commit cf57f9da60193074820fda597097461f8c8ec326 Author: ModelHub XC Date: Sun May 24 10:54:16 2026 +0800 初始化项目，由ModelHub XC社区提供模型 Model: HCY123902/llama-3-8b-inst-dpo-on-p-tw15-beta-1e-0 Source: Original Platform diff --git a/.gitattributes b/.gitattributes new file mode 100644 index 0000000..ea05d3b --- /dev/null +++ b/.gitattributes @@ -0,0 +1,40 @@ +*.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 +checkpoint-1000/tokenizer.json filter=lfs diff=lfs merge=lfs -text +checkpoint-1500/tokenizer.json filter=lfs diff=lfs merge=lfs -text +checkpoint-1872/tokenizer.json filter=lfs diff=lfs merge=lfs -text diff --git a/README.md b/README.md new file mode 100644 index 0000000..0b36521 --- /dev/null +++ b/README.md @@ -0,0 +1,69 @@ +--- +base_model: meta-llama/Meta-Llama-3-8B-Instruct +library_name: transformers +model_name: llama-3-8b-inst-dpo-on-p-tw15-beta-1e-0 +tags: +- generated_from_trainer +- trl +- dpo +licence: license +--- + +# Model Card for llama-3-8b-inst-dpo-on-p-tw15-beta-1e-0 + +This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct). +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-inst-dpo-on-p-tw15-beta-1e-0", 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/0g9xx1c9) + + +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..ceed8ae --- /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": 128009, + "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..38e1b41 --- /dev/null +++ b/checkpoint-1000/generation_config.json @@ -0,0 +1,12 @@ +{ + "bos_token_id": 128000, + "do_sample": true, + "eos_token_id": [ + 128001, + 128009 + ], + "max_length": 4096, + "temperature": 0.6, + "top_p": 0.9, + "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..c0e4329 --- /dev/null +++ b/checkpoint-1000/model-00001-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6650235aceb597b23b7c5872c795957e2dcb19b38568ff8044be60e89d58c1ed +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..c16df20 --- /dev/null +++ b/checkpoint-1000/model-00002-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:40a5293ae61119c44400715387e27bc1ec8f5b3369d57c0c99cf4a545564f6ff +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..9e53fff --- /dev/null +++ b/checkpoint-1000/model-00003-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fc127c47222d12f837b85ebbb98785f580bf3ac130b88211c1fc6a0f150f507e +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..e7f6697 --- /dev/null +++ b/checkpoint-1000/model-00004-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cd9a67a3c35c9f806334128707822ccd44c6b3755f5924418ae51d349f2ae434 +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..7f3a109 --- /dev/null +++ b/checkpoint-1000/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6d02245e75fa3caf978017a9ec84ec3281731afc6f194fa9f3f9d1519cbed572 +size 1465 diff --git a/checkpoint-1000/special_tokens_map.json b/checkpoint-1000/special_tokens_map.json new file mode 100644 index 0000000..b43be96 --- /dev/null +++ b/checkpoint-1000/special_tokens_map.json @@ -0,0 +1,17 @@ +{ + "bos_token": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "eos_token": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "pad_token": "<|eot_id|>" +} diff --git a/checkpoint-1000/tokenizer.json b/checkpoint-1000/tokenizer.json new file mode 100644 index 0000000..86a3394 --- /dev/null +++ b/checkpoint-1000/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3c5cf44023714fb39b05e71e425f8d7b92805ff73f7988b083b8c87f0bf87393 +size 17209961 diff --git a/checkpoint-1000/tokenizer_config.json b/checkpoint-1000/tokenizer_config.json new file mode 100644 index 0000000..34d134f --- /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": "<|eot_id|>", + "extra_special_tokens": {}, + "model_input_names": [ + "input_ids", + "attention_mask" + ], + "model_max_length": 1000000000000000019884624838656, + "pad_token": "<|eot_id|>", + "tokenizer_class": "PreTrainedTokenizerFast" +} diff --git a/checkpoint-1000/trainer_state.json b/checkpoint-1000/trainer_state.json new file mode 100644 index 0000000..96e6b47 --- /dev/null +++ b/checkpoint-1000/trainer_state.json @@ -0,0 +1,2199 @@ +{ + "best_global_step": null, + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 0.5344378381989445, + "eval_steps": 500, + "global_step": 1000, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.0005344378381989445, + "grad_norm": 63.52920150756836, + "learning_rate": 0.0, + "logits/chosen": -0.144775390625, + "logits/rejected": -0.1036376953125, + "logps/chosen": -110.828125, + "logps/rejected": -116.828125, + "logps/weighted_chosen": -0.2086181640625, + "logps/weighted_rejected": -1.271392822265625, + "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": 0.005344378381989445, + "grad_norm": 102.77810668945312, + "learning_rate": 4.787234042553191e-08, + "logits/chosen": -0.2549913227558136, + "logits/rejected": -0.24070316553115845, + "logps/chosen": -135.32899475097656, + "logps/rejected": -138.99305725097656, + "logps/weighted_chosen": -0.8547499179840088, + "logps/weighted_rejected": -0.9027608036994934, + "loss": 0.6928, + "rewards/accuracies": 0.2951388955116272, + "rewards/chosen": -0.0164930559694767, + "rewards/margins": -0.0349392369389534, + "rewards/rejected": 0.0184461809694767, + "rewards/weighted_accuracies": 0.4027777910232544, + "rewards/weighted_chosen": -0.002239651046693325, + "rewards/weighted_margins": 0.0028160943184047937, + "rewards/weighted_rejected": -0.0050794812850654125, + "step": 10 + }, + { + "epoch": 0.01068875676397889, + "grad_norm": 9.201544761657715, + "learning_rate": 1.0106382978723404e-07, + "logits/chosen": -0.19488906860351562, + "logits/rejected": -0.183399960398674, + "logps/chosen": -116.0374984741211, + "logps/rejected": -115.5054702758789, + "logps/weighted_chosen": -0.6929916143417358, + "logps/weighted_rejected": -0.6838119626045227, + "loss": 0.6917, + "rewards/accuracies": 0.38749998807907104, + "rewards/chosen": 0.01796874962747097, + "rewards/margins": 0.03007812425494194, + "rewards/rejected": -0.01210937462747097, + "rewards/weighted_accuracies": 0.4781250059604645, + "rewards/weighted_chosen": 0.0022293091751635075, + "rewards/weighted_margins": 0.00339679722674191, + "rewards/weighted_rejected": -0.0011629104847088456, + "step": 20 + }, + { + "epoch": 0.016033135145968335, + "grad_norm": 7.109442234039307, + "learning_rate": 1.5425531914893615e-07, + "logits/chosen": -0.1776607483625412, + "logits/rejected": -0.17884215712547302, + "logps/chosen": -123.2867202758789, + "logps/rejected": -129.75, + "logps/weighted_chosen": -0.7567901611328125, + "logps/weighted_rejected": -1.2207733392715454, + "loss": 0.6913, + "rewards/accuracies": 0.3968749940395355, + "rewards/chosen": -0.0074218749068677425, + "rewards/margins": 0.04843749850988388, + "rewards/rejected": -0.05585937574505806, + "rewards/weighted_accuracies": 0.5, + "rewards/weighted_chosen": 0.0007101058727130294, + "rewards/weighted_margins": 0.004384422209113836, + "rewards/weighted_rejected": -0.0036745071411132812, + "step": 30 + }, + { + "epoch": 0.02137751352795778, + "grad_norm": 6.0048956871032715, + "learning_rate": 2.074468085106383e-07, + "logits/chosen": -0.23842163383960724, + "logits/rejected": -0.2298881560564041, + "logps/chosen": -127.34062194824219, + "logps/rejected": -128.1710968017578, + "logps/weighted_chosen": -0.8051223754882812, + "logps/weighted_rejected": -0.5736129879951477, + "loss": 0.6934, + "rewards/accuracies": 0.34062498807907104, + "rewards/chosen": -0.01992187462747097, + "rewards/margins": -0.05351562425494194, + "rewards/rejected": 0.03359375149011612, + "rewards/weighted_accuracies": 0.4593749940395355, + "rewards/weighted_chosen": -7.83920258982107e-05, + "rewards/weighted_margins": -0.0009832382202148438, + "rewards/weighted_rejected": 0.000904083251953125, + "step": 40 + }, + { + "epoch": 0.026721891909947223, + "grad_norm": 32.7233772277832, + "learning_rate": 2.6063829787234044e-07, + "logits/chosen": -0.18835143744945526, + "logits/rejected": -0.170085147023201, + "logps/chosen": -120.88594055175781, + "logps/rejected": -121.3515625, + "logps/weighted_chosen": -0.6867736577987671, + "logps/weighted_rejected": -1.138249158859253, + "loss": 0.6903, + "rewards/accuracies": 0.3656249940395355, + "rewards/chosen": 0.03554687649011612, + "rewards/margins": -0.03164062649011612, + "rewards/rejected": 0.06718750298023224, + "rewards/weighted_accuracies": 0.515625, + "rewards/weighted_chosen": -0.006287765689194202, + "rewards/weighted_margins": 0.0064636231400072575, + "rewards/weighted_rejected": -0.01274185162037611, + "step": 50 + }, + { + "epoch": 0.03206627029193667, + "grad_norm": 55.560707092285156, + "learning_rate": 3.1382978723404253e-07, + "logits/chosen": -0.1308601349592209, + "logits/rejected": -0.1308906525373459, + "logps/chosen": -116.5875015258789, + "logps/rejected": -119.5406265258789, + "logps/weighted_chosen": -0.8019469976425171, + "logps/weighted_rejected": -0.8834930658340454, + "loss": 0.7002, + "rewards/accuracies": 0.3343749940395355, + "rewards/chosen": 0.03476562350988388, + "rewards/margins": -0.06171875074505806, + "rewards/rejected": 0.09648437798023224, + "rewards/weighted_accuracies": 0.4781250059604645, + "rewards/weighted_chosen": -0.01903533935546875, + "rewards/weighted_margins": -2.880096508306451e-05, + "rewards/weighted_rejected": -0.019007300958037376, + "step": 60 + }, + { + "epoch": 0.037410648673926114, + "grad_norm": 20.22484588623047, + "learning_rate": 3.6702127659574467e-07, + "logits/chosen": -0.20572929084300995, + "logits/rejected": -0.19457244873046875, + "logps/chosen": -114.4507827758789, + "logps/rejected": -118.63594055175781, + "logps/weighted_chosen": -0.6070801019668579, + "logps/weighted_rejected": -0.6768280267715454, + "loss": 0.6911, + "rewards/accuracies": 0.42500001192092896, + "rewards/chosen": 0.16484375298023224, + "rewards/margins": 0.07148437201976776, + "rewards/rejected": 0.09335937350988388, + "rewards/weighted_accuracies": 0.4749999940395355, + "rewards/weighted_chosen": -0.010281562805175781, + "rewards/weighted_margins": 0.007259177975356579, + "rewards/weighted_rejected": -0.017521286383271217, + "step": 70 + }, + { + "epoch": 0.04275502705591556, + "grad_norm": 15.305061340332031, + "learning_rate": 4.202127659574468e-07, + "logits/chosen": -0.19508513808250427, + "logits/rejected": -0.12093047797679901, + "logps/chosen": -112.57734680175781, + "logps/rejected": -112.77734375, + "logps/weighted_chosen": -0.4597320556640625, + "logps/weighted_rejected": -0.9711883664131165, + "loss": 0.6952, + "rewards/accuracies": 0.37812501192092896, + "rewards/chosen": 0.04570312425494194, + "rewards/margins": 0.01914062537252903, + "rewards/rejected": 0.02656250074505806, + "rewards/weighted_accuracies": 0.4625000059604645, + "rewards/weighted_chosen": -0.02759857103228569, + "rewards/weighted_margins": 0.021048355847597122, + "rewards/weighted_rejected": -0.04872245714068413, + "step": 80 + }, + { + "epoch": 0.048099405437905, + "grad_norm": 11.991273880004883, + "learning_rate": 4.734042553191489e-07, + "logits/chosen": -0.17391815781593323, + "logits/rejected": -0.11656417697668076, + "logps/chosen": -119.3070297241211, + "logps/rejected": -118.51249694824219, + "logps/weighted_chosen": -1.0416854619979858, + "logps/weighted_rejected": -1.5859466791152954, + "loss": 0.6736, + "rewards/accuracies": 0.4375, + "rewards/chosen": -0.08164062350988388, + "rewards/margins": 0.341796875, + "rewards/rejected": -0.4234375059604645, + "rewards/weighted_accuracies": 0.515625, + "rewards/weighted_chosen": -0.04084587097167969, + "rewards/weighted_margins": 0.10567589104175568, + "rewards/weighted_rejected": -0.14661255478858948, + "step": 90 + }, + { + "epoch": 0.053443783819894446, + "grad_norm": 17.195886611938477, + "learning_rate": 5.26595744680851e-07, + "logits/chosen": -0.204203799366951, + "logits/rejected": -0.14691162109375, + "logps/chosen": -125.05155944824219, + "logps/rejected": -125.3671875, + "logps/weighted_chosen": -0.7374237179756165, + "logps/weighted_rejected": -0.7348953485488892, + "loss": 0.7064, + "rewards/accuracies": 0.4281249940395355, + "rewards/chosen": -0.3832031190395355, + "rewards/margins": 0.15585938096046448, + "rewards/rejected": -0.5390625, + "rewards/weighted_accuracies": 0.46875, + "rewards/weighted_chosen": -0.060018159449100494, + "rewards/weighted_margins": 0.0024120330344885588, + "rewards/weighted_rejected": -0.06242714077234268, + "step": 100 + }, + { + "epoch": 0.058788162201883896, + "grad_norm": 9.103920936584473, + "learning_rate": 5.797872340425531e-07, + "logits/chosen": -0.2138931304216385, + "logits/rejected": -0.19011077284812927, + "logps/chosen": -118.6820297241211, + "logps/rejected": -121.28593444824219, + "logps/weighted_chosen": -0.4821228086948395, + "logps/weighted_rejected": -0.7759445309638977, + "loss": 0.6939, + "rewards/accuracies": 0.4281249940395355, + "rewards/chosen": -0.896484375, + "rewards/margins": 0.09531249850988388, + "rewards/rejected": -0.9917968511581421, + "rewards/weighted_accuracies": 0.528124988079071, + "rewards/weighted_chosen": -0.05413341522216797, + "rewards/weighted_margins": 0.04503212124109268, + "rewards/weighted_rejected": -0.09918193519115448, + "step": 110 + }, + { + "epoch": 0.06413254058387334, + "grad_norm": 8.539896011352539, + "learning_rate": 6.329787234042553e-07, + "logits/chosen": -0.17510680854320526, + "logits/rejected": -0.14726562798023224, + "logps/chosen": -122.58906555175781, + "logps/rejected": -128.4656219482422, + "logps/weighted_chosen": -0.9249496459960938, + "logps/weighted_rejected": -0.762683093547821, + "loss": 0.7325, + "rewards/accuracies": 0.4375, + "rewards/chosen": -1.068359375, + "rewards/margins": 0.28398436307907104, + "rewards/rejected": -1.3523437976837158, + "rewards/weighted_accuracies": 0.4781250059604645, + "rewards/weighted_chosen": -0.098027803003788, + "rewards/weighted_margins": -0.029563140124082565, + "rewards/weighted_rejected": -0.06836700439453125, + "step": 120 + }, + { + "epoch": 0.06947691896586278, + "grad_norm": 15.564044952392578, + "learning_rate": 6.861702127659574e-07, + "logits/chosen": -0.20233002305030823, + "logits/rejected": -0.18396148085594177, + "logps/chosen": -122.2671890258789, + "logps/rejected": -123.8382797241211, + "logps/weighted_chosen": -0.7501281499862671, + "logps/weighted_rejected": -0.821154773235321, + "loss": 0.7078, + "rewards/accuracies": 0.4000000059604645, + "rewards/chosen": -1.45703125, + "rewards/margins": 0.29179686307907104, + "rewards/rejected": -1.7488281726837158, + "rewards/weighted_accuracies": 0.5062500238418579, + "rewards/weighted_chosen": -0.007884788326919079, + "rewards/weighted_margins": 0.006314086727797985, + "rewards/weighted_rejected": -0.014200759120285511, + "step": 130 + }, + { + "epoch": 0.07482129734785223, + "grad_norm": 9.176558494567871, + "learning_rate": 7.393617021276596e-07, + "logits/chosen": -0.16455383598804474, + "logits/rejected": -0.14660033583641052, + "logps/chosen": -116.328125, + "logps/rejected": -119.9664077758789, + "logps/weighted_chosen": -0.9873703122138977, + "logps/weighted_rejected": -0.6730011105537415, + "loss": 0.7174, + "rewards/accuracies": 0.4749999940395355, + "rewards/chosen": -1.6007812023162842, + "rewards/margins": 0.533203125, + "rewards/rejected": -2.133984327316284, + "rewards/weighted_accuracies": 0.515625, + "rewards/weighted_chosen": 0.02084655687212944, + "rewards/weighted_margins": 0.05102996900677681, + "rewards/weighted_rejected": -0.030106354504823685, + "step": 140 + }, + { + "epoch": 0.08016567572984168, + "grad_norm": 10.172534942626953, + "learning_rate": 7.925531914893616e-07, + "logits/chosen": -0.1509498655796051, + "logits/rejected": -0.13156357407569885, + "logps/chosen": -120.0562515258789, + "logps/rejected": -117.85546875, + "logps/weighted_chosen": -0.8198806643486023, + "logps/weighted_rejected": -0.7782562375068665, + "loss": 0.7005, + "rewards/accuracies": 0.46562498807907104, + "rewards/chosen": -2.0234375, + "rewards/margins": 0.607421875, + "rewards/rejected": -2.630859375, + "rewards/weighted_accuracies": 0.543749988079071, + "rewards/weighted_chosen": -0.0781441181898117, + "rewards/weighted_margins": 0.034273337572813034, + "rewards/weighted_rejected": -0.11248989403247833, + "step": 150 + }, + { + "epoch": 0.08551005411183112, + "grad_norm": 25.879701614379883, + "learning_rate": 8.457446808510637e-07, + "logits/chosen": -0.17474937438964844, + "logits/rejected": -0.17009583115577698, + "logps/chosen": -121.2367172241211, + "logps/rejected": -121.62109375, + "logps/weighted_chosen": -0.8304061889648438, + "logps/weighted_rejected": -0.866925060749054, + "loss": 0.7072, + "rewards/accuracies": 0.4781250059604645, + "rewards/chosen": -2.219531297683716, + "rewards/margins": 0.716796875, + "rewards/rejected": -2.936328172683716, + "rewards/weighted_accuracies": 0.47187501192092896, + "rewards/weighted_chosen": -0.07932891696691513, + "rewards/weighted_margins": 0.017035294324159622, + "rewards/weighted_rejected": -0.09644661098718643, + "step": 160 + }, + { + "epoch": 0.09085443249382057, + "grad_norm": 9.220558166503906, + "learning_rate": 8.989361702127659e-07, + "logits/chosen": -0.22574539482593536, + "logits/rejected": -0.19882050156593323, + "logps/chosen": -120.62968444824219, + "logps/rejected": -124.78437805175781, + "logps/weighted_chosen": -0.9401000738143921, + "logps/weighted_rejected": -0.8424423336982727, + "loss": 0.7174, + "rewards/accuracies": 0.4937500059604645, + "rewards/chosen": -2.4033203125, + "rewards/margins": 0.613085925579071, + "rewards/rejected": -3.016406297683716, + "rewards/weighted_accuracies": 0.48750001192092896, + "rewards/weighted_chosen": -0.04837837070226669, + "rewards/weighted_margins": 0.04595222324132919, + "rewards/weighted_rejected": -0.09437103569507599, + "step": 170 + }, + { + "epoch": 0.09619881087581, + "grad_norm": 55.62498092651367, + "learning_rate": 9.52127659574468e-07, + "logits/chosen": -0.2216789275407791, + "logits/rejected": -0.20373383164405823, + "logps/chosen": -120.3851547241211, + "logps/rejected": -121.2328109741211, + "logps/weighted_chosen": -0.435232549905777, + "logps/weighted_rejected": -0.5373809933662415, + "loss": 0.7302, + "rewards/accuracies": 0.47187501192092896, + "rewards/chosen": -1.338281273841858, + "rewards/margins": 0.3658203184604645, + "rewards/rejected": -1.7041015625, + "rewards/weighted_accuracies": 0.53125, + "rewards/weighted_chosen": -0.08163909614086151, + "rewards/weighted_margins": -0.008893203921616077, + "rewards/weighted_rejected": -0.07281456142663956, + "step": 180 + }, + { + "epoch": 0.10154318925779945, + "grad_norm": 39.38788986206055, + "learning_rate": 9.99999129927477e-07, + "logits/chosen": -0.19243164360523224, + "logits/rejected": -0.15758666396141052, + "logps/chosen": -121.8960952758789, + "logps/rejected": -123.01094055175781, + "logps/weighted_chosen": -0.953173816204071, + "logps/weighted_rejected": -1.0853087902069092, + "loss": 0.6702, + "rewards/accuracies": 0.48750001192092896, + "rewards/chosen": -2.100390672683716, + "rewards/margins": 0.662890613079071, + "rewards/rejected": -2.7632813453674316, + "rewards/weighted_accuracies": 0.578125, + "rewards/weighted_chosen": -0.09999503940343857, + "rewards/weighted_margins": 0.12822990119457245, + "rewards/weighted_rejected": -0.22818031907081604, + "step": 190 + }, + { + "epoch": 0.10688756763978889, + "grad_norm": 12.883384704589844, + "learning_rate": 9.99894724888679e-07, + "logits/chosen": -0.17392578721046448, + "logits/rejected": -0.147491455078125, + "logps/chosen": -129.53750610351562, + "logps/rejected": -128.95703125, + "logps/weighted_chosen": -0.7322753667831421, + "logps/weighted_rejected": -1.0728164911270142, + "loss": 0.6777, + "rewards/accuracies": 0.4749999940395355, + "rewards/chosen": -3.215625047683716, + "rewards/margins": 0.9507812261581421, + "rewards/rejected": -4.166406154632568, + "rewards/weighted_accuracies": 0.5375000238418579, + "rewards/weighted_chosen": -0.1104557067155838, + "rewards/weighted_margins": 0.13128623366355896, + "rewards/weighted_rejected": -0.2415626496076584, + "step": 200 + }, + { + "epoch": 0.11223194602177834, + "grad_norm": 17.96660041809082, + "learning_rate": 9.996163469793475e-07, + "logits/chosen": -0.18083801865577698, + "logits/rejected": -0.12766647338867188, + "logps/chosen": -128.1953125, + "logps/rejected": -123.03047180175781, + "logps/weighted_chosen": -0.8699920773506165, + "logps/weighted_rejected": -0.8845611810684204, + "loss": 0.7572, + "rewards/accuracies": 0.5625, + "rewards/chosen": -3.5113282203674316, + "rewards/margins": 1.3683593273162842, + "rewards/rejected": -4.879296779632568, + "rewards/weighted_accuracies": 0.5531250238418579, + "rewards/weighted_chosen": -0.07919616997241974, + "rewards/weighted_margins": 0.04895629733800888, + "rewards/weighted_rejected": -0.12815475463867188, + "step": 210 + }, + { + "epoch": 0.11757632440376779, + "grad_norm": 63.790077209472656, + "learning_rate": 9.991640930802883e-07, + "logits/chosen": -0.14619751274585724, + "logits/rejected": -0.13860473036766052, + "logps/chosen": -126.61015319824219, + "logps/rejected": -129.9773406982422, + "logps/weighted_chosen": -1.0776779651641846, + "logps/weighted_rejected": -0.9309722781181335, + "loss": 0.7107, + "rewards/accuracies": 0.5531250238418579, + "rewards/chosen": -4.296484470367432, + "rewards/margins": 1.10546875, + "rewards/rejected": -5.40234375, + "rewards/weighted_accuracies": 0.5406249761581421, + "rewards/weighted_chosen": -0.013729858212172985, + "rewards/weighted_margins": 0.14467468857765198, + "rewards/weighted_rejected": -0.1582542359828949, + "step": 220 + }, + { + "epoch": 0.12292070278575723, + "grad_norm": 9.634794235229492, + "learning_rate": 9.98538120584459e-07, + "logits/chosen": -0.1792152374982834, + "logits/rejected": -0.1362098753452301, + "logps/chosen": -133.7140655517578, + "logps/rejected": -137.7546844482422, + "logps/weighted_chosen": -0.6676223874092102, + "logps/weighted_rejected": -0.9518234133720398, + "loss": 0.7614, + "rewards/accuracies": 0.515625, + "rewards/chosen": -4.493554592132568, + "rewards/margins": 1.0623047351837158, + "rewards/rejected": -5.555859565734863, + "rewards/weighted_accuracies": 0.565625011920929, + "rewards/weighted_chosen": -0.1120578795671463, + "rewards/weighted_margins": 0.031700897961854935, + "rewards/weighted_rejected": -0.14376182854175568, + "step": 230 + }, + { + "epoch": 0.12826508116774668, + "grad_norm": 8.717682838439941, + "learning_rate": 9.977386473421917e-07, + "logits/chosen": -0.09295501559972763, + "logits/rejected": -0.07836151123046875, + "logps/chosen": -121.78202819824219, + "logps/rejected": -126.39453125, + "logps/weighted_chosen": -0.8116515874862671, + "logps/weighted_rejected": -1.2165496349334717, + "loss": 0.6731, + "rewards/accuracies": 0.5874999761581421, + "rewards/chosen": -7.553515434265137, + "rewards/margins": 1.669921875, + "rewards/rejected": -9.224218368530273, + "rewards/weighted_accuracies": 0.534375011920929, + "rewards/weighted_chosen": -0.1755084991455078, + "rewards/weighted_margins": 0.15414047241210938, + "rewards/weighted_rejected": -0.32989272475242615, + "step": 240 + }, + { + "epoch": 0.13360945954973613, + "grad_norm": 35.197269439697266, + "learning_rate": 9.96765951585378e-07, + "logits/chosen": -0.021549224853515625, + "logits/rejected": 0.00345611572265625, + "logps/chosen": -123.0296859741211, + "logps/rejected": -127.9671859741211, + "logps/weighted_chosen": -1.186151146888733, + "logps/weighted_rejected": -1.386474609375, + "loss": 0.7026, + "rewards/accuracies": 0.5625, + "rewards/chosen": -9.133593559265137, + "rewards/margins": 1.150781273841858, + "rewards/rejected": -10.283984184265137, + "rewards/weighted_accuracies": 0.59375, + "rewards/weighted_chosen": -0.3282794952392578, + "rewards/weighted_margins": 0.1091407760977745, + "rewards/weighted_rejected": -0.4371795654296875, + "step": 250 + }, + { + "epoch": 0.13895383793172555, + "grad_norm": 84.2197494506836, + "learning_rate": 9.956203718306388e-07, + "logits/chosen": 0.025267411023378372, + "logits/rejected": 0.06335143744945526, + "logps/chosen": -128.8429718017578, + "logps/rejected": -133.50625610351562, + "logps/weighted_chosen": -0.9628280401229858, + "logps/weighted_rejected": -1.2008453607559204, + "loss": 0.7443, + "rewards/accuracies": 0.5, + "rewards/chosen": -10.861719131469727, + "rewards/margins": 0.871874988079071, + "rewards/rejected": -11.733593940734863, + "rewards/weighted_accuracies": 0.49687498807907104, + "rewards/weighted_chosen": -0.26259079575538635, + "rewards/weighted_margins": 0.06770706176757812, + "rewards/weighted_rejected": -0.3305191099643707, + "step": 260 + }, + { + "epoch": 0.144298216313715, + "grad_norm": 57.09628677368164, + "learning_rate": 9.943023067615136e-07, + "logits/chosen": -0.017973709851503372, + "logits/rejected": -0.0057998658157885075, + "logps/chosen": -137.60311889648438, + "logps/rejected": -140.91561889648438, + "logps/weighted_chosen": -0.8694183230400085, + "logps/weighted_rejected": -0.8910156488418579, + "loss": 0.7462, + "rewards/accuracies": 