted_chosen": -1.267511010169983, + "rewards/weighted_margins": 0.9030395746231079, + "rewards/weighted_rejected": -2.170910596847534, + "step": 1050 + }, + { + "epoch": 0.5548285789060455, + "grad_norm": 1079.8702392578125, + "learning_rate": 4.931468274802608e-07, + "logits/chosen": -1.070989966392517, + "logits/rejected": -1.1126830577850342, + "logps/chosen": -568.2453002929688, + "logps/rejected": -670.5187377929688, + "logps/weighted_chosen": -3.2323241233825684, + "logps/weighted_rejected": -4.560888767242432, + "loss": 0.5383, + "rewards/accuracies": 0.621874988079071, + "rewards/chosen": -285.74139404296875, + "rewards/margins": 123.85664367675781, + "rewards/rejected": -409.6675720214844, + "rewards/weighted_accuracies": 0.7250000238418579, + "rewards/weighted_chosen": -1.198522925376892, + "rewards/weighted_margins": 0.938366711139679, + "rewards/weighted_rejected": -2.136554002761841, + "step": 1060 + }, + { + "epoch": 0.5600628107825176, + "grad_norm": 56.90155792236328, + "learning_rate": 4.840114894407974e-07, + "logits/chosen": -1.10162353515625, + "logits/rejected": -1.1522216796875, + "logps/chosen": -528.5078125, + "logps/rejected": -583.03125, + "logps/weighted_chosen": -3.598315477371216, + "logps/weighted_rejected": -4.284228324890137, + "loss": 0.5674, + "rewards/accuracies": 0.6343749761581421, + "rewards/chosen": -252.3361358642578, + "rewards/margins": 79.1089859008789, + "rewards/rejected": -331.48748779296875, + "rewards/weighted_accuracies": 0.731249988079071, + "rewards/weighted_chosen": -1.137750267982483, + "rewards/weighted_margins": 0.870159924030304, + "rewards/weighted_rejected": -2.008807420730591, + "step": 1070 + }, + { + "epoch": 0.5652970426589898, + "grad_norm": 46.219932556152344, + "learning_rate": 4.748814914339811e-07, + "logits/chosen": -1.0657470226287842, + "logits/rejected": -1.1109619140625, + "logps/chosen": -569.3312377929688, + "logps/rejected": -623.2874755859375, + "logps/weighted_chosen": -3.5493407249450684, + "logps/weighted_rejected": -4.359619140625, + "loss": 0.5449, + "rewards/accuracies": 0.621874988079071, + "rewards/chosen": -284.59649658203125, + "rewards/margins": 83.34257507324219, + "rewards/rejected": -368.00311279296875, + "rewards/weighted_accuracies": 0.715624988079071, + "rewards/weighted_chosen": -1.1415588855743408, + "rewards/weighted_margins": 1.012670874595642, + "rewards/weighted_rejected": -2.1540770530700684, + "step": 1080 + }, + { + "epoch": 0.5705312745354619, + "grad_norm": 86.65474700927734, + "learning_rate": 4.657598828049801e-07, + "logits/chosen": -1.1116211414337158, + "logits/rejected": -1.1939513683319092, + "logps/chosen": -578.01953125, + "logps/rejected": -681.3382568359375, + "logps/weighted_chosen": -3.48370361328125, + "logps/weighted_rejected": -4.3099365234375, + "loss": 0.5293, + "rewards/accuracies": 0.612500011920929, + "rewards/chosen": -287.9331970214844, + "rewards/margins": 112.53828430175781, + "rewards/rejected": -400.390625, + "rewards/weighted_accuracies": 0.7281249761581421, + "rewards/weighted_chosen": -1.1633422374725342, + "rewards/weighted_margins": 0.9815673828125, + "rewards/weighted_rejected": -2.144360303878784, + "step": 1090 + }, + { + "epoch": 0.575765506411934, + "grad_norm": 419.50701904296875, + "learning_rate": 4.566497100969792e-07, + "logits/chosen": -1.1083495616912842, + "logits/rejected": -1.1515319347381592, + "logps/chosen": -681.3523559570312, + "logps/rejected": -761.9328002929688, + "logps/weighted_chosen": -3.8998045921325684, + "logps/weighted_rejected": -4.931884765625, + "loss": 0.5762, + "rewards/accuracies": 0.6625000238418579, + "rewards/chosen": -358.6919860839844, + "rewards/margins": 104.8218765258789, + "rewards/rejected": -463.4956970214844, + "rewards/weighted_accuracies": 0.734375, + "rewards/weighted_chosen": -1.351904273033142, + "rewards/weighted_margins": 1.048761010169983, + "rewards/weighted_rejected": -2.40142822265625, + "step": 1100 + }, + { + "epoch": 0.5809997382884062, + "grad_norm": 56.882720947265625, + "learning_rate": 4.475540160336576e-07, + "logits/chosen": -1.1287415027618408, + "logits/rejected": -1.183984398841858, + "logps/chosen": -593.2249755859375, + "logps/rejected": -657.9328002929688, + "logps/weighted_chosen": -3.99072265625, + "logps/weighted_rejected": -5.168993949890137, + "loss": 0.5249, + "rewards/accuracies": 0.628125011920929, + "rewards/chosen": -294.9878845214844, + "rewards/margins": 104.7953109741211, + "rewards/rejected": -399.80743408203125, + "rewards/weighted_accuracies": 0.7250000238418579, + "rewards/weighted_chosen": -1.250451683998108, + "rewards/weighted_margins": 1.0572631359100342, + "rewards/weighted_rejected": -2.306506395339966, + "step": 1110 + }, + { + "epoch": 0.5862339701648783, + "grad_norm": 633.436279296875, + "learning_rate": 4.3847583850294565e-07, + "logits/chosen": -1.109594702720642, + "logits/rejected": -1.14373779296875, + "logps/chosen": -646.6163940429688, + "logps/rejected": -697.4781494140625, + "logps/weighted_chosen": -4.158276557922363, + "logps/weighted_rejected": -4.803124904632568, + "loss": 0.5622, + "rewards/accuracies": 0.6656249761581421, + "rewards/chosen": -355.50311279296875, + "rewards/margins": 82.48612976074219, + "rewards/rejected": -437.83984375, + "rewards/weighted_accuracies": 0.7406250238418579, + "rewards/weighted_chosen": -1.350854516029358, + "rewards/weighted_margins": 1.000707983970642, + "rewards/weighted_rejected": -2.352795362472534, + "step": 1120 + }, + { + "epoch": 0.5914682020413504, + "grad_norm": 512.849853515625, + "learning_rate": 4.294182095423934e-07, + "logits/chosen": -1.0823333263397217, + "logits/rejected": -1.1519775390625, + "logps/chosen": -601.3250122070312, + "logps/rejected": -686.8515625, + "logps/weighted_chosen": -3.522167921066284, + "logps/weighted_rejected": -4.769873142242432, + "loss": 0.5542, + "rewards/accuracies": 0.590624988079071, + "rewards/chosen": -315.3882751464844, + "rewards/margins": 110.39375305175781, + "rewards/rejected": -425.638671875, + "rewards/weighted_accuracies": 0.703125, + "rewards/weighted_chosen": -1.290283203125, + "rewards/weighted_margins": 1.020105004310608, + "rewards/weighted_rejected": -2.3109498023986816, + "step": 1130 + }, + { + "epoch": 0.5967024339178225, + "grad_norm": 65.05388641357422, + "learning_rate": 4.20384154326496e-07, + "logits/chosen": -1.0805175304412842, + "logits/rejected": -1.14312744140625, + "logps/chosen": -516.7843627929688, + "logps/rejected": -553.4617309570312, + "logps/weighted_chosen": -3.43231201171875, + "logps/weighted_rejected": -4.573815822601318, + "loss": 0.5682, + "rewards/accuracies": 0.6187499761581421, + "rewards/chosen": -256.2328186035156, + "rewards/margins": 61.6025390625, + "rewards/rejected": -317.79412841796875, + "rewards/weighted_accuracies": 0.7124999761581421, + "rewards/weighted_chosen": -1.086877465248108, + "rewards/weighted_margins": 0.8528991937637329, + "rewards/weighted_rejected": -1.9411132335662842, + "step": 1140 + }, + { + "epoch": 0.6019366657942947, + "grad_norm": 75.07366943359375, + "learning_rate": 4.1137669015630863e-07, + "logits/chosen": -1.077880859375, + "logits/rejected": -1.1405029296875, + "logps/chosen": -537.2109375, + "logps/rejected": -624.4749755859375, + "logps/weighted_chosen": -3.0843262672424316, + "logps/weighted_rejected": -4.2431640625, + "loss": 0.5219, + "rewards/accuracies": 0.675000011920929, + "rewards/chosen": -241.1707000732422, + "rewards/margins": 99.6703109741211, + "rewards/rejected": -340.8062438964844, + "rewards/weighted_accuracies": 0.7124999761581421, + "rewards/weighted_chosen": -0.9316955804824829, + "rewards/weighted_margins": 0.957165539264679, + "rewards/weighted_rejected": -1.889074683189392, + "step": 1150 + }, + { + "epoch": 0.6071708976707668, + "grad_norm": 53.81990432739258, + "learning_rate": 4.023988254516943e-07, + "logits/chosen": -1.059973120689392, + "logits/rejected": -1.120214819908142, + "logps/chosen": -530.0851440429688, + "logps/rejected": -566.734375, + "logps/weighted_chosen": -3.7030029296875, + "logps/weighted_rejected": -4.251660346984863, + "loss": 0.5091, + "rewards/accuracies": 0.653124988079071, + "rewards/chosen": -229.01113891601562, + "rewards/margins": 70.63789367675781, + "rewards/rejected": -299.630859375, + "rewards/weighted_accuracies": 0.75, + "rewards/weighted_chosen": -0.8327713012695312, + "rewards/weighted_margins": 0.8830932378768921, + "rewards/weighted_rejected": -1.715429663658142, + "step": 1160 + }, + { + "epoch": 0.6124051295472389, + "grad_norm": 50.65428161621094, + "learning_rate": 3.9345355874653366e-07, + "logits/chosen": -1.058905005455017, + "logits/rejected": -1.0953247547149658, + "logps/chosen": -535.8616943359375, + "logps/rejected": -544.9773559570312, + "logps/weighted_chosen": -3.3579344749450684, + "logps/weighted_rejected": -4.205273628234863, + "loss": 0.5906, + "rewards/accuracies": 0.609375, + "rewards/chosen": -238.26016235351562, + "rewards/margins": 56.61601638793945, + "rewards/rejected": -294.7398376464844, + "rewards/weighted_accuracies": 0.6968749761581421, + "rewards/weighted_chosen": -0.988323986530304, + "rewards/weighted_margins": 0.763989269733429, + "rewards/weighted_rejected": -1.7522430419921875, + "step": 1170 + }, + { + "epoch": 0.6176393614237111, + "grad_norm": 44.16736602783203, + "learning_rate": 3.8454387768724157e-07, + "logits/chosen": -1.097143530845642, + "logits/rejected": -1.1111571788787842, + "logps/chosen": -475.8023376464844, + "logps/rejected": -494.05078125, + "logps/weighted_chosen": -3.5390625, + "logps/weighted_rejected": -4.489306449890137, + "loss": 0.5337, + "rewards/accuracies": 0.578125, + "rewards/chosen": -202.70858764648438, + "rewards/margins": 60.640235900878906, + "rewards/rejected": -263.4339904785156, + "rewards/weighted_accuracies": 0.7124999761581421, + "rewards/weighted_chosen": -0.8571746945381165, + "rewards/weighted_margins": 0.9136596918106079, + "rewards/weighted_rejected": -1.7705810070037842, + "step": 1180 + }, + { + "epoch": 0.6228735933001832, + "grad_norm": 402.7128601074219, + "learning_rate": 3.7567275803491525e-07, + "logits/chosen": -1.1082000732421875, + "logits/rejected": -1.155554175376892, + "logps/chosen": -559.3796997070312, + "logps/rejected": -586.2781372070312, + "logps/weighted_chosen": -3.2266845703125, + "logps/weighted_rejected": -4.550146579742432, + "loss": 0.5018, + "rewards/accuracies": 0.6312500238418579, + "rewards/chosen": -257.08673095703125, + "rewards/margins": 75.61640930175781, + "rewards/rejected": -332.78204345703125, + "rewards/weighted_accuracies": 0.746874988079071, + "rewards/weighted_chosen": -0.8839172124862671, + "rewards/weighted_margins": 1.0030639171600342, + "rewards/weighted_rejected": -1.886468529701233, + "step": 1190 + }, + { + "epoch": 0.6281078251766553, + "grad_norm": 393.51788330078125, + "learning_rate": 3.66843162671456e-07, + "logits/chosen": -1.1154296398162842, + "logits/rejected": -1.1315186023712158, + "logps/chosen": -552.2093505859375, + "logps/rejected": -680.3843994140625, + "logps/weighted_chosen": -3.9300293922424316, + "logps/weighted_rejected": -4.456152439117432, + "loss": 0.5959, + "rewards/accuracies": 0.6031249761581421, + "rewards/chosen": -278.14764404296875, + "rewards/margins": 130.2062530517578, + "rewards/rejected": -408.24530029296875, + "rewards/weighted_accuracies": 0.671875, + "rewards/weighted_chosen": -1.1631286144256592, + "rewards/weighted_margins": 0.8199828863143921, + "rewards/weighted_rejected": -1.9827392101287842, + "step": 1200 + }, + { + "epoch": 0.6333420570531274, + "grad_norm": 915.16943359375, + "learning_rate": 3.5805804060998924e-07, + "logits/chosen": -1.1075561046600342, + "logits/rejected": -1.1484863758087158, + "logps/chosen": -600.2609252929688, + "logps/rejected": -717.1453247070312, + "logps/weighted_chosen": -3.2332520484924316, + "logps/weighted_rejected": -4.710791110992432, + "loss": 0.499, + "rewards/accuracies": 0.668749988079071, + "rewards/chosen": -299.421875, + "rewards/margins": 147.1335906982422, + "rewards/rejected": -446.451171875, + "rewards/weighted_accuracies": 0.75, + "rewards/weighted_chosen": -1.162207007408142, + "rewards/weighted_margins": 1.1646606922149658, + "rewards/weighted_rejected": -2.326342821121216, + "step": 1210 + }, + { + "epoch": 0.6385762889295996, + "grad_norm": 685.1804809570312, + "learning_rate": 3.493203260099197e-07, + "logits/chosen": -1.1248352527618408, + "logits/rejected": -1.195556640625, + "logps/chosen": -622.3953247070312, + "logps/rejected": -694.5250244140625, + "logps/weighted_chosen": -3.321972608566284, + "logps/weighted_rejected": -4.685571193695068, + "loss": 0.5497, + "rewards/accuracies": 0.581250011920929, + "rewards/chosen": -334.587890625, + "rewards/margins": 87.31562805175781, + "rewards/rejected": -421.9072265625, + "rewards/weighted_accuracies": 0.706250011920929, + "rewards/weighted_chosen": -1.299523949623108, + "rewards/weighted_margins": 0.9338439702987671, + "rewards/weighted_rejected": -2.2326292991638184, + "step": 1220 + }, + { + "epoch": 0.6438105208060717, + "grad_norm": 740.0750732421875, + "learning_rate": 3.4063293719694407e-07, + "logits/chosen": -1.126245141029358, + "logits/rejected": -1.180517554283142, + "logps/chosen": -567.4812622070312, + "logps/rejected": -648.1539306640625, + "logps/weighted_chosen": -3.639111280441284, + "logps/weighted_rejected": -4.597058296203613, + "loss": 0.6, + "rewards/accuracies": 0.6312500238418579, + "rewards/chosen": -290.95684814453125, + "rewards/margins": 99.4990234375, + "rewards/rejected": -390.62872314453125, + "rewards/weighted_accuracies": 0.6812499761581421, + "rewards/weighted_chosen": -1.1769592761993408, + "rewards/weighted_margins": 0.8424926996231079, + "rewards/weighted_rejected": -2.0197691917419434, + "step": 1230 + }, + { + "epoch": 0.6490447526825438, + "grad_norm": 342.7829895019531, + "learning_rate": 3.319987756883559e-07, + "logits/chosen": -1.156945824623108, + "logits/rejected": -1.198333740234375, + "logps/chosen": -572.3328247070312, + "logps/rejected": -671.7633056640625, + "logps/weighted_chosen": -3.338793992996216, + "logps/weighted_rejected": -4.542822360992432, + "loss": 0.5046, + "rewards/accuracies": 0.65625, + "rewards/chosen": -294.57342529296875, + "rewards/margins": 108.07109069824219, + "rewards/rejected": -402.66680908203125, + "rewards/weighted_accuracies": 0.746874988079071, + "rewards/weighted_chosen": -1.004339575767517, + "rewards/weighted_margins": 1.094323754310608, + "rewards/weighted_rejected": -2.0980224609375, + "step": 1240 + }, + { + "epoch": 0.654278984559016, + "grad_norm": 406.2123107910156, + "learning_rate": 3.234207252239607e-07, + "logits/chosen": -1.1353332996368408, + "logits/rejected": -1.184594750404358, + "logps/chosen": -643.1171875, + "logps/rejected": -680.65625, + "logps/weighted_chosen": -3.9437499046325684, + "logps/weighted_rejected": -4.764257907867432, + "loss": 0.5738, + "rewards/accuracies": 0.6000000238418579, + "rewards/chosen": -350.52264404296875, + "rewards/margins": 77.30488586425781, + "rewards/rejected": -428.01641845703125, + "rewards/weighted_accuracies": 0.675000011920929, + "rewards/weighted_chosen": -1.3293838500976562, + "rewards/weighted_margins": 0.863757312297821, + "rewards/weighted_rejected": -2.193603515625, + "step": 1250 + }, + { + "epoch": 0.6595132164354881, + "grad_norm": 518.19775390625, + "learning_rate": 3.1490165080293175e-07, + "logits/chosen": -1.1296905279159546, + "logits/rejected": -1.1927978992462158, + "logps/chosen": -573.6226806640625, + "logps/rejected": -675.9562377929688, + "logps/weighted_chosen": -3.474902391433716, + "logps/weighted_rejected": -4.43798828125, + "loss": 0.5317, + "rewards/accuracies": 0.590624988079071, + "rewards/chosen": -306.15155029296875, + "rewards/margins": 106.2289047241211, + "rewards/rejected": -412.40234375, + "rewards/weighted_accuracies": 0.7124999761581421, + "rewards/weighted_chosen": -1.2220337390899658, + "rewards/weighted_margins": 0.9764038324356079, + "rewards/weighted_rejected": -2.197509765625, + "step": 1260 + }, + { + "epoch": 0.6647474483119602, + "grad_norm": 152.93478393554688, + "learning_rate": 3.06444397726922e-07, + "logits/chosen": -1.082611083984375, + "logits/rejected": -1.1652100086212158, + "logps/chosen": -609.8375244140625, + "logps/rejected": -677.9749755859375, + "logps/weighted_chosen": -3.4163575172424316, + "logps/weighted_rejected": -5.1240234375, + "loss": 0.5052, + "rewards/accuracies": 0.6187499761581421, + "rewards/chosen": -314.2923889160156, + "rewards/margins": 101.6371078491211, + "rewards/rejected": -415.8968811035156, + "rewards/weighted_accuracies": 0.71875, + "rewards/weighted_chosen": -1.135400414466858, + "rewards/weighted_margins": 1.23980712890625, + "rewards/weighted_rejected": -2.375744581222534, + "step": 1270 + }, + { + "epoch": 0.6699816801884323, + "grad_norm": 285.5711669921875, + "learning_rate": 2.980517906497586e-07, + "logits/chosen": -1.1207764148712158, + "logits/rejected": -1.1819336414337158, + "logps/chosen": -597.5718994140625, + "logps/rejected": -703.9953002929688, + "logps/weighted_chosen": -3.632373094558716, + "logps/weighted_rejected": -4.974999904632568, + "loss": 0.5807, + "rewards/accuracies": 0.6312500238418579, + "rewards/chosen": -304.43048095703125, + "rewards/margins": 127.4535140991211, + "rewards/rejected": -431.8851623535156, + "rewards/weighted_accuracies": 0.721875011920929, + "rewards/weighted_chosen": -1.202392578125, + "rewards/weighted_margins": 0.9914306402206421, + "rewards/weighted_rejected": -2.193042039871216, + "step": 1280 + }, + { + "epoch": 0.6752159120649045, + "grad_norm": 82.95504760742188, + "learning_rate": 2.89726632634029e-07, + "logits/chosen": -1.113470435142517, + "logits/rejected": -1.1529052257537842, + "logps/chosen": -599.0562744140625, + "logps/rejected": -677.3453369140625, + "logps/weighted_chosen": -3.367846727371216, + "logps/weighted_rejected": -4.500195503234863, + "loss": 0.574, + "rewards/accuracies": 0.590624988079071, + "rewards/chosen": -324.748046875, + "rewards/margins": 97.44921875, + "rewards/rejected": -422.11798095703125, + "rewards/weighted_accuracies": 0.6937500238418579, + "rewards/weighted_chosen": -1.2255462408065796, + "rewards/weighted_margins": 0.875323474407196, + "rewards/weighted_rejected": -2.1007933616638184, + "step": 1290 + }, + { + "epoch": 0.6804501439413766, + "grad_norm": 2586.47119140625, + "learning_rate": 2.814717042148827e-07, + "logits/chosen": -1.131799340248108, + "logits/rejected": -1.1629149913787842, + "logps/chosen": -566.9031372070312, + "logps/rejected": -659.9281005859375, + "logps/weighted_chosen": -4.212353706359863, + "logps/weighted_rejected": -4.885107517242432, + "loss": 0.606, + "rewards/accuracies": 0.5874999761581421, + "rewards/chosen": -292.85858154296875, + "rewards/margins": 106.29667663574219, + "rewards/rejected": -399.13201904296875, + "rewards/weighted_accuracies": 0.684374988079071, + "rewards/weighted_chosen": -1.2742125988006592, + "rewards/weighted_margins": 0.8754211664199829, + "rewards/weighted_rejected": -2.149639844894409, + "step": 1300 + }, + { + "epoch": 0.6856843758178487, + "grad_norm": 20.005109786987305, + "learning_rate": 2.7328976247135416e-07, + "logits/chosen": -1.156762719154358, + "logits/rejected": -1.1973998546600342, + "logps/chosen": -577.4781494140625, + "logps/rejected": -642.1609497070312, + "logps/weighted_chosen": -3.5555663108825684, + "logps/weighted_rejected": -4.547216892242432, + "loss": 0.6037, + "rewards/accuracies": 0.6187499761581421, + "rewards/chosen": -287.97772216796875, + "rewards/margins": 101.68867492675781, + "rewards/rejected": -389.75079345703125, + "rewards/weighted_accuracies": 0.690625011920929, + "rewards/weighted_chosen": -1.2163574695587158, + "rewards/weighted_margins": 0.806256115436554, + "rewards/weighted_rejected": -2.022289991378784, + "step": 1310 + }, + { + "epoch": 0.6909186076943209, + "grad_norm": 151.74427795410156, + "learning_rate": 2.651835401055217e-07, + "logits/chosen": -1.149621605873108, + "logits/rejected": -1.183782935142517, + "logps/chosen": -597.9265747070312, + "logps/rejected": -678.625, + "logps/weighted_chosen": -3.490771532058716, + "logps/weighted_rejected": -4.412158012390137, + "loss": 0.5855, + "rewards/accuracies": 0.609375, + "rewards/chosen": -320.44451904296875, + "rewards/margins": 92.2328109741211, + "rewards/rejected": -412.76953125, + "rewards/weighted_accuracies": 0.7124999761581421, + "rewards/weighted_chosen": -1.213354468345642, + "rewards/weighted_margins": 0.8388611078262329, + "rewards/weighted_rejected": -2.0516357421875, + "step": 1320 + }, + { + "epoch": 0.696152839570793, + "grad_norm": 47.340484619140625, + "learning_rate": 2.571557445298055e-07, + "logits/chosen": -1.135412573814392, + "logits/rejected": -1.1872680187225342, + "logps/chosen": -558.5109252929688, + "logps/rejected": -635.0374755859375, + "logps/weighted_chosen": -3.464648485183716, + "logps/weighted_rejected": -4.4168701171875, + "loss": 0.5666, + "rewards/accuracies": 0.625, + "rewards/chosen": -280.0796813964844, + "rewards/margins": 90.732421875, + "rewards/rejected": -370.8531188964844, + "rewards/weighted_accuracies": 0.675000011920929, + "rewards/weighted_chosen": -1.1060791015625, + "rewards/weighted_margins": 0.821246325969696, + "rewards/weighted_rejected": -1.9268798828125, + "step": 1330 + }, + { + "epoch": 0.7013870714472651, + "grad_norm": 49.26417922973633, + "learning_rate": 2.49209056962716e-07, + "logits/chosen": -1.146185278892517, + "logits/rejected": -1.186279296875, + "logps/chosen": -612.96875, + "logps/rejected": -650.0765380859375, + "logps/weighted_chosen": -3.656054735183716, + "logps/weighted_rejected": -4.784960746765137, + "loss": 0.5627, + "rewards/accuracies": 0.6031249761581421, + "rewards/chosen": -298.53436279296875, + "rewards/margins": 84.52030944824219, + "rewards/rejected": -383.0699157714844, + "rewards/weighted_accuracies": 0.706250011920929, + "rewards/weighted_chosen": -1.108679175376892, + "rewards/weighted_margins": 0.8104248046875, + "rewards/weighted_rejected": -1.9180908203125, + "step": 1340 + }, + { + "epoch": 0.7066213033237373, + "grad_norm": 43.23517990112305, + "learning_rate": 2.41346131533347e-07, + "logits/chosen": -1.1938354969024658, + "logits/rejected": -1.217199683189392, + "logps/chosen": -628.8203125, + "logps/rejected": -686.6749877929688, + "logps/weighted_chosen": -3.1535401344299316, + "logps/weighted_rejected": -4.464306831359863, + "loss": 0.546, + "rewards/accuracies": 0.625, + "rewards/chosen": -322.18829345703125, + "rewards/margins": 89.7154312133789, + "rewards/rejected": -411.8656311035156, + "rewards/weighted_accuracies": 0.737500011920929, + "rewards/weighted_chosen": -1.133204698562622, + "rewards/weighted_margins": 0.804534912109375, + "rewards/weighted_rejected": -1.9378173351287842, + "step": 1350 + }, + { + "epoch": 0.7118555352002094, + "grad_norm": 34.37676239013672, + "learning_rate": 2.3356959439491898e-07, + "logits/chosen": -1.1357300281524658, + "logits/rejected": -1.2062256336212158, + "logps/chosen": -595.59375, + "logps/rejected": -672.6422119140625, + "logps/weighted_chosen": -3.9969725608825684, + "logps/weighted_rejected": -4.689795017242432, + "loss": 0.5808, + "rewards/accuracies": 0.625, + "rewards/chosen": -325.7347717285156, + "rewards/margins": 104.75312805175781, + "rewards/rejected": -430.67108154296875, + "rewards/weighted_accuracies": 0.699999988079071, + "rewards/weighted_chosen": -1.2155777215957642, + "rewards/weighted_margins": 0.9368530511856079, + "rewards/weighted_rejected": -2.152294874191284, + "step": 1360 + }, + { + "epoch": 0.7170897670766815, + "grad_norm": 51.87490463256836, + "learning_rate": 2.258820428476645e-07, + "logits/chosen": -1.1657593250274658, + "logits/rejected": -1.2217528820037842, + "logps/chosen": -631.8890380859375, + "logps/rejected": -746.7828369140625, + "logps/weighted_chosen": -3.470263719558716, + "logps/weighted_rejected": -4.346484184265137, + "loss": 0.5501, + "rewards/accuracies": 0.612500011920929, + "rewards/chosen": -352.91015625, + "rewards/margins": 124.40898132324219, + "rewards/rejected": -477.31951904296875, + "rewards/weighted_accuracies": 0.734375, + "rewards/weighted_chosen": -1.186987280845642, + "rewards/weighted_margins": 0.836499035358429, + "rewards/weighted_rejected": -2.023852586746216, + "step": 1370 + }, + { + "epoch": 0.7223239989531536, + "grad_norm": 416.21136474609375, + "learning_rate": 2.1828604447135245e-07, + "logits/chosen": -1.11541748046875, + "logits/rejected": -1.1730468273162842, + "logps/chosen": -694.0234375, + "logps/rejected": -737.5452880859375, + "logps/weighted_chosen": -3.9507813453674316, + "logps/weighted_rejected": -5.173535346984863, + "loss": 