0.5625, + "rewards/chosen": -10.946093559265137, + "rewards/margins": 1.662109375, + "rewards/rejected": -12.608202934265137, + "rewards/weighted_accuracies": 0.565625011920929, + "rewards/weighted_chosen": -0.06702041625976562, + "rewards/weighted_margins": 0.13949890434741974, + "rewards/weighted_rejected": -0.20627327263355255, + "step": 270 + }, + { + "epoch": 0.14964259469570446, + "grad_norm": 45.0294303894043, + "learning_rate": 9.928122150897112e-07, + "logits/chosen": -0.045014191418886185, + "logits/rejected": 0.02437591552734375, + "logps/chosen": -130.1125030517578, + "logps/rejected": -131.6171875, + "logps/weighted_chosen": -0.7659690976142883, + "logps/weighted_rejected": -0.8499618768692017, + "loss": 0.7079, + "rewards/accuracies": 0.5406249761581421, + "rewards/chosen": -12.664453506469727, + "rewards/margins": 1.584375023841858, + "rewards/rejected": -14.249608993530273, + "rewards/weighted_accuracies": 0.543749988079071, + "rewards/weighted_chosen": -0.22265930473804474, + "rewards/weighted_margins": 0.07556991279125214, + "rewards/weighted_rejected": -0.298013299703598, + "step": 280 + }, + { + "epoch": 0.1549869730776939, + "grad_norm": 9.370684623718262, + "learning_rate": 9.9115061539547e-07, + "logits/chosen": -0.02069244347512722, + "logits/rejected": 0.01569976843893528, + "logps/chosen": -139.4453125, + "logps/rejected": -141.2578125, + "logps/weighted_chosen": -0.8775177001953125, + "logps/weighted_rejected": -1.2375701665878296, + "loss": 0.6949, + "rewards/accuracies": 0.5375000238418579, + "rewards/chosen": -13.07421875, + "rewards/margins": 1.177734375, + "rewards/rejected": -14.249608993530273, + "rewards/weighted_accuracies": 0.546875, + "rewards/weighted_chosen": -0.18484345078468323, + "rewards/weighted_margins": 0.12048187106847763, + "rewards/weighted_rejected": -0.30528029799461365, + "step": 290 + }, + { + "epoch": 0.16033135145968336, + "grad_norm": 30.68979263305664, + "learning_rate": 9.893180859470818e-07, + "logits/chosen": -0.0137786865234375, + "logits/rejected": 0.041573334485292435, + "logps/chosen": -130.22265625, + "logps/rejected": -132.06094360351562, + "logps/weighted_chosen": -1.1769134998321533, + "logps/weighted_rejected": -1.2438628673553467, + "loss": 0.7261, + "rewards/accuracies": 0.590624988079071, + "rewards/chosen": -13.093358993530273, + "rewards/margins": 2.108203172683716, + "rewards/rejected": -15.200780868530273, + "rewards/weighted_accuracies": 0.5843750238418579, + "rewards/weighted_chosen": -0.2848831117153168, + "rewards/weighted_margins": 0.07841148227453232, + "rewards/weighted_rejected": -0.36317139863967896, + "step": 300 + }, + { + "epoch": 0.16567572984167278, + "grad_norm": 71.56196594238281, + "learning_rate": 9.873152644996424e-07, + "logits/chosen": -0.11114044487476349, + "logits/rejected": -0.11599807441234589, + "logps/chosen": -131.01718139648438, + "logps/rejected": -132.28125, + "logps/weighted_chosen": -1.1032531261444092, + "logps/weighted_rejected": -0.8420807123184204, + "loss": 0.7872, + "rewards/accuracies": 0.5718749761581421, + "rewards/chosen": -12.979687690734863, + "rewards/margins": 2.407421827316284, + "rewards/rejected": -15.388280868530273, + "rewards/weighted_accuracies": 0.5218750238418579, + "rewards/weighted_chosen": 0.05532989650964737, + "rewards/weighted_margins": 0.2024696320295334, + "rewards/weighted_rejected": -0.1469375640153885, + "step": 310 + }, + { + "epoch": 0.17102010822366223, + "grad_norm": 12.055798530578613, + "learning_rate": 9.85142848073103e-07, + "logits/chosen": -0.10557708889245987, + "logits/rejected": -0.06234588474035263, + "logps/chosen": -133.39218139648438, + "logps/rejected": -134.2109375, + "logps/weighted_chosen": -0.700244128704071, + "logps/weighted_rejected": -1.4902985095977783, + "loss": 0.7558, + "rewards/accuracies": 0.534375011920929, + "rewards/chosen": -14.2265625, + "rewards/margins": 0.6207031011581421, + "rewards/rejected": -14.846094131469727, + "rewards/weighted_accuracies": 0.515625, + "rewards/weighted_chosen": -0.2170669585466385, + "rewards/weighted_margins": 0.03446922451257706, + "rewards/weighted_rejected": -0.25187987089157104, + "step": 320 + }, + { + "epoch": 0.17636448660565168, + "grad_norm": 41.381507873535156, + "learning_rate": 9.828015927096914e-07, + "logits/chosen": -0.085169218480587, + "logits/rejected": -0.05568351596593857, + "logps/chosen": -143.8234405517578, + "logps/rejected": -142.39608764648438, + "logps/weighted_chosen": -0.977459728717804, + "logps/weighted_rejected": -0.7958831787109375, + "loss": 0.6935, + "rewards/accuracies": 0.550000011920929, + "rewards/chosen": -14.643750190734863, + "rewards/margins": 1.578125, + "rewards/rejected": -16.222265243530273, + "rewards/weighted_accuracies": 0.581250011920929, + "rewards/weighted_chosen": -0.16206398606300354, + "rewards/weighted_margins": 0.07153511047363281, + "rewards/weighted_rejected": -0.23361511528491974, + "step": 330 + }, + { + "epoch": 0.18170886498764113, + "grad_norm": 7.313599586486816, + "learning_rate": 9.802923132107968e-07, + "logits/chosen": -0.10784301906824112, + "logits/rejected": -0.052645109593868256, + "logps/chosen": -143.91561889648438, + "logps/rejected": -147.7234344482422, + "logps/weighted_chosen": -0.7267853021621704, + "logps/weighted_rejected": -1.24786376953125, + "loss": 0.7102, + "rewards/accuracies": 0.5562499761581421, + "rewards/chosen": -16.782812118530273, + "rewards/margins": 1.23046875, + "rewards/rejected": -18.012500762939453, + "rewards/weighted_accuracies": 0.5687500238418579, + "rewards/weighted_chosen": -0.1999809294939041, + "rewards/weighted_margins": 0.056002043187618256, + "rewards/weighted_rejected": -0.2561977505683899, + "step": 340 + }, + { + "epoch": 0.18705324336963056, + "grad_norm": 66.15936279296875, + "learning_rate": 9.776158828534024e-07, + "logits/chosen": -0.12608642876148224, + "logits/rejected": -0.09711913764476776, + "logps/chosen": -143.48281860351562, + "logps/rejected": -188.25936889648438, + "logps/weighted_chosen": -1.260833740234375, + "logps/weighted_rejected": -0.9638397097587585, + "loss": 0.7338, + "rewards/accuracies": 0.612500011920929, + "rewards/chosen": -16.727733612060547, + "rewards/margins": 3.159374952316284, + "rewards/rejected": -19.883594512939453, + "rewards/weighted_accuracies": 0.559374988079071, + "rewards/weighted_chosen": -0.26622849702835083, + "rewards/weighted_margins": 0.05975799635052681, + "rewards/weighted_rejected": -0.3257743716239929, + "step": 350 + }, + { + "epoch": 0.19239762175162, + "grad_norm": 10.4098539352417, + "learning_rate": 9.747732330861695e-07, + "logits/chosen": -0.05247039720416069, + "logits/rejected": -0.024548912420868874, + "logps/chosen": -135.78750610351562, + "logps/rejected": -137.03671264648438, + "logps/weighted_chosen": -0.8440704345703125, + "logps/weighted_rejected": -0.8307411074638367, + "loss": 0.7415, + "rewards/accuracies": 0.578125, + "rewards/chosen": -16.864452362060547, + "rewards/margins": 2.5316405296325684, + "rewards/rejected": -19.396093368530273, + "rewards/weighted_accuracies": 0.5249999761581421, + "rewards/weighted_chosen": 0.005719375796616077, + "rewards/weighted_margins": 0.17883948981761932, + "rewards/weighted_rejected": -0.17323684692382812, + "step": 360 + }, + { + "epoch": 0.19774200013360946, + "grad_norm": 8.8024263381958, + "learning_rate": 9.717653532052742e-07, + "logits/chosen": -0.0410335548222065, + "logits/rejected": -0.049485016614198685, + "logps/chosen": -130.2781219482422, + "logps/rejected": -137.9343719482422, + "logps/weighted_chosen": -1.1705901622772217, + "logps/weighted_rejected": -0.7656036615371704, + "loss": 0.7402, + "rewards/accuracies": 0.6031249761581421, + "rewards/chosen": -17.155858993530273, + "rewards/margins": 2.863085985183716, + "rewards/rejected": -20.016016006469727, + "rewards/weighted_accuracies": 0.53125, + "rewards/weighted_chosen": -0.07308082282543182, + "rewards/weighted_margins": 0.04105224460363388, + "rewards/weighted_rejected": -0.114354707300663, + "step": 370 + }, + { + "epoch": 0.2030863785155989, + "grad_norm": 26.81167221069336, + "learning_rate": 9.685932900101146e-07, + "logits/chosen": -0.07604064792394638, + "logits/rejected": -0.06302948296070099, + "logps/chosen": -134.7859344482422, + "logps/rejected": -137.11093139648438, + "logps/weighted_chosen": -0.98638916015625, + "logps/weighted_rejected": -0.6243560910224915, + "loss": 0.7116, + "rewards/accuracies": 0.5625, + "rewards/chosen": -17.729297637939453, + "rewards/margins": 1.8025391101837158, + "rewards/rejected": -19.528514862060547, + "rewards/weighted_accuracies": 0.546875, + "rewards/weighted_chosen": -0.11945419013500214, + "rewards/weighted_margins": 0.07185058295726776, + "rewards/weighted_rejected": -0.1911766082048416, + "step": 380 + }, + { + "epoch": 0.20843075689758836, + "grad_norm": 40.89613342285156, + "learning_rate": 9.652581474390043e-07, + "logits/chosen": -0.09274902194738388, + "logits/rejected": -0.05452308803796768, + "logps/chosen": -139.3249969482422, + "logps/rejected": -141.53750610351562, + "logps/weighted_chosen": -0.9199386835098267, + "logps/weighted_rejected": -0.9217513799667358, + "loss": 0.7323, + "rewards/accuracies": 0.550000011920929, + "rewards/chosen": -19.379688262939453, + "rewards/margins": 2.321093797683716, + "rewards/rejected": -21.702733993530273, + "rewards/weighted_accuracies": 0.543749988079071, + "rewards/weighted_chosen": -0.05511016771197319, + "rewards/weighted_margins": 0.04057302325963974, + "rewards/weighted_rejected": -0.09564266353845596, + "step": 390 + }, + { + "epoch": 0.21377513527957778, + "grad_norm": 51.860347747802734, + "learning_rate": 9.61761086184981e-07, + "logits/chosen": -0.10268402099609375, + "logits/rejected": -0.05923309177160263, + "logps/chosen": -139.50936889648438, + "logps/rejected": -141.4015655517578, + "logps/weighted_chosen": -0.9449096918106079, + "logps/weighted_rejected": -1.177301049232483, + "loss": 0.7387, + "rewards/accuracies": 0.5625, + "rewards/chosen": -20.024608612060547, + "rewards/margins": 2.9205079078674316, + "rewards/rejected": -22.942577362060547, + "rewards/weighted_accuracies": 0.5625, + "rewards/weighted_chosen": -0.26060789823532104, + "rewards/weighted_margins": 0.017241667956113815, + "rewards/weighted_rejected": -0.2779647707939148, + "step": 400 + }, + { + "epoch": 0.21911951366156723, + "grad_norm": 7.539150238037109, + "learning_rate": 9.581033232918629e-07, + "logits/chosen": -0.015906523913145065, + "logits/rejected": 0.02000865899026394, + "logps/chosen": -138.39999389648438, + "logps/rejected": -141.68905639648438, + "logps/weighted_chosen": -1.0313812494277954, + "logps/weighted_rejected": -1.1196472644805908, + "loss": 0.7283, + "rewards/accuracies": 0.5874999761581421, + "rewards/chosen": -21.702342987060547, + "rewards/margins": 2.826953172683716, + "rewards/rejected": -24.53125, + "rewards/weighted_accuracies": 0.5249999761581421, + "rewards/weighted_chosen": -0.2783859372138977, + "rewards/weighted_margins": 0.06903991848230362, + "rewards/weighted_rejected": -0.3477470278739929, + "step": 410 + }, + { + "epoch": 0.22446389204355668, + "grad_norm": 6.638446807861328, + "learning_rate": 9.542861317306952e-07, + "logits/chosen": -0.05767669528722763, + "logits/rejected": -0.03759918361902237, + "logps/chosen": -140.64688110351562, + "logps/rejected": -141.9718780517578, + "logps/weighted_chosen": -1.5940430164337158, + "logps/weighted_rejected": -1.458154320716858, + "loss": 0.7371, + "rewards/accuracies": 0.5406249761581421, + "rewards/chosen": -22.33203125, + "rewards/margins": 1.1359374523162842, + "rewards/rejected": -23.46484375, + "rewards/weighted_accuracies": 0.515625, + "rewards/weighted_chosen": -0.299835205078125, + "rewards/weighted_margins": 0.032170869410037994, + "rewards/weighted_rejected": -0.3319748044013977, + "step": 420 + }, + { + "epoch": 0.22980827042554614, + "grad_norm": 24.621816635131836, + "learning_rate": 9.503108399567308e-07, + "logits/chosen": -0.15150222182273865, + "logits/rejected": -0.09339599311351776, + "logps/chosen": -149.8671875, + "logps/rejected": -153.6296844482422, + "logps/weighted_chosen": -0.7455719113349915, + "logps/weighted_rejected": -0.826556384563446, + "loss": 0.7125, + "rewards/accuracies": 0.550000011920929, + "rewards/chosen": -22.685546875, + "rewards/margins": 1.7355468273162842, + "rewards/rejected": -24.420312881469727, + "rewards/weighted_accuracies": 0.5406249761581421, + "rewards/weighted_chosen": -0.19403991103172302, + "rewards/weighted_margins": 0.02692871168255806, + "rewards/weighted_rejected": -0.2209148406982422, + "step": 430 + }, + { + "epoch": 0.23515264880753559, + "grad_norm": 10.726137161254883, + "learning_rate": 9.461788314471034e-07, + "logits/chosen": -0.08421440422534943, + "logits/rejected": -0.031351469457149506, + "logps/chosen": -143.1171875, + "logps/rejected": -146.890625, + "logps/weighted_chosen": -0.7524811029434204, + "logps/weighted_rejected": -0.8818389773368835, + "loss": 0.7062, + "rewards/accuracies": 0.550000011920929, + "rewards/chosen": -24.516796112060547, + "rewards/margins": 2.0640625953674316, + "rewards/rejected": -26.577733993530273, + "rewards/weighted_accuracies": 0.528124988079071, + "rewards/weighted_chosen": -0.22655639052391052, + "rewards/weighted_margins": 0.07337188720703125, + "rewards/weighted_rejected": -0.30005186796188354, + "step": 440 + }, + { + "epoch": 0.240497027189525, + "grad_norm": 8.586684226989746, + "learning_rate": 9.418915442193509e-07, + "logits/chosen": -0.03446502611041069, + "logits/rejected": 0.007968139834702015, + "logps/chosen": -144.7843780517578, + "logps/rejected": -155.96249389648438, + "logps/weighted_chosen": -1.1978485584259033, + "logps/weighted_rejected": -1.2447845935821533, + "loss": 0.6841, + "rewards/accuracies": 0.6312500238418579, + "rewards/chosen": -26.516016006469727, + "rewards/margins": 4.714453220367432, + "rewards/rejected": -31.236328125, + "rewards/weighted_accuracies": 0.5218750238418579, + "rewards/weighted_chosen": -0.25520628690719604, + "rewards/weighted_margins": 0.09325332939624786, + "rewards/weighted_rejected": -0.3483436703681946, + "step": 450 + }, + { + "epoch": 0.24584140557151446, + "grad_norm": 10.014894485473633, + "learning_rate": 9.374504703309579e-07, + "logits/chosen": -0.1379852294921875, + "logits/rejected": -0.0861412063241005, + "logps/chosen": -141.83828735351562, + "logps/rejected": -143.34530639648438, + "logps/weighted_chosen": -1.1420501470565796, + "logps/weighted_rejected": -1.055877685546875, + "loss": 0.6804, + "rewards/accuracies": 0.534375011920929, + "rewards/chosen": -25.53515625, + "rewards/margins": 2.038281202316284, + "rewards/rejected": -27.571874618530273, + "rewards/weighted_accuracies": 0.5218750238418579, + "rewards/weighted_chosen": -0.21983489394187927, + "rewards/weighted_margins": 0.17051848769187927, + "rewards/weighted_rejected": -0.3904357850551605, + "step": 460 + }, + { + "epoch": 0.2511857839535039, + "grad_norm": 10.024358749389648, + "learning_rate": 9.328571553600915e-07, + "logits/chosen": -0.1318313628435135, + "logits/rejected": -0.0828731507062912, + "logps/chosen": -144.39688110351562, + "logps/rejected": -147.41796875, + "logps/weighted_chosen": -0.9446808099746704, + "logps/weighted_rejected": -1.138859510421753, + "loss": 0.7137, + "rewards/accuracies": 0.5718749761581421, + "rewards/chosen": -25.697265625, + "rewards/margins": 1.9656250476837158, + "rewards/rejected": -27.665233612060547, + "rewards/weighted_accuracies": 0.4906249940395355, + "rewards/weighted_chosen": -0.2078697234392166, + "rewards/weighted_margins": 0.052256010472774506, + "rewards/weighted_rejected": -0.260000616312027, + "step": 470 + }, + { + "epoch": 0.25653016233549336, + "grad_norm": 13.618422508239746, + "learning_rate": 9.281131978677106e-07, + "logits/chosen": -0.16686096787452698, + "logits/rejected": -0.1246364563703537, + "logps/chosen": -149.2546844482422, + "logps/rejected": -152.47811889648438, + "logps/weighted_chosen": -0.686419665813446, + "logps/weighted_rejected": -1.0338714122772217, + "loss": 0.6805, + "rewards/accuracies": 0.59375, + "rewards/chosen": -25.559375762939453, + "rewards/margins": 3.1332030296325684, + "rewards/rejected": -28.689844131469727, + "rewards/weighted_accuracies": 0.534375011920929, + "rewards/weighted_chosen": -0.22538375854492188, + "rewards/weighted_margins": 0.13199158012866974, + "rewards/weighted_rejected": -0.35721856355667114, + "step": 480 + }, + { + "epoch": 0.2618745407174828, + "grad_norm": 30.852863311767578, + "learning_rate": 9.232202488412361e-07, + "logits/chosen": -0.10437698662281036, + "logits/rejected": -0.05600614473223686, + "logps/chosen": -143.5671844482422, + "logps/rejected": -149.11874389648438, + "logps/weighted_chosen": -1.236029028892517, + "logps/weighted_rejected": -1.6542327404022217, + "loss": 0.6937, + "rewards/accuracies": 0.5375000238418579, + "rewards/chosen": -28.171483993530273, + "rewards/margins": 2.311328172683716, + "rewards/rejected": -30.477733612060547, + "rewards/weighted_accuracies": 0.534375011920929, + "rewards/weighted_chosen": -0.27268296480178833, + "rewards/weighted_margins": 0.08454742282629013, + "rewards/weighted_rejected": -0.35735321044921875, + "step": 490 + }, + { + "epoch": 0.26721891909947226, + "grad_norm": 9.151970863342285, + "learning_rate": 9.181800111199766e-07, + "logits/chosen": -0.1230342835187912, + "logits/rejected": -0.0801507979631424, + "logps/chosen": -147.54843139648438, + "logps/rejected": -147.5437469482422, + "logps/weighted_chosen": -1.5909912586212158, + "logps/weighted_rejected": -1.286199927330017, + "loss": 0.7005, + "rewards/accuracies": 0.578125, + "rewards/chosen": -29.640233993530273, + "rewards/margins": 3.1507811546325684, + "rewards/rejected": -32.789451599121094, + "rewards/weighted_accuracies": 0.5874999761581421, + "rewards/weighted_chosen": -0.34203338623046875, + "rewards/weighted_margins": 0.12214355170726776, + "rewards/weighted_rejected": -0.4641265869140625, + "step": 500 + }, + { + "epoch": 0.26721891909947226, + "eval_logits/chosen": -0.29476672410964966, + "eval_logits/rejected": -0.26891082525253296, + "eval_logps/chosen": -149.5992889404297, + "eval_logps/rejected": -153.5249481201172, + "eval_logps/weighted_chosen": -0.8761399984359741, + "eval_logps/weighted_rejected": -1.120854377746582, + "eval_loss": 0.6999610066413879, + "eval_rewards/accuracies": 0.5784114003181458, + "eval_rewards/chosen": -28.267057418823242, + "eval_rewards/margins": 2.7842414379119873, + "eval_rewards/rejected": -31.05422592163086, + "eval_rewards/weighted_accuracies": 0.5264765620231628, + "eval_rewards/weighted_chosen": -0.23699289560317993, + "eval_rewards/weighted_margins": 0.08454981446266174, + "eval_rewards/weighted_rejected": -0.3215427100658417, + "eval_runtime": 1248.3443, + "eval_samples_per_second": 1.571, + "eval_steps_per_second": 0.393, + "step": 500 + }, + { + "epoch": 0.2725632974814617, + "grad_norm": 7.777375221252441, + "learning_rate": 9.129942388025066e-07, + "logits/chosen": -0.1614234894514084, + "logits/rejected": -0.12539061903953552, + "logps/chosen": -138.70468139648438, + "logps/rejected": -144.50936889648438, + "logps/weighted_chosen": -1.0583159923553467, + "logps/weighted_rejected": -1.0353851318359375, + "loss": 0.7783, + "rewards/accuracies": 0.543749988079071, + "rewards/chosen": -27.291797637939453, + "rewards/margins": 2.485156297683716, + "rewards/rejected": -29.778514862060547, + "rewards/weighted_accuracies": 0.5093749761581421, + "rewards/weighted_chosen": -0.21576309204101562, + "rewards/weighted_margins": 0.003093719482421875, + "rewards/weighted_rejected": -0.219013974070549, + "step": 510 + }, + { + "epoch": 0.2779076758634511, + "grad_norm": 42.880157470703125, + "learning_rate": 9.076647366362082e-07, + "logits/chosen": -0.0863800048828125, + "logits/rejected": -0.04528322070837021, + "logps/chosen": -148.21875, + "logps/rejected": -155.3046875, + "logps/weighted_chosen": -1.141107201576233, + "logps/weighted_rejected": -1.2995758056640625, + "loss": 0.7748, + "rewards/accuracies": 0.612500011920929, + "rewards/chosen": -33.478515625, + "rewards/margins": 4.057812690734863, + "rewards/rejected": -37.53515625, + "rewards/weighted_accuracies": 0.5531250238418579, + "rewards/weighted_chosen": -0.2628006041049957, + "rewards/weighted_margins": -0.00189971923828125, + "rewards/weighted_rejected": -0.26109999418258667, + "step": 520 + }, + { + "epoch": 0.2832520542454406, + "grad_norm": 24.728120803833008, + "learning_rate": 9.021933593891841e-07, + "logits/chosen": -0.020507430657744408, + "logits/rejected": 0.01460342388600111, + "logps/chosen": -154.15780639648438, + "logps/rejected": -156.328125, + "logps/weighted_chosen": -1.0978972911834717, + "logps/weighted_rejected": -0.9736785888671875, + "loss": 0.6998, + "rewards/accuracies": 0.578125, + "rewards/chosen": -40.849220275878906, + "rewards/margins": 2.678515672683716, + "rewards/rejected": -43.53046798706055, + "rewards/weighted_accuracies": 0.512499988079071, + "rewards/weighted_chosen": -0.2724204957485199, + "rewards/weighted_margins": 0.09940185397863388, + "rewards/weighted_rejected": -0.3718253970146179, + "step": 530 + }, + { + "epoch": 0.28859643262743, + "grad_norm": 11.090974807739258, + "learning_rate": 8.965820112047629e-07, + "logits/chosen": -0.07119598239660263, + "logits/rejected": -0.04578246921300888, + "logps/chosen": -163.8781280517578, + "logps/rejected": -172.9187469482422, + "logps/weighted_chosen": -0.9050750732421875, + "logps/weighted_rejected": -1.05401611328125, + "loss": 0.6621, + "rewards/accuracies": 0.559374988079071, + "rewards/chosen": -42.501564025878906, + "rewards/margins": 4.053515434265137, + "rewards/rejected": -46.54999923706055, + "rewards/weighted_accuracies": 0.6000000238418579, + "rewards/weighted_chosen": -0.1999252289533615, + "rewards/weighted_margins": 0.1526111662387848, + "rewards/weighted_rejected": -0.3521057069301605, + "step": 540 + }, + { + "epoch": 0.2939408110094195, + "grad_norm": 13.709502220153809, + "learning_rate": 8.908326449388213e-07, + "logits/chosen": -0.07627105712890625, + "logits/rejected": -0.03014526329934597, + "logps/chosen": -156.328125, + "logps/rejected": -161.06405639648438, + "logps/weighted_chosen": -1.228173851966858, + "logps/weighted_rejected": -1.3901275396347046, + "loss": 0.7368, + "rewards/accuracies": 0.6156250238418579, + "rewards/chosen": -41.07109451293945, + "rewards/margins": 4.337500095367432, + "rewards/rejected": -45.392967224121094, + "rewards/weighted_accuracies": 0.512499988079071, + "rewards/weighted_chosen": -0.172210693359375, + "rewards/weighted_margins": 0.07322921603918076, + "rewards/weighted_rejected": -0.2451316863298416, + "step": 550 + }, + { + "epoch": 0.2992851893914089, + "grad_norm": 10.929818153381348, + "learning_rate": 8.849472614801527e-07, + "logits/chosen": -0.07251434028148651, + "logits/rejected": -0.02443542517721653, + "logps/chosen": -161.86874389648438, + "logps/rejected": -165.0749969482422, + "logps/weighted_chosen": -0.752777099609375, + "logps/weighted_rejected": -1.0892517566680908, + "loss": 0.6642, + "rewards/accuracies": 0.606249988079071, + "rewards/chosen": -44.9609375, + "rewards/margins": 3.3421874046325684, + "rewards/rejected": -48.30546951293945, + "rewards/weighted_accuracies": 0.5874999761581421, + "rewards/weighted_chosen": -0.1927940398454666, + "rewards/weighted_margins": 0.19287796318531036, + "rewards/weighted_rejected": -0.38528138399124146, + "step": 560 + }, + { + "epoch": 0.30462956777339834, + "grad_norm": 11.730274200439453, + "learning_rate": 8.789279090541208e-07, + "logits/chosen": -0.05120544508099556, + "logits/rejected": 0.01478500384837389, + "logps/chosen": -161.8796844482422, + "logps/rejected": -172.51406860351562, + "logps/weighted_chosen": -0.8590911626815796, + "logps/weighted_rejected": -1.1839568614959717, + "loss": 0.6821, + "rewards/accuracies": 0.6187499761581421, + "rewards/chosen": -43.99609375, + "rewards/margins": 4.257031440734863, + "rewards/rejected": -48.24609375, + "rewards/weighted_accuracies": 0.515625, + "rewards/weighted_chosen": -0.29502803087234497, + "rewards/weighted_margins": 0.16868743300437927, + "rewards/weighted_rejected": -0.4639228880405426, + "step": 570 + }, + { + "epoch": 0.3099739461553878, + "grad_norm": 16.967859268188477, + "learning_rate": 8.72776682509837e-07, + "logits/chosen": 0.011944961734116077, + "logits/rejected": 0.08212585747241974, + "logps/chosen": -157.47500610351562, + "logps/rejected": -163.43594360351562, + "logps/weighted_chosen": -0.8815460205078125, + "logps/weighted_rejected": -1.220306396484375, + "loss": 0.6705, + "rewards/accuracies": 0.606249988079071, + "rewards/chosen": -43.369529724121094, + "rewards/margins": 4.614453315734863, + "rewards/rejected": -47.984375, + "rewards/weighted_accuracies": 0.565625011920929, + "rewards/weighted_chosen": -0.3169898986816406, + "rewards/weighted_margins": 0.19199447333812714, + "rewards/weighted_rejected": -0.5094040036201477, + "step": 580 + }, + { + "epoch": 0.31531832453737724, + "grad_norm": 9.289381980895996, + "learning_rate": 8.664957225911138e-07, + "logits/chosen": -0.04825134202837944, + "logits/rejected": -0.01284942589700222, + "logps/chosen": -163.6437530517578, + "logps/rejected": -184.9796905517578, + "logps/weighted_chosen": -0.775604248046875, + "logps/weighted_rejected": -1.1336090564727783, + "loss": 0.6889, + "rewards/accuracies": 0.596875011920929, + "rewards/chosen": -42.566795349121094, + "rewards/margins": 3.8046875, + "rewards/rejected": -46.36328125, + "rewards/weighted_accuracies": 0.565625011920929, + "rewards/weighted_chosen": -0.30945053696632385, + "rewards/weighted_margins": 0.116083525121212, + "rewards/weighted_rejected": -0.4253120422363281, + "step": 590 + }, + { + "epoch": 0.3206627029193667, + "grad_norm": 7.966893672943115, + "learning_rate": 8.600872151914451e-07, + "logits/chosen": 0.06866455078125, + "logits/rejected": 0.09470061957836151, + "logps/chosen": -158.2546844482422, + "logps/rejected": -167.48281860351562, + "logps/weighted_chosen": -1.3719971179962158, + "logps/weighted_rejected": -1.7499603033065796, + "loss": 0.7417, + "rewards/accuracies": 0.5843750238418579, + "rewards/chosen": -44.11249923706055, + "rewards/margins": 3.3988280296325684, + "rewards/rejected": -47.50468826293945, + "rewards/weighted_accuracies": 0.5375000238418579, + "rewards/weighted_chosen": -0.21634292602539062, + "rewards/weighted_margins": 0.19443893432617188, + "rewards/weighted_rejected": -0.41086500883102417, + "step": 600 + }, + { + "epoch": 0.32600708130135614, + "grad_norm": 49.298519134521484, + "learning_rate": 8.535533905932737e-07, + "logits/chosen": 0.06249542161822319, + "logits/rejected": 0.12158355861902237, + "logps/chosen": -173.4250030517578, + "logps/rejected": -183.015625, + "logps/weighted_chosen": -0.9283081293106079, + "logps/weighted_rejected": -1.2264496088027954, + "loss": 0.6778, + "rewards/accuracies": 0.590624988079071, + "rewards/chosen": -48.342185974121094, + "rewards/margins": 5.698437690734863, + "rewards/rejected": -54.049217224121094, + "rewards/weighted_accuracies": 0.578125, + "rewards/weighted_chosen": -0.3028602600097656, + "rewards/weighted_margins": 0.1688835173845291, + "rewards/weighted_rejected": -0.4717727601528168, + "step": 610 + }, + { + "epoch": 0.33135145968334556, + "grad_norm": 19.377832412719727, + "learning_rate": 8.468965226918105e-07, + "logits/chosen": 0.037828825414180756, + "logits/rejected": 0.0580875389277935, + "logps/chosen": -170.1281280517578, + "logps/rejected": -178.9093780517578, + "logps/weighted_chosen": -1.4841537475585938, + "logps/weighted_rejected": -1.118981957435608, + "loss": 0.7542, + "rewards/accuracies": 0.5874999761581421, + "rewards/chosen": -46.46210861206055, + "rewards/margins": 3.794921875, + "rewards/rejected": -50.26640701293945, + "rewards/weighted_accuracies": 0.596875011920929, + "rewards/weighted_chosen": -0.40849533677101135, + "rewards/weighted_margins": 0.0321044921875, + "rewards/weighted_rejected": -0.4405921995639801, + "step": 620 + }, + { + "epoch": 0.33669583806533504, + "grad_norm": 10.768741607666016, + "learning_rate": 8.40118928203676e-07, + "logits/chosen": 0.05730438232421875, + "logits/rejected": 0.07560119777917862, + "logps/chosen": -152.3718719482422, + "logps/rejected": -164.1218719482422, + "logps/weighted_chosen": -1.4566253423690796, + "logps/weighted_rejected": -1.3677704334259033, + "loss": 0.7025, + "rewards/accuracies": 0.643750011920929, + "rewards/chosen": -43.721092224121094, + "rewards/margins": 4.366015434265137, + "rewards/rejected": -48.07500076293945, + "rewards/weighted_accuracies": 0.590624988079071, + "rewards/weighted_chosen": -0.30116957426071167, + "rewards/weighted_margins": 0.09331168979406357, + "rewards/weighted_rejected": -0.39441680908203125, + "step": 630 + }, + { + "epoch": 0.34204021644732446, + "grad_norm": 8.60303020477295, + "learning_rate": 8.332229658606382e-07, + "logits/chosen": -0.055472563952207565, + "logits/rejected": -0.03988037258386612, + "logps/chosen": -175.6671905517578, + "logps/rejected": -180.06875610351562, + "logps/weighted_chosen": -0.8608856201171875, + "logps/weighted_rejected": -0.986541748046875, + "loss": 0.7297, + "rewards/accuracies": 0.596875011920929, + "rewards/chosen": -44.84062576293945, + "rewards/margins": 4.931250095367432, + "rewards/rejected": -49.77031326293945, + "rewards/weighted_accuracies": 0.515625, + "rewards/weighted_chosen": -0.23048630356788635, + "rewards/weighted_margins": 0.10724487155675888, + "rewards/weighted_rejected": -0.33786240220069885, + "step": 640 + }, + { + "epoch": 0.34738459482931394, + "grad_norm": 26.148893356323242, + "learning_rate": 8.262110355887302e-07, + "logits/chosen": -0.055730439722537994, + "logits/rejected": -0.00980529747903347, + "logps/chosen": -166.38125610351562, + "logps/rejected": -169.40469360351562, + "logps/weighted_chosen": -0.7286010980606079, + "logps/weighted_rejected": -0.8990936279296875, + "loss": 0.7103, + "rewards/accuracies": 0.609375, + "rewards/chosen": -46.994529724121094, + "rewards/margins": 3.8031249046325684, + "rewards/rejected": -50.78515625, + "rewards/weighted_accuracies": 0.518750011920929, + "rewards/weighted_chosen": -0.2977249026298523, + "rewards/weighted_margins": 0.08674392849206924, + "rewards/weighted_rejected": -0.3846077024936676, + "step": 650 + }, + { + "epoch": 0.35272897321130336, + "grad_norm": 11.823346138000488, + "learning_rate": 8.190855776730293e-07, + "logits/chosen": -0.07448883354663849, + "logits/rejected": -0.014659118838608265, + "logps/chosen": -171.80313110351562, + "logps/rejected": -182.2687530517578, + "logps/weighted_chosen": -0.9873107671737671, + "logps/weighted_rejected": -1.4951263666152954, + "loss": 0.7347, + "rewards/accuracies": 0.596875011920929, + "rewards/chosen": -50.564064025878906, + "rewards/margins": 4.532812595367432, + "rewards/rejected": -55.109375, + "rewards/weighted_accuracies": 0.5531250238418579, + "rewards/weighted_chosen": -0.28445130586624146, + "rewards/weighted_margins": 0.06389693915843964, + "rewards/weighted_rejected": -0.34831467270851135, + "step": 660 + }, + { + "epoch": 0.3580733515932928, + "grad_norm": 13.33145523071289, + "learning_rate": 8.118490719083917e-07, + "logits/chosen": -0.10270843654870987, + "logits/rejected": -0.062209319323301315, + "logps/chosen": -191.3000030517578, + "logps/rejected": -193.4109344482422, + "logps/weighted_chosen": -1.0586059093475342, + "logps/weighted_rejected": -1.3695876598358154, + "loss": 0.6575, + "rewards/accuracies": 0.546875, + "rewards/chosen": -58.06640625, + "rewards/margins": 4.108984470367432, + "rewards/rejected": -62.18359375, + "rewards/weighted_accuracies": 0.565625011920929, + "rewards/weighted_chosen": -0.2945011258125305, + "rewards/weighted_margins": 0.18829631805419922, + "rewards/weighted_rejected": -0.4826883375644684, + "step": 670 + }, + { + "epoch": 0.36341772997528227, + "grad_norm": 23.127918243408203, + "learning_rate": 8.045040367364368e-07, + "logits/chosen": 0.03829193115234375, + "logits/rejected": 0.055706024169921875, + "logps/chosen": -166.234375, + "logps/rejected": -169.96249389648438, + "logps/weighted_chosen": -1.025201439857483, + "logps/weighted_rejected": -1.302404761314392, + "loss": 0.7166, + "rewards/accuracies": 0.5843750238418579, + "rewards/chosen": -56.759376525878906, + "rewards/margins": 4.711718559265137, + "rewards/rejected": -61.46796798706055, + "rewards/weighted_accuracies": 0.590624988079071, + "rewards/weighted_chosen": -0.4052978456020355, + "rewards/weighted_margins": 0.09704723209142685, + "rewards/weighted_rejected": -0.5025444030761719, + "step": 680 + }, + { + "epoch": 0.3687621083572717, + "grad_norm": 7.137462615966797, + "learning_rate": 7.970530283690819e-07, + "logits/chosen": 0.04682769626379013, + "logits/rejected": 0.066925048828125, + "logps/chosen": -172.50155639648438, + "logps/rejected": -180.54061889648438, + "logps/weighted_chosen": -1.301055908203125, + "logps/weighted_rejected": -1.128961205482483, + "loss": 0.6864, + "rewards/accuracies": 0.6312500238418579, + "rewards/chosen": -58.544532775878906, + "rewards/margins": 5.397265434265137, + "rewards/rejected": -63.9453125, + "rewards/weighted_accuracies": 0.59375, + "rewards/weighted_chosen": -0.3540634214878082, + "rewards/weighted_margins": 0.1536247283220291, + "rewards/weighted_rejected": -0.5076873898506165, + "step": 690 + }, + { + "epoch": 0.3741064867392611, + "grad_norm": 7.699697971343994, + "learning_rate": 7.89498639898931e-07, + "logits/chosen": 0.05636443942785263, + "logits/rejected": 0.09223861992359161, + "logps/chosen": -169.8562469482422, + "logps/rejected": -173.65625, + "logps/weighted_chosen": -0.9024292230606079, + "logps/weighted_rejected": -1.1884033679962158, + "loss": 0.6642, + "rewards/accuracies": 0.6031249761581421, + "rewards/chosen": -59.00078201293945, + "rewards/margins": 4.887109279632568, + "rewards/rejected": -63.889060974121094, + "rewards/weighted_accuracies": 0.5562499761581421, + "rewards/weighted_chosen": -0.3017433285713196, + "rewards/weighted_margins": 0.20005035400390625, + "rewards/weighted_rejected": -0.5016593933105469, + "step": 700 + }, + { + "epoch": 0.3794508651212506, + "grad_norm": 18.41610336303711, + "learning_rate": 7.818435003968305e-07, + "logits/chosen": -0.03687286376953125, + "logits/rejected": -0.00025310515775345266, + "logps/chosen": -181.0812530517578, + "logps/rejected": -189.1828155517578, + "logps/weighted_chosen": -1.0574157238006592, + "logps/weighted_rejected": -1.2661011219024658, + "loss": 0.6918, + "rewards/accuracies": 0.590624988079071, + "rewards/chosen": -58.548439025878906, + "rewards/margins": 5.687890529632568, + "rewards/rejected": -64.234375, + "rewards/weighted_accuracies": 0.550000011920929, + "rewards/weighted_chosen": -0.33240431547164917, + "rewards/weighted_margins": 0.08642272651195526, + "rewards/weighted_rejected": -0.4188430905342102, + "step": 710 + }, + { + "epoch": 0.38479524350324, + "grad_norm": 26.664974212646484, + "learning_rate": 7.740902739969008e-07, + "logits/chosen": -0.08480529487133026, + "logits/rejected": -0.07062835991382599, + "logps/chosen": -171.16561889648438, + "logps/rejected": -175.68905639648438, + "logps/weighted_chosen": -0.828997790813446, + "logps/weighted_rejected": -0.9821502566337585, + "loss": 0.6819, + "rewards/accuracies": 0.6187499761581421, + "rewards/chosen": -49.982810974121094, + "rewards/margins": 4.688672065734863, + "rewards/rejected": -54.672264099121094, + "rewards/weighted_accuracies": 0.5531250238418579, + "rewards/weighted_chosen": -0.18975982069969177, + "rewards/weighted_margins": 0.1927967071533203, + "rewards/weighted_rejected": -0.38214111328125, + "step": 720 + }, + { + "epoch": 0.3901396218852295, + "grad_norm": 8.73801040649414, + "learning_rate": 7.662416589693695e-07, + "logits/chosen": -0.00596694927662611, + "logits/rejected": 0.019218826666474342, + "logps/chosen": -161.8249969482422, + "logps/rejected": -166.8312530517578, + "logps/weighted_chosen": -1.3982207775115967, + "logps/weighted_rejected": -1.7395598888397217, + "loss": 0.7772, + "rewards/accuracies": 0.6625000238418579, + "rewards/chosen": -46.8515625, + "rewards/margins": 5.096484184265137, + "rewards/rejected": -51.95000076293945, + "rewards/weighted_accuracies": 0.590624988079071, + "rewards/weighted_chosen": -0.23512879014015198, + "rewards/weighted_margins": 0.10187530517578125, + "rewards/weighted_rejected": -0.3369201719760895, + "step": 730 + }, + { + "epoch": 0.3954840002672189, + "grad_norm": 19.624563217163086, + "learning_rate": 7.583003867815192e-07, + "logits/chosen": -0.0639801025390625, + "logits/rejected": -0.05755653232336044, + "logps/chosen": -189.7468719482422, + "logps/rejected": -196.7062530517578, + "logps/weighted_chosen": -1.1809478998184204, + "logps/weighted_rejected": -1.3041900396347046, + "loss": 0.698, + "rewards/accuracies": 0.578125, + "rewards/chosen": -58.209373474121094, + "rewards/margins": 4.971875190734863, + "rewards/rejected": -63.1640625, + "rewards/weighted_accuracies": 0.518750011920929, + "rewards/weighted_chosen": -0.39410895109176636, + "rewards/weighted_margins": 0.08784560859203339, + "rewards/weighted_rejected": -0.4818374514579773, + "step": 740 + }, + { + "epoch": 0.40082837864920834, + "grad_norm": 19.04839324951172, + "learning_rate": 7.502692211470869e-07, + "logits/chosen": -0.05941009521484375, + "logits/rejected": -0.02816619910299778, + "logps/chosen": -182.8984375, + "logps/rejected": -190.1203155517578, + "logps/weighted_chosen": -0.853833019733429, + "logps/weighted_rejected": -1.1572723388671875, + "loss": 0.7079, + "rewards/accuracies": 0.574999988079071, + "rewards/chosen": -60.310157775878906, + "rewards/margins": 6.424609184265137, + "rewards/rejected": -66.73905944824219, + "rewards/weighted_accuracies": 0.546875, + "rewards/weighted_chosen": -0.4228218197822571, + "rewards/weighted_margins": 0.10426559299230576, + "rewards/weighted_rejected": -0.5270500183105469, + "step": 750 + }, + { + "epoch": 0.4061727570311978, + "grad_norm": 12.947752952575684, + "learning_rate": 7.421509570644387e-07, + "logits/chosen": -0.05351562425494194, + "logits/rejected": -0.0015235900646075606, + "logps/chosen": -180.66250610351562, + "logps/rejected": -185.6828155517578, + "logps/weighted_chosen": -1.2698882818222046, + "logps/weighted_rejected": -1.5226471424102783, + "loss": 0.6831, + "rewards/accuracies": 0.6343749761581421, + "rewards/chosen": -62.2109375, + "rewards/margins": 5.124218940734863, + "rewards/rejected": -67.33280944824219, + "rewards/weighted_accuracies": 0.59375, + "rewards/weighted_chosen": -0.32732313871383667, + "rewards/weighted_margins": 0.18027648329734802, + "rewards/weighted_rejected": -0.5074516534805298, + "step": 760 + }, + { + "epoch": 0.41151713541318724, + "grad_norm": 10.666881561279297, + "learning_rate": 7.339484198438566e-07, + "logits/chosen": 0.00444793701171875, + "logits/rejected": 0.03400421142578125, + "logps/chosen": -165.93124389648438, + "logps/rejected": -172.2937469482422, + "logps/weighted_chosen": -1.7782714366912842, + "logps/weighted_rejected": -1.320520043373108, + "loss": 0.6969, + "rewards/accuracies": 0.6000000238418579, + "rewards/chosen": -57.14374923706055, + "rewards/margins": 7.307421684265137, + "rewards/rejected": -64.4468765258789, + "rewards/weighted_accuracies": 0.5625, + "rewards/weighted_chosen": -0.3784065246582031, + "rewards/weighted_margins": 0.17845840752124786, + "rewards/weighted_rejected": -0.557085394859314, + "step": 770 + }, + { + "epoch": 0.4168615137951767, + "grad_norm": 11.188801765441895, + "learning_rate": 7.256644641242773e-07, + "logits/chosen": -0.03662109375, + "logits/rejected": 0.02340393140912056, + "logps/chosen": -170.4765625, + "logps/rejected": -175.390625, + "logps/weighted_chosen": -1.057519555091858, + "logps/weighted_rejected": -1.3154876232147217, + "loss": 0.7128, + "rewards/accuracies": 0.5562499761581421, + "rewards/chosen": -56.12578201293945, + "rewards/margins": 3.3226561546325684, + "rewards/rejected": -59.43828201293945, + "rewards/weighted_accuracies": 0.546875, + "rewards/weighted_chosen": -0.37098997831344604, + "rewards/weighted_margins": 0.07502517849206924, + "rewards/weighted_rejected": -0.4461258053779602, + "step": 780 + }, + { + "epoch": 0.42220589217716614, + "grad_norm": 9.07581615447998, + "learning_rate": 7.173019728798234e-07, + "logits/chosen": -0.03949751704931259, + "logits/rejected": -0.02464141882956028, + "logps/chosen": -169.55313110351562, + "logps/rejected": -177.18124389648438, + "logps/weighted_chosen": -0.8541290163993835, + "logps/weighted_rejected": -1.1333099603652954, + "loss": 0.6788, + "rewards/accuracies": 0.6625000238418579, + "rewards/chosen": -55.47734451293945, + "rewards/margins": 6.255859375, + "rewards/rejected": -61.71406173706055, + "rewards/weighted_accuracies": 0.574999988079071, + "rewards/weighted_chosen": -0.307028204202652, + "rewards/weighted_margins": 0.12036208808422089, + "rewards/weighted_rejected": -0.4274864196777344, + "step": 790 + }, + { + "epoch": 0.42755027055915557, + "grad_norm": 19.591575622558594, + "learning_rate": 7.088638564164738e-07, + "logits/chosen": -0.08669586479663849, + "logits/rejected": -0.05663032457232475, + "logps/chosen": -185.1593780517578, + "logps/rejected": -195.95938110351562, + "logps/weighted_chosen": -1.067468285560608, + "logps/weighted_rejected": -1.238824486732483, + "loss": 0.693, + "rewards/accuracies": 0.628125011920929, + "rewards/chosen": -63.07421875, + "rewards/margins": 6.429296970367432, + "rewards/rejected": -69.49296569824219, + "rewards/weighted_accuracies": 0.581250011920929, + "rewards/weighted_chosen": -0.3178604245185852, + "rewards/weighted_margins": 0.1114906296133995, + "rewards/weighted_rejected": -0.4294891357421875, + "step": 800 + }, + { + "epoch": 0.43289464894114504, + "grad_norm": 13.074841499328613, + "learning_rate": 7.003530513592215e-07, + "logits/chosen": -0.05083770677447319, + "logits/rejected": -0.0205841064453125, + "logps/chosen": -177.95156860351562, + "logps/rejected": -189.63436889648438, + "logps/weighted_chosen": -1.188201904296875, + "logps/weighted_rejected": -1.1756408214569092, + "loss": 0.6926, + "rewards/accuracies": 0.6187499761581421, + "rewards/chosen": -62.192970275878906, + "rewards/margins": 7.530859470367432, + "rewards/rejected": -69.73905944824219, + "rewards/weighted_accuracies": 0.543749988079071, + "rewards/weighted_chosen": -0.3717803955078125, + "rewards/weighted_margins": 0.12430572509765625, + "rewards/weighted_rejected": -0.49610137939453125, + "step": 810 + }, + { + "epoch": 0.43823902732313447, + "grad_norm": 51.0267448425293, + "learning_rate": 6.917725196300726e-07, + "logits/chosen": -0.1438644379377365, + "logits/rejected": -0.15563049912452698, + "logps/chosen": -191.125, + "logps/rejected": -203.52188110351562, + "logps/weighted_chosen": -1.097863793373108, + "logps/weighted_rejected": -0.971697986125946, + "loss": 0.6974, + "rewards/accuracies": 0.596875011920929, + "rewards/chosen": -63.764060974121094, + "rewards/margins": 7.415234565734863, + "rewards/rejected": -71.171875, + "rewards/weighted_accuracies": 0.5531250238418579, + "rewards/weighted_chosen": -0.3276786804199219, + "rewards/weighted_margins": 0.08215713500976562, + "rewards/weighted_rejected": -0.4100082516670227, + "step": 820 + }, + { + "epoch": 0.44358340570512395, + "grad_norm": 20.933467864990234, + "learning_rate": 6.831252474172411e-07, + "logits/chosen": -0.05628051608800888, + "logits/rejected": -0.03361053392291069, + "logps/chosen": -176.0031280517578, + "logps/rejected": -186.4343719482422, + "logps/weighted_chosen": -1.6239410638809204, + "logps/weighted_rejected": -1.3564453125, + "loss": 0.6817, + "rewards/accuracies": 0.609375, + "rewards/chosen": -60.45234298706055, + "rewards/margins": 6.214453220367432, + "rewards/rejected": -66.66015625, + "rewards/weighted_accuracies": 0.574999988079071, + "rewards/weighted_chosen": -0.2794052064418793, + "rewards/weighted_margins": 0.22185592353343964, + "rewards/weighted_rejected": -0.500841498374939, + "step": 830 + }, + { + "epoch": 0.44892778408711337, + "grad_norm": 11.393433570861816, + "learning_rate": 6.74414244135898e-07, + "logits/chosen": -0.15757599472999573, + "logits/rejected": -0.116943359375, + "logps/chosen": -177.0984344482422, + "logps/rejected": -185.13436889648438, + "logps/weighted_chosen": -0.666271984577179, + "logps/weighted_rejected": -0.7827666997909546, + "loss": 0.6919, + "rewards/accuracies": 0.625, + "rewards/chosen": -60.349998474121094, + "rewards/margins": 6.200390815734863, + "rewards/rejected": -66.5289077758789, + "rewards/weighted_accuracies": 0.581250011920929, + "rewards/weighted_chosen": -0.33733367919921875, + "rewards/weighted_margins": 0.08320007473230362, + "rewards/weighted_rejected": -0.42046278715133667, + "step": 840 + }, + { + "epoch": 0.4542721624691028, + "grad_norm": 9.733312606811523, + "learning_rate": 6.656425413808388e-07, + "logits/chosen": -0.12353897094726562, + "logits/rejected": -0.08713607490062714, + "logps/chosen": -179.66561889648438, + "logps/rejected": -183.109375, + "logps/weighted_chosen": -1.0826873779296875, + "logps/weighted_rejected": -1.106787085533142, + "loss": 0.6875, + "rewards/accuracies": 0.5562499761581421, + "rewards/chosen": -60.806251525878906, + "rewards/margins": 2.733203172683716, + "rewards/rejected": -63.51250076293945, + "rewards/weighted_accuracies": 0.5843750238418579, + "rewards/weighted_chosen": -0.22345809638500214, + "rewards/weighted_margins": 0.1886741667985916, + "rewards/weighted_rejected": -0.4119575619697571, + "step": 850 + }, + { + "epoch": 0.45961654085109227, + "grad_norm": 9.04237174987793, + "learning_rate": 6.568131918714294e-07, + "logits/chosen": -0.09643097221851349, + "logits/rejected": -0.08455581963062286, + "logps/chosen": -178.4453125, + "logps/rejected": -187.99063110351562, + "logps/weighted_chosen": -0.840087890625, + "logps/weighted_rejected": -1.158935546875, + "loss": 0.7138, + "rewards/accuracies": 0.596875011920929, + "rewards/chosen": -61.162498474121094, + "rewards/margins": 5.386328220367432, + "rewards/rejected": -66.54765319824219, + "rewards/weighted_accuracies": 0.550000011920929, + "rewards/weighted_chosen": -0.2982444763183594, + "rewards/weighted_margins": 0.08280716091394424, + "rewards/weighted_rejected": -0.38090819120407104, + "step": 860 + }, + { + "epoch": 0.4649609192330817, + "grad_norm": 7.623125076293945, + "learning_rate": 6.47929268389203e-07, + "logits/chosen": -0.06428680568933487, + "logits/rejected": -0.04297828674316406, + "logps/chosen": -190.86093139648438, + "logps/rejected": -205.9656219482422, + "logps/weighted_chosen": -0.9338012933731079, + "logps/weighted_rejected": -0.741748034954071, + "loss": 0.6808, + "rewards/accuracies": 0.6000000238418579, + "rewards/chosen": -65.76640319824219, + "rewards/margins": 7.994921684265137, + "rewards/rejected": -73.75703430175781, + "rewards/weighted_accuracies": 0.5, + "rewards/weighted_chosen": -0.28187790513038635, + "rewards/weighted_margins": 0.11656570434570312, + "rewards/weighted_rejected": -0.39813536405563354, + "step": 870 + }, + { + "epoch": 0.47030529761507117, + "grad_norm": 36.31256103515625, + "learning_rate": 6.389938627084732e-07, + "logits/chosen": -0.04478149488568306, + "logits/rejected": -0.054534912109375, + "logps/chosen": -181.5046844482422, + "logps/rejected": -192.84375, + "logps/weighted_chosen": -1.712792992591858, + "logps/weighted_rejected": -1.4381287097930908, + "loss": 0.7264, + "rewards/accuracies": 0.581250011920929, + "rewards/chosen": -64.04609680175781, + "rewards/margins": 6.507421970367432, + "rewards/rejected": -70.56953430175781, + "rewards/weighted_accuracies": 0.515625, + "rewards/weighted_chosen": -0.3980148434638977, + "rewards/weighted_margins": 0.10754700005054474, + "rewards/weighted_rejected": -0.5054702758789062, + "step": 880 + }, + { + "epoch": 0.4756496759970606, + "grad_norm": 6.0423126220703125, + "learning_rate": 6.300100845203373e-07, + "logits/chosen": -0.07687301933765411, + "logits/rejected": -0.04276275634765625, + "logps/chosen": -194.1609344482422, + "logps/rejected": -200.828125, + "logps/weighted_chosen": -0.6031554937362671, + "logps/weighted_rejected": -0.86151123046875, + "loss": 0.6793, + "rewards/accuracies": 0.59375, + "rewards/chosen": -63.5546875, + "rewards/margins": 5.604296684265137, + "rewards/rejected": -69.13984680175781, + "rewards/weighted_accuracies": 0.559374988079071, + "rewards/weighted_chosen": -0.27074965834617615, + "rewards/weighted_margins": 0.11608581244945526, + "rewards/weighted_rejected": -0.38665771484375, + "step": 890 + }, + { + "epoch": 0.48099405437905, + "grad_norm": 21.55549430847168, + "learning_rate": 6.20981060350445e-07, + "logits/chosen": -0.12602996826171875, + "logits/rejected": -0.08534087985754013, + "logps/chosen": -186.3312530517578, + "logps/rejected": -190.0625, + "logps/weighted_chosen": -0.8782455325126648, + "logps/weighted_rejected": -1.0832672119140625, + "loss": 0.6442, + "rewards/accuracies": 0.574999988079071, + "rewards/chosen": -65.2984390258789, + "rewards/margins": 6.321875095367432, + "rewards/rejected": -71.61796569824219, + "rewards/weighted_accuracies": 0.612500011920929, + "rewards/weighted_chosen": -0.3018638491630554, + "rewards/weighted_margins": 0.19889220595359802, + "rewards/weighted_rejected": -0.5003150701522827, + "step": 900 + }, + { + "epoch": 0.4863384327610395, + "grad_norm": 15.255428314208984, + "learning_rate": 6.119099324709087e-07, + "logits/chosen": -0.08440399169921875, + "logits/rejected": -0.03687591478228569, + "logps/chosen": -183.1640625, + "logps/rejected": -192.125, + "logps/weighted_chosen": -0.7069732546806335, + "logps/weighted_rejected": -1.348388671875, + "loss": 0.7239, + "rewards/accuracies": 0.609375, + "rewards/chosen": -65.2796859741211, + "rewards/margins": 7.408593654632568, + "rewards/rejected": -72.6890640258789, + "rewards/weighted_accuracies": 0.5093749761581421, + "rewards/weighted_chosen": -0.3756294250488281, + "rewards/weighted_margins": 0.05287017673254013, + "rewards/weighted_rejected": -0.4286758303642273, + "step": 