0.5413, + "rewards/accuracies": 0.5874999761581421, + "rewards/chosen": -423.78436279296875, + "rewards/margins": 56.84648513793945, + "rewards/rejected": -480.90936279296875, + "rewards/weighted_accuracies": 0.6875, + "rewards/weighted_chosen": -1.29058837890625, + "rewards/weighted_margins": 0.98492431640625, + "rewards/weighted_rejected": -2.275256395339966, + "step": 1380 + }, + { + "epoch": 0.7275582308296258, + "grad_norm": 68.63409423828125, + "learning_rate": 2.1078413626773545e-07, + "logits/chosen": -1.174198865890503, + "logits/rejected": -1.2014648914337158, + "logps/chosen": -640.5343627929688, + "logps/rejected": -758.4734497070312, + "logps/weighted_chosen": -3.5299744606018066, + "logps/weighted_rejected": -5.219091892242432, + "loss": 0.5755, + "rewards/accuracies": 0.6031249761581421, + "rewards/chosen": -360.80780029296875, + "rewards/margins": 128.4132843017578, + "rewards/rejected": -489.3277282714844, + "rewards/weighted_accuracies": 0.715624988079071, + "rewards/weighted_chosen": -1.248020887374878, + "rewards/weighted_margins": 0.9331451654434204, + "rewards/weighted_rejected": -2.181262254714966, + "step": 1390 + }, + { + "epoch": 0.7327924627060979, + "grad_norm": 323.93890380859375, + "learning_rate": 2.0337882381321347e-07, + "logits/chosen": -1.1679198741912842, + "logits/rejected": -1.1945312023162842, + "logps/chosen": -653.7203369140625, + "logps/rejected": -707.4640502929688, + "logps/weighted_chosen": -3.5541749000549316, + "logps/weighted_rejected": -4.433495998382568, + "loss": 0.5551, + "rewards/accuracies": 0.612500011920929, + "rewards/chosen": -362.28399658203125, + "rewards/margins": 91.1234359741211, + "rewards/rejected": -453.5562438964844, + "rewards/weighted_accuracies": 0.7093750238418579, + "rewards/weighted_chosen": -1.336692452430725, + "rewards/weighted_margins": 0.8907715082168579, + "rewards/weighted_rejected": -2.227795362472534, + "step": 1400 + }, + { + "epoch": 0.73802669458257, + "grad_norm": 74.13733673095703, + "learning_rate": 1.960725804219905e-07, + "logits/chosen": -1.149194359779358, + "logits/rejected": -1.2033812999725342, + "logps/chosen": -661.9656372070312, + "logps/rejected": -768.3421630859375, + "logps/weighted_chosen": -4.091479301452637, + "logps/weighted_rejected": -4.242919921875, + "loss": 0.5754, + "rewards/accuracies": 0.5718749761581421, + "rewards/chosen": -362.45428466796875, + "rewards/margins": 122.5601577758789, + "rewards/rejected": -484.8101501464844, + "rewards/weighted_accuracies": 0.6937500238418579, + "rewards/weighted_chosen": -1.2423064708709717, + "rewards/weighted_margins": 0.8276306390762329, + "rewards/weighted_rejected": -2.069873094558716, + "step": 1410 + }, + { + "epoch": 0.7432609264590422, + "grad_norm": 237.26011657714844, + "learning_rate": 1.8886784632000824e-07, + "logits/chosen": -1.1620361804962158, + "logits/rejected": -1.206933617591858, + "logps/chosen": -640.3624877929688, + "logps/rejected": -783.7437744140625, + "logps/weighted_chosen": -3.4447264671325684, + "logps/weighted_rejected": -4.929491996765137, + "loss": 0.5398, + "rewards/accuracies": 0.65625, + "rewards/chosen": -354.3374938964844, + "rewards/margins": 162.05233764648438, + "rewards/rejected": -516.314453125, + "rewards/weighted_accuracies": 0.7250000238418579, + "rewards/weighted_chosen": -1.1836121082305908, + "rewards/weighted_margins": 1.072717308998108, + "rewards/weighted_rejected": -2.256579637527466, + "step": 1420 + }, + { + "epoch": 0.7484951583355143, + "grad_norm": 482.7161560058594, + "learning_rate": 1.8176702782993025e-07, + "logits/chosen": -1.166711449623108, + "logits/rejected": -1.185546875, + "logps/chosen": -614.01953125, + "logps/rejected": -700.3484497070312, + "logps/weighted_chosen": -3.3925538063049316, + "logps/weighted_rejected": -4.749658107757568, + "loss": 0.5436, + "rewards/accuracies": 0.6031249761581421, + "rewards/chosen": -352.78125, + "rewards/margins": 93.58476257324219, + "rewards/rejected": -446.29296875, + "rewards/weighted_accuracies": 0.706250011920929, + "rewards/weighted_chosen": -1.2850220203399658, + "rewards/weighted_margins": 0.9663940668106079, + "rewards/weighted_rejected": -2.2514891624450684, + "step": 1430 + }, + { + "epoch": 0.7537293902119864, + "grad_norm": 48.64420700073242, + "learning_rate": 1.7477249656745034e-07, + "logits/chosen": -1.1055176258087158, + "logits/rejected": -1.157617211341858, + "logps/chosen": -559.9140625, + "logps/rejected": -613.4656372070312, + "logps/weighted_chosen": -3.6603026390075684, + "logps/weighted_rejected": -4.916943550109863, + "loss": 0.5473, + "rewards/accuracies": 0.5874999761581421, + "rewards/chosen": -309.0234375, + "rewards/margins": 77.943359375, + "rewards/rejected": -387.1015625, + "rewards/weighted_accuracies": 0.731249988079071, + "rewards/weighted_chosen": -1.15887451171875, + "rewards/weighted_margins": 1.0418579578399658, + "rewards/weighted_rejected": -2.2011961936950684, + "step": 1440 + }, + { + "epoch": 0.7589636220884585, + "grad_norm": 61.97262954711914, + "learning_rate": 1.6788658864919118e-07, + "logits/chosen": -1.111364722251892, + "logits/rejected": -1.2080078125, + "logps/chosen": -681.9781494140625, + "logps/rejected": -766.1875, + "logps/weighted_chosen": -3.5382080078125, + "logps/weighted_rejected": -4.320166110992432, + "loss": 0.5212, + "rewards/accuracies": 0.625, + "rewards/chosen": -367.3251953125, + "rewards/margins": 115.8160171508789, + "rewards/rejected": -482.9632873535156, + "rewards/weighted_accuracies": 0.715624988079071, + "rewards/weighted_chosen": -1.1493957042694092, + "rewards/weighted_margins": 1.0002562999725342, + "rewards/weighted_rejected": -2.149792432785034, + "step": 1450 + }, + { + "epoch": 0.7641978539649307, + "grad_norm": 33.67172622680664, + "learning_rate": 1.611116039124613e-07, + "logits/chosen": -1.0988738536834717, + "logits/rejected": -1.160314917564392, + "logps/chosen": -592.9827880859375, + "logps/rejected": -634.5718994140625, + "logps/weighted_chosen": -3.8994140625, + "logps/weighted_rejected": -4.845922946929932, + "loss": 0.5548, + "rewards/accuracies": 0.609375, + "rewards/chosen": -330.28790283203125, + "rewards/margins": 74.87812805175781, + "rewards/rejected": -405.23126220703125, + "rewards/weighted_accuracies": 0.6812499761581421, + "rewards/weighted_chosen": -1.151556372642517, + "rewards/weighted_margins": 0.903613269329071, + "rewards/weighted_rejected": -2.055248975753784, + "step": 1460 + }, + { + "epoch": 0.7694320858414028, + "grad_norm": 144.79293823242188, + "learning_rate": 1.5444980514712723e-07, + "logits/chosen": -1.196630835533142, + "logits/rejected": -1.222875952720642, + "logps/chosen": -668.71875, + "logps/rejected": -797.6375122070312, + "logps/weighted_chosen": -3.661572217941284, + "logps/weighted_rejected": -4.466870307922363, + "loss": 0.6131, + "rewards/accuracies": 0.621874988079071, + "rewards/chosen": -357.33673095703125, + "rewards/margins": 134.50155639648438, + "rewards/rejected": -491.7562561035156, + "rewards/weighted_accuracies": 0.6781250238418579, + "rewards/weighted_chosen": -1.314672827720642, + "rewards/weighted_margins": 0.78485107421875, + "rewards/weighted_rejected": -2.099261522293091, + "step": 1470 + }, + { + "epoch": 0.7746663177178749, + "grad_norm": 40.60256576538086, + "learning_rate": 1.4790341733986083e-07, + "logits/chosen": -1.150146484375, + "logits/rejected": -1.1887328624725342, + "logps/chosen": -620.7406005859375, + "logps/rejected": -698.5499877929688, + "logps/weighted_chosen": -3.9625487327575684, + "logps/weighted_rejected": -4.390429496765137, + "loss": 0.5719, + "rewards/accuracies": 0.6000000238418579, + "rewards/chosen": -333.4839782714844, + "rewards/margins": 102.4716796875, + "rewards/rejected": -436.12420654296875, + "rewards/weighted_accuracies": 0.7093750238418579, + "rewards/weighted_chosen": -1.234228491783142, + "rewards/weighted_margins": 0.851452648639679, + "rewards/weighted_rejected": -2.085034132003784, + "step": 1480 + }, + { + "epoch": 0.7799005495943471, + "grad_norm": 217.96365356445312, + "learning_rate": 1.4147462693101108e-07, + "logits/chosen": -1.131872534751892, + "logits/rejected": -1.179223656654358, + "logps/chosen": -645.7999877929688, + "logps/rejected": -766.2390747070312, + "logps/weighted_chosen": -3.494384765625, + "logps/weighted_rejected": -4.757568359375, + "loss": 0.5302, + "rewards/accuracies": 0.640625, + "rewards/chosen": -358.8677673339844, + "rewards/margins": 136.17617797851562, + "rewards/rejected": -495.09765625, + "rewards/weighted_accuracies": 0.734375, + "rewards/weighted_chosen": -1.171356201171875, + "rewards/weighted_margins": 1.06304931640625, + "rewards/weighted_rejected": -2.235211133956909, + "step": 1490 + }, + { + "epoch": 0.7851347814708192, + "grad_norm": 655.3441162109375, + "learning_rate": 1.3516558108435177e-07, + "logits/chosen": -1.133544921875, + "logits/rejected": -1.154302954673767, + "logps/chosen": -594.9124755859375, + "logps/rejected": -729.9781494140625, + "logps/weighted_chosen": -3.306835889816284, + "logps/weighted_rejected": -4.992211818695068, + "loss": 0.508, + "rewards/accuracies": 0.606249988079071, + "rewards/chosen": -315.57147216796875, + "rewards/margins": 149.4695281982422, + "rewards/rejected": -465.1382751464844, + "rewards/weighted_accuracies": 0.737500011920929, + "rewards/weighted_chosen": -1.0454895496368408, + "rewards/weighted_margins": 1.0130126476287842, + "rewards/weighted_rejected": -2.058490037918091, + "step": 1500 + }, + { + "epoch": 0.7851347814708192, + "eval_logits/chosen": -1.2306523323059082, + "eval_logits/rejected": -1.26123046875, + "eval_logps/chosen": -633.6630249023438, + "eval_logps/rejected": -713.4450073242188, + "eval_logps/weighted_chosen": -3.545902967453003, + "eval_logps/weighted_rejected": -4.605466842651367, + "eval_loss": 0.5521173477172852, + "eval_rewards/accuracies": 0.593999981880188, + "eval_rewards/chosen": -344.9159851074219, + "eval_rewards/margins": 100.7760009765625, + "eval_rewards/rejected": -445.6925048828125, + "eval_rewards/weighted_accuracies": 0.7149999737739563, + "eval_rewards/weighted_chosen": -1.159058928489685, + "eval_rewards/weighted_margins": 0.8911977410316467, + "eval_rewards/weighted_rejected": -2.0502567291259766, + "eval_runtime": 1316.3607, + "eval_samples_per_second": 1.519, + "eval_steps_per_second": 0.38, + "step": 1500 + }, + { + "epoch": 0.7903690133472913, + "grad_norm": 32.490413665771484, + "learning_rate": 1.2897838696994505e-07, + "logits/chosen": -1.1297607421875, + "logits/rejected": -1.168298363685608, + "logps/chosen": -585.390625, + "logps/rejected": -678.625, + "logps/weighted_chosen": -3.9292969703674316, + "logps/weighted_rejected": -4.704247951507568, + "loss": 0.5992, + "rewards/accuracies": 0.671875, + "rewards/chosen": -313.0980529785156, + "rewards/margins": 103.3773422241211, + "rewards/rejected": -416.39923095703125, + "rewards/weighted_accuracies": 0.715624988079071, + "rewards/weighted_chosen": -1.228363037109375, + "rewards/weighted_margins": 0.8051391839981079, + "rewards/weighted_rejected": -2.0337891578674316, + "step": 1510 + }, + { + "epoch": 0.7956032452237635, + "grad_norm": 103.49887084960938, + "learning_rate": 1.229151110603649e-07, + "logits/chosen": -1.1730270385742188, + "logits/rejected": -1.2250702381134033, + "logps/chosen": -639.2672119140625, + "logps/rejected": -710.4703369140625, + "logps/weighted_chosen": -3.5760254859924316, + "logps/weighted_rejected": -4.558545112609863, + "loss": 0.6109, + "rewards/accuracies": 0.6343749761581421, + "rewards/chosen": -342.1875, + "rewards/margins": 95.5367202758789, + "rewards/rejected": -437.6800842285156, + "rewards/weighted_accuracies": 0.721875011920929, + "rewards/weighted_chosen": -1.242523193359375, + "rewards/weighted_margins": 0.7712646722793579, + "rewards/weighted_rejected": -2.013842821121216, + "step": 1520 + }, + { + "epoch": 0.8008374771002356, + "grad_norm": 26.20205307006836, + "learning_rate": 1.1697777844051104e-07, + "logits/chosen": -1.1935913562774658, + "logits/rejected": -1.216577172279358, + "logps/chosen": -663.4562377929688, + "logps/rejected": -768.6937255859375, + "logps/weighted_chosen": -3.4478759765625, + "logps/weighted_rejected": -4.542138576507568, + "loss": 0.5215, + "rewards/accuracies": 0.653124988079071, + "rewards/chosen": -360.3539123535156, + "rewards/margins": 126.4839859008789, + "rewards/rejected": -486.7699279785156, + "rewards/weighted_accuracies": 0.7281249761581421, + "rewards/weighted_chosen": -1.057836890220642, + "rewards/weighted_margins": 0.954821765422821, + "rewards/weighted_rejected": -2.012655735015869, + "step": 1530 + }, + { + "epoch": 0.8060717089767077, + "grad_norm": 158.38113403320312, + "learning_rate": 1.111683721312477e-07, + "logits/chosen": -1.1692626476287842, + "logits/rejected": -1.199652075767517, + "logps/chosen": -622.1984252929688, + "logps/rejected": -711.6531372070312, + "logps/weighted_chosen": -3.277587890625, + "logps/weighted_rejected": -3.949267625808716, + "loss": 0.5668, + "rewards/accuracies": 0.6031249761581421, + "rewards/chosen": -346.7880859375, + "rewards/margins": 104.9453125, + "rewards/rejected": -451.7046813964844, + "rewards/weighted_accuracies": 0.706250011920929, + "rewards/weighted_chosen": -1.1675598621368408, + "rewards/weighted_margins": 0.843762218952179, + "rewards/weighted_rejected": -2.0111327171325684, + "step": 1540 + }, + { + "epoch": 0.8113059408531798, + "grad_norm": 189.9469757080078, + "learning_rate": 1.0548883242709033e-07, + "logits/chosen": -1.106359839439392, + "logits/rejected": -1.178131103515625, + "logps/chosen": -629.1640625, + "logps/rejected": -776.9000244140625, + "logps/weighted_chosen": -4.056591987609863, + "logps/weighted_rejected": -4.928466796875, + "loss": 0.4908, + "rewards/accuracies": 0.6781250238418579, + "rewards/chosen": -344.5941467285156, + "rewards/margins": 162.8507843017578, + "rewards/rejected": -507.3597717285156, + "rewards/weighted_accuracies": 0.7718750238418579, + "rewards/weighted_chosen": -1.1156005859375, + "rewards/weighted_margins": 1.0612304210662842, + "rewards/weighted_rejected": -2.17724609375, + "step": 1550 + }, + { + "epoch": 0.816540172729652, + "grad_norm": 301.8493957519531, + "learning_rate": 9.994105624816379e-08, + "logits/chosen": -1.164636254310608, + "logits/rejected": -1.2321898937225342, + "logps/chosen": -670.3968505859375, + "logps/rejected": -756.6937255859375, + "logps/weighted_chosen": -3.8463377952575684, + "logps/weighted_rejected": -5.091845512390137, + "loss": 0.5505, + "rewards/accuracies": 0.6187499761581421, + "rewards/chosen": -368.4644470214844, + "rewards/margins": 105.59648132324219, + "rewards/rejected": -473.984375, + "rewards/weighted_accuracies": 0.6968749761581421, + "rewards/weighted_chosen": -1.300726294517517, + "rewards/weighted_margins": 0.9333862066268921, + "rewards/weighted_rejected": -2.2337646484375, + "step": 1560 + }, + { + "epoch": 0.821774404606124, + "grad_norm": 212.7559051513672, + "learning_rate": 9.452689650664514e-08, + "logits/chosen": -1.166418433189392, + "logits/rejected": -1.244348168373108, + "logps/chosen": -682.6515502929688, + "logps/rejected": -709.1500244140625, + "logps/weighted_chosen": -3.459155321121216, + "logps/weighted_rejected": -4.23291015625, + "loss": 0.5405, + "rewards/accuracies": 0.590624988079071, + "rewards/chosen": -386.58514404296875, + "rewards/margins": 66.9886703491211, + "rewards/rejected": -453.6781311035156, + "rewards/weighted_accuracies": 0.7250000238418579, + "rewards/weighted_chosen": -1.2503662109375, + "rewards/weighted_margins": 0.901043713092804, + "rewards/weighted_rejected": -2.1515135765075684, + "step": 1570 + }, + { + "epoch": 0.8270086364825961, + "grad_norm": 280.51715087890625, + "learning_rate": 8.924816148790748e-08, + "logits/chosen": -1.174096703529358, + "logits/rejected": -1.233984351158142, + "logps/chosen": -679.6077880859375, + "logps/rejected": -780.9749755859375, + "logps/weighted_chosen": -3.776562452316284, + "logps/weighted_rejected": -4.682421684265137, + "loss": 0.5356, + "rewards/accuracies": 0.606249988079071, + "rewards/chosen": -385.0078125, + "rewards/margins": 117.716796875, + "rewards/rejected": -502.8359375, + "rewards/weighted_accuracies": 0.699999988079071, + "rewards/weighted_chosen": -1.1958191394805908, + "rewards/weighted_margins": 1.0562187433242798, + "rewards/weighted_rejected": -2.251452684402466, + "step": 1580 + }, + { + "epoch": 0.8322428683590684, + "grad_norm": 39.72044372558594, + "learning_rate": 8.410661424656607e-08, + "logits/chosen": -1.203576683998108, + "logits/rejected": -1.232690453529358, + "logps/chosen": -671.4718627929688, + "logps/rejected": -761.9312744140625, + "logps/weighted_chosen": -3.552441358566284, + "logps/weighted_rejected": -4.740527153015137, + "loss": 0.5263, + "rewards/accuracies": 0.606249988079071, + "rewards/chosen": -377.9898376464844, + "rewards/margins": 124.10469055175781, + "rewards/rejected": -501.8984375, + "rewards/weighted_accuracies": 0.731249988079071, + "rewards/weighted_chosen": -1.308996558189392, + "rewards/weighted_margins": 1.106896996498108, + "rewards/weighted_rejected": -2.4161620140075684, + "step": 1590 + }, + { + "epoch": 0.8374771002355405, + "grad_norm": 98.31045532226562, + "learning_rate": 7.910397201763308e-08, + "logits/chosen": -1.165185570716858, + "logits/rejected": -1.1953125, + "logps/chosen": -686.2750244140625, + "logps/rejected": -773.1312255859375, + "logps/weighted_chosen": -3.680419921875, + "logps/weighted_rejected": -4.370849609375, + "loss": 0.6126, + "rewards/accuracies": 0.565625011920929, + "rewards/chosen": -419.4292907714844, + "rewards/margins": 94.01953125, + "rewards/rejected": -513.4410400390625, + "rewards/weighted_accuracies": 0.7124999761581421, + "rewards/weighted_chosen": -1.337377905845642, + "rewards/weighted_margins": 0.9386047124862671, + "rewards/weighted_rejected": -2.2768921852111816, + "step": 1600 + }, + { + "epoch": 0.8427113321120125, + "grad_norm": 127.34908294677734, + "learning_rate": 7.424190564297489e-08, + "logits/chosen": -1.200341820716858, + "logits/rejected": -1.2474243640899658, + "logps/chosen": -723.3695068359375, + "logps/rejected": -836.8171997070312, + "logps/weighted_chosen": -3.5196776390075684, + "logps/weighted_rejected": -4.542773246765137, + "loss": 0.5396, + "rewards/accuracies": 0.612500011920929, + "rewards/chosen": -442.2417907714844, + "rewards/margins": 119.1753921508789, + "rewards/rejected": -561.5078125, + "rewards/weighted_accuracies": 0.7281249761581421, + "rewards/weighted_chosen": -1.3686096668243408, + "rewards/weighted_margins": 1.0560470819473267, + "rewards/weighted_rejected": -2.424060106277466, + "step": 1610 + }, + { + "epoch": 0.8479455639884846, + "grad_norm": 25.85185432434082, + "learning_rate": 6.952203901326464e-08, + "logits/chosen": -1.1488158702850342, + "logits/rejected": -1.197351098060608, + "logps/chosen": -701.828125, + "logps/rejected": -791.3828125, + "logps/weighted_chosen": -3.785595655441284, + "logps/weighted_rejected": -5.068505764007568, + "loss": 0.5519, + "rewards/accuracies": 0.596875011920929, + "rewards/chosen": -412.7074279785156, + "rewards/margins": 108.6156234741211, + "rewards/rejected": -521.37109375, + "rewards/weighted_accuracies": 0.731249988079071, + "rewards/weighted_chosen": -1.265100121498108, + "rewards/weighted_margins": 1.0268433094024658, + "rewards/weighted_rejected": -2.291638135910034, + "step": 1620 + }, + { + "epoch": 0.8531797958649568, + "grad_norm": 111.40572357177734, + "learning_rate": 6.494594852561558e-08, + "logits/chosen": -1.1355712413787842, + "logits/rejected": -1.18328857421875, + "logps/chosen": -710.8062744140625, + "logps/rejected": -790.7976684570312, + "logps/weighted_chosen": -3.600848436355591, + "logps/weighted_rejected": -4.999890327453613, + "loss": 0.5333, + "rewards/accuracies": 0.637499988079071, + "rewards/chosen": -419.97344970703125, + "rewards/margins": 99.8208999633789, + "rewards/rejected": -519.9085693359375, + "rewards/weighted_accuracies": 0.715624988079071, + "rewards/weighted_chosen": -1.1865936517715454, + "rewards/weighted_margins": 1.0258972644805908, + "rewards/weighted_rejected": -2.2139830589294434, + "step": 1630 + }, + { + "epoch": 0.8584140277414289, + "grad_norm": 33.190921783447266, + "learning_rate": 6.051516255707773e-08, + "logits/chosen": -1.163732886314392, + "logits/rejected": -1.213415503501892, + "logps/chosen": -656.5734252929688, + "logps/rejected": -773.390625, + "logps/weighted_chosen": -3.6455078125, + "logps/weighted_rejected": -4.722460746765137, + "loss": 0.539, + "rewards/accuracies": 0.609375, + "rewards/chosen": -383.56640625, + "rewards/margins": 137.61328125, + "rewards/rejected": -520.9671630859375, + "rewards/weighted_accuracies": 0.7437499761581421, + "rewards/weighted_chosen": -1.2532958984375, + "rewards/weighted_margins": 1.098425269126892, + "rewards/weighted_rejected": -2.3517823219299316, + "step": 1640 + }, + { + "epoch": 0.863648259617901, + "grad_norm": 214.31565856933594, + "learning_rate": 5.6231160954171796e-08, + "logits/chosen": -1.112097144126892, + "logits/rejected": -1.198706030845642, + "logps/chosen": -665.1546630859375, + "logps/rejected": -820.2890625, + "logps/weighted_chosen": -3.9293456077575684, + "logps/weighted_rejected": -4.075634956359863, + "loss": 0.5607, + "rewards/accuracies": 0.659375011920929, + "rewards/chosen": -395.1675720214844, + "rewards/margins": 151.1287078857422, + "rewards/rejected": -546.2628784179688, + "rewards/weighted_accuracies": 0.7093750238418579, + "rewards/weighted_chosen": -1.2908935546875, + "rewards/weighted_margins": 0.8940979242324829, + "rewards/weighted_rejected": -2.184948682785034, + "step": 1650 + }, + { + "epoch": 0.8688824914943732, + "grad_norm": 40.810096740722656, + "learning_rate": 5.209537453863289e-08, + "logits/chosen": -1.177636742591858, + "logits/rejected": -1.1999084949493408, + "logps/chosen": -580.1390380859375, + "logps/rejected": -695.2312622070312, + "logps/weighted_chosen": -3.4837646484375, + "logps/weighted_rejected": -4.563989162445068, + "loss": 0.5084, + "rewards/accuracies": 0.6312500238418579, + "rewards/chosen": -310.1695251464844, + "rewards/margins": 135.10391235351562, + "rewards/rejected": -445.3179626464844, + "rewards/weighted_accuracies": 0.737500011920929, + "rewards/weighted_chosen": -1.1034972667694092, + "rewards/weighted_margins": 1.097009301185608, + "rewards/weighted_rejected": -2.2013213634490967, + "step": 1660 + }, + { + "epoch": 0.8741167233708453, + "grad_norm": 198.37266540527344, + "learning_rate": 4.8109184629527344e-08, + "logits/chosen": -1.1480286121368408, + "logits/rejected": -1.174829125404358, + "logps/chosen": -629.2374877929688, + "logps/rejected": -765.0234375, + "logps/weighted_chosen": -3.574951171875, + "logps/weighted_rejected": -4.886132717132568, + "loss": 0.5108, + "rewards/accuracies": 0.621874988079071, + "rewards/chosen": -358.5101623535156, + "rewards/margins": 148.86758422851562, + "rewards/rejected": -507.44061279296875, + "rewards/weighted_accuracies": 0.737500011920929, + "rewards/weighted_chosen": -1.147027611732483, + "rewards/weighted_margins": 1.134619116783142, + "rewards/weighted_rejected": -2.281726121902466, + "step": 1670 + }, + { + "epoch": 0.8793509552473174, + "grad_norm": 21.939449310302734, + "learning_rate": 4.427392258190399e-08, + "logits/chosen": -1.1966369152069092, + "logits/rejected": -1.217309594154358, + "logps/chosen": -674.2453002929688, + "logps/rejected": -737.453125, + "logps/weighted_chosen": -3.573333740234375, + "logps/weighted_rejected": -5.062792778015137, + "loss": 0.5187, + "rewards/accuracies": 0.5718749761581421, + "rewards/chosen": -384.4058532714844, + "rewards/margins": 92.5804672241211, + "rewards/rejected": -477.02655029296875, + "rewards/weighted_accuracies": 0.7281249761581421, + "rewards/weighted_chosen": -1.1649169921875, + "rewards/weighted_margins": 0.9775329828262329, + "rewards/weighted_rejected": -2.1424317359924316, + "step": 1680 + }, + { + "epoch": 0.8845851871237895, + "grad_norm": 72.72052764892578, + "learning_rate": 4.059086934213141e-08, + "logits/chosen": -1.1529357433319092, + "logits/rejected": -1.1970093250274658, + "logps/chosen": -670.0140380859375, + "logps/rejected": -724.27734375, + "logps/weighted_chosen": -3.518237352371216, + "logps/weighted_rejected": -4.640625, + "loss": 0.5218, + "rewards/accuracies": 0.5874999761581421, + "rewards/chosen": -378.71990966796875, + "rewards/margins": 95.2548828125, + "rewards/rejected": -474.0453186035156, + "rewards/weighted_accuracies": 0.7437499761581421, + "rewards/weighted_chosen": -1.298101782798767, + "rewards/weighted_margins": 1.04913330078125, + "rewards/weighted_rejected": -2.347949266433716, + "step": 1690 + }, + { + "epoch": 0.8898194190002617, + "grad_norm": 320.4640197753906, + "learning_rate": 3.7061255020073346e-08, + "logits/chosen": -1.221380591392517, + "logits/rejected": -1.2476806640625, + "logps/chosen": -701.5734252929688, + "logps/rejected": -770.7281494140625, + "logps/weighted_chosen": -3.537548780441284, + "logps/weighted_rejected": -4.314111232757568, + "loss": 0.5509, + "rewards/accuracies": 0.590624988079071, + "rewards/chosen": -415.10858154296875, + "rewards/margins": 93.888671875, + "rewards/rejected": -509.1382751464844, + "rewards/weighted_accuracies": 0.7250000238418579, + "rewards/weighted_chosen": -1.25872802734375, + "rewards/weighted_margins": 0.9871581792831421, + "rewards/weighted_rejected": -2.246020555496216, + "step": 1700 + }, + { + "epoch": 0.8950536508767338, + "grad_norm": 21.794382095336914, + "learning_rate": 3.3686258478241027e-08, + "logits/chosen": -1.217187523841858, + "logits/rejected": -1.2429320812225342, + "logps/chosen": -727.0828247070312, + "logps/rejected": -782.7093505859375, + "logps/weighted_chosen": -3.617626905441284, + "logps/weighted_rejected": -4.555224418640137, + "loss": 0.5477, + "rewards/accuracies": 0.581250011920929, + "rewards/chosen": -415.97637939453125, + "rewards/margins": 87.7699203491211, + "rewards/rejected": -503.55938720703125, + "rewards/weighted_accuracies": 0.71875, + "rewards/weighted_chosen": -1.3385436534881592, + "rewards/weighted_margins": 0.872027575969696, + "rewards/weighted_rejected": -2.211010694503784, + "step": 1710 + }, + { + "epoch": 0.9002878827532059, + "grad_norm": 157.28964233398438, + "learning_rate": 3.0467006938063366e-08, + "logits/chosen": -1.1905517578125, + "logits/rejected": -1.2344787120819092, + "logps/chosen": -669.9632568359375, + "logps/rejected": -746.0343627929688, + "logps/weighted_chosen": -3.239794969558716, + "logps/weighted_rejected": -4.668139457702637, + "loss": 0.5525, + "rewards/accuracies": 0.606249988079071, + "rewards/chosen": -397.27459716796875, + "rewards/margins": 89.14570617675781, + "rewards/rejected": -486.55157470703125, + "rewards/weighted_accuracies": 0.699999988079071, + "rewards/weighted_chosen": -1.291589379310608, + "rewards/weighted_margins": 1.105200171470642, + "rewards/weighted_rejected": -2.396716356277466, + "step": 1720 + }, + { + "epoch": 0.9055221146296781, + "grad_norm": 295.63861083984375, + "learning_rate": 2.7404575603403646e-08, + "logits/chosen": -1.1640503406524658, + "logits/rejected": -1.215722680091858, + "logps/chosen": -654.3117065429688, + "logps/rejected": -804.9749755859375, + "logps/weighted_chosen": -3.811474561691284, + "logps/weighted_rejected": -5.056396484375, + "loss": 0.4448, + "rewards/accuracies": 0.659375011920929, + "rewards/chosen": -380.11505126953125, + "rewards/margins": 155.30078125, + "rewards/rejected": -535.3187255859375, + "rewards/weighted_accuracies": 0.778124988079071, + "rewards/weighted_chosen": -1.1834747791290283, + "rewards/weighted_margins": 1.2722076177597046, + "rewards/weighted_rejected": -2.4549317359924316, + "step": 1730 + }, + { + "epoch": 0.9107563465061502, + "grad_norm": 85.96134948730469, + "learning_rate": 2.4499987301450698e-08, + "logits/chosen": -1.1945922374725342, + "logits/rejected": -1.2708861827850342, + "logps/chosen": -713.2406005859375, + "logps/rejected": -868.5015869140625, + "logps/weighted_chosen": -3.6737060546875, + "logps/weighted_rejected": -4.847460746765137, + "loss": 0.4624, + "rewards/accuracies": 0.653124988079071, + "rewards/chosen": -393.1859436035156, + "rewards/margins": 189.7375030517578, + "rewards/rejected": -583.0999755859375, + "rewards/weighted_accuracies": 0.768750011920929, + "rewards/weighted_chosen": -1.1947753429412842, + "rewards/weighted_margins": 1.2012450695037842, + "rewards/weighted_rejected": -2.3964600563049316, + "step": 1740 + }, + { + "epoch": 0.9159905783826223, + "grad_norm": 99.16748046875, + "learning_rate": 2.1754212141102347e-08, + "logits/chosen": -1.1943480968475342, + "logits/rejected": -1.211828589439392, + "logps/chosen": -664.3336181640625, + "logps/rejected": -769.375, + "logps/weighted_chosen": -4.12176513671875, + "logps/weighted_rejected": -5.160546779632568, + "loss": 0.5123, + "rewards/accuracies": 0.653124988079071, + "rewards/chosen": -385.3828125, + "rewards/margins": 114.51953125, + "rewards/rejected": -500.05859375, + "rewards/weighted_accuracies": 0.762499988079071, + "rewards/weighted_chosen": -1.272180199623108, + "rewards/weighted_margins": 1.085974097251892, + "rewards/weighted_rejected": -2.3573365211486816, + "step": 1750 + }, + { + "epoch": 0.9212248102590945, + "grad_norm": 249.57980346679688, + "learning_rate": 1.9168167188957586e-08, + "logits/chosen": -1.141473412513733, + "logits/rejected": -1.218017578125, + "logps/chosen": -658.53125, + "logps/rejected": -720.2249755859375, + "logps/weighted_chosen": -3.7562012672424316, + "logps/weighted_rejected": -4.617724418640137, + "loss": 0.5498, + "rewards/accuracies": 0.59375, + "rewards/chosen": -377.919921875, + "rewards/margins": 95.85078430175781, + "rewards/rejected": -473.56719970703125, + "rewards/weighted_accuracies": 0.7124999761581421, + "rewards/weighted_chosen": -1.2995789051055908, + "rewards/weighted_margins": 1.051355004310608, + "rewards/weighted_rejected": -2.3507447242736816, + "step": 1760 + }, + { + "epoch": 0.9264590421355666, + "grad_norm": 64.01028442382812, + "learning_rate": 1.6742716163022865e-08, + "logits/chosen": -1.20599365234375, + "logits/rejected": -1.241949439048767, + "logps/chosen": -681.2359619140625, + "logps/rejected": -812.1702880859375, + "logps/weighted_chosen": -3.519580125808716, + "logps/weighted_rejected": -4.841210842132568, + "loss": 0.5064, + "rewards/accuracies": 0.6343749761581421, + "rewards/chosen": -385.01171875, + "rewards/margins": 150.9167938232422, + "rewards/rejected": -535.8961181640625, + "rewards/weighted_accuracies": 0.784375011920929, + "rewards/weighted_chosen": -1.3197021484375, + "rewards/weighted_margins": 1.097833275794983, + "rewards/weighted_rejected": -2.417858839035034, + "step": 1770 + }, + { + "epoch": 0.9316932740120387, + "grad_norm": 157.51995849609375, + "learning_rate": 1.4478669144238343e-08, + "logits/chosen": -1.164239525794983, + "logits/rejected": -1.182275414466858, + "logps/chosen": -622.8038940429688, + "logps/rejected": -767.3624877929688, + "logps/weighted_chosen": -3.6578125953674316, + "logps/weighted_rejected": -5.080908298492432, + "loss": 0.5548, + "rewards/accuracies": 0.637499988079071, + "rewards/chosen": -359.64569091796875, + "rewards/margins": 152.6183624267578, + "rewards/rejected": -512.1964721679688, + "rewards/weighted_accuracies": 0.734375, + "rewards/weighted_chosen": -1.2336914539337158, + "rewards/weighted_margins": 1.092431664466858, + "rewards/weighted_rejected": -2.3256287574768066, + "step": 1780 + }, + { + "epoch": 0.9369275058885108, + "grad_norm": 84.96482849121094, + "learning_rate": 1.23767823059166e-08, + "logits/chosen": -1.175811767578125, + "logits/rejected": -1.2118408679962158, + "logps/chosen": -703.2109375, + "logps/rejected": -821.1812744140625, + "logps/weighted_chosen": -3.453997850418091, + "logps/weighted_rejected": -4.739892482757568, + "loss": 0.5076, + "rewards/accuracies": 0.581250011920929, + "rewards/chosen": -412.1927795410156, + "rewards/margins": 128.1863250732422, + "rewards/rejected": -540.3038940429688, + "rewards/weighted_accuracies": 0.715624988079071, + "rewards/weighted_chosen": -1.3178520202636719, + "rewards/weighted_margins": 1.1170654296875, + "rewards/weighted_rejected": -2.4344725608825684, + "step": 1790 + }, + { + "epoch": 0.942161737764983, + "grad_norm": 24.094118118286133, + "learning_rate": 1.0437757661187486e-08, + "logits/chosen": -1.1822021007537842, + "logits/rejected": -1.2230346202850342, + "logps/chosen": -731.1359252929688, + "logps/rejected": -840.8406372070312, + "logps/weighted_chosen": -4.1541748046875, + "logps/weighted_rejected": -4.836865425109863, + "loss": 0.504, + "rewards/accuracies": 0.6031249761581421, + "rewards/chosen": -415.04376220703125, + "rewards/margins": 144.8796844482422, + "rewards/rejected": -559.7952880859375, + "rewards/weighted_accuracies": 0.731249988079071, + "rewards/weighted_chosen": -1.2489135265350342, + "rewards/weighted_margins": 1.05224609375, + "rewards/weighted_rejected": -2.301586866378784, + "step": 1800 + }, + { + "epoch": 0.9473959696414551, + "grad_norm": 365.9298400878906, + "learning_rate": 8.662242828530953e-09, + "logits/chosen": -1.1748535633087158, + "logits/rejected": -1.2249877452850342, + "logps/chosen": -639.7000122070312, + "logps/rejected": -771.2734375, + "logps/weighted_chosen": -4.110595703125, + "logps/weighted_rejected": -5.005419731140137, + "loss": 0.5637, + "rewards/accuracies": 0.6499999761581421, + "rewards/chosen": -378.4296875, + "rewards/margins": 139.9353485107422, + "rewards/rejected": -518.2789306640625, + "rewards/weighted_accuracies": 0.721875011920929, + "rewards/weighted_chosen": -1.3699219226837158, + "rewards/weighted_margins": 0.996142566204071, + "rewards/weighted_rejected": -2.3658814430236816, + "step": 1810 + }, + { + "epoch": 0.9526302015179272, + "grad_norm": 38.30413818359375, + "learning_rate": 7.050830815478082e-09, + "logits/chosen": -1.1999022960662842, + "logits/rejected": -1.227929711341858, + "logps/chosen": -652.6156005859375, + "logps/rejected": -764.3577880859375, + "logps/weighted_chosen": -3.982666015625, + "logps/weighted_rejected": -4.869433403015137, + "loss": 0.5582, + "rewards/accuracies": 0.609375, + "rewards/chosen": -392.23748779296875, + "rewards/margins": 127.1484375, + "rewards/rejected": -519.095703125, + "rewards/weighted_accuracies": 0.7281249761581421, + "rewards/weighted_chosen": -1.359582543373108, + "rewards/weighted_margins": 0.993237316608429, + "rewards/weighted_rejected": -2.352398633956909, + "step": 1820 + }, + { + "epoch": 0.9578644333943994, + "grad_norm": 314.2492980957031, + "learning_rate": 5.604059820551177e-09, + "logits/chosen": -1.1942627429962158, + "logits/rejected": -1.217626929283142, + "logps/chosen": -691.1734619140625, + "logps/rejected": -792.9508056640625, + "logps/weighted_chosen": -3.8807616233825684, + "logps/weighted_rejected": -4.9736328125, + "loss": 0.5307, + "rewards/accuracies": 0.612500011920929, + "rewards/chosen": -392.8941345214844, + "rewards/margins": 139.66659545898438, + "rewards/rejected": -532.2779541015625, + "rewards/weighted_accuracies": 0.7093750238418579, + "rewards/weighted_chosen": -1.2794067859649658, + "rewards/weighted_margins": 1.0887451171875, + "rewards/weighted_rejected": -2.3676390647888184, + "step": 1830 + }, + { + "epoch": 0.9630986652708715, + "grad_norm": 29.67596435546875, + "learning_rate": 4.322413053509943e-09, + "logits/chosen": -1.1745643615722656, + "logits/rejected": -1.2179076671600342, + "logps/chosen": -716.4593505859375, + "logps/rejected": -808.65625, + "logps/weighted_chosen": -3.5707030296325684, + "logps/weighted_rejected": -4.941259860992432, + "loss": 0.5277, + "rewards/accuracies": 0.606249988079071, + "rewards/chosen": -415.00274658203125, + "rewards/margins": 121.33085632324219, + "rewards/rejected": -536.318359375, + "rewards/weighted_accuracies": 0.734375, + "rewards/weighted_chosen": -1.2857787609100342, + "rewards/weighted_margins": 1.0569336414337158, + "rewards/weighted_rejected": -2.3438477516174316, + "step": 1840 + }, + { + "epoch": 0.9683328971473436, + "grad_norm": 43.18519592285156, + "learning_rate": 3.206318573963418e-09, + "logits/chosen": -1.18829345703125, + "logits/rejected": -1.2353515625, + "logps/chosen": -673.609375, + "logps/rejected": -773.1687622070312, + "logps/weighted_chosen": -3.8472657203674316, + "logps/weighted_rejected": -4.793212890625, + "loss": 0.5458, + "rewards/accuracies": 0.59375, + "rewards/chosen": -387.5249938964844, + "rewards/margins": 126.17655944824219, + "rewards/rejected": -513.7249755859375, + "rewards/weighted_accuracies": 0.703125, + "rewards/weighted_chosen": -1.34063720703125, + "rewards/weighted_margins": 0.9600464105606079, + "rewards/weighted_rejected": -2.3008790016174316, + "step": 1850 + }, + { + "epoch": 0.9735671290238157, + "grad_norm": 98.20772552490234, + "learning_rate": 2.256149148401387e-09, + "logits/chosen": -1.193994164466858, + "logits/rejected": -1.213659644126892, + "logps/chosen": -663.5203247070312, + "logps/rejected": -831.8812255859375, + "logps/weighted_chosen": -3.932324171066284, + "logps/weighted_rejected": -4.665478706359863, + "loss": 0.5257, + "rewards/accuracies": 0.625, + "rewards/chosen": -390.19647216796875, + "rewards/margins": 177.13241577148438, + "rewards/rejected": -567.4085693359375, + "rewards/weighted_accuracies": 0.7718750238418579, + "rewards/weighted_chosen": -1.37353515625, + "rewards/weighted_margins": 0.999591052532196, + "rewards/weighted_rejected": -2.3727660179138184, + "step": 1860 + }, + { + "epoch": 0.9788013609002879, + "grad_norm": 156.7181854248047, + "learning_rate": 1.4722221256933676e-09, + "logits/chosen": -1.210534691810608, + "logits/rejected": -1.2340819835662842, + "logps/chosen": -682.4734497070312, + "logps/rejected": -756.2828369140625, + "logps/weighted_chosen": -3.5858397483825684, + "logps/weighted_rejected": -5.135632514953613, + "loss": 0.6143, + "rewards/accuracies": 0.578125, + "rewards/chosen": -411.08319091796875, + "rewards/margins": 92.4906234741211, + "rewards/rejected": -503.4302673339844, + "rewards/weighted_accuracies": 0.7093750238418579, + "rewards/weighted_chosen": -1.425439476966858, + "rewards/weighted_margins": 0.88873291015625, + "rewards/weighted_rejected": -2.314349412918091, + "step": 1870 + }, + { + "epoch": 0.98403559277676, + "grad_norm": 43.11684799194336, + "learning_rate": 8.547993310970003e-10, + "logits/chosen": -1.1872589588165283, + "logits/rejected": -1.232177734375, + "logps/chosen": -659.8531494140625, + "logps/rejected": -765.3453369140625, + "logps/weighted_chosen": -4.0516357421875, + "logps/weighted_rejected": -4.990136623382568, + "loss": 0.6088, + "rewards/accuracies": 0.628125011920929, + "rewards/chosen": -378.859375, + "rewards/margins": 118.80351257324219, + "rewards/rejected": -497.828125, + "rewards/weighted_accuracies": 0.699999988079071, + "rewards/weighted_chosen": -1.2618834972381592, + "rewards/weighted_margins": 0.9063965082168579, + "rewards/weighted_rejected": -2.1686768531799316, + "step": 1880 + }, + { + "epoch": 0.9892698246532321, + "grad_norm": 290.6236877441406, + "learning_rate": 4.040869788100032e-10, + "logits/chosen": -1.1602783203125, + "logits/rejected": -1.1969115734100342, + "logps/chosen": -663.7374877929688, + "logps/rejected": -736.7015380859375, + "logps/weighted_chosen": -3.6440186500549316, + "logps/weighted_rejected": -4.590576171875, + "loss": 0.5181, + "rewards/accuracies": 0.6187499761581421, + "rewards/chosen": -384.9828186035156, + "rewards/margins": 106.361328125, + "rewards/rejected": -491.2445373535156, + "rewards/weighted_accuracies": 0.734375, + "rewards/weighted_chosen": -1.2502593994140625, + "rewards/weighted_margins": 1.019628882408142, + "rewards/weighted_rejected": -2.2702393531799316, + "step": 1890 + }, + { + "epoch": 0.9945040565297043, + "grad_norm": 531.091064453125, + "learning_rate": 1.202356030968743e-10, + "logits/chosen": -1.2035033702850342, + "logits/rejected": -1.2423584461212158, + "logps/chosen": -708.1984252929688, + "logps/rejected": -793.2374877929688, + "logps/weighted_chosen": -3.401611328125, + "logps/weighted_rejected": -4.605420112609863, + "loss": 0.5386, + "rewards/accuracies": 0.59375, + "rewards/chosen": -401.53399658203125, + "rewards/margins": 125.5843734741211, + "rewards/rejected": -527.2421875, + "rewards/weighted_accuracies": 0.734375, + "rewards/weighted_chosen": -1.2851684093475342, + "rewards/weighted_margins": 1.1517822742462158, + "rewards/weighted_rejected": -2.436840772628784, + "step": 1900 + }, + { + "epoch": 0.9997382884061764, + "grad_norm": 248.43289184570312, + "learning_rate": 3.3400080112211405e-12, + "logits/chosen": -1.171716332435608, + "logits/rejected": -1.211022973060608, + "logps/chosen": -732.8812255859375, + "logps/rejected": -844.5875244140625, + "logps/weighted_chosen": -3.752758741378784, + "logps/weighted_rejected": -4.8924560546875, + "loss": 0.4947, + "rewards/accuracies": 0.6468750238418579, + "rewards/chosen": -432.294921875, + "rewards/margins": 132.95858764648438, + "rewards/rejected": -565.1109619140625, + "rewards/weighted_accuracies": 0.778124988079071, + "rewards/weighted_chosen": -1.2966216802597046, + "rewards/weighted_margins": 1.052240014076233, + "rewards/weighted_rejected": -2.348583936691284, + "step": 1910 + } + ], + "logging_steps": 10, + "max_steps": 1911, + "num_input_tokens_seen": 0, + "num_train_epochs": 1, + "save_steps": 500, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": true + }, + "attributes": {} + } + }, + "total_flos": 0.0, + "train_batch_size": 1, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoint-1911/training_args.bin b/checkpoint-1911/training_args.bin new file mode 100644 index 0000000..eb22667 --- /dev/null +++ b/checkpoint-1911/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ded8eebbc35e0c697502403c9c9d2a0acb4a3f38810fc064a38066719f316edb +size 8785 diff --git a/checkpoint-1911/zero_to_fp32.py b/checkpoint-1911/zero_to_fp32.py new file mode 100644 index 0000000..0e75914 --- /dev/null +++ b/checkpoint-1911/zero_to_fp32.py @@ -0,0 +1,760 @@ +#!/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 . output_dir/ +# or +# python zero_to_fp32.py . output_dir/ --safe_serialization + +import argparse +import torch +import glob +import math +import os +import re +import gc +import json +import numpy as np +from tqdm import tqdm +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, weights_only=False) + + 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 tqdm(files, desc='Loading checkpoint shards'): + state_dict = torch.load(f, map_location=device, mmap=True, weights_only=False) + # 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}") + + fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] 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") + + +class GatheredTensor: + """ + A pseudo tensor that collects partitioned weights. + It is more memory efficient when there are multiple groups. + """ + + def __init__(self, flat_groups, flat_groups_offset, offset, partitioned_numel, shape): + self.flat_groups = flat_groups + self.flat_groups_offset = flat_groups_offset + self.offset = offset + self.partitioned_numel = partitioned_numel + self.shape = shape + self.dtype = self.flat_groups[0][0].dtype + + def contiguous(self): + """ + Merge partitioned weights from flat_groups into a single tensor. + """ + end_idx = self.offset + self.partitioned_numel + world_size = len(self.flat_groups) + pad_flat_param_chunks = [] + + for rank_i in range(world_size): + # for each rank, we need to collect weights from related group/groups + flat_groups_at_rank_i = self.flat_groups[rank_i] + start_group_id = None + end_group_id = None + for group_id in range(len(self.flat_groups_offset)): + if self.flat_groups_offset[group_id] <= self.offset < self.flat_groups_offset[group_id + 1]: + start_group_id = group_id + if self.flat_groups_offset[group_id] < end_idx <= self.flat_groups_offset[group_id + 1]: + end_group_id = group_id + break + # collect weights from related group/groups + for group_id in range(start_group_id, end_group_id + 1): + flat_tensor = flat_groups_at_rank_i[group_id] + start_offset = self.offset - self.flat_groups_offset[group_id] + end_offset = min(end_idx, self.flat_groups_offset[group_id + 1]) - self.flat_groups_offset[group_id] + pad_flat_param_chunks.append(flat_tensor[start_offset:end_offset]) + + # collect weights from all ranks + pad_flat_param = torch.cat(pad_flat_param_chunks, dim=0) + param = pad_flat_param[:self.shape.numel()].view(self.shape).contiguous() + return param + + +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 = sum([flat_group.numel() for flat_group in fp32_flat_groups[0]]) * 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 + flat_groups_offset = [0] + list(np.cumsum([flat_tensor.numel() for flat_tensor in fp32_flat_groups[0]])) + for name, shape in tqdm(param_shapes.items(), desc='Gathering sharded weights'): + 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}" + ) + + # memory efficient tensor + tensor = GatheredTensor(fp32_flat_groups, flat_groups_offset, offset, partitioned_numel, shape) + state_dict[name] = tensor + 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 to_torch_tensor(state_dict, return_empty_tensor=False): + """ + Convert state_dict of GatheredTensor to torch tensor + """ + torch_state_dict = {} + converted_tensors = {} + for name, tensor in state_dict.items(): + tensor_id = id(tensor) + if tensor_id in converted_tensors: # shared tensors + shared_tensor = torch_state_dict[converted_tensors[tensor_id]] + torch_state_dict[name] = shared_tensor + else: + converted_tensors[tensor_id] = name + if return_empty_tensor: + torch_state_dict[name] = torch.empty(tensor.shape, dtype=tensor.dtype) + else: + torch_state_dict[name] = tensor.contiguous() + return torch_state_dict + + +def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, + tag=None, + exclude_frozen_parameters=False, + lazy_mode=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 + - ``lazy_mode``: get state_dict in lazy mode. It returns a dict of pesduo tensor instead of torch tensor, which is more memory efficient. + Convert the pesduo tensor to torch tensor by ``.contiguous()`` + + Returns: + - pytorch ``state_dict`` + + 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. + + Note: the above usage may not work if your application doesn't have sufficient free CPU memory. + You may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with + the checkpoint. Or you can load state_dict in lazy mode :: + + from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, lazy_mode=True) # not on cpu + for name, lazy_tensor in state_dict.item(): + tensor = lazy_tensor.contiguous() # to cpu + print(name, tensor) + # del tensor to release memory if it no longer in use + """ + 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") + + state_dict = _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters) + if lazy_mode: + return state_dict + else: + return to_torch_tensor(state_dict) + + +def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, + output_dir, + max_shard_size="5GB", + safe_serialization=False, + 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_dir``: directory to the pytorch fp32 state_dict output files + - ``max_shard_size``: the maximum size for a checkpoint before being sharded, default value is 5GB + - ``safe_serialization``: whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`). + - ``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 + """ + + # Dependency pre-check + if safe_serialization: + try: + from safetensors.torch import save_file + except ImportError: + print('If you want to use `safe_serialization`, please `pip install safetensors`') + raise + if max_shard_size is not None: + try: + from huggingface_hub import split_torch_state_dict_into_shards + except ImportError: + print('If you want to use `max_shard_size`, please `pip install huggingface_hub`') + raise + + # Convert zero checkpoint to state_dict + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, + tag, + exclude_frozen_parameters, + lazy_mode=True) + + # Shard the model if it is too big. + weights_name = "model.safetensors" if safe_serialization else "pytorch_model.bin" + if max_shard_size is not None: + filename_pattern = weights_name.replace(".bin", "{suffix}.bin").replace(".safetensors", "{suffix}.safetensors") + # an memory-efficient approach for sharding + empty_state_dict = to_torch_tensor(state_dict, return_empty_tensor=True) + state_dict_split = split_torch_state_dict_into_shards(empty_state_dict, + filename_pattern=filename_pattern, + max_shard_size=max_shard_size) + else: + from collections import namedtuple + StateDictSplit = namedtuple("StateDictSplit", ["is_sharded", "filename_to_tensors"]) + state_dict_split = StateDictSplit(is_sharded=False, + filename_to_tensors={weights_name: list(state_dict.keys())}) + + # Save the model by shard + os.makedirs(output_dir, exist_ok=True) + filename_to_tensors = state_dict_split.filename_to_tensors.items() + for shard_file, tensors in tqdm(filename_to_tensors, desc="Saving checkpoint shards"): + shard_state_dict = {tensor_name: state_dict[tensor_name] for tensor_name in tensors} + shard_state_dict = to_torch_tensor(shard_state_dict) + output_path = os.path.join(output_dir, shard_file) + if safe_serialization: + save_file(shard_state_dict, output_path, metadata={"format": "pt"}) + else: + torch.save(shard_state_dict, output_path) + # release the memory of current shard + for tensor_name in list(shard_state_dict.keys()): + del state_dict[tensor_name] + del shard_state_dict[tensor_name] + del shard_state_dict + gc.collect() + + # Save index if sharded + if state_dict_split.is_sharded: + index = { + "metadata": state_dict_split.metadata, + "weight_map": state_dict_split.tensor_to_filename, + } + save_index_file = "model.safetensors.index.json" if safe_serialization else "pytorch_model.bin.index.json" + save_index_file = os.path.join(output_dir, save_index_file) + with open(save_index_file, "w", encoding="utf-8") as f: + content = json.dumps(index, indent=2, sort_keys=True) + "\n" + f.write(content) + + +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_dir", + type=str, + help="directory to the pytorch fp32 state_dict output files" + "(e.g. path/checkpoint-12-output/)") + parser.add_argument( + "--max_shard_size", + type=str, + default="5GB", + help="The maximum size for a checkpoint before being sharded. Checkpoints shard will then be each of size" + "lower than this size. If expressed as a string, needs to be digits followed by a unit (like `5MB`" + "We default it to 5GB in order for models to be able to run easily on free-tier google colab instances" + "without CPU OOM issues.") + parser.add_argument( + "--safe_serialization", + default=False, + action='store_true', + help="Whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).") + 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_dir, + max_shard_size=args.max_shard_size, + safe_serialization=args.safe_serialization, + tag=args.tag, + exclude_frozen_parameters=args.exclude_frozen_parameters) diff --git a/checkpoint-500/chat_template.jinja b/checkpoint-500/chat_template.jinja new file mode 100644 index 0000000..39bd0c9 --- /dev/null +++ b/checkpoint-500/chat_template.jinja @@ -0,0 +1,5 @@ +{% set loop_messages = messages %}{% for message in loop_messages %}{% set content = '<|start_header_id|>' + message['role'] + '<|end_header_id|> + +'+ message['content'] | trim + '<|eot_id|>' %}{% if loop.index0 == 0 %}{% set content = bos_token + content %}{% endif %}{{ content }}{% endfor %}{% if add_generation_prompt %}{{ '<|start_header_id|>assistant<|end_header_id|> + +' }}{% endif %} \ No newline at end of file diff --git a/checkpoint-500/config.json b/checkpoint-500/config.json new file mode 100644 index 0000000..3f8f5c0 --- /dev/null +++ b/checkpoint-500/config.json @@ -0,0 +1,29 @@ +{ + "architectures": [ + "LlamaForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "bos_token_id": 128000, + "eos_token_id": 128001, + "head_dim": 128, + "hidden_act": "silu", + "hidden_size": 4096, + "initializer_range": 0.02, + "intermediate_size": 14336, + "max_position_embeddings": 8192, + "mlp_bias": false, + "model_type": "llama", + "num_attention_heads": 32, + "num_hidden_layers": 32, + "num_key_value_heads": 8, + "pretraining_tp": 1, + "rms_norm_eps": 1e-05, + "rope_scaling": null, + "rope_theta": 500000.0, + "tie_word_embeddings": false, + "torch_dtype": "bfloat16", + "transformers_version": "4.54.1", + "use_cache": true, + "vocab_size": 128256 +} diff --git a/checkpoint-500/generation_config.json b/checkpoint-500/generation_config.json new file mode 100644 index 0000000..fc3c54a --- /dev/null +++ b/checkpoint-500/generation_config.json @@ -0,0 +1,6 @@ +{ + "_from_model_config": true, + "bos_token_id": 128000, + "eos_token_id": 128001, + "transformers_version": "4.54.1" +} diff --git a/checkpoint-500/latest b/checkpoint-500/latest new file mode 100644 index 0000000..f0b47ce --- /dev/null +++ b/checkpoint-500/latest @@ -0,0 +1 @@ +global_step500 \ No newline at end of file diff --git a/checkpoint-500/model-00001-of-00004.safetensors b/checkpoint-500/model-00001-of-00004.safetensors new file mode 100644 index 0000000..e0ffead --- /dev/null +++ b/checkpoint-500/model-00001-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5d0e0417a31e00007cec2ca3dcee9ee132ceccb2d085176f72f35652b94489f8 +size 4976698672 diff --git a/checkpoint-500/model-00002-of-00004.safetensors b/checkpoint-500/model-00002-of-00004.safetensors new file mode 100644 index 0000000..de27baa --- /dev/null +++ b/checkpoint-500/model-00002-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7da2742a734f33764b256d47b645026206cb2c4191051395186c6cb11252fae1 +size 4999802720 diff --git a/checkpoint-500/model-00003-of-00004.safetensors b/checkpoint-500/model-00003-of-00004.safetensors new file mode 100644 index 0000000..da363a1 --- /dev/null +++ b/checkpoint-500/model-00003-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a894a39c25440b4134993eea3ef53ef84961698f73597b9f0ad617a319aa07d4 +size 4915916176 diff --git a/checkpoint-500/model-00004-of-00004.safetensors b/checkpoint-500/model-00004-of-00004.safetensors new file mode 100644 index 0000000..39060e7 --- /dev/null +++ b/checkpoint-500/model-00004-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0e3d72580cffd44ebc7ce1633c81a89ca7f40d57f50c6d572af9c79cc76d174b +size 1168138808 diff --git a/checkpoint-500/model.safetensors.index.json b/checkpoint-500/model.safetensors.index.json new file mode 100644 index 0000000..58a7ef4 --- /dev/null +++ b/checkpoint-500/model.safetensors.index.json @@ -0,0 +1,299 @@ +{ + "metadata": { + "total_parameters": 266240, + "total_size": 16060522496 + }, + "weight_map": { + "lm_head.weight": "model-00004-of-00004.safetensors", + "model.embed_tokens.weight": "model-00001-of-00004.safetensors", + "model.layers.0.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.0.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.1.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.10.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.10.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.10.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.11.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.11.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.12.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.12.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.13.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.13.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.14.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.14.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.15.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.15.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.16.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.16.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.17.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.17.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.18.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.18.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.19.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.19.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.2.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.2.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.20.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.20.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.20.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.20.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.20.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.20.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.20.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.20.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.20.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.21.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.21.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.21.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.22.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.22.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.23.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.23.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.24.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.24.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.25.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.25.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.26.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.26.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.27.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.27.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.28.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.28.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.29.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.29.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.3.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.3.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.30.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.30.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.30.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.input_layernorm.weight": "model-00004-of-00004.safetensors", + "model.layers.31.mlp.down_proj.weight": "model-00004-of-00004.safetensors", + "model.layers.31.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.post_attention_layernorm.weight": "model-00004-of-00004.safetensors", + "model.layers.31.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.4.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.4.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.5.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.6.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.7.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.8.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.9.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.9.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.9.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.norm.weight": "model-00004-of-00004.safetensors" + } +} diff --git a/checkpoint-500/rng_state_0.pth b/checkpoint-500/rng_state_0.pth new file mode 100644 index 0000000..6ae1c3a --- /dev/null +++ b/checkpoint-500/rng_state_0.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:33fe1a45c0111b18df213058c73c3a4e717295b975e92faf7b2e048e6504b3f3 +size 14917 diff --git a/checkpoint-500/rng_state_1.pth b/checkpoint-500/rng_state_1.pth new file mode 100644 index 0000000..58bf4e2 --- /dev/null +++ b/checkpoint-500/rng_state_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4bf26da988f2f17ca0d366aece1dfdb5c3bcab91066168b7062b361b8c3ac2d6 +size 14917 diff --git a/checkpoint-500/scheduler.pt b/checkpoint-500/scheduler.pt new file mode 100644 index 0000000..a2af8e9 --- /dev/null +++ b/checkpoint-500/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:93f8728b8c285bed1ca96ea99a5e658a8a9c58f9dd1ce1805f1213195612503b +size 1465 diff --git a/checkpoint-500/special_tokens_map.json b/checkpoint-500/special_tokens_map.json new file mode 100644 index 0000000..e5b39b6 --- /dev/null +++ b/checkpoint-500/special_tokens_map.json @@ -0,0 +1,23 @@ +{ + "bos_token": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "eos_token": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "pad_token": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + } +} diff --git a/checkpoint-500/tokenizer.json b/checkpoint-500/tokenizer.json new file mode 100644 index 0000000..03aa64f --- /dev/null +++ b/checkpoint-500/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0968dcc0ee8e56c7dccd34a7f51f8065ea0cb9e2cc529e3243d1e5c0a4bdaa0c +size 17208754 diff --git a/checkpoint-500/tokenizer_config.json b/checkpoint-500/tokenizer_config.json new file mode 100644 index 0000000..877a9a9 --- /dev/null +++ b/checkpoint-500/tokenizer_config.json @@ -0,0 +1,2063 @@ +{ + "added_tokens_decoder": { + "128000": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128001": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128002": { + "content": "<|reserved_special_token_0|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128003": { + "content": "<|reserved_special_token_1|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128004": { + "content": "<|reserved_special_token_2|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128005": { + "content": "<|reserved_special_token_3|>", + "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": "<|reserved_special_token_4|>", + "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": "<|reserved_special_token_5|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128011": { + "content": "<|reserved_special_token_6|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128012": { + "content": "<|reserved_special_token_7|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128013": { + "content": "<|reserved_special_token_8|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128014": { + "content": "<|reserved_special_token_9|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128015": { + "content": "<|reserved_special_token_10|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128016": { + "content": "<|reserved_special_token_11|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128017": { + "content": "<|reserved_special_token_12|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128018": { + "content": "<|reserved_special_token_13|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128019": { + "content": "<|reserved_special_token_14|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128020": { + "content": "<|reserved_special_token_15|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128021": { + "content": "<|reserved_special_token_16|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128022": { + "content": "<|reserved_special_token_17|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128023": { + "content": "<|reserved_special_token_18|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128024": { + "content": "<|reserved_special_token_19|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128025": { + "content": "<|reserved_special_token_20|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128026": { + "content": "<|reserved_special_token_21|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128027": { + "content": "<|reserved_special_token_22|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128028": { + "content": "<|reserved_special_token_23|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128029": { + "content": "<|reserved_special_token_24|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128030": { + "content": "<|reserved_special_token_25|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128031": { + "content": "<|reserved_special_token_26|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128032": { + "content": "<|reserved_special_token_27|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128033": { + "content": "<|reserved_special_token_28|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128034": { + "content": "<|reserved_special_token_29|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128035": { + "content": "<|reserved_special_token_30|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128036": { + "content": "<|reserved_special_token_31|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128037": { + "content": "<|reserved_special_token_32|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128038": { + "content": "<|reserved_special_token_33|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128039": { + "content": "<|reserved_special_token_34|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128040": { + "content": "<|reserved_special_token_35|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128041": { + "content": "<|reserved_special_token_36|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128042": { + "content": "<|reserved_special_token_37|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128043": { + "content": "<|reserved_special_token_38|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128044": { + "content": "<|reserved_special_token_39|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128045": { + "content": "<|reserved_special_token_40|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128046": { + "content": "<|reserved_special_token_41|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128047": { + "content": "<|reserved_special_token_42|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128048": { + "content": "<|reserved_special_token_43|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128049": { + "content": "<|reserved_special_token_44|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128050": { + "content": "<|reserved_special_token_45|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128051": { + "content": "<|reserved_special_token_46|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128052": { + "content": "<|reserved_special_token_47|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128053": { + "content": "<|reserved_special_token_48|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128054": { + "content": "<|reserved_special_token_49|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128055": { + "content": "<|reserved_special_token_50|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128056": { + "content": "<|reserved_special_token_51|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128057": { + "content": "<|reserved_special_token_52|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128058": { + "content": "<|reserved_special_token_53|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128059": { + "content": "<|reserved_special_token_54|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128060": { + "content": "<|reserved_special_token_55|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128061": { + "content": "<|reserved_special_token_56|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128062": { + "content": "<|reserved_special_token_57|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128063": { + "content": "<|reserved_special_token_58|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128064": { + "content": "<|reserved_special_token_59|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128065": { + "content": "<|reserved_special_token_60|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128066": { + "content": "<|reserved_special_token_61|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128067": { + "content": "<|reserved_special_token_62|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128068": { + "content": "<|reserved_special_token_63|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128069": { + "content": "<|reserved_special_token_64|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128070": { + "content": "<|reserved_special_token_65|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128071": { + "content": "<|reserved_special_token_66|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128072": { + "content": "<|reserved_special_token_67|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128073": { + "content": "<|reserved_special_token_68|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128074": { + "content": "<|reserved_special_token_69|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128075": { + "content": "<|reserved_special_token_70|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128076": { + "content": "<|reserved_special_token_71|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128077": { + "content": "<|reserved_special_token_72|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128078": { + "content": "<|reserved_special_token_73|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128079": { + "content": "<|reserved_special_token_74|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128080": { + "content": "<|reserved_special_token_75|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128081": { + "content": "<|reserved_special_token_76|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128082": { + "content": "<|reserved_special_token_77|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128083": { + "content": "<|reserved_special_token_78|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128084": { + "content": "<|reserved_special_token_79|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128085": { + "content": "<|reserved_special_token_80|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128086": { + "content": "<|reserved_special_token_81|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128087": { + "content": "<|reserved_special_token_82|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128088": { + "content": "<|reserved_special_token_83|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128089": { + "content": "<|reserved_special_token_84|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128090": { + "content": "<|reserved_special_token_85|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128091": { + "content": "<|reserved_special_token_86|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128092": { + "content": "<|reserved_special_token_87|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128093": { + "content": "<|reserved_special_token_88|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128094": { + "content": "<|reserved_special_token_89|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128095": { + "content": "<|reserved_special_token_90|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128096": { + "content": "<|reserved_special_token_91|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128097": { + "content": "<|reserved_special_token_92|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128098": { + "content": "<|reserved_special_token_93|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128099": { + "content": "<|reserved_special_token_94|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128100": { + "content": "<|reserved_special_token_95|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128101": { + "content": "<|reserved_special_token_96|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128102": { + "content": "<|reserved_special_token_97|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128103": { + "content": "<|reserved_special_token_98|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128104": { + "content": "<|reserved_special_token_99|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128105": { + "content": "<|reserved_special_token_100|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128106": { + "content": "<|reserved_special_token_101|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128107": { + "content": "<|reserved_special_token_102|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128108": { + "content": "<|reserved_special_token_103|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128109": { + "content": "<|reserved_special_token_104|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128110": { + "content": "<|reserved_special_token_105|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128111": { + "content": "<|reserved_special_token_106|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128112": { + "content": "<|reserved_special_token_107|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128113": { + "content": "<|reserved_special_token_108|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128114": { + "content": "<|reserved_special_token_109|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128115": { + "content": "<|reserved_special_token_110|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128116": { + "content": "<|reserved_special_token_111|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128117": { + "content": "<|reserved_special_token_112|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128118": { + "content": "<|reserved_special_token_113|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128119": { + "content": "<|reserved_special_token_114|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128120": { + "content": "<|reserved_special_token_115|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128121": { + "content": "<|reserved_special_token_116|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128122": { + "content": "<|reserved_special_token_117|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128123": { + "content": "<|reserved_special_token_118|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128124": { + "content": "<|reserved_special_token_119|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128125": { + "content": "<|reserved_special_token_120|