910 + }, + { + "epoch": 0.4916828111430289, + "grad_norm": 10.355165481567383, + "learning_rate": 6.027998578067316e-07, + "logits/chosen": -0.06318511813879013, + "logits/rejected": -0.02981414832174778, + "logps/chosen": -199.36563110351562, + "logps/rejected": -208.70938110351562, + "logps/weighted_chosen": -0.6963897943496704, + "logps/weighted_rejected": -0.969342052936554, + "loss": 0.676, + "rewards/accuracies": 0.6031249761581421, + "rewards/chosen": -71.73905944824219, + "rewards/margins": 6.481640815734863, + "rewards/rejected": -78.2171859741211, + "rewards/weighted_accuracies": 0.550000011920929, + "rewards/weighted_chosen": -0.2880393862724304, + "rewards/weighted_margins": 0.1389778107404709, + "rewards/weighted_rejected": -0.426931768655777, + "step": 920 + }, + { + "epoch": 0.4970271895250184, + "grad_norm": 5.530470371246338, + "learning_rate": 5.936540068371394e-07, + "logits/chosen": 0.006648254580795765, + "logits/rejected": 0.005427551455795765, + "logps/chosen": -193.99374389648438, + "logps/rejected": -202.4406280517578, + "logps/weighted_chosen": -1.107049584388733, + "logps/weighted_rejected": -1.3529266119003296, + "loss": 0.6926, + "rewards/accuracies": 0.612500011920929, + "rewards/chosen": -74.79765319824219, + "rewards/margins": 5.872656345367432, + "rewards/rejected": -80.6624984741211, + "rewards/weighted_accuracies": 0.559374988079071, + "rewards/weighted_chosen": -0.43673935532569885, + "rewards/weighted_margins": 0.12621307373046875, + "rewards/weighted_rejected": -0.5626624822616577, + "step": 930 + }, + { + "epoch": 0.5023715679070078, + "grad_norm": 5.006593704223633, + "learning_rate": 5.844755624921918e-07, + "logits/chosen": -0.005947875790297985, + "logits/rejected": 0.02864990197122097, + "logps/chosen": -187.453125, + "logps/rejected": -192.7375030517578, + "logps/weighted_chosen": -1.132598876953125, + "logps/weighted_rejected": -1.0424163341522217, + "loss": 0.7378, + "rewards/accuracies": 0.609375, + "rewards/chosen": -70.77812194824219, + "rewards/margins": 5.782812595367432, + "rewards/rejected": -76.56562805175781, + "rewards/weighted_accuracies": 0.512499988079071, + "rewards/weighted_chosen": -0.44052886962890625, + "rewards/weighted_margins": 0.07312469184398651, + "rewards/weighted_rejected": -0.5137832760810852, + "step": 940 + }, + { + "epoch": 0.5077159462889973, + "grad_norm": 16.502933502197266, + "learning_rate": 5.752677190450634e-07, + "logits/chosen": -0.021286392584443092, + "logits/rejected": -0.00705032330006361, + "logps/chosen": -186.49374389648438, + "logps/rejected": -196.24844360351562, + "logps/weighted_chosen": -1.068359375, + "logps/weighted_rejected": -1.3232848644256592, + "loss": 0.7251, + "rewards/accuracies": 0.625, + "rewards/chosen": -67.70234680175781, + "rewards/margins": 7.561327934265137, + "rewards/rejected": -75.28047180175781, + "rewards/weighted_accuracies": 0.528124988079071, + "rewards/weighted_chosen": -0.390481561422348, + "rewards/weighted_margins": 0.058994293212890625, + "rewards/weighted_rejected": -0.44965285062789917, + "step": 950 + }, + { + "epoch": 0.5130603246709867, + "grad_norm": 12.000792503356934, + "learning_rate": 5.66033681000375e-07, + "logits/chosen": -0.03893890231847763, + "logits/rejected": 0.01745758019387722, + "logps/chosen": -189.5234375, + "logps/rejected": -201.2624969482422, + "logps/weighted_chosen": -1.0573517084121704, + "logps/weighted_rejected": -1.4250671863555908, + "loss": 0.6656, + "rewards/accuracies": 0.637499988079071, + "rewards/chosen": -69.40625, + "rewards/margins": 8.408203125, + "rewards/rejected": -77.78593444824219, + "rewards/weighted_accuracies": 0.5718749761581421, + "rewards/weighted_chosen": -0.2941024899482727, + "rewards/weighted_margins": 0.1710365265607834, + "rewards/weighted_rejected": -0.4653762876987457, + "step": 960 + }, + { + "epoch": 0.5184047030529761, + "grad_norm": 8.829608917236328, + "learning_rate": 5.567766619789658e-07, + "logits/chosen": -0.03634033352136612, + "logits/rejected": 0.0003803253057412803, + "logps/chosen": -190.27969360351562, + "logps/rejected": -199.0500030517578, + "logps/weighted_chosen": -1.1300811767578125, + "logps/weighted_rejected": -1.3106689453125, + "loss": 0.6931, + "rewards/accuracies": 0.565625011920929, + "rewards/chosen": -67.55078125, + "rewards/margins": 7.454297065734863, + "rewards/rejected": -74.9703140258789, + "rewards/weighted_accuracies": 0.606249988079071, + "rewards/weighted_chosen": -0.3190757632255554, + "rewards/weighted_margins": 0.11787338554859161, + "rewards/weighted_rejected": -0.4370986819267273, + "step": 970 + }, + { + "epoch": 0.5237490814349656, + "grad_norm": 9.326337814331055, + "learning_rate": 5.474998835994916e-07, + "logits/chosen": 0.009844208136200905, + "logits/rejected": 0.051239013671875, + "logps/chosen": -195.78125, + "logps/rejected": -208.5343780517578, + "logps/weighted_chosen": -1.179107666015625, + "logps/weighted_rejected": -1.546759009361267, + "loss": 0.6677, + "rewards/accuracies": 0.6343749761581421, + "rewards/chosen": -70.6781234741211, + "rewards/margins": 6.912109375, + "rewards/rejected": -77.5609359741211, + "rewards/weighted_accuracies": 0.5625, + "rewards/weighted_chosen": -0.28845518827438354, + "rewards/weighted_margins": 0.1378173828125, + "rewards/weighted_rejected": -0.42629241943359375, + "step": 980 + }, + { + "epoch": 0.529093459816955, + "grad_norm": 10.308549880981445, + "learning_rate": 5.382065743572411e-07, + "logits/chosen": -0.025909423828125, + "logits/rejected": 0.01560287456959486, + "logps/chosen": -192.8093719482422, + "logps/rejected": -205.2843780517578, + "logps/weighted_chosen": -0.8877533078193665, + "logps/weighted_rejected": -1.2557891607284546, + "loss": 0.6811, + "rewards/accuracies": 0.671875, + "rewards/chosen": -70.57890319824219, + "rewards/margins": 8.194921493530273, + "rewards/rejected": -78.80000305175781, + "rewards/weighted_accuracies": 0.534375011920929, + "rewards/weighted_chosen": -0.3638969361782074, + "rewards/weighted_margins": 0.13902893662452698, + "rewards/weighted_rejected": -0.5028846859931946, + "step": 990 + }, + { + "epoch": 0.5344378381989445, + "grad_norm": 6.612018585205078, + "learning_rate": 5.288999685005571e-07, + "logits/chosen": -0.03016967698931694, + "logits/rejected": -0.0050643919967114925, + "logps/chosen": -192.640625, + "logps/rejected": -201.4656219482422, + "logps/weighted_chosen": -0.626147449016571, + "logps/weighted_rejected": -0.8550964593887329, + "loss": 0.672, + "rewards/accuracies": 0.5843750238418579, + "rewards/chosen": -72.01640319824219, + "rewards/margins": 6.489062309265137, + "rewards/rejected": -78.5062484741211, + "rewards/weighted_accuracies": 0.581250011920929, + "rewards/weighted_chosen": -0.2895965576171875, + "rewards/weighted_margins": 0.1136070266366005, + "rewards/weighted_rejected": -0.4031356871128082, + "step": 1000 + }, + { + "epoch": 0.5344378381989445, + "eval_logits/chosen": -0.1623995155096054, + "eval_logits/rejected": -0.13676175475120544, + "eval_logps/chosen": -193.06211853027344, + "eval_logps/rejected": -200.78919982910156, + "eval_logps/weighted_chosen": -1.0030580759048462, + "eval_logps/weighted_rejected": -1.3398514986038208, + "eval_loss": 0.6716485619544983, + "eval_rewards/accuracies": 0.5982688665390015, + "eval_rewards/chosen": -71.73777770996094, + "eval_rewards/margins": 6.556771755218506, + "eval_rewards/rejected": -78.29378509521484, + "eval_rewards/weighted_accuracies": 0.5570265054702759, + "eval_rewards/weighted_chosen": -0.36391088366508484, + "eval_rewards/weighted_margins": 0.17662912607192993, + "eval_rewards/weighted_rejected": -0.5405399203300476, + "eval_runtime": 1144.2369, + "eval_samples_per_second": 1.714, + "eval_steps_per_second": 0.429, + "step": 1000 + } + ], + "logging_steps": 10, + "max_steps": 1872, + "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..7b414b2 --- /dev/null +++ b/checkpoint-1000/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fa2dcf6fe65fd94a0fde469e207f237b0752c6e558745acec2daa0276a1175ce +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..ceed8ae --- /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": 128009, + "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..38e1b41 --- /dev/null +++ b/checkpoint-1500/generation_config.json @@ -0,0 +1,12 @@ +{ + "bos_token_id": 128000, + "do_sample": true, + "eos_token_id": [ + 128001, + 128009 + ], + "max_length": 4096, + "temperature": 0.6, + "top_p": 0.9, + "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..c69c2b3 --- /dev/null +++ b/checkpoint-1500/model-00001-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3540c5ad481e40d8cc01aaf92ff518c466482b964a505216eccdc14b404cf000 +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..7adb592 --- /dev/null +++ b/checkpoint-1500/model-00002-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:60b49b13f2ace7ebf151412d9a3a12b4db378bbcfe8dcf00a7eaffa77af05b92 +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..9bf8c75 --- /dev/null +++ b/checkpoint-1500/model-00003-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:646e860466ca994c9e673ae4a2666186c7494ca7023633bbc120cf5aff103e08 +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..afe74a3 --- /dev/null +++ b/checkpoint-1500/model-00004-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e07911476889abbd8bfd0d873421c64e1d33c30c15c38320d9513da702b0353a +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..ef5650d --- /dev/null +++ b/checkpoint-1500/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f53922f0fb6da4e5c5c6e6c579d18979967d65d494cb8faee4b2651a5778c1d1 +size 1465 diff --git a/checkpoint-1500/special_tokens_map.json b/checkpoint-1500/special_tokens_map.json new file mode 100644 index 0000000..b43be96 --- /dev/null +++ b/checkpoint-1500/special_tokens_map.json @@ -0,0 +1,17 @@ +{ + "bos_token": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "eos_token": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "pad_token": "<|eot_id|>" +} diff --git a/checkpoint-1500/tokenizer.json b/checkpoint-1500/tokenizer.json new file mode 100644 index 0000000..86a3394 --- /dev/null +++ b/checkpoint-1500/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3c5cf44023714fb39b05e71e425f8d7b92805ff73f7988b083b8c87f0bf87393 +size 17209961 diff --git a/checkpoint-1500/tokenizer_config.json b/checkpoint-1500/tokenizer_config.json new file mode 100644 index 0000000..34d134f --- /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": "<|eot_id|>", + "extra_special_tokens": {}, + "model_input_names": [ + "input_ids", + "attention_mask" + ], + "model_max_length": 1000000000000000019884624838656, + "pad_token": "<|eot_id|>", + "tokenizer_class": "PreTrainedTokenizerFast" +} diff --git a/checkpoint-1500/trainer_state.json b/checkpoint-1500/trainer_state.json new file mode 100644 index 0000000..ee3f1a5 --- /dev/null +++ b/checkpoint-1500/trainer_state.json @@ -0,0 +1,3271 @@ +{ + "best_global_step": null, + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 0.8016567572984167, + "eval_steps": 500, + "global_step": 1500, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.0005344378381989445, + "grad_norm": 63.52920150756836, + "learning_rate": 0.0, + "logits/chosen": -0.144775390625, + "logits/rejected": -0.1036376953125, + "logps/chosen": -110.828125, + "logps/rejected": -116.828125, + "logps/weighted_chosen": -0.2086181640625, + "logps/weighted_rejected": -1.271392822265625, + "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": 0.005344378381989445, + "grad_norm": 102.77810668945312, + "learning_rate": 4.787234042553191e-08, + "logits/chosen": -0.2549913227558136, + "logits/rejected": -0.24070316553115845, + "logps/chosen": -135.32899475097656, + "logps/rejected": -138.99305725097656, + "logps/weighted_chosen": -0.8547499179840088, + "logps/weighted_rejected": -0.9027608036994934, + "loss": 0.6928, + "rewards/accuracies": 0.2951388955116272, + "rewards/chosen": -0.0164930559694767, + "rewards/margins": -0.0349392369389534, + "rewards/rejected": 0.0184461809694767, + "rewards/weighted_accuracies": 0.4027777910232544, + "rewards/weighted_chosen": -0.002239651046693325, + "rewards/weighted_margins": 0.0028160943184047937, + "rewards/weighted_rejected": -0.0050794812850654125, + "step": 10 + }, + { + "epoch": 0.01068875676397889, + "grad_norm": 9.201544761657715, + "learning_rate": 1.0106382978723404e-07, + "logits/chosen": -0.19488906860351562, + "logits/rejected": -0.183399960398674, + "logps/chosen": -116.0374984741211, + "logps/rejected": -115.5054702758789, + "logps/weighted_chosen": -0.6929916143417358, + "logps/weighted_rejected": -0.6838119626045227, + "loss": 0.6917, + "rewards/accuracies": 0.38749998807907104, + "rewards/chosen": 0.01796874962747097, + "rewards/margins": 0.03007812425494194, + "rewards/rejected": -0.01210937462747097, + "rewards/weighted_accuracies": 0.4781250059604645, + "rewards/weighted_chosen": 0.0022293091751635075, + "rewards/weighted_margins": 0.00339679722674191, + "rewards/weighted_rejected": -0.0011629104847088456, + "step": 20 + }, + { + "epoch": 0.016033135145968335, + "grad_norm": 7.109442234039307, + "learning_rate": 1.5425531914893615e-07, + "logits/chosen": -0.1776607483625412, + "logits/rejected": -0.17884215712547302, + "logps/chosen": -123.2867202758789, + "logps/rejected": -129.75, + "logps/weighted_chosen": -0.7567901611328125, + "logps/weighted_rejected": -1.2207733392715454, + "loss": 0.6913, + "rewards/accuracies": 0.3968749940395355, + "rewards/chosen": -0.0074218749068677425, + "rewards/margins": 0.04843749850988388, + "rewards/rejected": -0.05585937574505806, + "rewards/weighted_accuracies": 0.5, + "rewards/weighted_chosen": 0.0007101058727130294, + "rewards/weighted_margins": 0.004384422209113836, + "rewards/weighted_rejected": -0.0036745071411132812, + "step": 30 + }, + { + "epoch": 0.02137751352795778, + "grad_norm": 6.0048956871032715, + "learning_rate": 2.074468085106383e-07, + "logits/chosen": -0.23842163383960724, + "logits/rejected": -0.2298881560564041, + "logps/chosen": -127.34062194824219, + "logps/rejected": -128.1710968017578, + "logps/weighted_chosen": -0.8051223754882812, + "logps/weighted_rejected": -0.5736129879951477, + "loss": 0.6934, + "rewards/accuracies": 0.34062498807907104, + "rewards/chosen": -0.01992187462747097, + "rewards/margins": -0.05351562425494194, + "rewards/rejected": 0.03359375149011612, + "rewards/weighted_accuracies": 0.4593749940395355, + "rewards/weighted_chosen": -7.83920258982107e-05, + "rewards/weighted_margins": -0.0009832382202148438, + "rewards/weighted_rejected": 0.000904083251953125, + "step": 40 + }, + { + "epoch": 0.026721891909947223, + "grad_norm": 32.7233772277832, + "learning_rate": 2.6063829787234044e-07, + "logits/chosen": -0.18835143744945526, + "logits/rejected": -0.170085147023201, + "logps/chosen": -120.88594055175781, + "logps/rejected": -121.3515625, + "logps/weighted_chosen": -0.6867736577987671, + "logps/weighted_rejected": -1.138249158859253, + "loss": 0.6903, + "rewards/accuracies": 0.3656249940395355, + "rewards/chosen": 0.03554687649011612, + "rewards/margins": -0.03164062649011612, + "rewards/rejected": 0.06718750298023224, + "rewards/weighted_accuracies": 0.515625, + "rewards/weighted_chosen": -0.006287765689194202, + "rewards/weighted_margins": 0.0064636231400072575, + "rewards/weighted_rejected": -0.01274185162037611, + "step": 50 + }, + { + "epoch": 0.03206627029193667, + "grad_norm": 55.560707092285156, + "learning_rate": 3.1382978723404253e-07, + "logits/chosen": -0.1308601349592209, + "logits/rejected": -0.1308906525373459, + "logps/chosen": -116.5875015258789, + "logps/rejected": -119.5406265258789, + "logps/weighted_chosen": -0.8019469976425171, + "logps/weighted_rejected": -0.8834930658340454, + "loss": 0.7002, + "rewards/accuracies": 0.3343749940395355, + "rewards/chosen": 0.03476562350988388, + "rewards/margins": -0.06171875074505806, + "rewards/rejected": 0.09648437798023224, + "rewards/weighted_accuracies": 0.4781250059604645, + "rewards/weighted_chosen": -0.01903533935546875, + "rewards/weighted_margins": -2.880096508306451e-05, + "rewards/weighted_rejected": -0.019007300958037376, + "step": 60 + }, + { + "epoch": 0.037410648673926114, + "grad_norm": 20.22484588623047, + "learning_rate": 3.6702127659574467e-07, + "logits/chosen": -0.20572929084300995, + "logits/rejected": -0.19457244873046875, + "logps/chosen": -114.4507827758789, + "logps/rejected": -118.63594055175781, + "logps/weighted_chosen": -0.6070801019668579, + "logps/weighted_rejected": -0.6768280267715454, + "loss": 0.6911, + "rewards/accuracies": 0.42500001192092896, + "rewards/chosen": 0.16484375298023224, + "rewards/margins": 0.07148437201976776, + "rewards/rejected": 0.09335937350988388, + "rewards/weighted_accuracies": 0.4749999940395355, + "rewards/weighted_chosen": -0.010281562805175781, + "rewards/weighted_margins": 0.007259177975356579, + "rewards/weighted_rejected": -0.017521286383271217, + "step": 70 + }, + { + "epoch": 0.04275502705591556, + "grad_norm": 15.305061340332031, + "learning_rate": 4.202127659574468e-07, + "logits/chosen": -0.19508513808250427, + "logits/rejected": -0.12093047797679901, + "logps/chosen": -112.57734680175781, + "logps/rejected": -112.77734375, + "logps/weighted_chosen": -0.4597320556640625, + "logps/weighted_rejected": -0.9711883664131165, + "loss": 0.6952, + "rewards/accuracies": 0.37812501192092896, + "rewards/chosen": 0.04570312425494194, + "rewards/margins": 0.01914062537252903, + "rewards/rejected": 0.02656250074505806, + "rewards/weighted_accuracies": 0.4625000059604645, + "rewards/weighted_chosen": -0.02759857103228569, + "rewards/weighted_margins": 0.021048355847597122, + "rewards/weighted_rejected": -0.04872245714068413, + "step": 80 + }, + { + "epoch": 0.048099405437905, + "grad_norm": 11.991273880004883, + "learning_rate": 4.734042553191489e-07, + "logits/chosen": -0.17391815781593323, + "logits/rejected": -0.11656417697668076, + "logps/chosen": -119.3070297241211, + "logps/rejected": -118.51249694824219, + "logps/weighted_chosen": -1.0416854619979858, + "logps/weighted_rejected": -1.5859466791152954, + "loss": 0.6736, + "rewards/accuracies": 0.4375, + "rewards/chosen": -0.08164062350988388, + "rewards/margins": 0.341796875, + "rewards/rejected": -0.4234375059604645, + "rewards/weighted_accuracies": 0.515625, + "rewards/weighted_chosen": -0.04084587097167969, + "rewards/weighted_margins": 0.10567589104175568, + "rewards/weighted_rejected": -0.14661255478858948, + "step": 90 + }, + { + "epoch": 0.053443783819894446, + "grad_norm": 17.195886611938477, + "learning_rate": 5.26595744680851e-07, + "logits/chosen": -0.204203799366951, + "logits/rejected": -0.14691162109375, + "logps/chosen": -125.05155944824219, + "logps/rejected": -125.3671875, + "logps/weighted_chosen": -0.7374237179756165, + "logps/weighted_rejected": -0.7348953485488892, + "loss": 0.7064, + "rewards/accuracies": 0.4281249940395355, + "rewards/chosen": -0.3832031190395355, + "rewards/margins": 0.15585938096046448, + "rewards/rejected": -0.5390625, + "rewards/weighted_accuracies": 0.46875, + "rewards/weighted_chosen": -0.060018159449100494, + "rewards/weighted_margins": 0.0024120330344885588, + "rewards/weighted_rejected": -0.06242714077234268, + "step": 100 + }, + { + "epoch": 0.058788162201883896, + "grad_norm": 9.103920936584473, + "learning_rate": 5.797872340425531e-07, + "logits/chosen": -0.2138931304216385, + "logits/rejected": -0.19011077284812927, + "logps/chosen": -118.6820297241211, + "logps/rejected": -121.28593444824219, + "logps/weighted_chosen": -0.4821228086948395, + "logps/weighted_rejected": -0.7759445309638977, + "loss": 0.6939, + "rewards/accuracies": 0.4281249940395355, + "rewards/chosen": -0.896484375, + "rewards/margins": 0.09531249850988388, + "rewards/rejected": -0.9917968511581421, + "rewards/weighted_accuracies": 0.528124988079071, + "rewards/weighted_chosen": -0.05413341522216797, + "rewards/weighted_margins": 0.04503212124109268, + "rewards/weighted_rejected": -0.09918193519115448, + "step": 110 + }, + { + "epoch": 0.06413254058387334, + "grad_norm": 8.539896011352539, + "learning_rate": 6.329787234042553e-07, + "logits/chosen": -0.17510680854320526, + "logits/rejected": -0.14726562798023224, + "logps/chosen": -122.58906555175781, + "logps/rejected": -128.4656219482422, + "logps/weighted_chosen": -0.9249496459960938, + "logps/weighted_rejected": -0.762683093547821, + "loss": 0.7325, + "rewards/accuracies": 0.4375, + "rewards/chosen": -1.068359375, + "rewards/margins": 0.28398436307907104, + "rewards/rejected": -1.3523437976837158, + "rewards/weighted_accuracies": 0.4781250059604645, + "rewards/weighted_chosen": -0.098027803003788, + "rewards/weighted_margins": -0.029563140124082565, + "rewards/weighted_rejected": -0.06836700439453125, + "step": 120 + }, + { + "epoch": 0.06947691896586278, + "grad_norm": 15.564044952392578, + "learning_rate": 6.861702127659574e-07, + "logits/chosen": -0.20233002305030823, + "logits/rejected": -0.18396148085594177, + "logps/chosen": -122.2671890258789, + "logps/rejected": -123.8382797241211, + "logps/weighted_chosen": -0.7501281499862671, + "logps/weighted_rejected": -0.821154773235321, + "loss": 0.7078, + "rewards/accuracies": 0.4000000059604645, + "rewards/chosen": -1.45703125, + "rewards/margins": 0.29179686307907104, + "rewards/rejected": -1.7488281726837158, + "rewards/weighted_accuracies": 0.5062500238418579, + "rewards/weighted_chosen": -0.007884788326919079, + "rewards/weighted_margins": 0.006314086727797985, + "rewards/weighted_rejected": -0.014200759120285511, + "step": 130 + }, + { + "epoch": 0.07482129734785223, + "grad_norm": 9.176558494567871, + "learning_rate": 7.393617021276596e-07, + "logits/chosen": -0.16455383598804474, + "logits/rejected": -0.14660033583641052, + "logps/chosen": -116.328125, + "logps/rejected": -119.9664077758789, + "logps/weighted_chosen": -0.9873703122138977, + "logps/weighted_rejected": -0.6730011105537415, + "loss": 0.7174, + "rewards/accuracies": 0.4749999940395355, + "rewards/chosen": -1.6007812023162842, + "rewards/margins": 0.533203125, + "rewards/rejected": -2.133984327316284, + "rewards/weighted_accuracies": 0.515625, + "rewards/weighted_chosen": 0.02084655687212944, + "rewards/weighted_margins": 0.05102996900677681, + "rewards/weighted_rejected": -0.030106354504823685, + "step": 140 + }, + { + "epoch": 0.08016567572984168, + "grad_norm": 10.172534942626953, + "learning_rate": 7.925531914893616e-07, + "logits/chosen": -0.1509498655796051, + "logits/rejected": -0.13156357407569885, + "logps/chosen": -120.0562515258789, + "logps/rejected": -117.85546875, + "logps/weighted_chosen": -0.8198806643486023, + "logps/weighted_rejected": -0.7782562375068665, + "loss": 0.7005, + "rewards/accuracies": 0.46562498807907104, + "rewards/chosen": -2.0234375, + "rewards/margins": 0.607421875, + "rewards/rejected": -2.630859375, + "rewards/weighted_accuracies": 0.543749988079071, + "rewards/weighted_chosen": -0.0781441181898117, + "rewards/weighted_margins": 0.034273337572813034, + "rewards/weighted_rejected": -0.11248989403247833, + "step": 150 + }, + { + "epoch": 0.08551005411183112, + "grad_norm": 25.879701614379883, + "learning_rate": 8.457446808510637e-07, + "logits/chosen": -0.17474937438964844, + "logits/rejected": -0.17009583115577698, + "logps/chosen": -121.2367172241211, + "logps/rejected": -121.62109375, + "logps/weighted_chosen": -0.8304061889648438, + "logps/weighted_rejected": -0.866925060749054, + "loss": 0.7072, + "rewards/accuracies": 0.4781250059604645, + "rewards/chosen": -2.219531297683716, + "rewards/margins": 0.716796875, + "rewards/rejected": -2.936328172683716, + "rewards/weighted_accuracies": 0.47187501192092896, + "rewards/weighted_chosen": -0.07932891696691513, + "rewards/weighted_margins": 0.017035294324159622, + "rewards/weighted_rejected": -0.09644661098718643, + "step": 160 + }, + { + "epoch": 0.09085443249382057, + "grad_norm": 9.220558166503906, + "learning_rate": 8.989361702127659e-07, + "logits/chosen": -0.22574539482593536, + "logits/rejected": -0.19882050156593323, + "logps/chosen": -120.62968444824219, + "logps/rejected": -124.78437805175781, + "logps/weighted_chosen": -0.9401000738143921, + "logps/weighted_rejected": -0.8424423336982727, + "loss": 0.7174, + "rewards/accuracies": 0.4937500059604645, + "rewards/chosen": -2.4033203125, + "rewards/margins": 0.613085925579071, + "rewards/rejected": -3.016406297683716, + "rewards/weighted_accuracies": 0.48750001192092896, + "rewards/weighted_chosen": -0.04837837070226669, + "rewards/weighted_margins": 0.04595222324132919, + "rewards/weighted_rejected": -0.09437103569507599, + "step": 170 + }, + { + "epoch": 0.09619881087581, + "grad_norm": 55.62498092651367, + "learning_rate": 9.52127659574468e-07, + "logits/chosen": -0.2216789275407791, + "logits/rejected": -0.20373383164405823, + "logps/chosen": -120.3851547241211, + "logps/rejected": -121.2328109741211, + "logps/weighted_chosen": -0.435232549905777, + "logps/weighted_rejected": -0.5373809933662415, + "loss": 0.7302, + "rewards/accuracies": 0.47187501192092896, + "rewards/chosen": -1.338281273841858, + "rewards/margins": 