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128126": { + "content": "<|reserved_special_token_121|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128127": { + "content": "<|reserved_special_token_122|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128128": { + "content": "<|reserved_special_token_123|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128129": { + "content": "<|reserved_special_token_124|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128130": { + "content": "<|reserved_special_token_125|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128131": { + "content": "<|reserved_special_token_126|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128132": { + "content": "<|reserved_special_token_127|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128133": { + "content": "<|reserved_special_token_128|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128134": { + "content": "<|reserved_special_token_129|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128135": { + "content": "<|reserved_special_token_130|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128136": { + "content": "<|reserved_special_token_131|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128137": { + "content": "<|reserved_special_token_132|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128138": { + "content": "<|reserved_special_token_133|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128139": { + "content": "<|reserved_special_token_134|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128140": { + "content": "<|reserved_special_token_135|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128141": { + "content": "<|reserved_special_token_136|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128142": { + "content": "<|reserved_special_token_137|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128143": { + "content": "<|reserved_special_token_138|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128144": { + "content": "<|reserved_special_token_139|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128145": { + "content": "<|reserved_special_token_140|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128146": { + "content": "<|reserved_special_token_141|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128147": { + "content": "<|reserved_special_token_142|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128148": { + "content": "<|reserved_special_token_143|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128149": { + "content": "<|reserved_special_token_144|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128150": { + "content": "<|reserved_special_token_145|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128151": { + "content": "<|reserved_special_token_146|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128152": { + "content": "<|reserved_special_token_147|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128153": { + "content": "<|reserved_special_token_148|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128154": { + "content": "<|reserved_special_token_149|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128155": { + "content": "<|reserved_special_token_150|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128156": { + "content": "<|reserved_special_token_151|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128157": { + "content": "<|reserved_special_token_152|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128158": { + "content": "<|reserved_special_token_153|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128159": { + "content": "<|reserved_special_token_154|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128160": { + "content": "<|reserved_special_token_155|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128161": { + "content": "<|reserved_special_token_156|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128162": { + "content": "<|reserved_special_token_157|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128163": { + "content": "<|reserved_special_token_158|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128164": { + "content": "<|reserved_special_token_159|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128165": { + "content": "<|reserved_special_token_160|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128166": { + "content": "<|reserved_special_token_161|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128167": { + "content": "<|reserved_special_token_162|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128168": { + "content": "<|reserved_special_token_163|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128169": { + "content": "<|reserved_special_token_164|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128170": { + "content": "<|reserved_special_token_165|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128171": { + "content": "<|reserved_special_token_166|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128172": { + "content": "<|reserved_special_token_167|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128173": { + "content": "<|reserved_special_token_168|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128174": { + "content": "<|reserved_special_token_169|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128175": { + "content": "<|reserved_special_token_170|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128176": { + "content": "<|reserved_special_token_171|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128177": { + "content": "<|reserved_special_token_172|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128178": { + "content": "<|reserved_special_token_173|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128179": { + "content": "<|reserved_special_token_174|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128180": { + "content": "<|reserved_special_token_175|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128181": { + "content": "<|reserved_special_token_176|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128182": { + "content": "<|reserved_special_token_177|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128183": { + "content": "<|reserved_special_token_178|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128184": { + "content": "<|reserved_special_token_179|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128185": { + "content": "<|reserved_special_token_180|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128186": { + "content": "<|reserved_special_token_181|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128187": { + "content": "<|reserved_special_token_182|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128188": { + "content": "<|reserved_special_token_183|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128189": { + "content": "<|reserved_special_token_184|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128190": { + "content": "<|reserved_special_token_185|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128191": { + "content": "<|reserved_special_token_186|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128192": { + "content": "<|reserved_special_token_187|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128193": { + "content": "<|reserved_special_token_188|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128194": { + "content": "<|reserved_special_token_189|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128195": { + "content": "<|reserved_special_token_190|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128196": { + "content": "<|reserved_special_token_191|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128197": { + "content": "<|reserved_special_token_192|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128198": { + "content": "<|reserved_special_token_193|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128199": { + "content": "<|reserved_special_token_194|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128200": { + "content": "<|reserved_special_token_195|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128201": { + "content": "<|reserved_special_token_196|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128202": { + "content": "<|reserved_special_token_197|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128203": { + "content": "<|reserved_special_token_198|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128204": { + "content": "<|reserved_special_token_199|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128205": { + "content": "<|reserved_special_token_200|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128206": { + "content": "<|reserved_special_token_201|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128207": { + "content": "<|reserved_special_token_202|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128208": { + "content": "<|reserved_special_token_203|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128209": { + "content": "<|reserved_special_token_204|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128210": { + "content": "<|reserved_special_token_205|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128211": { + "content": "<|reserved_special_token_206|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128212": { + "content": "<|reserved_special_token_207|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128213": { + "content": "<|reserved_special_token_208|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128214": { + "content": "<|reserved_special_token_209|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128215": { + "content": "<|reserved_special_token_210|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128216": { + "content": "<|reserved_special_token_211|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128217": { + "content": "<|reserved_special_token_212|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128218": { + "content": "<|reserved_special_token_213|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128219": { + "content": "<|reserved_special_token_214|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128220": { + "content": "<|reserved_special_token_215|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128221": { + "content": "<|reserved_special_token_216|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128222": { + "content": "<|reserved_special_token_217|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128223": { + "content": "<|reserved_special_token_218|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128224": { + "content": "<|reserved_special_token_219|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128225": { + "content": "<|reserved_special_token_220|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128226": { + "content": "<|reserved_special_token_221|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128227": { + "content": "<|reserved_special_token_222|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128228": { + "content": "<|reserved_special_token_223|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128229": { + "content": "<|reserved_special_token_224|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128230": { + "content": "<|reserved_special_token_225|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128231": { + "content": "<|reserved_special_token_226|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128232": { + "content": "<|reserved_special_token_227|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128233": { + "content": "<|reserved_special_token_228|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128234": { + "content": "<|reserved_special_token_229|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128235": { + "content": "<|reserved_special_token_230|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128236": { + "content": "<|reserved_special_token_231|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128237": { + "content": "<|reserved_special_token_232|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128238": { + "content": "<|reserved_special_token_233|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128239": { + "content": "<|reserved_special_token_234|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128240": { + "content": "<|reserved_special_token_235|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128241": { + "content": "<|reserved_special_token_236|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128242": { + "content": "<|reserved_special_token_237|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128243": { + "content": "<|reserved_special_token_238|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128244": { + "content": "<|reserved_special_token_239|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128245": { + "content": "<|reserved_special_token_240|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128246": { + "content": "<|reserved_special_token_241|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128247": { + "content": "<|reserved_special_token_242|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128248": { + "content": "<|reserved_special_token_243|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128249": { + "content": "<|reserved_special_token_244|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128250": { + "content": "<|reserved_special_token_245|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128251": { + "content": "<|reserved_special_token_246|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128252": { + "content": "<|reserved_special_token_247|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128253": { + "content": "<|reserved_special_token_248|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128254": { + "content": "<|reserved_special_token_249|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128255": { + "content": "<|reserved_special_token_250|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + } + }, + "bos_token": "<|begin_of_text|>", + "clean_up_tokenization_spaces": true, + "eos_token": "<|end_of_text|>", + "extra_special_tokens": {}, + "model_input_names": [ + "input_ids", + "attention_mask" + ], + "model_max_length": 32768, + "pad_token": "<|end_of_text|>", + "tokenizer_class": "PreTrainedTokenizerFast" +} diff --git a/checkpoint-500/trainer_state.json b/checkpoint-500/trainer_state.json new file mode 100644 index 0000000..86f0a0e --- /dev/null +++ b/checkpoint-500/trainer_state.json @@ -0,0 +1,1127 @@ +{ + "best_global_step": null, + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 0.26171159382360637, + "eval_steps": 500, + "global_step": 500, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.0005234231876472127, + "grad_norm": 136.4956817626953, + "learning_rate": 0.0, + "logits/chosen": -0.40118408203125, + "logits/rejected": -0.41802978515625, + "logps/chosen": -297.609375, + "logps/rejected": -247.84375, + "logps/weighted_chosen": -4.50634765625, + "logps/weighted_rejected": -3.43408203125, + "loss": 0.6914, + "rewards/accuracies": 0.0, + "rewards/chosen": 0.0, + "rewards/margins": 0.0, + "rewards/rejected": 0.0, + "rewards/weighted_accuracies": 0.0, + "rewards/weighted_chosen": 0.0, + "rewards/weighted_margins": 0.0, + "rewards/weighted_rejected": 0.0, + "step": 1 + }, + { + "epoch": 0.005234231876472127, + "grad_norm": 269.8279113769531, + "learning_rate": 4.6875e-08, + "logits/chosen": -0.3176456093788147, + "logits/rejected": -0.3530849814414978, + "logps/chosen": -275.5694580078125, + "logps/rejected": -255.875, + "logps/weighted_chosen": -2.478325843811035, + "logps/weighted_rejected": -2.635009765625, + "loss": 0.6911, + "rewards/accuracies": 0.2604166567325592, + "rewards/chosen": -0.0588107630610466, + "rewards/margins": -0.0334201380610466, + "rewards/rejected": -0.025390625, + "rewards/weighted_accuracies": 0.3229166567325592, + "rewards/weighted_chosen": 0.0010899438057094812, + "rewards/weighted_margins": 0.0014837053604424, + "rewards/weighted_rejected": -0.00039333768654614687, + "step": 10 + }, + { + "epoch": 0.010468463752944255, + "grad_norm": 12.950738906860352, + "learning_rate": 9.895833333333332e-08, + "logits/chosen": -0.2977302670478821, + "logits/rejected": -0.3139175474643707, + "logps/chosen": -294.37579345703125, + "logps/rejected": -272.7203063964844, + "logps/weighted_chosen": -2.30804443359375, + "logps/weighted_rejected": -2.247570753097534, + "loss": 0.692, + "rewards/accuracies": 0.296875, + "rewards/chosen": -0.05908203125, + "rewards/margins": -0.01513671875, + "rewards/rejected": -0.0439453125, + "rewards/weighted_accuracies": 0.34687501192092896, + "rewards/weighted_chosen": 7.171630568336695e-05, + "rewards/weighted_margins": 3.662109520519152e-05, + "rewards/weighted_rejected": 3.509521411615424e-05, + "step": 20 + }, + { + "epoch": 0.015702695629416383, + "grad_norm": 65.14871215820312, + "learning_rate": 1.5104166666666664e-07, + "logits/chosen": -0.2917991578578949, + "logits/rejected": -0.3373458981513977, + "logps/chosen": -297.9375, + "logps/rejected": -268.0062561035156, + "logps/weighted_chosen": -2.3156495094299316, + "logps/weighted_rejected": -2.529247999191284, + "loss": 0.693, + "rewards/accuracies": 0.28125, + "rewards/chosen": 0.02177734300494194, + "rewards/margins": -0.06005859375, + "rewards/rejected": 0.08183594048023224, + "rewards/weighted_accuracies": 0.34062498807907104, + "rewards/weighted_chosen": 0.0012060165172442794, + "rewards/weighted_margins": -0.0021120072342455387, + "rewards/weighted_rejected": 0.0033180236350744963, + "step": 30 + }, + { + "epoch": 0.02093692750588851, + "grad_norm": 30.699859619140625, + "learning_rate": 2.03125e-07, + "logits/chosen": -0.301095575094223, + "logits/rejected": -0.3433547914028168, + "logps/chosen": -278.7124938964844, + "logps/rejected": -253.84530639648438, + "logps/weighted_chosen": -2.3508667945861816, + "logps/weighted_rejected": -2.55126953125, + "loss": 0.6909, + "rewards/accuracies": 0.2718749940395355, + "rewards/chosen": -0.014892578125, + "rewards/margins": -0.105224609375, + "rewards/rejected": 0.09033203125, + "rewards/weighted_accuracies": 0.359375, + "rewards/weighted_chosen": 0.0031726837623864412, + "rewards/weighted_margins": 0.0010789871448650956, + "rewards/weighted_rejected": 0.002093696501106024, + "step": 40 + }, + { + "epoch": 0.02617115938236064, + "grad_norm": 15.275498390197754, + "learning_rate": 2.552083333333333e-07, + "logits/chosen": -0.28184205293655396, + "logits/rejected": -0.32042425870895386, + "logps/chosen": -280.19140625, + "logps/rejected": -267.4398498535156, + "logps/weighted_chosen": -2.281604051589966, + "logps/weighted_rejected": -2.278125047683716, + "loss": 0.691, + "rewards/accuracies": 0.3125, + "rewards/chosen": 0.009082031436264515, + "rewards/margins": -0.11552734673023224, + "rewards/rejected": 0.12460937350988388, + "rewards/weighted_accuracies": 0.375, + "rewards/weighted_chosen": 0.0036834715865552425, + "rewards/weighted_margins": 0.0022117614280432463, + "rewards/weighted_rejected": 0.0014717101585119963, + "step": 50 + }, + { + "epoch": 0.031405391258832765, + "grad_norm": 42.36075973510742, + "learning_rate": 3.0729166666666665e-07, + "logits/chosen": -0.3147949278354645, + "logits/rejected": -0.3081672787666321, + "logps/chosen": -277.5492248535156, + "logps/rejected": -261.63201904296875, + "logps/weighted_chosen": -2.6925292015075684, + "logps/weighted_rejected": -2.5440917015075684, + "loss": 0.6888, + "rewards/accuracies": 0.39375001192092896, + "rewards/chosen": 0.15625, + "rewards/margins": 0.06132812425494194, + "rewards/rejected": 0.09492187201976776, + "rewards/weighted_accuracies": 0.43437498807907104, + "rewards/weighted_chosen": 0.0014663696056231856, + "rewards/weighted_margins": 0.0070289610885083675, + "rewards/weighted_rejected": -0.005612182430922985, + "step": 60 + }, + { + "epoch": 0.036639623135304895, + "grad_norm": 70.38953399658203, + "learning_rate": 3.59375e-07, + "logits/chosen": -0.31798094511032104, + "logits/rejected": -0.32500457763671875, + "logps/chosen": -289.6695251464844, + "logps/rejected": -244.7351531982422, + "logps/weighted_chosen": -2.19110107421875, + "logps/weighted_rejected": -2.427661180496216, + "loss": 0.6869, + "rewards/accuracies": 0.4375, + "rewards/chosen": 0.36865234375, + "rewards/margins": 0.34619140625, + "rewards/rejected": 0.0224609375, + "rewards/weighted_accuracies": 0.5375000238418579, + "rewards/weighted_chosen": 0.02108154259622097, + "rewards/weighted_margins": 0.011628913693130016, + "rewards/weighted_rejected": 0.009454727172851562, + "step": 70 + }, + { + "epoch": 0.04187385501177702, + "grad_norm": 54.81600570678711, + "learning_rate": 4.114583333333333e-07, + "logits/chosen": -0.2848251461982727, + "logits/rejected": -0.3061889708042145, + "logps/chosen": -289.08203125, + "logps/rejected": -269.95782470703125, + "logps/weighted_chosen": -2.3393311500549316, + "logps/weighted_rejected": -2.5093994140625, + "loss": 0.6827, + "rewards/accuracies": 0.44999998807907104, + "rewards/chosen": 0.679980456829071, + "rewards/margins": 0.3672851622104645, + "rewards/rejected": 0.31269532442092896, + "rewards/weighted_accuracies": 0.5062500238418579, + "rewards/weighted_chosen": 0.048927437514066696, + "rewards/weighted_margins": 0.025808846578001976, + "rewards/weighted_rejected": 0.02311401441693306, + "step": 80 + }, + { + "epoch": 0.04710808688824915, + "grad_norm": 30.603567123413086, + "learning_rate": 4.6354166666666664e-07, + "logits/chosen": -0.312631219625473, + "logits/rejected": -0.30586013197898865, + "logps/chosen": -279.92578125, + "logps/rejected": -257.5234375, + "logps/weighted_chosen": -2.5583739280700684, + "logps/weighted_rejected": -2.6037840843200684, + "loss": 0.6752, + "rewards/accuracies": 0.49687498807907104, + "rewards/chosen": 0.757617175579071, + "rewards/margins": 0.5347656011581421, + "rewards/rejected": 0.22285155951976776, + "rewards/weighted_accuracies": 0.546875, + "rewards/weighted_chosen": 0.08651771396398544, + "rewards/weighted_margins": 0.05735473707318306, + "rewards/weighted_rejected": 0.02917785570025444, + "step": 90 + }, + { + "epoch": 0.05234231876472128, + "grad_norm": 99.52488708496094, + "learning_rate": 5.156249999999999e-07, + "logits/chosen": -0.2793869078159332, + "logits/rejected": -0.3509994447231293, + "logps/chosen": -290.19061279296875, + "logps/rejected": -285.94842529296875, + "logps/weighted_chosen": -2.013537645339966, + "logps/weighted_rejected": -2.5231690406799316, + "loss": 0.6984, + "rewards/accuracies": 0.512499988079071, + "rewards/chosen": 0.72607421875, + "rewards/margins": 0.80712890625, + "rewards/rejected": -0.0810546875, + "rewards/weighted_accuracies": 0.6187499761581421, + "rewards/weighted_chosen": 0.07289886474609375, + "rewards/weighted_margins": 0.019171524792909622, + "rewards/weighted_rejected": 0.05372200161218643, + "step": 100 + }, + { + "epoch": 0.05757655064119341, + "grad_norm": 55.06239318847656, + "learning_rate": 5.677083333333333e-07, + "logits/chosen": -0.3277145326137543, + "logits/rejected": -0.3189544677734375, + "logps/chosen": -296.17889404296875, + "logps/rejected": -261.51953125, + "logps/weighted_chosen": -2.647631883621216, + "logps/weighted_rejected": -2.6458740234375, + "loss": 0.6844, + "rewards/accuracies": 0.518750011920929, + "rewards/chosen": 0.1376953125, + "rewards/margins": 0.789746105670929, + "rewards/rejected": -0.6522461175918579, + "rewards/weighted_accuracies": 0.565625011920929, + "rewards/weighted_chosen": 0.05432138592004776, + "rewards/weighted_margins": 0.03696594387292862, + "rewards/weighted_rejected": 0.01735992357134819, + "step": 110 + }, + { + "epoch": 0.06281078251766553, + "grad_norm": 42.45036315917969, + "learning_rate": 6.197916666666666e-07, + "logits/chosen": -0.33232802152633667, + "logits/rejected": -0.35865744948387146, + "logps/chosen": -294.6078186035156, + "logps/rejected": -255.01327514648438, + "logps/weighted_chosen": -1.970422387123108, + "logps/weighted_rejected": -2.1345458030700684, + "loss": 0.6837, + "rewards/accuracies": 0.609375, + "rewards/chosen": 0.0013671874767169356, + "rewards/margins": 1.2595703601837158, + "rewards/rejected": -1.258203148841858, + "rewards/weighted_accuracies": 0.59375, + "rewards/weighted_chosen": 0.048947714269161224, + "rewards/weighted_margins": 0.03674011304974556, + "rewards/weighted_rejected": 0.012198830023407936, + "step": 120 + }, + { + "epoch": 0.06804501439413765, + "grad_norm": 28.637327194213867, + "learning_rate": 6.718749999999999e-07, + "logits/chosen": -0.2933105528354645, + "logits/rejected": -0.29020920395851135, + "logps/chosen": -305.86602783203125, + "logps/rejected": -279.09844970703125, + "logps/weighted_chosen": -1.9946777820587158, + "logps/weighted_rejected": -2.2933349609375, + "loss": 0.6754, + "rewards/accuracies": 0.625, + "rewards/chosen": 0.285400390625, + "rewards/margins": 1.7852051258087158, + "rewards/rejected": -1.4998047351837158, + "rewards/weighted_accuracies": 0.6187499761581421, + "rewards/weighted_chosen": 0.06223602220416069, + "rewards/weighted_margins": 0.05499267578125, + "rewards/weighted_rejected": 0.0072036744095385075, + "step": 130 + }, + { + "epoch": 0.07327924627060979, + "grad_norm": 20.866477966308594, + "learning_rate": 7.239583333333333e-07, + "logits/chosen": -0.33005982637405396, + "logits/rejected": -0.34638214111328125, + "logps/chosen": -300.7281188964844, + "logps/rejected": -276.9156188964844, + "logps/weighted_chosen": -2.294506788253784, + "logps/weighted_rejected": -2.561084032058716, + "loss": 0.6606, + "rewards/accuracies": 0.628125011920929, + "rewards/chosen": -1.998437523841858, + "rewards/margins": 2.3173828125, + "rewards/rejected": -4.315820217132568, + "rewards/weighted_accuracies": 0.671875, + "rewards/weighted_chosen": 0.04616241529583931, + "rewards/weighted_margins": 0.08864898979663849, + "rewards/weighted_rejected": -0.04253540188074112, + "step": 140 + }, + { + "epoch": 0.07851347814708191, + "grad_norm": 43.497623443603516, + "learning_rate": 7.760416666666666e-07, + "logits/chosen": -0.2890525758266449, + "logits/rejected": -0.30130767822265625, + "logps/chosen": -287.3109436035156, + "logps/rejected": -256.72967529296875, + "logps/weighted_chosen": -2.883862257003784, + "logps/weighted_rejected": -2.796630859375, + "loss": 0.664, + "rewards/accuracies": 0.628125011920929, + "rewards/chosen": -3.108203172683716, + "rewards/margins": 2.7275390625, + "rewards/rejected": -5.835741996765137, + "rewards/weighted_accuracies": 0.6343749761581421, + "rewards/weighted_chosen": 0.04738159105181694, + "rewards/weighted_margins": 0.09773864597082138, + "rewards/weighted_rejected": -0.05034027248620987, + "step": 150 + }, + { + "epoch": 0.08374771002355404, + "grad_norm": 12.989988327026367, + "learning_rate": 8.28125e-07, + "logits/chosen": -0.3334007263183594, + "logits/rejected": -0.3831237852573395, + "logps/chosen": -308.2320251464844, + "logps/rejected": -281.5335998535156, + "logps/weighted_chosen": -2.378857374191284, + "logps/weighted_rejected": -2.3824095726013184, + "loss": 0.6464, + "rewards/accuracies": 0.6499999761581421, + "rewards/chosen": -4.347460746765137, + "rewards/margins": 3.98388671875, + "rewards/rejected": -8.331347465515137, + "rewards/weighted_accuracies": 0.65625, + "rewards/weighted_chosen": 0.04017028957605362, + "rewards/weighted_margins": 0.15716858208179474, + "rewards/weighted_rejected": -0.11710510402917862, + "step": 160 + }, + { + "epoch": 0.08898194190002617, + "grad_norm": 691.15625, + "learning_rate": 8.802083333333333e-07, + "logits/chosen": -0.37783128023147583, + "logits/rejected": -0.37789613008499146, + "logps/chosen": -310.76953125, + "logps/rejected": -270.8374938964844, + "logps/weighted_chosen": -2.6180419921875, + "logps/weighted_rejected": -2.8051514625549316, + "loss": 0.683, + "rewards/accuracies": 0.6156250238418579, + "rewards/chosen": -7.780859470367432, + "rewards/margins": 4.9013671875, + "rewards/rejected": -12.683398246765137, + "rewards/weighted_accuracies": 0.643750011920929, + "rewards/weighted_chosen": 0.06108245998620987, + "rewards/weighted_margins": 0.17230224609375, + "rewards/weighted_rejected": -0.11125946044921875, + "step": 170 + }, + { + "epoch": 0.0942161737764983, + "grad_norm": 65.56312561035156, + "learning_rate": 9.322916666666666e-07, + "logits/chosen": -0.3433074951171875, + "logits/rejected": -0.363290399312973, + "logps/chosen": -285.8812561035156, + "logps/rejected": -274.94842529296875, + "logps/weighted_chosen": -2.50494384765625, + "logps/weighted_rejected": -2.9246826171875, + "loss": 0.6821, + "rewards/accuracies": 0.6343749761581421, + "rewards/chosen": -12.179101943969727, + "rewards/margins": 6.432421684265137, + "rewards/rejected": -18.608592987060547, + "rewards/weighted_accuracies": 0.59375, + "rewards/weighted_chosen": -0.0216827392578125, + "rewards/weighted_margins": 0.19813232123851776, + "rewards/weighted_rejected": -0.2197723388671875, + "step": 180 + }, + { + "epoch": 0.09945040565297043, + "grad_norm": 26.828750610351562, + "learning_rate": 9.84375e-07, + "logits/chosen": -0.36560744047164917, + "logits/rejected": -0.40380173921585083, + "logps/chosen": -321.97967529296875, + "logps/rejected": -287.5570373535156, + "logps/weighted_chosen": -2.3037109375, + "logps/weighted_rejected": -2.64697265625, + "loss": 0.6489, + "rewards/accuracies": 0.596875011920929, + "rewards/chosen": -16.246679306030273, + "rewards/margins": 7.440039157867432, + "rewards/rejected": -23.685937881469727, + "rewards/weighted_accuracies": 0.640625, + "rewards/weighted_chosen": -0.11536254733800888, + "rewards/weighted_margins": 0.19368895888328552, + "rewards/weighted_rejected": -0.309121698141098, + "step": 190 + }, + { + "epoch": 0.10468463752944256, + "grad_norm": 117.84883880615234, + "learning_rate": 9.99959085414323e-07, + "logits/chosen": -0.4036270081996918, + "logits/rejected": -0.43666380643844604, + "logps/chosen": -327.8265686035156, + "logps/rejected": -284.4234313964844, + "logps/weighted_chosen": -2.6368165016174316, + "logps/weighted_rejected": -3.0156006813049316, + "loss": 0.6383, + "rewards/accuracies": 0.6156250238418579, + "rewards/chosen": -18.538379669189453, + "rewards/margins": 8.212695121765137, + "rewards/rejected": -26.748046875, + "rewards/weighted_accuracies": 0.6468750238418579, + "rewards/weighted_chosen": -0.12391586601734161, + "rewards/weighted_margins": 0.28039854764938354, + "rewards/weighted_rejected": -0.404205322265625, + "step": 200 + }, + { + "epoch": 0.10991886940591468, + "grad_norm": 15.525572776794434, + "learning_rate": 9.997587035630105e-07, + "logits/chosen": -0.41632765531539917, + "logits/rejected": -0.4597106873989105, + "logps/chosen": -308.75469970703125, + "logps/rejected": -316.69842529296875, + "logps/weighted_chosen": -2.439282178878784, + "logps/weighted_rejected": -3.152050733566284, + "loss": 0.6437, + "rewards/accuracies": 0.653124988079071, + "rewards/chosen": -21.632617950439453, + "rewards/margins": 10.560155868530273, + "rewards/rejected": -32.18867111206055, + "rewards/weighted_accuracies": 0.671875, + "rewards/weighted_chosen": -0.2110244780778885, + "rewards/weighted_margins": 0.23976440727710724, + "rewards/weighted_rejected": -0.4508819580078125, + "step": 210 + }, + { + "epoch": 0.11515310128238682, + "grad_norm": 31.583986282348633, + "learning_rate": 9.99391406364405e-07, + "logits/chosen": -0.37994080781936646, + "logits/rejected": -0.384744256734848, + "logps/chosen": -318.63592529296875, + "logps/rejected": -305.0, + "logps/weighted_chosen": -2.9573731422424316, + "logps/weighted_rejected": -3.207348585128784, + "loss": 0.7259, + "rewards/accuracies": 0.606249988079071, + "rewards/chosen": -26.522266387939453, + "rewards/margins": 9.657129287719727, + "rewards/rejected": -36.18437576293945, + "rewards/weighted_accuracies": 0.625, + "rewards/weighted_chosen": -0.3494918942451477, + "rewards/weighted_margins": 0.17980042099952698, + "rewards/weighted_rejected": -0.5295776128768921, + "step": 220 + }, + { + "epoch": 0.12038733315885894, + "grad_norm": 611.2006225585938, + "learning_rate": 9.988573164927884e-07, + "logits/chosen": -0.3254447877407074, + "logits/rejected": -0.36442261934280396, + "logps/chosen": -297.16717529296875, + "logps/rejected": -296.48907470703125, + "logps/weighted_chosen": -2.35888671875, + "logps/weighted_rejected": -2.804126024246216, + "loss": 0.6637, + "rewards/accuracies": 0.6656249761581421, + "rewards/chosen": -28.293750762939453, + "rewards/margins": 16.522851943969727, + "rewards/rejected": -44.817970275878906, + "rewards/weighted_accuracies": 0.6812499761581421, + "rewards/weighted_chosen": -0.1478603333234787, + "rewards/weighted_margins": 0.3882904052734375, + "rewards/weighted_rejected": -0.5360336303710938, + "step": 230 + }, + { + "epoch": 0.12562156503533106, + "grad_norm": 34.738059997558594, + "learning_rate": 9.98156612329838e-07, + "logits/chosen": -0.3876676559448242, + "logits/rejected": -0.4364303648471832, + "logps/chosen": -312.5921936035156, + "logps/rejected": -348.60626220703125, + "logps/weighted_chosen": -2.594775438308716, + "logps/weighted_rejected": -3.1068115234375, + "loss": 0.7482, + "rewards/accuracies": 0.637499988079071, + "rewards/chosen": -47.75507736206055, + "rewards/margins": 18.762109756469727, + "rewards/rejected": -66.5289077758789, + "rewards/weighted_accuracies": 0.6499999761581421, + "rewards/weighted_chosen": -0.319406121969223, + "rewards/weighted_margins": 0.204498291015625, + "rewards/weighted_rejected": -0.5241454839706421, + "step": 240 + }, + { + "epoch": 0.13085579691180318, + "grad_norm": 141.2784423828125, + "learning_rate": 9.97289527905053e-07, + "logits/chosen": -0.4159797728061676, + "logits/rejected": -0.4382568299770355, + "logps/chosen": -308.44842529296875, + "logps/rejected": -313.49530029296875, + "logps/weighted_chosen": -2.802807569503784, + "logps/weighted_rejected": -2.9874024391174316, + "loss": 0.7583, + "rewards/accuracies": 0.609375, + "rewards/chosen": -44.16621017456055, + "rewards/margins": 13.382226943969727, + "rewards/rejected": -57.54296875, + "rewards/weighted_accuracies": 0.6187499761581421, + "rewards/weighted_chosen": -0.20551452040672302, + "rewards/weighted_margins": 0.22150878608226776, + "rewards/weighted_rejected": -0.42730408906936646, + "step": 250 + }, + { + "epoch": 0.1360900287882753, + "grad_norm": 75.63247680664062, + "learning_rate": 9.962563528175875e-07, + "logits/chosen": -0.4156021177768707, + "logits/rejected": -0.4493362307548523, + "logps/chosen": -342.1312561035156, + "logps/rejected": -317.84063720703125, + "logps/weighted_chosen": -2.56585693359375, + "logps/weighted_rejected": -3.4276366233825684, + "loss": 0.6673, + "rewards/accuracies": 0.5625, + "rewards/chosen": -43.44648361206055, + "rewards/margins": 13.459375381469727, + "rewards/rejected": -56.90898513793945, + "rewards/weighted_accuracies": 0.6156250238418579, + "rewards/weighted_chosen": -0.310464471578598, + "rewards/weighted_margins": 0.32973939180374146, + "rewards/weighted_rejected": -0.6404876708984375, + "step": 260 + }, + { + "epoch": 0.14132426066474746, + "grad_norm": 32.991607666015625, + "learning_rate": 9.950574321395277e-07, + "logits/chosen": -0.507153332233429, + "logits/rejected": -0.5264068841934204, + "logps/chosen": -317.0335998535156, + "logps/rejected": -300.5718688964844, + "logps/weighted_chosen": -2.669018507003784, + "logps/weighted_rejected": -2.8846678733825684, + "loss": 0.6774, + "rewards/accuracies": 0.609375, + "rewards/chosen": -27.629491806030273, + "rewards/margins": 9.762109756469727, + "rewards/rejected": -37.39765548706055, + "rewards/weighted_accuracies": 0.6000000238418579, + "rewards/weighted_chosen": -0.23316803574562073, + "rewards/weighted_margins": 0.2321929931640625, + "rewards/weighted_rejected": -0.46527099609375, + "step": 270 + }, + { + "epoch": 0.14655849254121958, + "grad_norm": 41.87471389770508, + "learning_rate": 9.936931663006413e-07, + "logits/chosen": -0.5015045404434204, + "logits/rejected": -0.49525564908981323, + "logps/chosen": -319.9296875, + "logps/rejected": -312.28826904296875, + "logps/weighted_chosen": -2.667797803878784, + "logps/weighted_rejected": -3.121386766433716, + "loss": 0.6458, + "rewards/accuracies": 0.6625000238418579, + "rewards/chosen": -17.077342987060547, + "rewards/margins": 15.751367568969727, + "rewards/rejected": -32.82636642456055, + "rewards/weighted_accuracies": 0.703125, + "rewards/weighted_chosen": -0.10239715874195099, + "rewards/weighted_margins": 0.3067169189453125, + "rewards/weighted_rejected": -0.409027099609375, + "step": 280 + }, + { + "epoch": 0.1517927244176917, + "grad_norm": 22.655261993408203, + "learning_rate": 9.921640109546357e-07, + "logits/chosen": -0.4936012327671051, + "logits/rejected": -0.5640270113945007, + "logps/chosen": -286.4117126464844, + "logps/rejected": -286.15313720703125, + "logps/weighted_chosen": -2.640380859375, + "logps/weighted_rejected": -3.774829149246216, + "loss": 0.6369, + "rewards/accuracies": 0.6343749761581421, + "rewards/chosen": -19.629491806030273, + "rewards/margins": 14.98291015625, + "rewards/rejected": -34.61503982543945, + "rewards/weighted_accuracies": 0.668749988079071, + "rewards/weighted_chosen": -0.21005859971046448, + "rewards/weighted_margins": 0.3854110836982727, + "rewards/weighted_rejected": -0.5955780148506165, + "step": 290 + }, + { + "epoch": 0.15702695629416383, + "grad_norm": 33.83458709716797, + "learning_rate": 9.90470476826975e-07, + "logits/chosen": -0.5687958002090454, + "logits/rejected": -0.5700439214706421, + "logps/chosen": -298.06561279296875, + "logps/rejected": -311.8359375, + "logps/weighted_chosen": -2.555248975753784, + "logps/weighted_rejected": -3.0333251953125, + "loss": 0.6326, + "rewards/accuracies": 0.6625000238418579, + "rewards/chosen": -28.754297256469727, + "rewards/margins": 15.813085556030273, + "rewards/rejected": -44.567771911621094, + "rewards/weighted_accuracies": 0.6468750238418579, + "rewards/weighted_chosen": -0.25576478242874146, + "rewards/weighted_margins": 0.3327178955078125, + "rewards/weighted_rejected": -0.5886474847793579, + "step": 300 + }, + { + "epoch": 0.16226118817063595, + "grad_norm": 29.578954696655273, + "learning_rate": 9.886131295443002e-07, + "logits/chosen": -0.6218292117118835, + "logits/rejected": -0.6808746457099915, + "logps/chosen": -313.44061279296875, + "logps/rejected": -297.9156188964844, + "logps/weighted_chosen": -2.719403028488159, + "logps/weighted_rejected": -3.1142334938049316, + "loss": 0.6221, + "rewards/accuracies": 0.653124988079071, + "rewards/chosen": -32.1875, + "rewards/margins": 15.571874618530273, + "rewards/rejected": -47.7666015625, + "rewards/weighted_accuracies": 0.6499999761581421, + "rewards/weighted_chosen": -0.19769592583179474, + "rewards/weighted_margins": 0.521374523639679, + "rewards/weighted_rejected": -0.7188171148300171, + "step": 310 + }, + { + "epoch": 0.16749542004710807, + "grad_norm": 15.948070526123047, + "learning_rate": 9.865925894455166e-07, + "logits/chosen": -0.6636382937431335, + "logits/rejected": -0.6817939877510071, + "logps/chosen": -336.1507873535156, + "logps/rejected": -306.62890625, + "logps/weighted_chosen": -2.83941650390625, + "logps/weighted_rejected": -3.273486375808716, + "loss": 0.7046, + "rewards/accuracies": 0.637499988079071, + "rewards/chosen": -36.21074295043945, + "rewards/margins": 12.9765625, + "rewards/rejected": -49.17656326293945, + "rewards/weighted_accuracies": 0.6343749761581421, + "rewards/weighted_chosen": -0.3679946959018707, + "rewards/weighted_margins": 0.278738409280777, + "rewards/weighted_rejected": -0.6464904546737671, + "step": 320 + }, + { + "epoch": 0.17272965192358022, + "grad_norm": 53.10931396484375, + "learning_rate": 9.84409531374603e-07, + "logits/chosen": -0.6222923398017883, + "logits/rejected": -0.607867419719696, + "logps/chosen": -344.19061279296875, + "logps/rejected": -315.84686279296875, + "logps/weighted_chosen": -2.915234327316284, + "logps/weighted_rejected": -3.346630811691284, + "loss": 0.6334, + "rewards/accuracies": 0.668749988079071, + "rewards/chosen": -39.39531326293945, + "rewards/margins": 13.606640815734863, + "rewards/rejected": -53.00273513793945, + "rewards/weighted_accuracies": 0.65625, + "rewards/weighted_chosen": -0.33662718534469604, + "rewards/weighted_margins": 0.3510375916957855, + "rewards/weighted_rejected": -0.6879855990409851, + "step": 330 + }, + { + "epoch": 0.17796388380005235, + "grad_norm": 13.994344711303711, + "learning_rate": 9.820646844552219e-07, + "logits/chosen": -0.6615959405899048, + "logits/rejected": -0.714038074016571, + "logps/chosen": -312.7328186035156, + "logps/rejected": -323.3359375, + "logps/weighted_chosen": -2.910736083984375, + "logps/weighted_rejected": -3.2298340797424316, + "loss": 0.6376, + "rewards/accuracies": 0.675000011920929, + "rewards/chosen": -36.327537536621094, + "rewards/margins": 21.465625762939453, + "rewards/rejected": -57.79179763793945, + "rewards/weighted_accuracies": 0.675000011920929, + "rewards/weighted_chosen": -0.34371185302734375, + "rewards/weighted_margins": 0.390716552734375, + "rewards/weighted_rejected": -0.734301745891571, + "step": 340 + }, + { + "epoch": 0.18319811567652447, + "grad_norm": 13.682384490966797, + "learning_rate": 9.795588318471964e-07, + "logits/chosen": -0.774444580078125, + "logits/rejected": -0.7887184023857117, + "logps/chosen": -302.76251220703125, + "logps/rejected": -336.90313720703125, + "logps/weighted_chosen": -2.718432664871216, + "logps/weighted_rejected": -3.0622315406799316, + "loss": 0.6451, + "rewards/accuracies": 0.6343749761581421, + "rewards/chosen": -43.29413986206055, + "rewards/margins": 17.229297637939453, + "rewards/rejected": -60.52812576293945, + "rewards/weighted_accuracies": 0.637499988079071, + "rewards/weighted_chosen": -0.33645325899124146, + "rewards/weighted_margins": 0.33460694551467896, + "rewards/weighted_rejected": -0.6709426641464233, + "step": 350 + }, + { + "epoch": 0.1884323475529966, + "grad_norm": 19.35496711730957, + "learning_rate": 9.768928104849415e-07, + "logits/chosen": -0.815447986125946, + "logits/rejected": -0.8216186761856079, + "logps/chosen": -330.66015625, + "logps/rejected": -313.8687438964844, + "logps/weighted_chosen": -2.970752000808716, + "logps/weighted_rejected": -3.1889405250549316, + "loss": 0.6763, + "rewards/accuracies": 0.621874988079071, + "rewards/chosen": -47.90937423706055, + "rewards/margins": 17.064844131469727, + "rewards/rejected": -64.98515319824219, + "rewards/weighted_accuracies": 0.6499999761581421, + "rewards/weighted_chosen": -0.3897460997104645, + "rewards/weighted_margins": 0.384918212890625, + "rewards/weighted_rejected": -0.7745300531387329, + "step": 360 + }, + { + "epoch": 0.19366657942946872, + "grad_norm": 21.70383071899414, + "learning_rate": 9.740675107979355e-07, + "logits/chosen": -0.7939224243164062, + "logits/rejected": -0.827624499797821, + "logps/chosen": -373.0406188964844, + "logps/rejected": -349.3343811035156, + "logps/weighted_chosen": -2.506054639816284, + "logps/weighted_rejected": -3.43896484375, + "loss": 0.6215, + "rewards/accuracies": 0.596875011920929, + "rewards/chosen": -58.082420349121094, + "rewards/margins": 16.594335556030273, + "rewards/rejected": -74.671875, + "rewards/weighted_accuracies": 0.671875, + "rewards/weighted_chosen": -0.48696595430374146, + "rewards/weighted_margins": 0.412750244140625, + "rewards/weighted_rejected": -0.90020751953125, + "step": 370 + }, + { + "epoch": 0.19890081130594087, + "grad_norm": 21.682790756225586, + "learning_rate": 9.71083876413323e-07, + "logits/chosen": -0.77825927734375, + "logits/rejected": -0.7921401858329773, + "logps/chosen": -367.0062561035156, + "logps/rejected": -355.78594970703125, + "logps/weighted_chosen": -2.644775390625, + "logps/weighted_rejected": -3.5374755859375, + "loss": 0.6609, + "rewards/accuracies": 0.628125011920929, + "rewards/chosen": -64.59492492675781, + "rewards/margins": 21.614843368530273, + "rewards/rejected": -86.1988296508789, + "rewards/weighted_accuracies": 0.6625000238418579, + "rewards/weighted_chosen": -0.5685027837753296, + "rewards/weighted_margins": 0.3586792051792145, + "rewards/weighted_rejected": -0.9269439578056335, + "step": 380 + }, + { + "epoch": 0.204135043182413, + "grad_norm": 30.26679229736328, + "learning_rate": 9.67942903840751e-07, + "logits/chosen": -0.8401764035224915, + "logits/rejected": -0.898754894733429, + "logps/chosen": -362.0328063964844, + "logps/rejected": -362.3687438964844, + "logps/weighted_chosen": -2.8179688453674316, + "logps/weighted_rejected": -3.4515624046325684, + "loss": 0.5992, + "rewards/accuracies": 0.668749988079071, + "rewards/chosen": -57.87226486206055, + "rewards/margins": 30.196680068969727, + "rewards/rejected": -88.0804672241211, + "rewards/weighted_accuracies": 0.715624988079071, + "rewards/weighted_chosen": -0.456369012594223, + "rewards/weighted_margins": 0.495849609375, + "rewards/weighted_rejected": -0.951983630657196, + "step": 390 + }, + { + "epoch": 0.2093692750588851, + "grad_norm": 37.81236267089844, + "learning_rate": 9.646456421395447e-07, + "logits/chosen": -0.8902359008789062, + "logits/rejected": -0.9107666015625, + "logps/chosen": -380.6343688964844, + "logps/rejected": -399.7281188964844, + "logps/weighted_chosen": -2.709228515625, + "logps/weighted_rejected": -3.455004930496216, + "loss": 0.6557, + "rewards/accuracies": 0.6812499761581421, + "rewards/chosen": -62.43437576293945, + "rewards/margins": 33.634765625, + "rewards/rejected": -96.0941390991211, + "rewards/weighted_accuracies": 0.653124988079071, + "rewards/weighted_chosen": -0.4688262939453125, + "rewards/weighted_margins": 0.3546081483364105, + "rewards/weighted_rejected": -0.8232513666152954, + "step": 400 + }, + { + "epoch": 0.21460350693535724, + "grad_norm": 21.761716842651367, + "learning_rate": 9.611931925683266e-07, + "logits/chosen": -0.849993109703064, + "logits/rejected": -0.8886367678642273, + "logps/chosen": -368.8374938964844, + "logps/rejected": -353.73126220703125, + "logps/weighted_chosen": -2.533374071121216, + "logps/weighted_rejected": -3.233935594558716, + "loss": 0.591, + "rewards/accuracies": 0.6656249761581421, + "rewards/chosen": -62.575782775878906, + "rewards/margins": 27.423242568969727, + "rewards/rejected": -90.0132827758789, + "rewards/weighted_accuracies": 0.6937500238418579, + "rewards/weighted_chosen": -0.4288879334926605, + "rewards/weighted_margins": 0.507708728313446, + "rewards/weighted_rejected": -0.9364013671875, + "step": 410 + }, + { + "epoch": 0.21983773881182936, + "grad_norm": 28.842378616333008, + "learning_rate": 9.575867082172085e-07, + "logits/chosen": -0.836225152015686, + "logits/rejected": -0.8854034543037415, + "logps/chosen": -371.90625, + "logps/rejected": -369.4125061035156, + "logps/weighted_chosen": -2.9190917015075684, + "logps/weighted_rejected": -3.118579149246216, + "loss": 0.6198, + "rewards/accuracies": 0.668749988079071, + "rewards/chosen": -66.98905944824219, + "rewards/margins": 32.337890625, + "rewards/rejected": -99.3402328491211, + "rewards/weighted_accuracies": 0.6625000238418579, + "rewards/weighted_chosen": -0.4911743104457855, + "rewards/weighted_margins": 0.5335143804550171, + "rewards/weighted_rejected": -1.0242736339569092, + "step": 420 + }, + { + "epoch": 0.22507197068830148, + "grad_norm": 16.805343627929688, + "learning_rate": 9.538273936226673e-07, + "logits/chosen": -0.8646026849746704, + "logits/rejected": -0.898516833782196, + "logps/chosen": -328.94921875, + "logps/rejected": -353.66717529296875, + "logps/weighted_chosen": -3.1983399391174316, + "logps/weighted_rejected": -3.6509766578674316, + "loss": 0.6173, + "rewards/accuracies": 0.625, + "rewards/chosen": -63.52910232543945, + "rewards/margins": 25.6201171875, + "rewards/rejected": -89.1246109008789, + "rewards/weighted_accuracies": 0.6499999761581421, + "rewards/weighted_chosen": -0.436279296875, + "rewards/weighted_margins": 0.438516229391098, + "rewards/weighted_rejected": -0.8747711181640625, + "step": 430 + }, + { + "epoch": 0.23030620256477363, + "grad_norm": 23.78999137878418, + "learning_rate": 9.499165043652391e-07, + "logits/chosen": -0.8759521245956421, + "logits/rejected": -0.8865814208984375, + "logps/chosen": -358.69140625, + "logps/rejected": -358.76251220703125, + "logps/weighted_chosen": -3.2625732421875, + "logps/weighted_rejected": -3.359301805496216, + "loss": 0.6366, + "rewards/accuracies": 0.6187499761581421, + "rewards/chosen": -67.4326171875, + "rewards/margins": 24.610157012939453, + "rewards/rejected": -92.03788757324219, + "rewards/weighted_accuracies": 0.668749988079071, + "rewards/weighted_chosen": -0.5896576046943665, + "rewards/weighted_margins": 0.34770506620407104, + "rewards/weighted_rejected": -0.9376128911972046, + "step": 440 + }, + { + "epoch": 0.23554043444124576, + "grad_norm": 49.646263122558594, + "learning_rate": 9.458553466501665e-07, + "logits/chosen": -0.9158996343612671, + "logits/rejected": -0.9590820074081421, + "logps/chosen": -355.3109436035156, + "logps/rejected": -341.9765625, + "logps/weighted_chosen": -3.231823682785034, + "logps/weighted_rejected": -3.4568848609924316, + "loss": 0.657, + "rewards/accuracies": 0.6312500238418579, + "rewards/chosen": -69.14453125, + "rewards/margins": 28.884180068969727, + "rewards/rejected": -98.0492172241211, + "rewards/weighted_accuracies": 