0.3658203184604645, + "rewards/rejected": -1.7041015625, + "rewards/weighted_accuracies": 0.53125, + "rewards/weighted_chosen": -0.08163909614086151, + "rewards/weighted_margins": -0.008893203921616077, + "rewards/weighted_rejected": -0.07281456142663956, + "step": 180 + }, + { + "epoch": 0.10154318925779945, + "grad_norm": 39.38788986206055, + "learning_rate": 9.99999129927477e-07, + "logits/chosen": -0.19243164360523224, + "logits/rejected": -0.15758666396141052, + "logps/chosen": -121.8960952758789, + "logps/rejected": -123.01094055175781, + "logps/weighted_chosen": -0.953173816204071, + "logps/weighted_rejected": -1.0853087902069092, + "loss": 0.6702, + "rewards/accuracies": 0.48750001192092896, + "rewards/chosen": -2.100390672683716, + "rewards/margins": 0.662890613079071, + "rewards/rejected": -2.7632813453674316, + "rewards/weighted_accuracies": 0.578125, + "rewards/weighted_chosen": -0.09999503940343857, + "rewards/weighted_margins": 0.12822990119457245, + "rewards/weighted_rejected": -0.22818031907081604, + "step": 190 + }, + { + "epoch": 0.10688756763978889, + "grad_norm": 12.883384704589844, + "learning_rate": 9.99894724888679e-07, + "logits/chosen": -0.17392578721046448, + "logits/rejected": -0.147491455078125, + "logps/chosen": -129.53750610351562, + "logps/rejected": -128.95703125, + "logps/weighted_chosen": -0.7322753667831421, + "logps/weighted_rejected": -1.0728164911270142, + "loss": 0.6777, + "rewards/accuracies": 0.4749999940395355, + "rewards/chosen": -3.215625047683716, + "rewards/margins": 0.9507812261581421, + "rewards/rejected": -4.166406154632568, + "rewards/weighted_accuracies": 0.5375000238418579, + "rewards/weighted_chosen": -0.1104557067155838, + "rewards/weighted_margins": 0.13128623366355896, + "rewards/weighted_rejected": -0.2415626496076584, + "step": 200 + }, + { + "epoch": 0.11223194602177834, + "grad_norm": 17.96660041809082, + "learning_rate": 9.996163469793475e-07, + "logits/chosen": -0.18083801865577698, + "logits/rejected": -0.12766647338867188, + "logps/chosen": -128.1953125, + "logps/rejected": -123.03047180175781, + "logps/weighted_chosen": -0.8699920773506165, + "logps/weighted_rejected": -0.8845611810684204, + "loss": 0.7572, + "rewards/accuracies": 0.5625, + "rewards/chosen": -3.5113282203674316, + "rewards/margins": 1.3683593273162842, + "rewards/rejected": -4.879296779632568, + "rewards/weighted_accuracies": 0.5531250238418579, + "rewards/weighted_chosen": -0.07919616997241974, + "rewards/weighted_margins": 0.04895629733800888, + "rewards/weighted_rejected": -0.12815475463867188, + "step": 210 + }, + { + "epoch": 0.11757632440376779, + "grad_norm": 63.790077209472656, + "learning_rate": 9.991640930802883e-07, + "logits/chosen": -0.14619751274585724, + "logits/rejected": -0.13860473036766052, + "logps/chosen": -126.61015319824219, + "logps/rejected": -129.9773406982422, + "logps/weighted_chosen": -1.0776779651641846, + "logps/weighted_rejected": -0.9309722781181335, + "loss": 0.7107, + "rewards/accuracies": 0.5531250238418579, + "rewards/chosen": -4.296484470367432, + "rewards/margins": 1.10546875, + "rewards/rejected": -5.40234375, + "rewards/weighted_accuracies": 0.5406249761581421, + "rewards/weighted_chosen": -0.013729858212172985, + "rewards/weighted_margins": 0.14467468857765198, + "rewards/weighted_rejected": -0.1582542359828949, + "step": 220 + }, + { + "epoch": 0.12292070278575723, + "grad_norm": 9.634794235229492, + "learning_rate": 9.98538120584459e-07, + "logits/chosen": -0.1792152374982834, + "logits/rejected": -0.1362098753452301, + "logps/chosen": -133.7140655517578, + "logps/rejected": -137.7546844482422, + "logps/weighted_chosen": -0.6676223874092102, + "logps/weighted_rejected": -0.9518234133720398, + "loss": 0.7614, + "rewards/accuracies": 0.515625, + "rewards/chosen": -4.493554592132568, + "rewards/margins": 1.0623047351837158, + "rewards/rejected": -5.555859565734863, + "rewards/weighted_accuracies": 0.565625011920929, + "rewards/weighted_chosen": -0.1120578795671463, + "rewards/weighted_margins": 0.031700897961854935, + "rewards/weighted_rejected": -0.14376182854175568, + "step": 230 + }, + { + "epoch": 0.12826508116774668, + "grad_norm": 8.717682838439941, + "learning_rate": 9.977386473421917e-07, + "logits/chosen": -0.09295501559972763, + "logits/rejected": -0.07836151123046875, + "logps/chosen": -121.78202819824219, + "logps/rejected": -126.39453125, + "logps/weighted_chosen": -0.8116515874862671, + "logps/weighted_rejected": -1.2165496349334717, + "loss": 0.6731, + "rewards/accuracies": 0.5874999761581421, + "rewards/chosen": -7.553515434265137, + "rewards/margins": 1.669921875, + "rewards/rejected": -9.224218368530273, + "rewards/weighted_accuracies": 0.534375011920929, + "rewards/weighted_chosen": -0.1755084991455078, + "rewards/weighted_margins": 0.15414047241210938, + "rewards/weighted_rejected": -0.32989272475242615, + "step": 240 + }, + { + "epoch": 0.13360945954973613, + "grad_norm": 35.197269439697266, + "learning_rate": 9.96765951585378e-07, + "logits/chosen": -0.021549224853515625, + "logits/rejected": 0.00345611572265625, + "logps/chosen": -123.0296859741211, + "logps/rejected": -127.9671859741211, + "logps/weighted_chosen": -1.186151146888733, + "logps/weighted_rejected": -1.386474609375, + "loss": 0.7026, + "rewards/accuracies": 0.5625, + "rewards/chosen": -9.133593559265137, + "rewards/margins": 1.150781273841858, + "rewards/rejected": -10.283984184265137, + "rewards/weighted_accuracies": 0.59375, + "rewards/weighted_chosen": -0.3282794952392578, + "rewards/weighted_margins": 0.1091407760977745, + "rewards/weighted_rejected": -0.4371795654296875, + "step": 250 + }, + { + "epoch": 0.13895383793172555, + "grad_norm": 84.2197494506836, + "learning_rate": 9.956203718306388e-07, + "logits/chosen": 0.025267411023378372, + "logits/rejected": 0.06335143744945526, + "logps/chosen": -128.8429718017578, + "logps/rejected": -133.50625610351562, + "logps/weighted_chosen": -0.9628280401229858, + "logps/weighted_rejected": -1.2008453607559204, + "loss": 0.7443, + "rewards/accuracies": 0.5, + "rewards/chosen": -10.861719131469727, + "rewards/margins": 0.871874988079071, + "rewards/rejected": -11.733593940734863, + "rewards/weighted_accuracies": 0.49687498807907104, + "rewards/weighted_chosen": -0.26259079575538635, + "rewards/weighted_margins": 0.06770706176757812, + "rewards/weighted_rejected": -0.3305191099643707, + "step": 260 + }, + { + "epoch": 0.144298216313715, + "grad_norm": 57.09628677368164, + "learning_rate": 9.943023067615136e-07, + "logits/chosen": -0.017973709851503372, + "logits/rejected": -0.0057998658157885075, + "logps/chosen": -137.60311889648438, + "logps/rejected": -140.91561889648438, + "logps/weighted_chosen": -0.8694183230400085, + "logps/weighted_rejected": -0.8910156488418579, + "loss": 0.7462, + "rewards/accuracies": 0.5625, + "rewards/chosen": -10.946093559265137, + "rewards/margins": 1.662109375, + "rewards/rejected": -12.608202934265137, + "rewards/weighted_accuracies": 0.565625011920929, + "rewards/weighted_chosen": -0.06702041625976562, + "rewards/weighted_margins": 0.13949890434741974, + "rewards/weighted_rejected": -0.20627327263355255, + "step": 270 + }, + { + "epoch": 0.14964259469570446, + "grad_norm": 45.0294303894043, + "learning_rate": 9.928122150897112e-07, + "logits/chosen": -0.045014191418886185, + "logits/rejected": 0.02437591552734375, + "logps/chosen": -130.1125030517578, + "logps/rejected": -131.6171875, + "logps/weighted_chosen": -0.7659690976142883, + "logps/weighted_rejected": -0.8499618768692017, + "loss": 0.7079, + "rewards/accuracies": 0.5406249761581421, + "rewards/chosen": -12.664453506469727, + "rewards/margins": 1.584375023841858, + "rewards/rejected": -14.249608993530273, + "rewards/weighted_accuracies": 0.543749988079071, + "rewards/weighted_chosen": -0.22265930473804474, + "rewards/weighted_margins": 0.07556991279125214, + "rewards/weighted_rejected": -0.298013299703598, + "step": 280 + }, + { + "epoch": 0.1549869730776939, + "grad_norm": 9.370684623718262, + "learning_rate": 9.9115061539547e-07, + "logits/chosen": -0.02069244347512722, + "logits/rejected": 0.01569976843893528, + "logps/chosen": -139.4453125, + "logps/rejected": -141.2578125, + "logps/weighted_chosen": -0.8775177001953125, + "logps/weighted_rejected": -1.2375701665878296, + "loss": 0.6949, + "rewards/accuracies": 0.5375000238418579, + "rewards/chosen": -13.07421875, + "rewards/margins": 1.177734375, + "rewards/rejected": -14.249608993530273, + "rewards/weighted_accuracies": 0.546875, + "rewards/weighted_chosen": -0.18484345078468323, + "rewards/weighted_margins": 0.12048187106847763, + "rewards/weighted_rejected": -0.30528029799461365, + "step": 290 + }, + { + "epoch": 0.16033135145968336, + "grad_norm": 30.68979263305664, + "learning_rate": 9.893180859470818e-07, + "logits/chosen": -0.0137786865234375, + "logits/rejected": 0.041573334485292435, + "logps/chosen": -130.22265625, + "logps/rejected": -132.06094360351562, + "logps/weighted_chosen": -1.1769134998321533, + "logps/weighted_rejected": -1.2438628673553467, + "loss": 0.7261, + "rewards/accuracies": 0.590624988079071, + "rewards/chosen": -13.093358993530273, + "rewards/margins": 2.108203172683716, + "rewards/rejected": -15.200780868530273, + "rewards/weighted_accuracies": 0.5843750238418579, + "rewards/weighted_chosen": -0.2848831117153168, + "rewards/weighted_margins": 0.07841148227453232, + "rewards/weighted_rejected": -0.36317139863967896, + "step": 300 + }, + { + "epoch": 0.16567572984167278, + "grad_norm": 71.56196594238281, + "learning_rate": 9.873152644996424e-07, + "logits/chosen": -0.11114044487476349, + "logits/rejected": -0.11599807441234589, + "logps/chosen": -131.01718139648438, + "logps/rejected": -132.28125, + "logps/weighted_chosen": -1.1032531261444092, + "logps/weighted_rejected": -0.8420807123184204, + "loss": 0.7872, + "rewards/accuracies": 0.5718749761581421, + "rewards/chosen": -12.979687690734863, + "rewards/margins": 2.407421827316284, + "rewards/rejected": -15.388280868530273, + "rewards/weighted_accuracies": 0.5218750238418579, + "rewards/weighted_chosen": 0.05532989650964737, + "rewards/weighted_margins": 0.2024696320295334, + "rewards/weighted_rejected": -0.1469375640153885, + "step": 310 + }, + { + "epoch": 0.17102010822366223, + "grad_norm": 12.055798530578613, + "learning_rate": 9.85142848073103e-07, + "logits/chosen": -0.10557708889245987, + "logits/rejected": -0.06234588474035263, + "logps/chosen": -133.39218139648438, + "logps/rejected": -134.2109375, + "logps/weighted_chosen": -0.700244128704071, + "logps/weighted_rejected": -1.4902985095977783, + "loss": 0.7558, + "rewards/accuracies": 0.534375011920929, + "rewards/chosen": -14.2265625, + "rewards/margins": 0.6207031011581421, + "rewards/rejected": -14.846094131469727, + "rewards/weighted_accuracies": 0.515625, + "rewards/weighted_chosen": -0.2170669585466385, + "rewards/weighted_margins": 0.03446922451257706, + "rewards/weighted_rejected": -0.25187987089157104, + "step": 320 + }, + { + "epoch": 0.17636448660565168, + "grad_norm": 41.381507873535156, + "learning_rate": 9.828015927096914e-07, + "logits/chosen": -0.085169218480587, + "logits/rejected": -0.05568351596593857, + "logps/chosen": -143.8234405517578, + "logps/rejected": -142.39608764648438, + "logps/weighted_chosen": -0.977459728717804, + "logps/weighted_rejected": -0.7958831787109375, + "loss": 0.6935, + "rewards/accuracies": 0.550000011920929, + "rewards/chosen": -14.643750190734863, + "rewards/margins": 1.578125, + "rewards/rejected": -16.222265243530273, + "rewards/weighted_accuracies": 0.581250011920929, + "rewards/weighted_chosen": -0.16206398606300354, + "rewards/weighted_margins": 0.07153511047363281, + "rewards/weighted_rejected": -0.23361511528491974, + "step": 330 + }, + { + "epoch": 0.18170886498764113, + "grad_norm": 7.313599586486816, + "learning_rate": 9.802923132107968e-07, + "logits/chosen": -0.10784301906824112, + "logits/rejected": -0.052645109593868256, + "logps/chosen": -143.91561889648438, + "logps/rejected": -147.7234344482422, + "logps/weighted_chosen": -0.7267853021621704, + "logps/weighted_rejected": -1.24786376953125, + "loss": 0.7102, + "rewards/accuracies": 0.5562499761581421, + "rewards/chosen": -16.782812118530273, + "rewards/margins": 1.23046875, + "rewards/rejected": -18.012500762939453, + "rewards/weighted_accuracies": 0.5687500238418579, + "rewards/weighted_chosen": -0.1999809294939041, + "rewards/weighted_margins": 0.056002043187618256, + "rewards/weighted_rejected": -0.2561977505683899, + "step": 340 + }, + { + "epoch": 0.18705324336963056, + "grad_norm": 66.15936279296875, + "learning_rate": 9.776158828534024e-07, + "logits/chosen": -0.12608642876148224, + "logits/rejected": -0.09711913764476776, + "logps/chosen": -143.48281860351562, + "logps/rejected": -188.25936889648438, + "logps/weighted_chosen": -1.260833740234375, + "logps/weighted_rejected": -0.9638397097587585, + "loss": 0.7338, + "rewards/accuracies": 0.612500011920929, + "rewards/chosen": -16.727733612060547, + "rewards/margins": 3.159374952316284, + "rewards/rejected": -19.883594512939453, + "rewards/weighted_accuracies": 0.559374988079071, + "rewards/weighted_chosen": -0.26622849702835083, + "rewards/weighted_margins": 0.05975799635052681, + "rewards/weighted_rejected": -0.3257743716239929, + "step": 350 + }, + { + "epoch": 0.19239762175162, + "grad_norm": 10.4098539352417, + "learning_rate": 9.747732330861695e-07, + "logits/chosen": -0.05247039720416069, + "logits/rejected": -0.024548912420868874, + "logps/chosen": -135.78750610351562, + "logps/rejected": -137.03671264648438, + "logps/weighted_chosen": -0.8440704345703125, + "logps/weighted_rejected": -0.8307411074638367, + "loss": 0.7415, + "rewards/accuracies": 0.578125, + "rewards/chosen": -16.864452362060547, + "rewards/margins": 2.5316405296325684, + "rewards/rejected": -19.396093368530273, + "rewards/weighted_accuracies": 0.5249999761581421, + "rewards/weighted_chosen": 0.005719375796616077, + "rewards/weighted_margins": 0.17883948981761932, + "rewards/weighted_rejected": -0.17323684692382812, + "step": 360 + }, + { + "epoch": 0.19774200013360946, + "grad_norm": 8.8024263381958, + "learning_rate": 9.717653532052742e-07, + "logits/chosen": -0.0410335548222065, + "logits/rejected": -0.049485016614198685, + "logps/chosen": -130.2781219482422, + "logps/rejected": -137.9343719482422, + "logps/weighted_chosen": -1.1705901622772217, + "logps/weighted_rejected": -0.7656036615371704, + "loss": 0.7402, + "rewards/accuracies": 0.6031249761581421, + "rewards/chosen": -17.155858993530273, + "rewards/margins": 2.863085985183716, + "rewards/rejected": -20.016016006469727, + "rewards/weighted_accuracies": 0.53125, + "rewards/weighted_chosen": -0.07308082282543182, + "rewards/weighted_margins": 0.04105224460363388, + "rewards/weighted_rejected": -0.114354707300663, + "step": 370 + }, + { + "epoch": 0.2030863785155989, + "grad_norm": 26.81167221069336, + "learning_rate": 9.685932900101146e-07, + "logits/chosen": -0.07604064792394638, + "logits/rejected": -0.06302948296070099, + "logps/chosen": -134.7859344482422, + "logps/rejected": -137.11093139648438, + "logps/weighted_chosen": -0.98638916015625, + "logps/weighted_rejected": -0.6243560910224915, + "loss": 0.7116, + "rewards/accuracies": 0.5625, + "rewards/chosen": -17.729297637939453, + "rewards/margins": 1.8025391101837158, + "rewards/rejected": -19.528514862060547, + "rewards/weighted_accuracies": 0.546875, + "rewards/weighted_chosen": -0.11945419013500214, + "rewards/weighted_margins": 0.07185058295726776, + "rewards/weighted_rejected": -0.1911766082048416, + "step": 380 + }, + { + "epoch": 0.20843075689758836, + "grad_norm": 40.89613342285156, + "learning_rate": 9.652581474390043e-07, + "logits/chosen": -0.09274902194738388, + "logits/rejected": -0.05452308803796768, + "logps/chosen": -139.3249969482422, + "logps/rejected": -141.53750610351562, + "logps/weighted_chosen": -0.9199386835098267, + "logps/weighted_rejected": -0.9217513799667358, + "loss": 0.7323, + "rewards/accuracies": 0.550000011920929, + "rewards/chosen": -19.379688262939453, + "rewards/margins": 2.321093797683716, + "rewards/rejected": -21.702733993530273, + "rewards/weighted_accuracies": 0.543749988079071, + "rewards/weighted_chosen": -0.05511016771197319, + "rewards/weighted_margins": 0.04057302325963974, + "rewards/weighted_rejected": -0.09564266353845596, + "step": 390 + }, + { + "epoch": 0.21377513527957778, + "grad_norm": 51.860347747802734, + "learning_rate": 9.61761086184981e-07, + "logits/chosen": -0.10268402099609375, + "logits/rejected": -0.05923309177160263, + "logps/chosen": -139.50936889648438, + "logps/rejected": -141.4015655517578, + "logps/weighted_chosen": -0.9449096918106079, + "logps/weighted_rejected": -1.177301049232483, + "loss": 0.7387, + "rewards/accuracies": 0.5625, + "rewards/chosen": -20.024608612060547, + "rewards/margins": 2.9205079078674316, + "rewards/rejected": -22.942577362060547, + "rewards/weighted_accuracies": 0.5625, + "rewards/weighted_chosen": -0.26060789823532104, + "rewards/weighted_margins": 0.017241667956113815, + "rewards/weighted_rejected": -0.2779647707939148, + "step": 400 + }, + { + "epoch": 0.21911951366156723, + "grad_norm": 7.539150238037109, + "learning_rate": 9.581033232918629e-07, + "logits/chosen": -0.015906523913145065, + "logits/rejected": 0.02000865899026394, + "logps/chosen": -138.39999389648438, + "logps/rejected": -141.68905639648438, + "logps/weighted_chosen": -1.0313812494277954, + "logps/weighted_rejected": -1.1196472644805908, + "loss": 0.7283, + "rewards/accuracies": 0.5874999761581421, + "rewards/chosen": -21.702342987060547, + "rewards/margins": 2.826953172683716, + "rewards/rejected": -24.53125, + "rewards/weighted_accuracies": 0.5249999761581421, + "rewards/weighted_chosen": -0.2783859372138977, + "rewards/weighted_margins": 0.06903991848230362, + "rewards/weighted_rejected": -0.3477470278739929, + "step": 410 + }, + { + "epoch": 0.22446389204355668, + "grad_norm": 6.638446807861328, + "learning_rate": 9.542861317306952e-07, + "logits/chosen": -0.05767669528722763, + "logits/rejected": -0.03759918361902237, + "logps/chosen": -140.64688110351562, + "logps/rejected": -141.9718780517578, + "logps/weighted_chosen": -1.5940430164337158, + "logps/weighted_rejected": -1.458154320716858, + "loss": 0.7371, + "rewards/accuracies": 0.5406249761581421, + "rewards/chosen": -22.33203125, + "rewards/margins": 1.1359374523162842, + "rewards/rejected": -23.46484375, + "rewards/weighted_accuracies": 0.515625, + "rewards/weighted_chosen": -0.299835205078125, + "rewards/weighted_margins": 0.032170869410037994, + "rewards/weighted_rejected": -0.3319748044013977, + "step": 420 + }, + { + "epoch": 0.22980827042554614, + "grad_norm": 24.621816635131836, + "learning_rate": 9.503108399567308e-07, + "logits/chosen": -0.15150222182273865, + "logits/rejected": -0.09339599311351776, + "logps/chosen": -149.8671875, + "logps/rejected": -153.6296844482422, + "logps/weighted_chosen": -0.7455719113349915, + "logps/weighted_rejected": -0.826556384563446, + "loss": 0.7125, + "rewards/accuracies": 0.550000011920929, + "rewards/chosen": -22.685546875, + "rewards/margins": 1.7355468273162842, + "rewards/rejected": -24.420312881469727, + "rewards/weighted_accuracies": 0.5406249761581421, + "rewards/weighted_chosen": -0.19403991103172302, + "rewards/weighted_margins": 0.02692871168255806, + "rewards/weighted_rejected": -0.2209148406982422, + "step": 430 + }, + { + "epoch": 0.23515264880753559, + "grad_norm": 10.726137161254883, + "learning_rate": 9.461788314471034e-07, + "logits/chosen": -0.08421440422534943, + "logits/rejected": -0.031351469457149506, + "logps/chosen": -143.1171875, + "logps/rejected": -146.890625, + "logps/weighted_chosen": -0.7524811029434204, + "logps/weighted_rejected": -0.8818389773368835, + "loss": 0.7062, + "rewards/accuracies": 0.550000011920929, + "rewards/chosen": -24.516796112060547, + "rewards/margins": 2.0640625953674316, + "rewards/rejected": -26.577733993530273, + "rewards/weighted_accuracies": 0.528124988079071, + "rewards/weighted_chosen": -0.22655639052391052, + "rewards/weighted_margins": 0.07337188720703125, + "rewards/weighted_rejected": -0.30005186796188354, + "step": 440 + }, + { + "epoch": 0.240497027189525, + "grad_norm": 8.586684226989746, + "learning_rate": 9.418915442193509e-07, + "logits/chosen": -0.03446502611041069, + "logits/rejected": 0.007968139834702015, + "logps/chosen": -144.7843780517578, + "logps/rejected": -155.96249389648438, + "logps/weighted_chosen": -1.1978485584259033, + "logps/weighted_rejected": -1.2447845935821533, + "loss": 0.6841, + "rewards/accuracies": 0.6312500238418579, + "rewards/chosen": -26.516016006469727, + "rewards/margins": 4.714453220367432, + "rewards/rejected": -31.236328125, + "rewards/weighted_accuracies": 0.5218750238418579, + "rewards/weighted_chosen": -0.25520628690719604, + "rewards/weighted_margins": 0.09325332939624786, + "rewards/weighted_rejected": -0.3483436703681946, + "step": 450 + }, + { + "epoch": 0.24584140557151446, + "grad_norm": 10.014894485473633, + "learning_rate": 9.374504703309579e-07, + "logits/chosen": -0.1379852294921875, + "logits/rejected": -0.0861412063241005, + "logps/chosen": -141.83828735351562, + "logps/rejected": -143.34530639648438, + "logps/weighted_chosen": -1.1420501470565796, + "logps/weighted_rejected": -1.055877685546875, + "loss": 0.6804, + "rewards/accuracies": 0.534375011920929, + "rewards/chosen": -25.53515625, + "rewards/margins": 2.038281202316284, + "rewards/rejected": -27.571874618530273, + "rewards/weighted_accuracies": 0.5218750238418579, + "rewards/weighted_chosen": -0.21983489394187927, + "rewards/weighted_margins": 0.17051848769187927, + "rewards/weighted_rejected": -0.3904357850551605, + "step": 460 + }, + { + "epoch": 0.2511857839535039, + "grad_norm": 10.024358749389648, + "learning_rate": 9.328571553600915e-07, + "logits/chosen": -0.1318313628435135, + "logits/rejected": -0.0828731507062912, + "logps/chosen": -144.39688110351562, + "logps/rejected": -147.41796875, + "logps/weighted_chosen": -0.9446808099746704, + "logps/weighted_rejected": -1.138859510421753, + "loss": 0.7137, + "rewards/accuracies": 0.5718749761581421, + "rewards/chosen": -25.697265625, + "rewards/margins": 1.9656250476837158, + "rewards/rejected": -27.665233612060547, + "rewards/weighted_accuracies": 0.4906249940395355, + "rewards/weighted_chosen": -0.2078697234392166, + "rewards/weighted_margins": 0.052256010472774506, + "rewards/weighted_rejected": -0.260000616312027, + "step": 470 + }, + { + "epoch": 0.25653016233549336, + "grad_norm": 13.618422508239746, + "learning_rate": 9.281131978677106e-07, + "logits/chosen": -0.16686096787452698, + "logits/rejected": -0.1246364563703537, + "logps/chosen": -149.2546844482422, + "logps/rejected": -152.47811889648438, + "logps/weighted_chosen": -0.686419665813446, + "logps/weighted_rejected": -1.0338714122772217, + "loss": 0.6805, + "rewards/accuracies": 0.59375, + "rewards/chosen": -25.559375762939453, + "rewards/margins": 3.1332030296325684, + "rewards/rejected": -28.689844131469727, + "rewards/weighted_accuracies": 0.534375011920929, + "rewards/weighted_chosen": -0.22538375854492188, + "rewards/weighted_margins": 0.13199158012866974, + "rewards/weighted_rejected": -0.35721856355667114, + "step": 480 + }, + { + "epoch": 0.2618745407174828, + "grad_norm": 30.852863311767578, + "learning_rate": 9.232202488412361e-07, + "logits/chosen": -0.10437698662281036, + "logits/rejected": -0.05600614473223686, + "logps/chosen": -143.5671844482422, + "logps/rejected": -149.11874389648438, + "logps/weighted_chosen": -1.236029028892517, + "logps/weighted_rejected": -1.6542327404022217, + "loss": 0.6937, + "rewards/accuracies": 0.5375000238418579, + "rewards/chosen": -28.171483993530273, + "rewards/margins": 2.311328172683716, + "rewards/rejected": -30.477733612060547, + "rewards/weighted_accuracies": 0.534375011920929, + "rewards/weighted_chosen": -0.27268296480178833, + "rewards/weighted_margins": 0.08454742282629013, + "rewards/weighted_rejected": -0.35735321044921875, + "step": 490 + }, + { + "epoch": 0.26721891909947226, + "grad_norm": 9.151970863342285, + "learning_rate": 9.181800111199766e-07, + "logits/chosen": -0.1230342835187912, + "logits/rejected": -0.0801507979631424, + "logps/chosen": -147.54843139648438, + "logps/rejected": -147.5437469482422, + "logps/weighted_chosen": -1.5909912586212158, + "logps/weighted_rejected": -1.286199927330017, + "loss": 0.7005, + "rewards/accuracies": 0.578125, + "rewards/chosen": -29.640233993530273, + "rewards/margins": 3.1507811546325684, + "rewards/rejected": -32.789451599121094, + "rewards/weighted_accuracies": 0.5874999761581421, + "rewards/weighted_chosen": -0.34203338623046875, + "rewards/weighted_margins": 0.12214355170726776, + "rewards/weighted_rejected": -0.4641265869140625, + "step": 500 + }, + { + "epoch": 0.26721891909947226, + "eval_logits/chosen": -0.29476672410964966, + "eval_logits/rejected": -0.26891082525253296, + "eval_logps/chosen": -149.5992889404297, + "eval_logps/rejected": -153.5249481201172, + "eval_logps/weighted_chosen": -0.8761399984359741, + "eval_logps/weighted_rejected": -1.120854377746582, + "eval_loss": 