0.6875, + "rewards/weighted_chosen": -0.6470184326171875, + "rewards/weighted_margins": 0.4533935487270355, + "rewards/weighted_rejected": -1.100683569908142, + "step": 450 + }, + { + "epoch": 0.24077466631771788, + "grad_norm": 38.45325469970703, + "learning_rate": 9.416452768711366e-07, + "logits/chosen": -0.9281005859375, + "logits/rejected": -0.9545837640762329, + "logps/chosen": -375.6390686035156, + "logps/rejected": -370.203125, + "logps/weighted_chosen": -2.9860596656799316, + "logps/weighted_rejected": -3.635498046875, + "loss": 0.6457, + "rewards/accuracies": 0.612500011920929, + "rewards/chosen": -83.3871078491211, + "rewards/margins": 26.872264862060547, + "rewards/rejected": -110.26094055175781, + "rewards/weighted_accuracies": 0.6625000238418579, + "rewards/weighted_chosen": -0.6367782354354858, + "rewards/weighted_margins": 0.515301525592804, + "rewards/weighted_rejected": -1.152197241783142, + "step": 460 + }, + { + "epoch": 0.24600889819419, + "grad_norm": 59.67642593383789, + "learning_rate": 9.372877011572557e-07, + "logits/chosen": -0.8797394037246704, + "logits/rejected": -0.904949963092804, + "logps/chosen": -398.8531188964844, + "logps/rejected": -388.22186279296875, + "logps/weighted_chosen": -3.0103516578674316, + "logps/weighted_rejected": -3.409374952316284, + "loss": 0.6181, + "rewards/accuracies": 0.6031249761581421, + "rewards/chosen": -88.2437515258789, + "rewards/margins": 25.542577743530273, + "rewards/rejected": -113.7890625, + "rewards/weighted_accuracies": 0.6781250238418579, + "rewards/weighted_chosen": -0.6517273187637329, + "rewards/weighted_margins": 0.482818603515625, + "rewards/weighted_rejected": -1.1340011358261108, + "step": 470 + }, + { + "epoch": 0.2512431300706621, + "grad_norm": 27.465656280517578, + "learning_rate": 9.327840749034141e-07, + "logits/chosen": -0.910723865032196, + "logits/rejected": -0.9439147710800171, + "logps/chosen": -364.4750061035156, + "logps/rejected": -392.03436279296875, + "logps/weighted_chosen": -2.8880372047424316, + "logps/weighted_rejected": -4.069311618804932, + "loss": 0.6276, + "rewards/accuracies": 0.6812499761581421, + "rewards/chosen": -77.44355773925781, + "rewards/margins": 38.147850036621094, + "rewards/rejected": -115.59648132324219, + "rewards/weighted_accuracies": 0.684374988079071, + "rewards/weighted_chosen": -0.5768798589706421, + "rewards/weighted_margins": 0.541027843952179, + "rewards/weighted_rejected": -1.1178162097930908, + "step": 480 + }, + { + "epoch": 0.2564773619471343, + "grad_norm": 34.35914993286133, + "learning_rate": 9.281359022841965e-07, + "logits/chosen": -0.80584716796875, + "logits/rejected": -0.8291991949081421, + "logps/chosen": -349.86798095703125, + "logps/rejected": -354.75701904296875, + "logps/weighted_chosen": -2.914257764816284, + "logps/weighted_rejected": -4.178680419921875, + "loss": 0.63, + "rewards/accuracies": 0.625, + "rewards/chosen": -74.94511413574219, + "rewards/margins": 34.957618713378906, + "rewards/rejected": -109.90644836425781, + "rewards/weighted_accuracies": 0.690625011920929, + "rewards/weighted_chosen": -0.564099133014679, + "rewards/weighted_margins": 0.658795177936554, + "rewards/weighted_rejected": -1.2229950428009033, + "step": 490 + }, + { + "epoch": 0.26171159382360637, + "grad_norm": 41.01088333129883, + "learning_rate": 9.233447357514989e-07, + "logits/chosen": -0.775347888469696, + "logits/rejected": -0.8226684331893921, + "logps/chosen": -369.1148376464844, + "logps/rejected": -373.8374938964844, + "logps/weighted_chosen": -3.288012742996216, + "logps/weighted_rejected": -3.7696776390075684, + "loss": 0.6225, + "rewards/accuracies": 0.643750011920929, + "rewards/chosen": -71.1283187866211, + "rewards/margins": 32.867576599121094, + "rewards/rejected": -104.0218734741211, + "rewards/weighted_accuracies": 0.6937500238418579, + "rewards/weighted_chosen": -0.5540069341659546, + "rewards/weighted_margins": 0.598828136920929, + "rewards/weighted_rejected": -1.1535155773162842, + "step": 500 + }, + { + "epoch": 0.26171159382360637, + "eval_logits/chosen": -0.8767929673194885, + "eval_logits/rejected": -0.8968105316162109, + "eval_logps/chosen": -369.21099853515625, + "eval_logps/rejected": -379.3609924316406, + "eval_logps/weighted_chosen": -2.957455635070801, + "eval_logps/weighted_rejected": -3.6658689975738525, + "eval_loss": 0.6115986108779907, + "eval_rewards/accuracies": 0.6294999718666077, + "eval_rewards/chosen": -80.43468475341797, + "eval_rewards/margins": 31.297624588012695, + "eval_rewards/rejected": -111.73987579345703, + "eval_rewards/weighted_accuracies": 0.6759999990463257, + "eval_rewards/weighted_chosen": -0.5706117153167725, + "eval_rewards/weighted_margins": 0.5400464534759521, + "eval_rewards/weighted_rejected": -1.1106581687927246, + "eval_runtime": 1361.4222, + "eval_samples_per_second": 1.469, + "eval_steps_per_second": 0.367, + "step": 500 + } + ], + "logging_steps": 10, + "max_steps": 1911, + "num_input_tokens_seen": 0, + "num_train_epochs": 1, + "save_steps": 500, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 0.0, + "train_batch_size": 1, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoint-500/training_args.bin b/checkpoint-500/training_args.bin new file mode 100644 index 0000000..eb22667 --- /dev/null +++ b/checkpoint-500/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ded8eebbc35e0c697502403c9c9d2a0acb4a3f38810fc064a38066719f316edb +size 8785 diff --git a/checkpoint-500/zero_to_fp32.py b/checkpoint-500/zero_to_fp32.py new file mode 100644 index 0000000..0e75914 --- /dev/null +++ b/checkpoint-500/zero_to_fp32.py @@ -0,0 +1,760 @@ +#!/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 . output_dir/ +# or +# python zero_to_fp32.py . output_dir/ --safe_serialization + +import argparse +import torch +import glob +import math +import os +import re +import gc +import json +import numpy as np +from tqdm import tqdm +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, weights_only=False) + + 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 tqdm(files, desc='Loading checkpoint shards'): + state_dict = torch.load(f, map_location=device, mmap=True, weights_only=False) + # 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}") + + fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] 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") + + +class GatheredTensor: + """ + A pseudo tensor that collects partitioned weights. + It is more memory efficient when there are multiple groups. + """ + + def __init__(self, flat_groups, flat_groups_offset, offset, partitioned_numel, shape): + self.flat_groups = flat_groups + self.flat_groups_offset = flat_groups_offset + self.offset = offset + self.partitioned_numel = partitioned_numel + self.shape = shape + self.dtype = self.flat_groups[0][0].dtype + + def contiguous(self): + """ + Merge partitioned weights from flat_groups into a single tensor. + """ + end_idx = self.offset + self.partitioned_numel + world_size = len(self.flat_groups) + pad_flat_param_chunks = [] + + for rank_i in range(world_size): + # for each rank, we need to collect weights from related group/groups + flat_groups_at_rank_i = self.flat_groups[rank_i] + start_group_id = None + end_group_id = None + for group_id in range(len(self.flat_groups_offset)): + if self.flat_groups_offset[group_id] <= self.offset < self.flat_groups_offset[group_id + 1]: + start_group_id = group_id + if self.flat_groups_offset[group_id] < end_idx <= self.flat_groups_offset[group_id + 1]: + end_group_id = group_id + break + # collect weights from related group/groups + for group_id in range(start_group_id, end_group_id + 1): + flat_tensor = flat_groups_at_rank_i[group_id] + start_offset = self.offset - self.flat_groups_offset[group_id] + end_offset = min(end_idx, self.flat_groups_offset[group_id + 1]) - self.flat_groups_offset[group_id] + pad_flat_param_chunks.append(flat_tensor[start_offset:end_offset]) + + # collect weights from all ranks + pad_flat_param = torch.cat(pad_flat_param_chunks, dim=0) + param = pad_flat_param[:self.shape.numel()].view(self.shape).contiguous() + return param + + +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 = sum([flat_group.numel() for flat_group in fp32_flat_groups[0]]) * 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 + flat_groups_offset = [0] + list(np.cumsum([flat_tensor.numel() for flat_tensor in fp32_flat_groups[0]])) + for name, shape in tqdm(param_shapes.items(), desc='Gathering sharded weights'): + 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}" + ) + + # memory efficient tensor + tensor = GatheredTensor(fp32_flat_groups, flat_groups_offset, offset, partitioned_numel, shape) + state_dict[name] = tensor + 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 to_torch_tensor(state_dict, return_empty_tensor=False): + """ + Convert state_dict of GatheredTensor to torch tensor + """ + torch_state_dict = {} + converted_tensors = {} + for name, tensor in state_dict.items(): + tensor_id = id(tensor) + if tensor_id in converted_tensors: # shared tensors + shared_tensor = torch_state_dict[converted_tensors[tensor_id]] + torch_state_dict[name] = shared_tensor + else: + converted_tensors[tensor_id] = name + if return_empty_tensor: + torch_state_dict[name] = torch.empty(tensor.shape, dtype=tensor.dtype) + else: + torch_state_dict[name] = tensor.contiguous() + return torch_state_dict + + +def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, + tag=None, + exclude_frozen_parameters=False, + lazy_mode=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 + - ``lazy_mode``: get state_dict in lazy mode. It returns a dict of pesduo tensor instead of torch tensor, which is more memory efficient. + Convert the pesduo tensor to torch tensor by ``.contiguous()`` + + Returns: + - pytorch ``state_dict`` + + 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. + + Note: the above usage may not work if your application doesn't have sufficient free CPU memory. + You may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with + the checkpoint. Or you can load state_dict in lazy mode :: + + from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, lazy_mode=True) # not on cpu + for name, lazy_tensor in state_dict.item(): + tensor = lazy_tensor.contiguous() # to cpu + print(name, tensor) + # del tensor to release memory if it no longer in use + """ + 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") + + state_dict = _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters) + if lazy_mode: + return state_dict + else: + return to_torch_tensor(state_dict) + + +def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, + output_dir, + max_shard_size="5GB", + safe_serialization=False, + 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_dir``: directory to the pytorch fp32 state_dict output files + - ``max_shard_size``: the maximum size for a checkpoint before being sharded, default value is 5GB + - ``safe_serialization``: whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`). + - ``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 + """ + + # Dependency pre-check + if safe_serialization: + try: + from safetensors.torch import save_file + except ImportError: + print('If you want to use `safe_serialization`, please `pip install safetensors`') + raise + if max_shard_size is not None: + try: + from huggingface_hub import split_torch_state_dict_into_shards + except ImportError: + print('If you want to use `max_shard_size`, please `pip install huggingface_hub`') + raise + + # Convert zero checkpoint to state_dict + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, + tag, + exclude_frozen_parameters, + lazy_mode=True) + + # Shard the model if it is too big. + weights_name = "model.safetensors" if safe_serialization else "pytorch_model.bin" + if max_shard_size is not None: + filename_pattern = weights_name.replace(".bin", "{suffix}.bin").replace(".safetensors", "{suffix}.safetensors") + # an memory-efficient approach for sharding + empty_state_dict = to_torch_tensor(state_dict, return_empty_tensor=True) + state_dict_split = split_torch_state_dict_into_shards(empty_state_dict, + filename_pattern=filename_pattern, + max_shard_size=max_shard_size) + else: + from collections import namedtuple + StateDictSplit = namedtuple("StateDictSplit", ["is_sharded", "filename_to_tensors"]) + state_dict_split = StateDictSplit(is_sharded=False, + filename_to_tensors={weights_name: list(state_dict.keys())}) + + # Save the model by shard + os.makedirs(output_dir, exist_ok=True) + filename_to_tensors = state_dict_split.filename_to_tensors.items() + for shard_file, tensors in tqdm(filename_to_tensors, desc="Saving checkpoint shards"): + shard_state_dict = {tensor_name: state_dict[tensor_name] for tensor_name in tensors} + shard_state_dict = to_torch_tensor(shard_state_dict) + output_path = os.path.join(output_dir, shard_file) + if safe_serialization: + save_file(shard_state_dict, output_path, metadata={"format": "pt"}) + else: + torch.save(shard_state_dict, output_path) + # release the memory of current shard + for tensor_name in list(shard_state_dict.keys()): + del state_dict[tensor_name] + del shard_state_dict[tensor_name] + del shard_state_dict + gc.collect() + + # Save index if sharded + if state_dict_split.is_sharded: + index = { + "metadata": state_dict_split.metadata, + "weight_map": state_dict_split.tensor_to_filename, + } + save_index_file = "model.safetensors.index.json" if safe_serialization else "pytorch_model.bin.index.json" + save_index_file = os.path.join(output_dir, save_index_file) + with open(save_index_file, "w", encoding="utf-8") as f: + content = json.dumps(index, indent=2, sort_keys=True) + "\n" + f.write(content) + + +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_dir", + type=str, + help="directory to the pytorch fp32 state_dict output files" + "(e.g. path/checkpoint-12-output/)") + parser.add_argument( + "--max_shard_size", + type=str, + default="5GB", + help="The maximum size for a checkpoint before being sharded. Checkpoints shard will then be each of size" + "lower than this size. If expressed as a string, needs to be digits followed by a unit (like `5MB`" + "We default it to 5GB in order for models to be able to run easily on free-tier google colab instances" + "without CPU OOM issues.") + parser.add_argument( + "--safe_serialization", + default=False, + action='store_true', + help="Whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).") + 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_dir, + max_shard_size=args.max_shard_size, + safe_serialization=args.safe_serialization, + tag=args.tag, + exclude_frozen_parameters=args.exclude_frozen_parameters) diff --git a/config.json b/config.json new file mode 100644 index 0000000..3f8f5c0 --- /dev/null +++ b/config.json @@ -0,0 +1,29 @@ +{ + "architectures": [ + "LlamaForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "bos_token_id": 128000, + "eos_token_id": 128001, + "head_dim": 128, + "hidden_act": "silu", + "hidden_size": 4096, + "initializer_range": 0.02, + "intermediate_size": 14336, + "max_position_embeddings": 8192, + "mlp_bias": false, + "model_type": "llama", + "num_attention_heads": 32, + "num_hidden_layers": 32, + "num_key_value_heads": 8, + "pretraining_tp": 1, + "rms_norm_eps": 1e-05, + "rope_scaling": null, + "rope_theta": 500000.0, + "tie_word_embeddings": false, + "torch_dtype": "bfloat16", + "transformers_version": "4.54.1", + "use_cache": true, + "vocab_size": 128256 +} diff --git a/model-00001-of-00004.safetensors b/model-00001-of-00004.safetensors new file mode 100644 index 0000000..fe3fa95 --- /dev/null +++ b/model-00001-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8209180ff91c21e5237d8af84b424786491a3be181c7fdd56dfbbf618026f5ab +size 4976698672 diff --git a/model-00002-of-00004.safetensors b/model-00002-of-00004.safetensors new file mode 100644 index 0000000..b1ca729 --- /dev/null +++ b/model-00002-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:057704766c2933c8127ef9ff0f2d95990eaad4c77689f3ca38fc64e15b61462a +size 4999802720 diff --git a/model-00003-of-00004.safetensors b/model-00003-of-00004.safetensors new file mode 100644 index 0000000..62a518a --- /dev/null +++ b/model-00003-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dd5511a86272c495069570a333effca3e14edab8427d25d9f9d62034005354ad +size 4915916176 diff --git a/model-00004-of-00004.safetensors b/model-00004-of-00004.safetensors new file mode 100644 index 0000000..0d5a6a1 --- /dev/null +++ b/model-00004-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7107756fe86b7f389d0887d3517180fc28a3d56fc2e46efcc95cb8aadc7dc253 +size 1168138808 diff --git a/model.safetensors.index.json b/model.safetensors.index.json new file mode 100644 index 0000000..58a7ef4 --- /dev/null +++ b/model.safetensors.index.json @@ -0,0 +1,299 @@ +{ + "metadata": { + "total_parameters": 266240, + "total_size": 16060522496 + }, + "weight_map": { + "lm_head.weight": "model-00004-of-00004.safetensors", + "model.embed_tokens.weight": "model-00001-of-00004.safetensors", + "model.layers.0.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.0.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.1.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.10.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.10.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.10.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.11.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.11.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.12.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.12.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.13.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.13.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.14.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.14.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.15.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.15.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.16.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.16.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.17.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.17.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.18.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.18.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.19.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.19.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.2.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.2.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.20.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.20.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.20.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.20.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.20.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.20.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.20.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.20.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.20.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.21.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.21.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.21.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.22.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.22.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.23.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.23.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.24.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.24.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.25.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.25.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.26.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.26.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.27.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.27.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.28.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.28.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.29.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.29.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.3.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.3.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.30.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.30.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.30.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.input_layernorm.weight": "model-00004-of-00004.safetensors", + "model.layers.31.mlp.down_proj.weight": "model-00004-of-00004.safetensors", + "model.layers.31.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.post_attention_layernorm.weight": "model-00004-of-00004.safetensors", + "model.layers.31.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.4.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.4.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.5.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.6.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.7.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.8.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.9.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.9.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.9.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.norm.weight": "model-00004-of-00004.safetensors" + } +} diff --git a/special_tokens_map.json b/special_tokens_map.json new file mode 100644 index 0000000..e5b39b6 --- /dev/null +++ b/special_tokens_map.json @@ -0,0 +1,23 @@ +{ + "bos_token": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "eos_token": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "pad_token": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + } +} diff --git a/tokenizer.json b/tokenizer.json new file mode 100644 index 0000000..03aa64f --- /dev/null +++ b/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0968dcc0ee8e56c7dccd34a7f51f8065ea0cb9e2cc529e3243d1e5c0a4bdaa0c +size 17208754 diff --git a/tokenizer_config.json b/tokenizer_config.json new file mode 100644 index 0000000..877a9a9 --- /dev/null +++ b/tokenizer_config.json @@ -0,0 +1,2063 @@ +{ + "added_tokens_decoder": { + "128000": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128001": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128002": { + "content": "<|reserved_special_token_0|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128003": { + "content": "<|reserved_special_token_1|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128004": { + "content": "<|reserved_special_token_2|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128005": { + "content": "<|reserved_special_token_3|>", + "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": "<|reserved_special_token_4|>", + "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": "<|reserved_special_token_5|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128011": { + "content": "<|reserved_special_token_6|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128012": { + "content": "<|reserved_special_token_7|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128013": { + "content": "<|reserved_special_token_8|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128014": { + "content": "<|reserved_special_token_9|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128015": { + "content": "<|reserved_special_token_10|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128016": { + "content": "<|reserved_special_token_11|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128017": { + "content": "<|reserved_special_token_12|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128018": { + "content": "<|reserved_special_token_13|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128019": { + "content": "<|reserved_special_token_14|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128020": { + "content": "<|reserved_special_token_15|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128021": { + "content": "<|reserved_special_token_16|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128022": { + "content": "<|reserved_special_token_17|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128023": { + "content": "<|reserved_special_token_18|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128024": { + "content": "<|reserved_special_token_19|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128025": { + "content": "<|reserved_special_token_20|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128026": { + "content": "<|reserved_special_token_21|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128027": { + "content": "<|reserved_special_token_22|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128028": { + "content": "<|reserved_special_token_23|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128029": { + "content": "<|reserved_special_token_24|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128030": { + "content": "<|reserved_special_token_25|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128031": { + "content": "<|reserved_special_token_26|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128032": { + "content": "<|reserved_special_token_27|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128033": { + "content": "<|reserved_special_token_28|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128034": { + "content": "<|reserved_special_token_29|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128035": { + "content": "<|reserved_special_token_30|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128036": { + "content": "<|reserved_special_token_31|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128037": { + "content": "<|reserved_special_token_32|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128038": { + "content": "<|reserved_special_token_33|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128039": { + "content": "<|reserved_special_token_34|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128040": { + "content": "<|reserved_special_token_35|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128041": { + "content": "<|reserved_special_token_36|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128042": { + "content": "<|reserved_special_token_37|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128043": { + "content": "<|reserved_special_token_38|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128044": { + "content": "<|reserved_special_token_39|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128045": { + "content": "<|reserved_special_token_40|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128046": { + "content": "<|reserved_special_token_41|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128047": { + "content": "<|reserved_special_token_42|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128048": { + "content": "<|reserved_special_token_43|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128049": { + "content": "<|reserved_special_token_44|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128050": { + "content": "<|reserved_special_token_45|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128051": { + "content": "<|reserved_special_token_46|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128052": { + "content": "<|reserved_special_token_47|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128053": { + "content": "<|reserved_special_token_48|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128054": { + "content": "<|reserved_special_token_49|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128055": { + "content": "<|reserved_special_token_50|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128056": { + "content": "<|reserved_special_token_51|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128057": { + "content": "<|reserved_special_token_52|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128058": { + "content": "<|reserved_special_token_53|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128059": { + "content": "<|reserved_special_token_54|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128060": { + "content": "<|reserved_special_token_55|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128061": { + "content": "<|reserved_special_token_56|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128062": { + "content": "<|reserved_special_token_57|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128063": { + "content": "<|reserved_special_token_58|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128064": { + "content": "<|reserved_special_token_59|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128065": { + "content": "<|reserved_special_token_60|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128066": { + "content": "<|reserved_special_token_61|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128067": { + "content": "<|reserved_special_token_62|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128068": { + "content": "<|reserved_special_token_63|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128069": { + "content": "<|reserved_special_token_64|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128070": { + "content": "<|reserved_special_token_65|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128071": { + "content": "<|reserved_special_token_66|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128072": { + "content": "<|reserved_special_token_67|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128073": { + "content": "<|reserved_special_token_68|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128074": { + "content": "<|reserved_special_token_69|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128075": { + "content": "<|reserved_special_token_70|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128076": { + "content": "<|reserved_special_token_71|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128077": { + "content": "<|reserved_special_token_72|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128078": { + "content": "<|reserved_special_token_73|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128079": { + "content": "<|reserved_special_token_74|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128080": { + "content": "<|reserved_special_token_75|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128081": { + "content": "<|reserved_special_token_76|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128082": { + "content": "<|reserved_special_token_77|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128083": { + "content": "<|reserved_special_token_78|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128084": { + "content": "<|reserved_special_token_79|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128085": { + "content": "<|reserved_special_token_80|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128086": { + "content": "<|reserved_special_token_81|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128087": { + "content": "<|reserved_special_token_82|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128088": { + "content": "<|reserved_special_token_83|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128089": { + "content": "<|reserved_special_token_84|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128090": { + "content": "<|reserved_special_token_85|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128091": { + "content": "<|reserved_special_token_86|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128092": { + "content": "<|reserved_special_token_87|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128093": { + "content": "<|reserved_special_token_88|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128094": { + "content": "<|reserved_special_token_89|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128095": { + "content": "<|reserved_special_token_90|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128096": { + "content": "<|reserved_special_token_91|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128097": { + "content": "<|reserved_special_token_92|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128098": { + "content": "<|reserved_special_token_93|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128099": { + "content": "<|reserved_special_token_94|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128100": { + "content": "<|reserved_special_token_95|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128101": { + "content": "<|reserved_special_token_96|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128102": { + "content": "<|reserved_special_token_97|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128103": { + "content": "<|reserved_special_token_98|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128104": { + "content": "<|reserved_special_token_99|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128105": { + "content": "<|reserved_special_token_100|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128106": { + "content": "<|reserved_special_token_101|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128107": { + "content": "<|reserved_special_token_102|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128108": { + "content": "<|reserved_special_token_103|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128109": { + "content": "<|reserved_special_token_104|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128110": { + "content": "<|reserved_special_token_105|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128111": { + "content": "<|reserved_special_token_106|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128112": { + "content": "<|reserved_special_token_107|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128113": { + "content": "<|reserved_special_token_108|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128114": { + "content": "<|reserved_special_token_109|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128115": { + "content": "<|reserved_special_token_110|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128116": { + "content": "<|reserved_special_token_111|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128117": { + "content": "<|reserved_special_token_112|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128118": { + "content": "<|reserved_special_token_113|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128119": { + "content": "<|reserved_special_token_114|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128120": { + "content": "<|reserved_special_token_115|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128121": { + "content": "<|reserved_special_token_116|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128122": { + "content": "<|reserved_special_token_117|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128123": { + "content": "<|reserved_special_token_118|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128124": { + "content": "<|reserved_special_token_119|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128125": { + "content": "<|reserved_special_token_120|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128126": { + "content": "<|reserved_special_token_121|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128127": { + "content": "<|reserved_special_token_122|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128128": { + "content": "<|reserved_special_token_123|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128129": { + "content": "<|reserved_special_token_124|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128130": { + "content": "<|reserved_special_token_125|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128131": { + "content": "<|reserved_special_token_126|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128132": { + "content": "<|reserved_special_token_127|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128133": { + "content": "<|reserved_special_token_128|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128134": { + "content": "<|reserved_special_token_129|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128135": { + "content": "<|reserved_special_token_130|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128136": { + "content": "<|reserved_special_token_131|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128137": { + "content": "<|reserved_special_token_132|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128138": { + "content": "<|reserved_special_token_133|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128139": { + "content": "<|reserved_special_token_134|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128140": { + "content": "<|reserved_special_token_135|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128141": { + "content": "<|reserved_special_token_136|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128142": { + "content": "<|reserved_special_token_137|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128143": { + "content": "<|reserved_special_token_138|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128144": { + "content": "<|reserved_special_token_139|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128145": { + "content": "<|reserved_special_token_140|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128146": { + "content": "<|reserved_special_token_141|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128147": { + "content": "<|reserved_special_token_142|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128148": { + "content": "<|reserved_special_token_143|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128149": { + "content": "<|reserved_special_token_144|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128150": { + "content": "<|reserved_special_token_145|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128151": { + "content": "<|reserved_special_token_146|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128152": { + "content": "<|reserved_special_token_147|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128153": { + "content": "<|reserved_special_token_148|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128154": { + "content": "<|reserved_special_token_149|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128155": { + "content": "<|reserved_special_token_150|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128156": { + "content": "<|reserved_special_token_151|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128157": { + "content": "<|reserved_special_token_152|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128158": { + "content": "<|reserved_special_token_153|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128159": { + "content": "<|reserved_special_token_154|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128160": { + "content": "<|reserved_special_token_155|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128161": { + "content": "<|reserved_special_token_156|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128162": { + "content": "<|reserved_special_token_157|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128163": { + "content": "<|reserved_special_token_158|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128164": { + "content": "<|reserved_special_token_159|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128165": { + "content": "<|reserved_special_token_160|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128166": { + "content": "<|reserved_special_token_161|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128167": { + "content": "<|reserved_special_token_162|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128168": { + "content": "<|reserved_special_token_163|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128169": { + "content": "<|reserved_special_token_164|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128170": { + "content": "<|reserved_special_token_165|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128171": { + "content": "<|reserved_special_token_166|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128172": { + "content": "<|reserved_special_token_167|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128173": { + "content": "<|reserved_special_token_168|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128174": { + "content": "<|reserved_special_token_169|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128175": { + "content": "<|reserved_special_token_170|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128176": { + "content": "<|reserved_special_token_171|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128177": { + "content": "<|reserved_special_token_172|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128178": { + "content": "<|reserved_special_token_173|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128179": { + "content": "<|reserved_special_token_174|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128180": { + "content": "<|reserved_special_token_175|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128181": { + "content": "<|reserved_special_token_176|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128182": { + "content": "<|reserved_special_token_177|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128183": { + "content": "<|reserved_special_token_178|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128184": { + "content": "<|reserved_special_token_179|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128185": { + "content": "<|reserved_special_token_180|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128186": { + "content": "<|reserved_special_token_181|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128187": { + "content": "<|reserved_special_token_182|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128188": { + "content": "<|reserved_special_token_183|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128189": { + "content": "<|reserved_special_token_184|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128190": { + "content": "<|reserved_special_token_185|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128191": { + "content": "<|reserved_special_token_186|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128192": { + "content": "<|reserved_special_token_187|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128193": { + "content": "<|reserved_special_token_188|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128194": { + "content": "<|reserved_special_token_189|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128195": { + "content": "<|reserved_special_token_190|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128196": { + "content": "<|reserved_special_token_191|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128197": { + "content": "<|reserved_special_token_192|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128198": { + "content": "<|reserved_special_token_193|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128199": { + "content": "<|reserved_special_token_194|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128200": { + "content": "<|reserved_special_token_195|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128201": { + "content": "<|reserved_special_token_196|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128202": { + "content": "<|reserved_special_token_197|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128203": { + "content": "<|reserved_special_token_198|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128204": { + "content": "<|reserved_special_token_199|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128205": { + "content": "<|reserved_special_token_200|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128206": { + "content": "<|reserved_special_token_201|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128207": { + "content": "<|reserved_special_token_202|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128208": { + "content": "<|reserved_special_token_203|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128209": { + "content": "<|reserved_special_token_204|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128210": { + "content": "<|reserved_special_token_205|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128211": { + "content": "<|reserved_special_token_206|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128212": { + "content": "<|reserved_special_token_207|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128213": { + "content": "<|reserved_special_token_208|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128214": { + "content": "<|reserved_special_token_209|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128215": { + "content": "<|reserved_special_token_210|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128216": { + "content": "<|reserved_special_token_211|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128217": { + "content": "<|reserved_special_token_212|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128218": { + "content": "<|reserved_special_token_213|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128219": { + "content": "<|reserved_special_token_214|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128220": { + "content": "<|reserved_special_token_215|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128221": { + "content": "<|reserved_special_token_216|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128222": { + "content": "<|reserved_special_token_217|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128223": { + "content": "<|reserved_special_token_218|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128224": { + "content": "<|reserved_special_token_219|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128225": { + "content": "<|reserved_special_token_220|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128226": { + "content": "<|reserved_special_token_221|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128227": { + "content": "<|reserved_special_token_222|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128228": { + "content": "<|reserved_special_token_223|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128229": { + "content": "<|reserved_special_token_224|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128230": { + "content": "<|reserved_special_token_225|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128231": { + "content": "<|reserved_special_token_226|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128232": { + "content": "<|reserved_special_token_227|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128233": { + "content": "<|reserved_special_token_228|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128234": { + "content": "<|reserved_special_token_229|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128235": { + "content": "<|reserved_special_token_230|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128236": { + "content": "<|reserved_special_token_231|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128237": { + "content": "<|reserved_special_token_232|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128238": { + "content": "<|reserved_special_token_233|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128239": { + "content": "<|reserved_special_token_234|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128240": { + "content": "<|reserved_special_token_235|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128241": { + "content": "<|reserved_special_token_236|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128242": { + "content": "<|reserved_special_token_237|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128243": { + "content": "<|reserved_special_token_238|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128244": { + "content": "<|reserved_special_token_239|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128245": { + "content": "<|reserved_special_token_240|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128246": { + "content": "<|reserved_special_token_241|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128247": { + "content": "<|reserved_special_token_242|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128248": { + "content": "<|reserved_special_token_243|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128249": { + "content": "<|reserved_special_token_244|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128250": { + "content": "<|reserved_special_token_245|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128251": { + "content": "<|reserved_special_token_246|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128252": { + "content": "<|reserved_special_token_247|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128253": { + "content": "<|reserved_special_token_248|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128254": { + "content": "<|reserved_special_token_249|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128255": { + "content": "<|reserved_special_token_250|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + } + }, + "bos_token": "<|begin_of_text|>", + "clean_up_tokenization_spaces": true, + "eos_token": "<|end_of_text|>", + "extra_special_tokens": {}, + "model_input_names": [ + "input_ids", + "attention_mask" + ], + "model_max_length": 32768, + "pad_token": "<|end_of_text|>", + "tokenizer_class": "PreTrainedTokenizerFast" +} diff --git a/training_args.bin b/training_args.bin new file mode 100644 index 0000000..eb22667 --- /dev/null +++ b/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ded8eebbc35e0c697502403c9c9d2a0acb4a3f38810fc064a38066719f316edb +size 8785