0.6999610066413879, + "eval_rewards/accuracies": 0.5784114003181458, + "eval_rewards/chosen": -28.267057418823242, + "eval_rewards/margins": 2.7842414379119873, + "eval_rewards/rejected": -31.05422592163086, + "eval_rewards/weighted_accuracies": 0.5264765620231628, + "eval_rewards/weighted_chosen": -0.23699289560317993, + "eval_rewards/weighted_margins": 0.08454981446266174, + "eval_rewards/weighted_rejected": -0.3215427100658417, + "eval_runtime": 1248.3443, + "eval_samples_per_second": 1.571, + "eval_steps_per_second": 0.393, + "step": 500 + }, + { + "epoch": 0.2725632974814617, + "grad_norm": 7.777375221252441, + "learning_rate": 9.129942388025066e-07, + "logits/chosen": -0.1614234894514084, + "logits/rejected": -0.12539061903953552, + "logps/chosen": -138.70468139648438, + "logps/rejected": -144.50936889648438, + "logps/weighted_chosen": -1.0583159923553467, + "logps/weighted_rejected": -1.0353851318359375, + "loss": 0.7783, + "rewards/accuracies": 0.543749988079071, + "rewards/chosen": -27.291797637939453, + "rewards/margins": 2.485156297683716, + "rewards/rejected": -29.778514862060547, + "rewards/weighted_accuracies": 0.5093749761581421, + "rewards/weighted_chosen": -0.21576309204101562, + "rewards/weighted_margins": 0.003093719482421875, + "rewards/weighted_rejected": -0.219013974070549, + "step": 510 + }, + { + "epoch": 0.2779076758634511, + "grad_norm": 42.880157470703125, + "learning_rate": 9.076647366362082e-07, + "logits/chosen": -0.0863800048828125, + "logits/rejected": -0.04528322070837021, + "logps/chosen": -148.21875, + "logps/rejected": -155.3046875, + "logps/weighted_chosen": -1.141107201576233, + "logps/weighted_rejected": -1.2995758056640625, + "loss": 0.7748, + "rewards/accuracies": 0.612500011920929, + "rewards/chosen": -33.478515625, + "rewards/margins": 4.057812690734863, + "rewards/rejected": -37.53515625, + "rewards/weighted_accuracies": 0.5531250238418579, + "rewards/weighted_chosen": -0.2628006041049957, + "rewards/weighted_margins": -0.00189971923828125, + "rewards/weighted_rejected": -0.26109999418258667, + "step": 520 + }, + { + "epoch": 0.2832520542454406, + "grad_norm": 24.728120803833008, + "learning_rate": 9.021933593891841e-07, + "logits/chosen": -0.020507430657744408, + "logits/rejected": 0.01460342388600111, + "logps/chosen": -154.15780639648438, + "logps/rejected": -156.328125, + "logps/weighted_chosen": -1.0978972911834717, + "logps/weighted_rejected": -0.9736785888671875, + "loss": 0.6998, + "rewards/accuracies": 0.578125, + "rewards/chosen": -40.849220275878906, + "rewards/margins": 2.678515672683716, + "rewards/rejected": -43.53046798706055, + "rewards/weighted_accuracies": 0.512499988079071, + "rewards/weighted_chosen": -0.2724204957485199, + "rewards/weighted_margins": 0.09940185397863388, + "rewards/weighted_rejected": -0.3718253970146179, + "step": 530 + }, + { + "epoch": 0.28859643262743, + "grad_norm": 11.090974807739258, + "learning_rate": 8.965820112047629e-07, + "logits/chosen": -0.07119598239660263, + "logits/rejected": -0.04578246921300888, + "logps/chosen": -163.8781280517578, + "logps/rejected": -172.9187469482422, + "logps/weighted_chosen": -0.9050750732421875, + "logps/weighted_rejected": -1.05401611328125, + "loss": 0.6621, + "rewards/accuracies": 0.559374988079071, + "rewards/chosen": -42.501564025878906, + "rewards/margins": 4.053515434265137, + "rewards/rejected": -46.54999923706055, + "rewards/weighted_accuracies": 0.6000000238418579, + "rewards/weighted_chosen": -0.1999252289533615, + "rewards/weighted_margins": 0.1526111662387848, + "rewards/weighted_rejected": -0.3521057069301605, + "step": 540 + }, + { + "epoch": 0.2939408110094195, + "grad_norm": 13.709502220153809, + "learning_rate": 8.908326449388213e-07, + "logits/chosen": -0.07627105712890625, + "logits/rejected": -0.03014526329934597, + "logps/chosen": -156.328125, + "logps/rejected": -161.06405639648438, + "logps/weighted_chosen": -1.228173851966858, + "logps/weighted_rejected": -1.3901275396347046, + "loss": 0.7368, + "rewards/accuracies": 0.6156250238418579, + "rewards/chosen": -41.07109451293945, + "rewards/margins": 4.337500095367432, + "rewards/rejected": -45.392967224121094, + "rewards/weighted_accuracies": 0.512499988079071, + "rewards/weighted_chosen": -0.172210693359375, + "rewards/weighted_margins": 0.07322921603918076, + "rewards/weighted_rejected": -0.2451316863298416, + "step": 550 + }, + { + "epoch": 0.2992851893914089, + "grad_norm": 10.929818153381348, + "learning_rate": 8.849472614801527e-07, + "logits/chosen": -0.07251434028148651, + "logits/rejected": -0.02443542517721653, + "logps/chosen": -161.86874389648438, + "logps/rejected": -165.0749969482422, + "logps/weighted_chosen": -0.752777099609375, + "logps/weighted_rejected": -1.0892517566680908, + "loss": 0.6642, + "rewards/accuracies": 0.606249988079071, + "rewards/chosen": -44.9609375, + "rewards/margins": 3.3421874046325684, + "rewards/rejected": -48.30546951293945, + "rewards/weighted_accuracies": 0.5874999761581421, + "rewards/weighted_chosen": -0.1927940398454666, + "rewards/weighted_margins": 0.19287796318531036, + "rewards/weighted_rejected": -0.38528138399124146, + "step": 560 + }, + { + "epoch": 0.30462956777339834, + "grad_norm": 11.730274200439453, + "learning_rate": 8.789279090541208e-07, + "logits/chosen": -0.05120544508099556, + "logits/rejected": 0.01478500384837389, + "logps/chosen": -161.8796844482422, + "logps/rejected": -172.51406860351562, + "logps/weighted_chosen": -0.8590911626815796, + "logps/weighted_rejected": -1.1839568614959717, + "loss": 0.6821, + "rewards/accuracies": 0.6187499761581421, + "rewards/chosen": -43.99609375, + "rewards/margins": 4.257031440734863, + "rewards/rejected": -48.24609375, + "rewards/weighted_accuracies": 0.515625, + "rewards/weighted_chosen": -0.29502803087234497, + "rewards/weighted_margins": 0.16868743300437927, + "rewards/weighted_rejected": -0.4639228880405426, + "step": 570 + }, + { + "epoch": 0.3099739461553878, + "grad_norm": 16.967859268188477, + "learning_rate": 8.72776682509837e-07, + "logits/chosen": 0.011944961734116077, + "logits/rejected": 0.08212585747241974, + "logps/chosen": -157.47500610351562, + "logps/rejected": -163.43594360351562, + "logps/weighted_chosen": -0.8815460205078125, + "logps/weighted_rejected": -1.220306396484375, + "loss": 0.6705, + "rewards/accuracies": 0.606249988079071, + "rewards/chosen": -43.369529724121094, + "rewards/margins": 4.614453315734863, + "rewards/rejected": -47.984375, + "rewards/weighted_accuracies": 0.565625011920929, + "rewards/weighted_chosen": -0.3169898986816406, + "rewards/weighted_margins": 0.19199447333812714, + "rewards/weighted_rejected": -0.5094040036201477, + "step": 580 + }, + { + "epoch": 0.31531832453737724, + "grad_norm": 9.289381980895996, + "learning_rate": 8.664957225911138e-07, + "logits/chosen": -0.04825134202837944, + "logits/rejected": -0.01284942589700222, + "logps/chosen": -163.6437530517578, + "logps/rejected": -184.9796905517578, + "logps/weighted_chosen": -0.775604248046875, + "logps/weighted_rejected": -1.1336090564727783, + "loss": 0.6889, + "rewards/accuracies": 0.596875011920929, + "rewards/chosen": -42.566795349121094, + "rewards/margins": 3.8046875, + "rewards/rejected": -46.36328125, + "rewards/weighted_accuracies": 0.565625011920929, + "rewards/weighted_chosen": -0.30945053696632385, + "rewards/weighted_margins": 0.116083525121212, + "rewards/weighted_rejected": -0.4253120422363281, + "step": 590 + }, + { + "epoch": 0.3206627029193667, + "grad_norm": 7.966893672943115, + "learning_rate": 8.600872151914451e-07, + "logits/chosen": 0.06866455078125, + "logits/rejected": 0.09470061957836151, + "logps/chosen": -158.2546844482422, + "logps/rejected": -167.48281860351562, + "logps/weighted_chosen": -1.3719971179962158, + "logps/weighted_rejected": -1.7499603033065796, + "loss": 0.7417, + "rewards/accuracies": 0.5843750238418579, + "rewards/chosen": -44.11249923706055, + "rewards/margins": 3.3988280296325684, + "rewards/rejected": -47.50468826293945, + "rewards/weighted_accuracies": 0.5375000238418579, + "rewards/weighted_chosen": -0.21634292602539062, + "rewards/weighted_margins": 0.19443893432617188, + "rewards/weighted_rejected": -0.41086500883102417, + "step": 600 + }, + { + "epoch": 0.32600708130135614, + "grad_norm": 49.298519134521484, + "learning_rate": 8.535533905932737e-07, + "logits/chosen": 0.06249542161822319, + "logits/rejected": 0.12158355861902237, + "logps/chosen": -173.4250030517578, + "logps/rejected": -183.015625, + "logps/weighted_chosen": -0.9283081293106079, + "logps/weighted_rejected": -1.2264496088027954, + "loss": 0.6778, + "rewards/accuracies": 0.590624988079071, + "rewards/chosen": -48.342185974121094, + "rewards/margins": 5.698437690734863, + "rewards/rejected": -54.049217224121094, + "rewards/weighted_accuracies": 0.578125, + "rewards/weighted_chosen": -0.3028602600097656, + "rewards/weighted_margins": 0.1688835173845291, + "rewards/weighted_rejected": -0.4717727601528168, + "step": 610 + }, + { + "epoch": 0.33135145968334556, + "grad_norm": 19.377832412719727, + "learning_rate": 8.468965226918105e-07, + "logits/chosen": 0.037828825414180756, + "logits/rejected": 0.0580875389277935, + "logps/chosen": -170.1281280517578, + "logps/rejected": -178.9093780517578, + "logps/weighted_chosen": -1.4841537475585938, + "logps/weighted_rejected": -1.118981957435608, + "loss": 0.7542, + "rewards/accuracies": 0.5874999761581421, + "rewards/chosen": -46.46210861206055, + "rewards/margins": 3.794921875, + "rewards/rejected": -50.26640701293945, + "rewards/weighted_accuracies": 0.596875011920929, + "rewards/weighted_chosen": -0.40849533677101135, + "rewards/weighted_margins": 0.0321044921875, + "rewards/weighted_rejected": -0.4405921995639801, + "step": 620 + }, + { + "epoch": 0.33669583806533504, + "grad_norm": 10.768741607666016, + "learning_rate": 8.40118928203676e-07, + "logits/chosen": 0.05730438232421875, + "logits/rejected": 0.07560119777917862, + "logps/chosen": -152.3718719482422, + "logps/rejected": -164.1218719482422, + "logps/weighted_chosen": -1.4566253423690796, + "logps/weighted_rejected": -1.3677704334259033, + "loss": 0.7025, + "rewards/accuracies": 0.643750011920929, + "rewards/chosen": -43.721092224121094, + "rewards/margins": 4.366015434265137, + "rewards/rejected": -48.07500076293945, + "rewards/weighted_accuracies": 0.590624988079071, + "rewards/weighted_chosen": -0.30116957426071167, + "rewards/weighted_margins": 0.09331168979406357, + "rewards/weighted_rejected": -0.39441680908203125, + "step": 630 + }, + { + "epoch": 0.34204021644732446, + "grad_norm": 8.60303020477295, + "learning_rate": 8.332229658606382e-07, + "logits/chosen": -0.055472563952207565, + "logits/rejected": -0.03988037258386612, + "logps/chosen": -175.6671905517578, + "logps/rejected": -180.06875610351562, + "logps/weighted_chosen": -0.8608856201171875, + "logps/weighted_rejected": -0.986541748046875, + "loss": 0.7297, + "rewards/accuracies": 0.596875011920929, + "rewards/chosen": -44.84062576293945, + "rewards/margins": 4.931250095367432, + "rewards/rejected": -49.77031326293945, + "rewards/weighted_accuracies": 0.515625, + "rewards/weighted_chosen": -0.23048630356788635, + "rewards/weighted_margins": 0.10724487155675888, + "rewards/weighted_rejected": -0.33786240220069885, + "step": 640 + }, + { + "epoch": 0.34738459482931394, + "grad_norm": 26.148893356323242, + "learning_rate": 8.262110355887302e-07, + "logits/chosen": -0.055730439722537994, + "logits/rejected": -0.00980529747903347, + "logps/chosen": -166.38125610351562, + "logps/rejected": -169.40469360351562, + "logps/weighted_chosen": -0.7286010980606079, + "logps/weighted_rejected": -0.8990936279296875, + "loss": 0.7103, + "rewards/accuracies": 0.609375, + "rewards/chosen": -46.994529724121094, + "rewards/margins": 3.8031249046325684, + "rewards/rejected": -50.78515625, + "rewards/weighted_accuracies": 0.518750011920929, + "rewards/weighted_chosen": -0.2977249026298523, + "rewards/weighted_margins": 0.08674392849206924, + "rewards/weighted_rejected": -0.3846077024936676, + "step": 650 + }, + { + "epoch": 0.35272897321130336, + "grad_norm": 11.823346138000488, + "learning_rate": 8.190855776730293e-07, + "logits/chosen": -0.07448883354663849, + "logits/rejected": -0.014659118838608265, + "logps/chosen": -171.80313110351562, + "logps/rejected": -182.2687530517578, + "logps/weighted_chosen": -0.9873107671737671, + "logps/weighted_rejected": -1.4951263666152954, + "loss": 0.7347, + "rewards/accuracies": 0.596875011920929, + "rewards/chosen": -50.564064025878906, + "rewards/margins": 4.532812595367432, + "rewards/rejected": -55.109375, + "rewards/weighted_accuracies": 0.5531250238418579, + "rewards/weighted_chosen": -0.28445130586624146, + "rewards/weighted_margins": 0.06389693915843964, + "rewards/weighted_rejected": -0.34831467270851135, + "step": 660 + }, + { + "epoch": 0.3580733515932928, + "grad_norm": 13.33145523071289, + "learning_rate": 8.118490719083917e-07, + "logits/chosen": -0.10270843654870987, + "logits/rejected": -0.062209319323301315, + "logps/chosen": -191.3000030517578, + "logps/rejected": -193.4109344482422, + "logps/weighted_chosen": -1.0586059093475342, + "logps/weighted_rejected": -1.3695876598358154, + "loss": 0.6575, + "rewards/accuracies": 0.546875, + "rewards/chosen": -58.06640625, + "rewards/margins": 4.108984470367432, + "rewards/rejected": -62.18359375, + "rewards/weighted_accuracies": 0.565625011920929, + "rewards/weighted_chosen": -0.2945011258125305, + "rewards/weighted_margins": 0.18829631805419922, + "rewards/weighted_rejected": -0.4826883375644684, + "step": 670 + }, + { + "epoch": 0.36341772997528227, + "grad_norm": 23.127918243408203, + "learning_rate": 8.045040367364368e-07, + "logits/chosen": 0.03829193115234375, + "logits/rejected": 0.055706024169921875, + "logps/chosen": -166.234375, + "logps/rejected": -169.96249389648438, + "logps/weighted_chosen": -1.025201439857483, + "logps/weighted_rejected": -1.302404761314392, + "loss": 0.7166, + "rewards/accuracies": 0.5843750238418579, + "rewards/chosen": -56.759376525878906, + "rewards/margins": 4.711718559265137, + "rewards/rejected": -61.46796798706055, + "rewards/weighted_accuracies": 0.590624988079071, + "rewards/weighted_chosen": -0.4052978456020355, + "rewards/weighted_margins": 0.09704723209142685, + "rewards/weighted_rejected": -0.5025444030761719, + "step": 680 + }, + { + "epoch": 0.3687621083572717, + "grad_norm": 7.137462615966797, + "learning_rate": 7.970530283690819e-07, + "logits/chosen": 0.04682769626379013, + "logits/rejected": 0.066925048828125, + "logps/chosen": -172.50155639648438, + "logps/rejected": -180.54061889648438, + "logps/weighted_chosen": -1.301055908203125, + "logps/weighted_rejected": -1.128961205482483, + "loss": 0.6864, + "rewards/accuracies": 0.6312500238418579, + "rewards/chosen": -58.544532775878906, + "rewards/margins": 5.397265434265137, + "rewards/rejected": -63.9453125, + "rewards/weighted_accuracies": 0.59375, + "rewards/weighted_chosen": -0.3540634214878082, + "rewards/weighted_margins": 0.1536247283220291, + "rewards/weighted_rejected": -0.5076873898506165, + "step": 690 + }, + { + "epoch": 0.3741064867392611, + "grad_norm": 7.699697971343994, + "learning_rate": 7.89498639898931e-07, + "logits/chosen": 0.05636443942785263, + "logits/rejected": 0.09223861992359161, + "logps/chosen": -169.8562469482422, + "logps/rejected": -173.65625, + "logps/weighted_chosen": -0.9024292230606079, + "logps/weighted_rejected": -1.1884033679962158, + "loss": 0.6642, + "rewards/accuracies": 0.6031249761581421, + "rewards/chosen": -59.00078201293945, + "rewards/margins": 4.887109279632568, + "rewards/rejected": -63.889060974121094, + "rewards/weighted_accuracies": 0.5562499761581421, + "rewards/weighted_chosen": -0.3017433285713196, + "rewards/weighted_margins": 0.20005035400390625, + "rewards/weighted_rejected": -0.5016593933105469, + "step": 700 + }, + { + "epoch": 0.3794508651212506, + "grad_norm": 18.41610336303711, + "learning_rate": 7.818435003968305e-07, + "logits/chosen": -0.03687286376953125, + "logits/rejected": -0.00025310515775345266, + "logps/chosen": -181.0812530517578, + "logps/rejected": -189.1828155517578, + "logps/weighted_chosen": -1.0574157238006592, + "logps/weighted_rejected": -1.2661011219024658, + "loss": 0.6918, + "rewards/accuracies": 0.590624988079071, + "rewards/chosen": -58.548439025878906, + "rewards/margins": 5.687890529632568, + "rewards/rejected": -64.234375, + "rewards/weighted_accuracies": 0.550000011920929, + "rewards/weighted_chosen": -0.33240431547164917, + "rewards/weighted_margins": 0.08642272651195526, + "rewards/weighted_rejected": -0.4188430905342102, + "step": 710 + }, + { + "epoch": 0.38479524350324, + "grad_norm": 26.664974212646484, + "learning_rate": 7.740902739969008e-07, + "logits/chosen": -0.08480529487133026, + "logits/rejected": -0.07062835991382599, + "logps/chosen": -171.16561889648438, + "logps/rejected": -175.68905639648438, + "logps/weighted_chosen": -0.828997790813446, + "logps/weighted_rejected": -0.9821502566337585, + "loss": 0.6819, + "rewards/accuracies": 0.6187499761581421, + "rewards/chosen": -49.982810974121094, + "rewards/margins": 4.688672065734863, + "rewards/rejected": -54.672264099121094, + "rewards/weighted_accuracies": 0.5531250238418579, + "rewards/weighted_chosen": -0.18975982069969177, + "rewards/weighted_margins": 0.1927967071533203, + "rewards/weighted_rejected": -0.38214111328125, + "step": 720 + }, + { + "epoch": 0.3901396218852295, + "grad_norm": 8.73801040649414, + "learning_rate": 7.662416589693695e-07, + "logits/chosen": -0.00596694927662611, + "logits/rejected": 0.019218826666474342, + "logps/chosen": -161.8249969482422, + "logps/rejected": -166.8312530517578, + "logps/weighted_chosen": -1.3982207775115967, + "logps/weighted_rejected": -1.7395598888397217, + "loss": 0.7772, + "rewards/accuracies": 0.6625000238418579, + "rewards/chosen": -46.8515625, + "rewards/margins": 5.096484184265137, + "rewards/rejected": -51.95000076293945, + "rewards/weighted_accuracies": 0.590624988079071, + "rewards/weighted_chosen": -0.23512879014015198, + "rewards/weighted_margins": 0.10187530517578125, + "rewards/weighted_rejected": -0.3369201719760895, + "step": 730 + }, + { + "epoch": 0.3954840002672189, + "grad_norm": 19.624563217163086, + "learning_rate": 7.583003867815192e-07, + "logits/chosen": -0.0639801025390625, + "logits/rejected": -0.05755653232336044, + "logps/chosen": -189.7468719482422, + "logps/rejected": -196.7062530517578, + "logps/weighted_chosen": -1.1809478998184204, + "logps/weighted_rejected": -1.3041900396347046, + "loss": 0.698, + "rewards/accuracies": 0.578125, + "rewards/chosen": -58.209373474121094, + "rewards/margins": 4.971875190734863, + "rewards/rejected": -63.1640625, + "rewards/weighted_accuracies": 0.518750011920929, + "rewards/weighted_chosen": -0.39410895109176636, + "rewards/weighted_margins": 0.08784560859203339, + "rewards/weighted_rejected": -0.4818374514579773, + "step": 740 + }, + { + "epoch": 0.40082837864920834, + "grad_norm": 19.04839324951172, + "learning_rate": 7.502692211470869e-07, + "logits/chosen": -0.05941009521484375, + "logits/rejected": -0.02816619910299778, + "logps/chosen": -182.8984375, + "logps/rejected": -190.1203155517578, + "logps/weighted_chosen": -0.853833019733429, + "logps/weighted_rejected": -1.1572723388671875, + "loss": 0.7079, + "rewards/accuracies": 0.574999988079071, + "rewards/chosen": -60.310157775878906, + "rewards/margins": 6.424609184265137, + "rewards/rejected": -66.73905944824219, + "rewards/weighted_accuracies": 0.546875, + "rewards/weighted_chosen": -0.4228218197822571, + "rewards/weighted_margins": 0.10426559299230576, + "rewards/weighted_rejected": -0.5270500183105469, + "step": 750 + }, + { + "epoch": 0.4061727570311978, + "grad_norm": 12.947752952575684, + "learning_rate": 7.421509570644387e-07, + "logits/chosen": -0.05351562425494194, + "logits/rejected": -0.0015235900646075606, + "logps/chosen": -180.66250610351562, + "logps/rejected": -185.6828155517578, + "logps/weighted_chosen": -1.2698882818222046, + "logps/weighted_rejected": -1.5226471424102783, + "loss": 0.6831, + "rewards/accuracies": 0.6343749761581421, + "rewards/chosen": -62.2109375, + "rewards/margins": 5.124218940734863, + "rewards/rejected": -67.33280944824219, + "rewards/weighted_accuracies": 0.59375, + "rewards/weighted_chosen": -0.32732313871383667, + "rewards/weighted_margins": 0.18027648329734802, + "rewards/weighted_rejected": -0.5074516534805298, + "step": 760 + }, + { + "epoch": 0.41151713541318724, + "grad_norm": 10.666881561279297, + "learning_rate": 7.339484198438566e-07, + "logits/chosen": 0.00444793701171875, + "logits/rejected": 0.03400421142578125, + "logps/chosen": -165.93124389648438, + "logps/rejected": -172.2937469482422, + "logps/weighted_chosen": -1.7782714366912842, + "logps/weighted_rejected": -1.320520043373108, + "loss": 0.6969, + "rewards/accuracies": 0.6000000238418579, + "rewards/chosen": -57.14374923706055, + "rewards/margins": 7.307421684265137, + "rewards/rejected": -64.4468765258789, + "rewards/weighted_accuracies": 0.5625, + "rewards/weighted_chosen": -0.3784065246582031, + "rewards/weighted_margins": 0.17845840752124786, + "rewards/weighted_rejected": -0.557085394859314, + "step": 770 + }, + { + "epoch": 0.4168615137951767, + "grad_norm": 11.188801765441895, + "learning_rate": 7.256644641242773e-07, + "logits/chosen": -0.03662109375, + "logits/rejected": 0.02340393140912056, + "logps/chosen": -170.4765625, + "logps/rejected": -175.390625, + "logps/weighted_chosen": -1.057519555091858, + "logps/weighted_rejected": -1.3154876232147217, + "loss": 0.7128, + "rewards/accuracies": 0.5562499761581421, + "rewards/chosen": -56.12578201293945, + "rewards/margins": 3.3226561546325684, + "rewards/rejected": -59.43828201293945, + "rewards/weighted_accuracies": 0.546875, + "rewards/weighted_chosen": -0.37098997831344604, + "rewards/weighted_margins": 0.07502517849206924, + "rewards/weighted_rejected": -0.4461258053779602, + "step": 780 + }, + { + "epoch": 0.42220589217716614, + "grad_norm": 9.07581615447998, + "learning_rate": 7.173019728798234e-07, + "logits/chosen": -0.03949751704931259, + "logits/rejected": -0.02464141882956028, + "logps/chosen": -169.55313110351562, + "logps/rejected": -177.18124389648438, + "logps/weighted_chosen": -0.8541290163993835, + "logps/weighted_rejected": -1.1333099603652954, + "loss": 0.6788, + "rewards/accuracies": 0.6625000238418579, + "rewards/chosen": -55.47734451293945, + "rewards/margins": 6.255859375, + "rewards/rejected": -61.71406173706055, + "rewards/weighted_accuracies": 0.574999988079071, + "rewards/weighted_chosen": -0.307028204202652, + "rewards/weighted_margins": 0.12036208808422089, + "rewards/weighted_rejected": -0.4274864196777344, + "step": 790 + }, + { + "epoch": 0.42755027055915557, + "grad_norm": 19.591575622558594, + "learning_rate": 7.088638564164738e-07, + "logits/chosen": -0.08669586479663849, + "logits/rejected": -0.05663032457232475, + "logps/chosen": -185.1593780517578, + "logps/rejected": -195.95938110351562, + "logps/weighted_chosen": -1.067468285560608, + "logps/weighted_rejected": -1.238824486732483, + "loss": 0.693, + "rewards/accuracies": 0.628125011920929, + "rewards/chosen": -63.07421875, + "rewards/margins": 6.429296970367432, + "rewards/rejected": -69.49296569824219, + "rewards/weighted_accuracies": 0.581250011920929, + "rewards/weighted_chosen": -0.3178604245185852, + "rewards/weighted_margins": 0.1114906296133995, + "rewards/weighted_rejected": -0.4294891357421875, + "step": 800 + }, + { + "epoch": 0.43289464894114504, + "grad_norm": 13.074841499328613, + "learning_rate": 7.003530513592215e-07, + "logits/chosen": -0.05083770677447319, + "logits/rejected": -0.0205841064453125, + "logps/chosen": -177.95156860351562, + "logps/rejected": -189.63436889648438, + "logps/weighted_chosen": -1.188201904296875, + "logps/weighted_rejected": -1.1756408214569092, + "loss": 0.6926, + "rewards/accuracies": 0.6187499761581421, + "rewards/chosen": -62.192970275878906, + "rewards/margins": 7.530859470367432, + "rewards/rejected": -69.73905944824219, + "rewards/weighted_accuracies": 0.543749988079071, + "rewards/weighted_chosen": -0.3717803955078125, + "rewards/weighted_margins": 0.12430572509765625, + "rewards/weighted_rejected": -0.49610137939453125, + "step": 810 + }, + { + "epoch": 0.43823902732313447, + "grad_norm": 51.0267448425293, + "learning_rate": 6.917725196300726e-07, + "logits/chosen": -0.1438644379377365, + "logits/rejected": -0.15563049912452698, + "logps/chosen": -191.125, + "logps/rejected": -203.52188110351562, + "logps/weighted_chosen": -1.097863793373108, + "logps/weighted_rejected": -0.971697986125946, + "loss": 0.6974, + "rewards/accuracies": 0.596875011920929, + "rewards/chosen": -63.764060974121094, + "rewards/margins": 7.415234565734863, + "rewards/rejected": -71.171875, + "rewards/weighted_accuracies": 0.5531250238418579, + "rewards/weighted_chosen": -0.3276786804199219, + "rewards/weighted_margins": 0.08215713500976562, + "rewards/weighted_rejected": -0.4100082516670227, + "step": 820 + }, + { + "epoch": 0.44358340570512395, + "grad_norm": 20.933467864990234, + "learning_rate": 6.831252474172411e-07, + "logits/chosen": -0.05628051608800888, + "logits/rejected": -0.03361053392291069, + "logps/chosen": -176.0031280517578, + "logps/rejected": -186.4343719482422, + "logps/weighted_chosen": -1.6239410638809204, + "logps/weighted_rejected": -1.3564453125, + "loss": 0.6817, + "rewards/accuracies": 0.609375, + "rewards/chosen": -60.45234298706055, + "rewards/margins": 6.214453220367432, + "rewards/rejected": -66.66015625, + "rewards/weighted_accuracies": 0.574999988079071, + "rewards/weighted_chosen": -0.2794052064418793, + "rewards/weighted_margins": 0.22185592353343964, + "rewards/weighted_rejected": -0.500841498374939, + "step": 830 + }, + { + "epoch": 0.44892778408711337, + "grad_norm": 11.393433570861816, + "learning_rate": 6.74414244135898e-07, + "logits/chosen": -0.15757599472999573, + "logits/rejected": -0.116943359375, + "logps/chosen": -177.0984344482422, + "logps/rejected": -185.13436889648438, + "logps/weighted_chosen": -0.666271984577179, + "logps/weighted_rejected": -0.7827666997909546, + "loss": 0.6919, + "rewards/accuracies": 0.625, + "rewards/chosen": -60.349998474121094, + "rewards/margins": 6.200390815734863, + "rewards/rejected": -66.5289077758789, + "rewards/weighted_accuracies": 0.581250011920929, + "rewards/weighted_chosen": -0.33733367919921875, + "rewards/weighted_margins": 0.08320007473230362, + "rewards/weighted_rejected": -0.42046278715133667, + "step": 840 + }, + { + "epoch": 0.4542721624691028, + "grad_norm": 9.733312606811523, + "learning_rate": 6.656425413808388e-07, + "logits/chosen": -0.12353897094726562, + "logits/rejected": -0.08713607490062714, + "logps/chosen": -179.66561889648438, + "logps/rejected": -183.109375, + "logps/weighted_chosen": -1.0826873779296875, + "logps/weighted_rejected": -1.106787085533142, + "loss": 0.6875, + "rewards/accuracies": 0.5562499761581421, + "rewards/chosen": -60.806251525878906, + "rewards/margins": 2.733203172683716, + "rewards/rejected": -63.51250076293945, + "rewards/weighted_accuracies": 0.5843750238418579, + "rewards/weighted_chosen": -0.22345809638500214, + "rewards/weighted_margins": 0.1886741667985916, + "rewards/weighted_rejected": -0.4119575619697571, + "step": 850 + }, + { + "epoch": 0.45961654085109227, + "grad_norm": 9.04237174987793, + "learning_rate": 6.568131918714294e-07, + "logits/chosen": -0.09643097221851349, + "logits/rejected": -0.08455581963062286, + "logps/chosen": -178.4453125, + "logps/rejected": -187.99063110351562, + "logps/weighted_chosen": -0.840087890625, + "logps/weighted_rejected": -1.158935546875, + "loss": 0.7138, + "rewards/accuracies": 0.596875011920929, + "rewards/chosen": -61.162498474121094, + "rewards/margins": 5.386328220367432, + "rewards/rejected": -66.54765319824219, + "rewards/weighted_accuracies": 0.550000011920929, + "rewards/weighted_chosen": -0.2982444763183594, + "rewards/weighted_margins": 0.08280716091394424, + "rewards/weighted_rejected": -0.38090819120407104, + "step": 860 + }, + { + "epoch": 0.4649609192330817, + "grad_norm": 7.623125076293945, + "learning_rate": 6.47929268389203e-07, + "logits/chosen": -0.06428680568933487, + "logits/rejected": -0.04297828674316406, + "logps/chosen": -190.86093139648438, + "logps/rejected": -205.9656219482422, + "logps/weighted_chosen": -0.9338012933731079, + "logps/weighted_rejected": -0.741748034954071, + "loss": 0.6808, + "rewards/accuracies": 0.6000000238418579, + "rewards/chosen": -65.76640319824219, + "rewards/margins": 7.994921684265137, + "rewards/rejected": -73.75703430175781, + "rewards/weighted_accuracies": 0.5, + "rewards/weighted_chosen": -0.28187790513038635, + "rewards/weighted_margins": 0.11656570434570312, + "rewards/weighted_rejected": -0.39813536405563354, + "step": 870 + }, + { + "epoch": 0.47030529761507117, + "grad_norm": 36.31256103515625, + "learning_rate": 6.389938627084732e-07, + "logits/chosen": -0.04478149488568306, + "logits/rejected": -0.054534912109375, + "logps/chosen": -181.5046844482422, + "logps/rejected": -192.84375, + "logps/weighted_chosen": -1.712792992591858, + "logps/weighted_rejected": -1.4381287097930908, + "loss": 0.7264, + "rewards/accuracies": 0.581250011920929, + "rewards/chosen": -64.04609680175781, + "rewards/margins": 6.507421970367432, + "rewards/rejected": -70.56953430175781, + "rewards/weighted_accuracies": 0.515625, + "rewards/weighted_chosen": -0.3980148434638977, + "rewards/weighted_margins": 0.10754700005054474, + "rewards/weighted_rejected": -0.5054702758789062, + "step": 880 + }, + { + "epoch": 0.4756496759970606, + "grad_norm": 6.0423126220703125, + "learning_rate": 6.300100845203373e-07, + "logits/chosen": -0.07687301933765411, + "logits/rejected": -0.04276275634765625, + "logps/chosen": -194.1609344482422, + "logps/rejected": -200.828125, + "logps/weighted_chosen": -0.6031554937362671, + "logps/weighted_rejected": -0.86151123046875, + "loss": 0.6793, + "rewards/accuracies": 0.59375, + "rewards/chosen": -63.5546875, + "rewards/margins": 5.604296684265137, + "rewards/rejected": -69.13984680175781, + "rewards/weighted_accuracies": 0.559374988079071, + "rewards/weighted_chosen": -0.27074965834617615, + "rewards/weighted_margins": 0.11608581244945526, + "rewards/weighted_rejected": -0.38665771484375, + "step": 890 + }, + { + "epoch": 0.48099405437905, + "grad_norm": 21.55549430847168, + "learning_rate": 6.20981060350445e-07, + "logits/chosen": -0.12602996826171875, + "logits/rejected": -0.08534087985754013, + "logps/chosen": -186.3312530517578, + "logps/rejected": -190.0625, + "logps/weighted_chosen": -0.8782455325126648, + "logps/weighted_rejected": -1.0832672119140625, + "loss": 0.6442, + "rewards/accuracies": 0.574999988079071, + "rewards/chosen": -65.2984390258789, + "rewards/margins": 6.321875095367432, + "rewards/rejected": -71.61796569824219, + "rewards/weighted_accuracies": 0.612500011920929, + "rewards/weighted_chosen": -0.3018638491630554, + "rewards/weighted_margins": 0.19889220595359802, + "rewards/weighted_rejected": -0.5003150701522827, + "step": 900 + }, + { + "epoch": 0.4863384327610395, + "grad_norm": 15.255428314208984, + "learning_rate": 6.119099324709087e-07, + "logits/chosen": -0.08440399169921875, + "logits/rejected": -0.03687591478228569, + "logps/chosen": -183.1640625, + "logps/rejected": -192.125, + "logps/weighted_chosen": -0.7069732546806335, + "logps/weighted_rejected": -1.348388671875, + "loss": 0.7239, + "rewards/accuracies": 0.609375, + "rewards/chosen": -65.2796859741211, + "rewards/margins": 7.408593654632568, + "rewards/rejected": -72.6890640258789, + "rewards/weighted_accuracies": 0.5093749761581421, + "rewards/weighted_chosen": -0.3756294250488281, + "rewards/weighted_margins": 0.05287017673254013, + "rewards/weighted_rejected": -0.4286758303642273, + "step": 910 + }, + { + "epoch": 0.4916828111430289, + "grad_norm": 10.355165481567383, + "learning_rate": 6.027998578067316e-07, + "logits/chosen": -0.06318511813879013, + "logits/rejected": -0.02981414832174778, + "logps/chosen": -199.36563110351562, + "logps/rejected": -208.70938110351562, + "logps/weighted_chosen": -0.6963897943496704, + "logps/weighted_rejected": -0.969342052936554, + "loss": 0.676, + "rewards/accuracies": 0.6031249761581421, + "rewards/chosen": -71.73905944824219, + "rewards/margins": 6.481640815734863, + "rewards/rejected": -78.2171859741211, + "rewards/weighted_accuracies": 0.550000011920929, + "rewards/weighted_chosen": -0.2880393862724304, + "rewards/weighted_margins": 0.1389778107404709, + "rewards/weighted_rejected": -0.426931768655777, + "step": 920 + }, + { + "epoch": 0.4970271895250184, + "grad_norm": 5.530470371246338, + "learning_rate": 5.936540068371394e-07, + "logits/chosen": 0.006648254580795765, + "logits/rejected": 0.005427551455795765, + "logps/chosen": -193.99374389648438, + "logps/rejected": -202.4406280517578, + "logps/weighted_chosen": -1.107049584388733, + "logps/weighted_rejected": -1.3529266119003296, + "loss": 0.6926, + "rewards/accuracies": 0.612500011920929, + "rewards/chosen": -74.79765319824219, + "rewards/margins": 5.872656345367432, + "rewards/rejected": -80.6624984741211, + "rewards/weighted_accuracies": 0.559374988079071, + "rewards/weighted_chosen": -0.43673935532569885, + "rewards/weighted_margins": 0.12621307373046875, + "rewards/weighted_rejected": -0.5626624822616577, + "step": 930 + }, + { + "epoch": 0.5023715679070078, + "grad_norm": 5.006593704223633, + "learning_rate": 5.844755624921918e-07, + "logits/chosen": -0.005947875790297985, + "logits/rejected": 0.02864990197122097, + "logps/chosen": -187.453125, + "logps/rejected": -192.7375030517578, + "logps/weighted_chosen": -1.132598876953125, + "logps/weighted_rejected": -1.0424163341522217, + "loss": 0.7378, + "rewards/accuracies": 0.609375, + "rewards/chosen": -70.77812194824219, + "rewards/margins": 5.782812595367432, + "rewards/rejected": -76.56562805175781, + "rewards/weighted_accuracies": 0.512499988079071, + "rewards/weighted_chosen": -0.44052886962890625, + "rewards/weighted_margins": 0.07312469184398651, + "rewards/weighted_rejected": -0.5137832760810852, + "step": 940 + }, + { + "epoch": 0.5077159462889973, + "grad_norm": 16.502933502197266, + "learning_rate": 5.752677190450634e-07, + "logits/chosen": -0.021286392584443092, + "logits/rejected": -0.00705032330006361, + "logps/chosen": -186.49374389648438, + "logps/rejected": -196.24844360351562, + "logps/weighted_chosen": -1.068359375, + "logps/weighted_rejected": -1.3232848644256592, + "loss": 0.7251, + "rewards/accuracies": 0.625, + "rewards/chosen": -67.70234680175781, + "rewards/margins": 7.561327934265137, + "rewards/rejected": -75.28047180175781, + "rewards/weighted_accuracies": 0.528124988079071, + "rewards/weighted_chosen": -0.390481561422348, + "rewards/weighted_margins": 0.058994293212890625, + "rewards/weighted_rejected": -0.44965285062789917, + "step": 950 + }, + { + "epoch": 0.5130603246709867, + "grad_norm": 12.000792503356934, + "learning_rate": 5.66033681000375e-07, + "logits/chosen": -0.03893890231847763, + "logits/rejected": 0.01745758019387722, + "logps/chosen": -189.5234375, + "logps/rejected": -201.2624969482422, + "logps/weighted_chosen": -1.0573517084121704, + "logps/weighted_rejected": -1.4250671863555908, + "loss": 0.6656, + "rewards/accuracies": 0.637499988079071, + "rewards/chosen": -69.40625, + "rewards/margins": 8.408203125, + "rewards/rejected": -77.78593444824219, + "rewards/weighted_accuracies": 0.5718749761581421, + "rewards/weighted_chosen": -0.2941024899482727, + "rewards/weighted_margins": 0.1710365265607834, + "rewards/weighted_rejected": -0.4653762876987457, + "step": 960 + }, + { + "epoch": 0.5184047030529761, + "grad_norm": 8.829608917236328, + "learning_rate": 5.567766619789658e-07, + "logits/chosen": -0.03634033352136612, + "logits/rejected": 0.0003803253057412803, + "logps/chosen": -190.27969360351562, + "logps/rejected": -199.0500030517578, + "logps/weighted_chosen": -1.1300811767578125, + "logps/weighted_rejected": -1.3106689453125, + "loss": 0.6931, + "rewards/accuracies": 0.565625011920929, + "rewards/chosen": -67.55078125, + "rewards/margins": 7.454297065734863, + "rewards/rejected": -74.9703140258789, + "rewards/weighted_accuracies": 0.606249988079071, + "rewards/weighted_chosen": -0.3190757632255554, + "rewards/weighted_margins": 0.11787338554859161, + "rewards/weighted_rejected": -0.4370986819267273, + "step": 970 + }, + { + "epoch": 0.5237490814349656, + "grad_norm": 9.326337814331055, + "learning_rate": 5.474998835994916e-07, + "logits/chosen": 0.009844208136200905, + "logits/rejected": 0.051239013671875, + "logps/chosen": -195.78125, + "logps/rejected": -208.5343780517578, + "logps/weighted_chosen": -1.179107666015625, + "logps/weighted_rejected": -1.546759009361267, + "loss": 0.6677, + "rewards/accuracies": 0.6343749761581421, + "rewards/chosen": -70.6781234741211, + "rewards/margins": 6.912109375, + "rewards/rejected": -77.5609359741211, + "rewards/weighted_accuracies": 0.5625, + "rewards/weighted_chosen": -0.28845518827438354, + "rewards/weighted_margins": 0.1378173828125, + "rewards/weighted_rejected": -0.42629241943359375, + "step": 980 + }, + { + "epoch": 0.529093459816955, + "grad_norm": 10.308549880981445, + "learning_rate": 5.382065743572411e-07, + "logits/chosen": -0.025909423828125, + "logits/rejected": 0.01560287456959486, + "logps/chosen": -192.8093719482422, + "logps/rejected": -205.2843780517578, + "logps/weighted_chosen": -0.8877533078193665, + "logps/weighted_rejected": -1.2557891607284546, + "loss": 0.6811, + "rewards/accuracies": 0.671875, + "rewards/chosen": -70.57890319824219, + "rewards/margins": 8.194921493530273, + "rewards/rejected": -78.80000305175781, + "rewards/weighted_accuracies": 0.534375011920929, + "rewards/weighted_chosen": -0.3638969361782074, + "rewards/weighted_margins": 0.13902893662452698, + "rewards/weighted_rejected": -0.5028846859931946, + "step": 990 + }, + { + "epoch": 0.5344378381989445, + "grad_norm": 6.612018585205078, + "learning_rate": 5.288999685005571e-07, + "logits/chosen": -0.03016967698931694, + "logits/rejected": -0.0050643919967114925, + "logps/chosen": -192.640625, + "logps/rejected": -201.4656219482422, + "logps/weighted_chosen": -0.626147449016571, + "logps/weighted_rejected": -0.8550964593887329, + "loss": 0.672, + "rewards/accuracies": 0.5843750238418579, + "rewards/chosen": -72.01640319824219, + "rewards/margins": 6.489062309265137, + "rewards/rejected": -78.5062484741211, + "rewards/weighted_accuracies": 0.581250011920929, + "rewards/weighted_chosen": -0.2895965576171875, + "rewards/weighted_margins": 0.1136070266366005, + "rewards/weighted_rejected": -0.4031356871128082, + "step": 1000 + }, + { + "epoch": 0.5344378381989445, + "eval_logits/chosen": -0.1623995155096054, + "eval_logits/rejected": -0.13676175475120544, + "eval_logps/chosen": -193.06211853027344, + "eval_logps/rejected": -200.78919982910156, + "eval_logps/weighted_chosen": -1.0030580759048462, + "eval_logps/weighted_rejected": -1.3398514986038208, + "eval_loss": 0.6716485619544983, + "eval_rewards/accuracies": 0.5982688665390015, + "eval_rewards/chosen": -71.73777770996094, + "eval_rewards/margins": 6.556771755218506, + "eval_rewards/rejected": -78.29378509521484, + "eval_rewards/weighted_accuracies": 0.5570265054702759, + "eval_rewards/weighted_chosen": -0.36391088366508484, + "eval_rewards/weighted_margins": 0.17662912607192993, + "eval_rewards/weighted_rejected": -0.5405399203300476, + "eval_runtime": 1144.2369, + "eval_samples_per_second": 1.714, + "eval_steps_per_second": 0.429, + "step": 1000 + }, + { + "epoch": 0.539782216580934, + "grad_norm": 6.273863315582275, + "learning_rate": 5.19583304905257e-07, + "logits/chosen": -0.04050292819738388, + "logits/rejected": -0.010775757022202015, + "logps/chosen": -187.6999969482422, + "logps/rejected": -195.6999969482422, + "logps/weighted_chosen": -0.898364245891571, + "logps/weighted_rejected": -1.051977515220642, + "loss": 0.699, + "rewards/accuracies": 0.606249988079071, + "rewards/chosen": -72.2406234741211, + "rewards/margins": 6.910546779632568, + "rewards/rejected": -79.1578140258789, + "rewards/weighted_accuracies": 0.574999988079071, + "rewards/weighted_chosen": -0.3900245726108551, + "rewards/weighted_margins": 0.1439811736345291, + "rewards/weighted_rejected": -0.5338424444198608, + "step": 1010 + }, + { + "epoch": 0.5451265949629234, + "grad_norm": 8.971517562866211, + "learning_rate": 5.102598259474434e-07, + "logits/chosen": -0.005509376525878906, + "logits/rejected": 0.02565765380859375, + "logps/chosen": -184.39999389648438, + "logps/rejected": -187.3781280517578, + "logps/weighted_chosen": -0.8080657720565796, + "logps/weighted_rejected": -1.043609619140625, + "loss": 0.7008, + "rewards/accuracies": 0.621874988079071, + "rewards/chosen": -72.9703140258789, + "rewards/margins": 5.461718559265137, + "rewards/rejected": -78.44844055175781, + "rewards/weighted_accuracies": 0.5562499761581421, + "rewards/weighted_chosen": -0.3826606869697571, + "rewards/weighted_margins": 0.062392424792051315, + "rewards/weighted_rejected": -0.44501036405563354, + "step": 1020 + }, + { + "epoch": 0.5504709733449128, + "grad_norm": 12.744133949279785, + "learning_rate": 5.009327763750932e-07, + "logits/chosen": -0.049317169934511185, + "logits/rejected": -0.0027618408203125, + "logps/chosen": -192.65625, + "logps/rejected": -202.55624389648438, + "logps/weighted_chosen": -0.8524872064590454, + "logps/weighted_rejected": -1.3228943347930908, + "loss": 0.6779, + "rewards/accuracies": 0.6156250238418579, + "rewards/chosen": -74.0992202758789, + "rewards/margins": 7.507031440734863, + "rewards/rejected": -81.62187194824219, + "rewards/weighted_accuracies": 0.5531250238418579, + "rewards/weighted_chosen": -0.3904098570346832, + "rewards/weighted_margins": 0.14227446913719177, + "rewards/weighted_rejected": -0.5326492190361023, + "step": 1030 + }, + { + "epoch": 0.5558153517269022, + "grad_norm": 6.275015354156494, + "learning_rate": 4.916054021788265e-07, + "logits/chosen": -0.04349822923541069, + "logits/rejected": -0.005797195248305798, + "logps/chosen": -192.04061889648438, + "logps/rejected": -196.96875, + "logps/weighted_chosen": -1.060455322265625, + "logps/weighted_rejected": -1.1041381359100342, + "loss": 0.676, + "rewards/accuracies": 0.621874988079071, + "rewards/chosen": -73.9984359741211, + "rewards/margins": 6.270703315734863, + "rewards/rejected": -80.27656555175781, + "rewards/weighted_accuracies": 0.5687500238418579, + "rewards/weighted_chosen": -0.3287033140659332, + "rewards/weighted_margins": 0.17382964491844177, + "rewards/weighted_rejected": -0.5024040341377258, + "step": 1040 + }, + { + "epoch": 0.5611597301088918, + "grad_norm": 7.003974437713623, + "learning_rate": 4.822809494622379e-07, + "logits/chosen": -0.07877655327320099, + "logits/rejected": -0.06013183668255806, + "logps/chosen": -189.9453125, + "logps/rejected": -201.99374389648438, + "logps/weighted_chosen": -1.28973388671875, + "logps/weighted_rejected": -1.1621277332305908, + "loss": 0.7014, + "rewards/accuracies": 0.6000000238418579, + "rewards/chosen": -69.9781265258789, + "rewards/margins": 8.531641006469727, + "rewards/rejected": -78.4828109741211, + "rewards/weighted_accuracies": 0.5218750238418579, + "rewards/weighted_chosen": -0.38208693265914917, + "rewards/weighted_margins": 0.09144439548254013, + "rewards/weighted_rejected": -0.4734359681606293, + "step": 1050 + }, + { + "epoch": 0.5665041084908812, + "grad_norm": 7.428668975830078, + "learning_rate": 4.729626633121914e-07, + "logits/chosen": -0.065679170191288, + "logits/rejected": -0.02408447302877903, + "logps/chosen": -191.80313110351562, + "logps/rejected": -202.32186889648438, + "logps/weighted_chosen": -1.075415015220642, + "logps/weighted_rejected": -1.3513977527618408, + "loss": 0.6657, + "rewards/accuracies": 0.5718749761581421, + "rewards/chosen": -73.8218765258789, + "rewards/margins": 7.512499809265137, + "rewards/rejected": -81.32499694824219, + "rewards/weighted_accuracies": 0.590624988079071, + "rewards/weighted_chosen": -0.32668304443359375, + "rewards/weighted_margins": 0.21370239555835724, + "rewards/weighted_rejected": -0.5402389764785767, + "step": 1060 + }, + { + "epoch": 0.5718484868728706, + "grad_norm": 10.470946311950684, + "learning_rate": 4.636537866694692e-07, + "logits/chosen": -0.023516083136200905, + "logits/rejected": 0.0013282776344567537, + "logps/chosen": -193.6531219482422, + "logps/rejected": -200.671875, + "logps/weighted_chosen": -1.336614966392517, + "logps/weighted_rejected": -1.549737572669983, + "loss": 0.6482, + "rewards/accuracies": 0.6312500238418579, + "rewards/chosen": -75.12890625, + "rewards/margins": 6.398828029632568, + "rewards/rejected": -81.51094055175781, + "rewards/weighted_accuracies": 0.612500011920929, + "rewards/weighted_chosen": -0.3157196044921875, + "rewards/weighted_margins": 0.2204277068376541, + "rewards/weighted_rejected": -0.5359970331192017, + "step": 1070 + }, + { + "epoch": 0.57719286525486, + "grad_norm": 5.486422061920166, + "learning_rate": 4.5435755920016557e-07, + "logits/chosen": -0.15928420424461365, + "logits/rejected": -0.16242370009422302, + "logps/chosen": -213.3874969482422, + "logps/rejected": -219.72500610351562, + "logps/weighted_chosen": -0.9821227788925171, + "logps/weighted_rejected": -0.865551769733429, + "loss": 0.6778, + "rewards/accuracies": 0.6000000238418579, + "rewards/chosen": -76.74687194824219, + "rewards/margins": 8.090624809265137, + "rewards/rejected": -84.828125, + "rewards/weighted_accuracies": 0.5375000238418579, + "rewards/weighted_chosen": -0.32687264680862427, + "rewards/weighted_margins": 0.08376693725585938, + "rewards/weighted_rejected": -0.4107864499092102, + "step": 1080 + }, + { + "epoch": 0.5825372436368494, + "grad_norm": 8.151789665222168, + "learning_rate": 4.450772161682221e-07, + "logits/chosen": -0.068756103515625, + "logits/rejected": -0.015399170108139515, + "logps/chosen": -199.9187469482422, + "logps/rejected": -204.18124389648438, + "logps/weighted_chosen": -1.588891625404358, + "logps/weighted_rejected": -1.757287621498108, + "loss": 0.6788, + "rewards/accuracies": 0.628125011920929, + "rewards/chosen": -74.18281555175781, + "rewards/margins": 8.891406059265137, + "rewards/rejected": -83.09687805175781, + "rewards/weighted_accuracies": 0.5218750238418579, + "rewards/weighted_chosen": -0.30588531494140625, + "rewards/weighted_margins": 0.1938468962907791, + "rewards/weighted_rejected": -0.49979859590530396, + "step": 1090 + }, + { + "epoch": 0.587881622018839, + "grad_norm": 6.956281661987305, + "learning_rate": 4.3581598730949376e-07, + "logits/chosen": -0.064672090113163, + "logits/rejected": -0.038168907165527344, + "logps/chosen": -203.828125, + "logps/rejected": -218.46249389648438, + "logps/weighted_chosen": -1.4421265125274658, + "logps/weighted_rejected": -1.3359496593475342, + "loss": 0.6969, + "rewards/accuracies": 0.659375011920929, + "rewards/chosen": -78.1781234741211, + "rewards/margins": 10.614843368530273, + "rewards/rejected": -88.7796859741211, + "rewards/weighted_accuracies": 0.5625, + "rewards/weighted_chosen": -0.38915711641311646, + "rewards/weighted_margins": 0.0950775146484375, + "rewards/weighted_rejected": -0.4842666685581207, + "step": 1100 + }, + { + "epoch": 0.5932260004008284, + "grad_norm": 9.076947212219238, + "learning_rate": 4.2657709570774007e-07, + "logits/chosen": -0.06801605224609375, + "logits/rejected": -0.05637931823730469, + "logps/chosen": -195.67813110351562, + "logps/rejected": -204.66250610351562, + "logps/weighted_chosen": -1.285638451576233, + "logps/weighted_rejected": -1.3720200061798096, + "loss": 0.6571, + "rewards/accuracies": 0.625, + "rewards/chosen": -76.2890625, + "rewards/margins": 8.073827743530273, + "rewards/rejected": -84.35624694824219, + "rewards/weighted_accuracies": 0.5687500238418579, + "rewards/weighted_chosen": -0.4137374758720398, + "rewards/weighted_margins": 0.19571837782859802, + "rewards/weighted_rejected": -0.6095176935195923, + "step": 1110 + }, + { + "epoch": 0.5985703787828178, + "grad_norm": 7.4237751960754395, + "learning_rate": 4.173637566729304e-07, + "logits/chosen": -0.09175796806812286, + "logits/rejected": -0.07670821994543076, + "logps/chosen": -196.52188110351562, + "logps/rejected": -210.00625610351562, + "logps/weighted_chosen": -1.169805884361267, + "logps/weighted_rejected": -1.150994896888733, + "loss": 0.6803, + "rewards/accuracies": 0.6499999761581421, + "rewards/chosen": -76.2906265258789, + "rewards/margins": 10.744921684265137, + "rewards/rejected": -87.03125, + "rewards/weighted_accuracies": 0.59375, + "rewards/weighted_chosen": -0.39191436767578125, + "rewards/weighted_margins": 0.15792007744312286, + "rewards/weighted_rejected": -0.549652099609375, + "step": 1120 + }, + { + "epoch": 0.6039147571648072, + "grad_norm": 6.803237438201904, + "learning_rate": 4.0817917662225467e-07, + "logits/chosen": -0.14934463798999786, + "logits/rejected": -0.13103675842285156, + "logps/chosen": -204.88436889648438, + "logps/rejected": -215.25155639648438, + "logps/weighted_chosen": -1.14501953125, + "logps/weighted_rejected": -1.147515892982483, + "loss": 0.6706, + "rewards/accuracies": 0.659375011920929, + "rewards/chosen": -80.02812194824219, + "rewards/margins": 9.226953506469727, + "rewards/rejected": -89.2734375, + "rewards/weighted_accuracies": 0.565625011920929, + "rewards/weighted_chosen": -0.3231521546840668, + "rewards/weighted_margins": 0.16764679551124573, + "rewards/weighted_rejected": -0.49060362577438354, + "step": 1130 + }, + { + "epoch": 0.6092591355467967, + "grad_norm": 7.308501720428467, + "learning_rate": 3.9902655196422957e-07, + "logits/chosen": -0.17415618896484375, + "logits/rejected": -0.13456878066062927, + "logps/chosen": -204.93594360351562, + "logps/rejected": -214.28750610351562, + "logps/weighted_chosen": -0.799731433391571, + "logps/weighted_rejected": -1.1109497547149658, + "loss": 0.691, + "rewards/accuracies": 0.596875011920929, + "rewards/chosen": -82.1039047241211, + "rewards/margins": 7.87109375, + "rewards/rejected": -89.9781265258789, + "rewards/weighted_accuracies": 0.5625, + "rewards/weighted_chosen": -0.4164085388183594, + "rewards/weighted_margins": 0.125599667429924, + "rewards/weighted_rejected": -0.5417296886444092, + "step": 1140 + }, + { + "epoch": 0.6146035139287862, + "grad_norm": 13.811458587646484, + "learning_rate": 3.8990906798628763e-07, + "logits/chosen": -0.08937682956457138, + "logits/rejected": -0.06486892700195312, + "logps/chosen": -210.97500610351562, + "logps/rejected": -214.6125030517578, + "logps/weighted_chosen": -1.324133276939392, + "logps/weighted_rejected": -1.3257598876953125, + "loss": 0.7048, + "rewards/accuracies": 0.528124988079071, + "rewards/chosen": -80.96875, + "rewards/margins": 6.654687404632568, + "rewards/rejected": -87.6578140258789, + "rewards/weighted_accuracies": 0.543749988079071, + "rewards/weighted_chosen": -0.3775787353515625, + "rewards/weighted_margins": 0.11986999213695526, + "rewards/weighted_rejected": -0.4972671568393707, + "step": 1150 + }, + { + "epoch": 0.6199478923107756, + "grad_norm": 10.00770378112793, + "learning_rate": 3.8082989774623587e-07, + "logits/chosen": -0.01934986189007759, + "logits/rejected": -0.015110015869140625, + "logps/chosen": -188.74063110351562, + "logps/rejected": -204.4187469482422, + "logps/weighted_chosen": -1.20220947265625, + "logps/weighted_rejected": -1.109490990638733, + "loss": 0.6717, + "rewards/accuracies": 0.637499988079071, + "rewards/chosen": -77.26094055175781, + "rewards/margins": 9.564844131469727, + "rewards/rejected": -86.8343734741211, + "rewards/weighted_accuracies": 0.578125, + "rewards/weighted_chosen": -0.3926025331020355, + "rewards/weighted_margins": 0.13916321098804474, + "rewards/weighted_rejected": -0.5318969488143921, + "step": 1160 + }, + { + "epoch": 0.625292270692765, + "grad_norm": 28.483694076538086, + "learning_rate": 3.717922009679725e-07, + "logits/chosen": -0.053440093994140625, + "logits/rejected": -0.0072647095657885075, + "logps/chosen": -201.890625, + "logps/rejected": -210.5187530517578, + "logps/weighted_chosen": -1.0479736328125, + "logps/weighted_rejected": -1.5367584228515625, + "loss": 0.6686, + "rewards/accuracies": 0.621874988079071, + "rewards/chosen": -78.7554702758789, + "rewards/margins": 8.385546684265137, + "rewards/rejected": -87.1624984741211, + "rewards/weighted_accuracies": 0.5687500238418579, + "rewards/weighted_chosen": -0.352639764547348, + "rewards/weighted_margins": 0.159870907664299, + "rewards/weighted_rejected": -0.5122512578964233, + "step": 1170 + }, + { + "epoch": 0.6306366490747545, + "grad_norm": 7.887178897857666, + "learning_rate": 3.6279912294184207e-07, + "logits/chosen": -0.06793518364429474, + "logits/rejected": -0.03975982591509819, + "logps/chosen": -208.45938110351562, + "logps/rejected": -216.97500610351562, + "logps/weighted_chosen": -1.1531188488006592, + "logps/weighted_rejected": -1.3199646472930908, + "loss": 0.664, + "rewards/accuracies": 0.625, + "rewards/chosen": -83.76875305175781, + "rewards/margins": 8.121874809265137, + "rewards/rejected": -91.8890609741211, + "rewards/weighted_accuracies": 0.6000000238418579, + "rewards/weighted_chosen": -0.3757583498954773, + "rewards/weighted_margins": 0.18969421088695526, + "rewards/weighted_rejected": -0.5654312372207642, + "step": 1180 + }, + { + "epoch": 0.6359810274567439, + "grad_norm": 21.66887664794922, + "learning_rate": 3.5385379343001507e-07, + "logits/chosen": -0.041289519518613815, + "logits/rejected": -0.01831817626953125, + "logps/chosen": -198.00936889648438, + "logps/rejected": -204.0437469482422, + "logps/weighted_chosen": -1.02520751953125, + "logps/weighted_rejected": -1.4494140148162842, + "loss": 0.6816, + "rewards/accuracies": 0.596875011920929, + "rewards/chosen": -79.15156555175781, + "rewards/margins": 7.596484184265137, + "rewards/rejected": -86.73515319824219, + "rewards/weighted_accuracies": 0.6031249761581421, + "rewards/weighted_chosen": -0.3985244631767273, + "rewards/weighted_margins": 0.12162628024816513, + "rewards/weighted_rejected": -0.5204399228096008, + "step": 1190 + }, + { + "epoch": 0.6413254058387334, + "grad_norm": 11.715490341186523, + "learning_rate": 3.4495932557727114e-07, + "logits/chosen": -0.0670623779296875, + "logits/rejected": -0.04256897047162056, + "logps/chosen": -200.75936889648438, + "logps/rejected": -214.21875, + "logps/weighted_chosen": -1.1181519031524658, + "logps/weighted_rejected": -1.4523468017578125, + "loss": 0.6876, + "rewards/accuracies": 0.6468750238418579, + "rewards/chosen": -81.3304672241211, + "rewards/margins": 10.844531059265137, + "rewards/rejected": -92.1703109741211, + "rewards/weighted_accuracies": 0.550000011920929, + "rewards/weighted_chosen": -0.4529586732387543, + "rewards/weighted_margins": 0.178059384226799, + "rewards/weighted_rejected": -0.6310364007949829, + "step": 1200 + }, + { + "epoch": 0.6466697842207229, + "grad_norm": 5.799643516540527, + "learning_rate": 3.3611881482756464e-07, + "logits/chosen": -0.02250823937356472, + "logits/rejected": -0.02404480054974556, + "logps/chosen": -209.1593780517578, + "logps/rejected": -213.9921875, + "logps/weighted_chosen": -1.55322265625, + "logps/weighted_rejected": -1.4918396472930908, + "loss": 0.6919, + "rewards/accuracies": 0.574999988079071, + "rewards/chosen": -84.61796569824219, + "rewards/margins": 6.630078315734863, + "rewards/rejected": -91.2593765258789, + "rewards/weighted_accuracies": 0.528124988079071, + "rewards/weighted_chosen": -0.4175216555595398, + "rewards/weighted_margins": 0.15537413954734802, + "rewards/weighted_rejected": -0.5732803344726562, + "step": 1210 + }, + { + "epoch": 0.6520141626027123, + "grad_norm": 10.090937614440918, + "learning_rate": 3.2733533784675273e-07, + "logits/chosen": 0.013554382137954235, + "logits/rejected": 0.04766502231359482, + "logps/chosen": -197.359375, + "logps/rejected": -209.89688110351562, + "logps/weighted_chosen": -1.3024734258651733, + "logps/weighted_rejected": -1.784857153892517, + "loss": 0.69, + "rewards/accuracies": 0.621874988079071, + "rewards/chosen": -79.140625, + "rewards/margins": 10.651171684265137, + "rewards/rejected": -89.7906265258789, + "rewards/weighted_accuracies": 0.5531250238418579, + "rewards/weighted_chosen": -0.4760070741176605, + "rewards/weighted_margins": 0.15030670166015625, + "rewards/weighted_rejected": -0.6262580752372742, + "step": 1220 + }, + { + "epoch": 0.6573585409847017, + "grad_norm": 7.221394062042236, + "learning_rate": 3.1861195145185603e-07, + "logits/chosen": -0.03308219835162163, + "logits/rejected": 0.00255584716796875, + "logps/chosen": -206.4031219482422, + "logps/rejected": -215.953125, + "logps/weighted_chosen": -1.181341528892517, + "logps/weighted_rejected": -1.657598853111267, + "loss": 0.6826, + "rewards/accuracies": 0.6000000238418579, + "rewards/chosen": -83.6734390258789, + "rewards/margins": 7.490234375, + "rewards/rejected": -91.1578140258789, + "rewards/weighted_accuracies": 0.5531250238418579, + "rewards/weighted_chosen": -0.3481903076171875, + "rewards/weighted_margins": 0.16850128769874573, + "rewards/weighted_rejected": -0.5166076421737671, + "step": 1230 + }, + { + "epoch": 0.6627029193666911, + "grad_norm": 50.77132797241211, + "learning_rate": 3.099516915472289e-07, + "logits/chosen": 0.006022644229233265, + "logits/rejected": 0.08475188910961151, + "logps/chosen": -200.8328094482422, + "logps/rejected": -205.4609375, + "logps/weighted_chosen": -0.924468994140625, + "logps/weighted_rejected": -1.4624541997909546, + "loss": 0.6753, + "rewards/accuracies": 0.6031249761581421, + "rewards/chosen": -80.05937194824219, + "rewards/margins": 5.724609375, + "rewards/rejected": -85.7828140258789, + "rewards/weighted_accuracies": 0.5874999761581421, + "rewards/weighted_chosen": -0.3424621522426605, + "rewards/weighted_margins": 0.1885635405778885, + "rewards/weighted_rejected": -0.5309646725654602, + "step": 1240 + }, + { + "epoch": 0.6680472977486807, + "grad_norm": 7.760387897491455, + "learning_rate": 3.013575720680062e-07, + "logits/chosen": 0.020127106457948685, + "logits/rejected": 0.042375944554805756, + "logps/chosen": -202.19375610351562, + "logps/rejected": -208.9968719482422, + "logps/weighted_chosen": -1.1934722661972046, + "logps/weighted_rejected": -1.6453979015350342, + "loss": 0.665, + "rewards/accuracies": 0.6187499761581421, + "rewards/chosen": -81.3656234741211, + "rewards/margins": 8.227343559265137, + "rewards/rejected": -89.56718444824219, + "rewards/weighted_accuracies": 0.550000011920929, + "rewards/weighted_chosen": -0.3932647705078125, + "rewards/weighted_margins": 0.21837767958641052, + "rewards/weighted_rejected": -0.6111129522323608, + "step": 1250 + }, + { + "epoch": 0.6733916761306701, + "grad_norm": 16.958906173706055, + "learning_rate": 2.928325839311974e-07, + "logits/chosen": -0.02686614915728569, + "logits/rejected": -0.016692351549863815, + "logps/chosen": -194.46719360351562, + "logps/rejected": -203.3093719482422, + "logps/weighted_chosen": -1.3333160877227783, + "logps/weighted_rejected": -1.2732360363006592, + "loss": 0.7271, + "rewards/accuracies": 0.6031249761581421, + "rewards/chosen": -79.31562805175781, + "rewards/margins": 8.610547065734863, + "rewards/rejected": -87.9281234741211, + "rewards/weighted_accuracies": 0.5531250238418579, + "rewards/weighted_chosen": -0.4805351197719574, + "rewards/weighted_margins": 0.173136904835701, + "rewards/weighted_rejected": -0.6535751223564148, + "step": 1260 + }, + { + "epoch": 0.6787360545126595, + "grad_norm": 19.331541061401367, + "learning_rate": 2.843796939947896e-07, + "logits/chosen": -0.015142058953642845, + "logits/rejected": 0.009578704833984375, + "logps/chosen": -199.8125, + "logps/rejected": -213.27499389648438, + "logps/weighted_chosen": -1.1416137218475342, + "logps/weighted_rejected": -1.472558617591858, + "loss": 0.6635, + "rewards/accuracies": 0.6187499761581421, + "rewards/chosen": -82.5484390258789, + "rewards/margins": 10.644140243530273, + "rewards/rejected": -93.1968765258789, + "rewards/weighted_accuracies": 0.550000011920929, + "rewards/weighted_chosen": -0.4067550599575043, + "rewards/weighted_margins": 0.1748603880405426, + "rewards/weighted_rejected": -0.5817291140556335, + "step": 1270 + }, + { + "epoch": 0.6840804328946489, + "grad_norm": 15.563817977905273, + "learning_rate": 2.7600184402522575e-07, + "logits/chosen": 0.02512359619140625, + "logits/rejected": 0.06499175727367401, + "logps/chosen": -196.04061889648438, + "logps/rejected": -209.2218780517578, + "logps/weighted_chosen": -1.2925841808319092, + "logps/weighted_rejected": -1.1955718994140625, + "loss": 0.6787, + "rewards/accuracies": 0.668749988079071, + "rewards/chosen": -79.7046890258789, + "rewards/margins": 11.271875381469727, + "rewards/rejected": -90.94921875, + "rewards/weighted_accuracies": 0.596875011920929, + "rewards/weighted_chosen": -0.4067886471748352, + "rewards/weighted_margins": 0.17460021376609802, + "rewards/weighted_rejected": -0.5814231634140015, + "step": 1280 + }, + { + "epoch": 0.6894248112766383, + "grad_norm": 13.546245574951172, + "learning_rate": 2.6770194967361303e-07, + "logits/chosen": -0.0179290771484375, + "logits/rejected": 0.0178680419921875, + "logps/chosen": -213.72500610351562, + "logps/rejected": -227.05313110351562, + "logps/weighted_chosen": -1.001013159751892, + "logps/weighted_rejected": -1.686248779296875, + "loss": 0.6712, + "rewards/accuracies": 0.6031249761581421, + "rewards/chosen": -86.60624694824219, + "rewards/margins": 8.127344131469727, + "rewards/rejected": -94.73124694824219, + "rewards/weighted_accuracies": 0.5406249761581421, + "rewards/weighted_chosen": -0.3357299864292145, + "rewards/weighted_margins": 0.20237883925437927, + "rewards/weighted_rejected": -0.5380722284317017, + "step": 1290 + }, + { + "epoch": 0.6947691896586279, + "grad_norm": 14.806809425354004, + "learning_rate": 2.594828994610211e-07, + "logits/chosen": 0.027711868286132812, + "logits/rejected": 0.05919818952679634, + "logps/chosen": -193.265625, + "logps/rejected": -206.9031219482422, + "logps/weighted_chosen": -1.0292694568634033, + "logps/weighted_rejected": -1.295262098312378, + "loss": 0.6787, + "rewards/accuracies": 0.6000000238418579, + "rewards/chosen": -80.3687515258789, + "rewards/margins": 8.969141006469727, + "rewards/rejected": -89.3671875, + "rewards/weighted_accuracies": 0.5562499761581421, + "rewards/weighted_chosen": -0.3460655212402344, + "rewards/weighted_margins": 0.17020568251609802, + "rewards/weighted_rejected": -0.5160125494003296, + "step": 1300 + }, + { + "epoch": 0.7001135680406173, + "grad_norm": 12.198771476745605, + "learning_rate": 2.513475537732219e-07, + "logits/chosen": 0.0022941590286791325, + "logits/rejected": 0.025787353515625, + "logps/chosen": -204.703125, + "logps/rejected": -217.66250610351562, + "logps/weighted_chosen": -1.215307593345642, + "logps/weighted_rejected": -1.44464111328125, + "loss": 0.665, + "rewards/accuracies": 0.609375, + "rewards/chosen": -84.4281234741211, + "rewards/margins": 8.751562118530273, + "rewards/rejected": -93.18281555175781, + "rewards/weighted_accuracies": 0.559374988079071, + "rewards/weighted_chosen": -0.3498634397983551, + "rewards/weighted_margins": 0.19211502373218536, + "rewards/weighted_rejected": -0.541577160358429, + "step": 1310 + }, + { + "epoch": 0.7054579464226067, + "grad_norm": 18.498014450073242, + "learning_rate": 2.4329874386522113e-07, + "logits/chosen": 0.0073295594193041325, + "logits/rejected": 0.055358123034238815, + "logps/chosen": -199.6437530517578, + "logps/rejected": -208.70156860351562, + "logps/weighted_chosen": -1.0841186046600342, + "logps/weighted_rejected": -1.4295532703399658, + "loss": 0.6551, + "rewards/accuracies": 0.59375, + "rewards/chosen": -83.765625, + "rewards/margins": 8.8515625, + "rewards/rejected": -92.6343765258789, + "rewards/weighted_accuracies": 0.596875011920929, + "rewards/weighted_chosen": -0.4359336793422699, + "rewards/weighted_margins": 0.21386031806468964, + "rewards/weighted_rejected": -0.6501365900039673, + "step": 1320 + }, + { + "epoch": 0.7108023248045962, + "grad_norm": 15.962686538696289, + "learning_rate": 2.3533927087592713e-07, + "logits/chosen": -0.000156402587890625, + "logits/rejected": 0.01469268836081028, + "logps/chosen": -206.29061889648438, + "logps/rejected": -216.75, + "logps/weighted_chosen": -1.1605316400527954, + "logps/weighted_rejected": -1.276606798171997, + "loss": 0.7058, + "rewards/accuracies": 0.59375, + "rewards/chosen": -83.1468734741211, + "rewards/margins": 7.533203125, + "rewards/rejected": -90.6796875, + "rewards/weighted_accuracies": 0.528124988079071, + "rewards/weighted_chosen": -0.422750860452652, + "rewards/weighted_margins": 0.102025605738163, + "rewards/weighted_rejected": -0.5250183343887329, + "step": 1330 + }, + { + "epoch": 0.7161467031865856, + "grad_norm": 10.239470481872559, + "learning_rate": 2.2747190485330193e-07, + "logits/chosen": 0.02880706824362278, + "logits/rejected": 0.07036437839269638, + "logps/chosen": -199.8859405517578, + "logps/rejected": -209.57186889648438, + "logps/weighted_chosen": -1.2549285888671875, + "logps/weighted_rejected": -1.352288842201233, + "loss": 0.6718, + "rewards/accuracies": 0.637499988079071, + "rewards/chosen": -80.2109375, + "rewards/margins": 10.519922256469727, + "rewards/rejected": -90.71875, + "rewards/weighted_accuracies": 0.612500011920929, + "rewards/weighted_chosen": -0.3369941711425781, + "rewards/weighted_margins": 0.2133018523454666, + "rewards/weighted_rejected": -0.5501846075057983, + "step": 1340 + }, + { + "epoch": 0.7214910815685751, + "grad_norm": 12.942761421203613, + "learning_rate": 2.1969938379033166e-07, + "logits/chosen": -0.0024871826171875, + "logits/rejected": -0.01717834547162056, + "logps/chosen": -206.4250030517578, + "logps/rejected": -220.47030639648438, + "logps/weighted_chosen": -1.2035309076309204, + "logps/weighted_rejected": -1.3001312017440796, + "loss": 0.7016, + "rewards/accuracies": 0.596875011920929, + "rewards/chosen": -82.8765640258789, + "rewards/margins": 8.98828125, + "rewards/rejected": -91.8671875, + "rewards/weighted_accuracies": 0.5718749761581421, + "rewards/weighted_chosen": -0.37746506929397583, + "rewards/weighted_margins": 0.11672210693359375, + "rewards/weighted_rejected": -0.4940811097621918, + "step": 1350 + }, + { + "epoch": 0.7268354599505645, + "grad_norm": 16.300827026367188, + "learning_rate": 2.1202441267215265e-07, + "logits/chosen": 0.04172973707318306, + "logits/rejected": 0.06365661323070526, + "logps/chosen": -183.5437469482422, + "logps/rejected": -198.49063110351562, + "logps/weighted_chosen": -1.542272925376892, + "logps/weighted_rejected": -1.678741455078125, + "loss": 0.7155, + "rewards/accuracies": 0.643750011920929, + "rewards/chosen": -78.47343444824219, + "rewards/margins": 10.018359184265137, + "rewards/rejected": -88.47343444824219, + "rewards/weighted_accuracies": 0.5625, + "rewards/weighted_chosen": -0.4637313783168793, + "rewards/weighted_margins": 0.1779930144548416, + "rewards/weighted_rejected": -0.6415374875068665, + "step": 1360 + }, + { + "epoch": 0.732179838332554, + "grad_norm": 17.129104614257812, + "learning_rate": 2.0444966253466572e-07, + "logits/chosen": -0.01758270338177681, + "logits/rejected": 0.049257658421993256, + "logps/chosen": -198.3562469482422, + "logps/rejected": -212.31875610351562, + "logps/weighted_chosen": -1.0854461193084717, + "logps/weighted_rejected": -1.6400238275527954, + "loss": 0.6718, + "rewards/accuracies": 0.606249988079071, + "rewards/chosen": -81.81718444824219, + "rewards/margins": 8.70703125, + "rewards/rejected": -90.52812194824219, + "rewards/weighted_accuracies": 0.5874999761581421, + "rewards/weighted_chosen": -0.3319801390171051, + "rewards/weighted_margins": 0.17666855454444885, + "rewards/weighted_rejected": -0.5086410641670227, + "step": 1370 + }, + { + "epoch": 0.7375242167145434, + "grad_norm": 15.24010181427002, + "learning_rate": 1.9697776953496397e-07, + "logits/chosen": -0.0036972046364098787, + "logits/rejected": 0.04438323900103569, + "logps/chosen": -204.8406219482422, + "logps/rejected": -215.5343780517578, + "logps/weighted_chosen": -1.2466247081756592, + "logps/weighted_rejected": -1.588470458984375, + "loss": 0.6659, + "rewards/accuracies": 0.6187499761581421, + "rewards/chosen": -82.30078125, + "rewards/margins": 10.021875381469727, + "rewards/rejected": -92.33515930175781, + "rewards/weighted_accuracies": 0.574999988079071, + "rewards/weighted_chosen": -0.3886352479457855, + "rewards/weighted_margins": 0.18134918808937073, + "rewards/weighted_rejected": -0.5699905157089233, + "step": 1380 + }, + { + "epoch": 0.7428685950965328, + "grad_norm": 13.349143028259277, + "learning_rate": 1.8961133403390113e-07, + "logits/chosen": -0.03137054294347763, + "logits/rejected": 0.01921997033059597, + "logps/chosen": -205.91250610351562, + "logps/rejected": -215.74374389648438, + "logps/weighted_chosen": -0.754425048828125, + "logps/weighted_rejected": -1.830108642578125, + "loss": 0.6648, + "rewards/accuracies": 0.596875011920929, + "rewards/chosen": -82.6187515258789, + "rewards/margins": 9.654296875, + "rewards/rejected": -92.2750015258789, + "rewards/weighted_accuracies": 0.590624988079071, + "rewards/weighted_chosen": -0.38862380385398865, + "rewards/weighted_margins": 0.23016509413719177, + "rewards/weighted_rejected": -0.618792712688446, + "step": 1390 + }, + { + "epoch": 0.7482129734785222, + "grad_norm": 12.553123474121094, + "learning_rate": 1.823529196911156e-07, + "logits/chosen": -0.07255706936120987, + "logits/rejected": -0.03428497165441513, + "logps/chosen": -207.0109405517578, + "logps/rejected": -213.7937469482422, + "logps/weighted_chosen": -1.1017853021621704, + "logps/weighted_rejected": -1.335351586341858, + "loss": 0.6721, + "rewards/accuracies": 0.609375, + "rewards/chosen": -83.64375305175781, + "rewards/margins": 7.791796684265137, + "rewards/rejected": -91.4296875, + "rewards/weighted_accuracies": 0.59375, + "rewards/weighted_chosen": -0.35135191679000854, + "rewards/weighted_margins": 0.21256867051124573, + "rewards/weighted_rejected": -0.5637038946151733, + "step": 1400 + }, + { + "epoch": 0.7535573518605118, + "grad_norm": 10.907644271850586, + "learning_rate": 1.7520505257282886e-07, + "logits/chosen": 0.02479248121380806, + "logits/rejected": 0.05216522142291069, + "logps/chosen": -198.2531280517578, + "logps/rejected": -207.4499969482422, + "logps/weighted_chosen": -1.016326904296875, + "logps/weighted_rejected": -1.4452117681503296, + "loss": 0.6724, + "rewards/accuracies": 0.6312500238418579, + "rewards/chosen": -80.75312805175781, + "rewards/margins": 9.733593940734863, + "rewards/rejected": -90.4671859741211, + "rewards/weighted_accuracies": 0.578125, + "rewards/weighted_chosen": -0.3796447813510895, + "rewards/weighted_margins": 0.1787368804216385, + "rewards/weighted_rejected": -0.5583381652832031, + "step": 1410 + }, + { + "epoch": 0.7589017302425012, + "grad_norm": 9.989439964294434, + "learning_rate": 1.6817022027272572e-07, + "logits/chosen": 0.03133545070886612, + "logits/rejected": 0.03775329515337944, + "logps/chosen": -207.06405639648438, + "logps/rejected": -224.08749389648438, + "logps/weighted_chosen": -1.253021240234375, + "logps/weighted_rejected": -1.788061499595642, + "loss": 0.703, + "rewards/accuracies": 0.640625, + "rewards/chosen": -82.796875, + "rewards/margins": 11.674609184265137, + "rewards/rejected": -94.50468444824219, + "rewards/weighted_accuracies": 0.5687500238418579, + "rewards/weighted_chosen": -0.3869224488735199, + "rewards/weighted_margins": 0.14797210693359375, + "rewards/weighted_rejected": -0.5350815057754517, + "step": 1420 + }, + { + "epoch": 0.7642461086244906, + "grad_norm": 12.953351974487305, + "learning_rate": 1.6125087104622502e-07, + "logits/chosen": -0.03018951416015625, + "logits/rejected": 0.010332489386200905, + "logps/chosen": -204.82186889648438, + "logps/rejected": -211.2375030517578, + "logps/weighted_chosen": -1.2213348150253296, + "logps/weighted_rejected": -1.4504730701446533, + "loss": 0.679, + "rewards/accuracies": 0.6187499761581421, + "rewards/chosen": -81.8109359741211, + "rewards/margins": 8.600781440734863, + "rewards/rejected": -90.39375305175781, + "rewards/weighted_accuracies": 0.6000000238418579, + "rewards/weighted_chosen": -0.43824997544288635, + "rewards/weighted_margins": 0.12977370619773865, + "rewards/weighted_rejected": -0.56793212890625, + "step": 1430 + }, + { + "epoch": 0.76959048700648, + "grad_norm": 51.06248474121094, + "learning_rate": 1.5444941295843945e-07, + "logits/chosen": -0.0010368346702307463, + "logits/rejected": -0.0032970428001135588, + "logps/chosen": -206.72500610351562, + "logps/rejected": -225.6125030517578, + "logps/weighted_chosen": -0.7966338992118835, + "logps/weighted_rejected": -0.8868865966796875, + "loss": 0.6564, + "rewards/accuracies": 0.640625, + "rewards/chosen": -83.40937805175781, + "rewards/margins": 12.608983993530273, + "rewards/rejected": -96.03125, + "rewards/weighted_accuracies": 0.5625, + "rewards/weighted_chosen": -0.2895980775356293, + "rewards/weighted_margins": 0.20346298813819885, + "rewards/weighted_rejected": -0.49294739961624146, + "step": 1440 + }, + { + "epoch": 0.7749348653884695, + "grad_norm": 7.5457658767700195, + "learning_rate": 1.4776821304612392e-07, + "logits/chosen": -0.01905670203268528, + "logits/rejected": 0.01137466449290514, + "logps/chosen": -196.77969360351562, + "logps/rejected": -210.0859375, + "logps/weighted_chosen": -0.9277313351631165, + "logps/weighted_rejected": -1.508001685142517, + "loss": 0.6648, + "rewards/accuracies": 0.625, + "rewards/chosen": -79.80155944824219, + "rewards/margins": 9.823827743530273, + "rewards/rejected": -89.63984680175781, + "rewards/weighted_accuracies": 0.578125, + "rewards/weighted_chosen": -0.3523406982421875, + "rewards/weighted_margins": 0.22584763169288635, + "rewards/weighted_rejected": -0.5784717798233032, + "step": 1450 + }, + { + "epoch": 0.780279243770459, + "grad_norm": 22.205232620239258, + "learning_rate": 1.4120959649390136e-07, + "logits/chosen": -0.06232147291302681, + "logits/rejected": -0.025969695299863815, + "logps/chosen": -205.7578125, + "logps/rejected": -216.8156280517578, + "logps/weighted_chosen": -1.046392798423767, + "logps/weighted_rejected": -1.1339843273162842, + "loss": 0.6717, + "rewards/accuracies": 0.643750011920929, + "rewards/chosen": -82.03125, + "rewards/margins": 10.355077743530273, + "rewards/rejected": -92.41093444824219, + "rewards/weighted_accuracies": 0.6000000238418579, + "rewards/weighted_chosen": -0.39169007539749146, + "rewards/weighted_margins": 0.185791015625, + "rewards/weighted_rejected": -0.5777603387832642, + "step": 1460 + }, + { + "epoch": 0.7856236221524484, + "grad_norm": 15.942556381225586, + "learning_rate": 1.347758458250541e-07, + "logits/chosen": 0.012713623233139515, + "logits/rejected": 0.04683647304773331, + "logps/chosen": -203.40469360351562, + "logps/rejected": -219.640625, + "logps/weighted_chosen": -1.0474426746368408, + "logps/weighted_rejected": -1.3139832019805908, + "loss": 0.6875, + "rewards/accuracies": 0.5843750238418579, + "rewards/chosen": -82.4453125, + "rewards/margins": 7.944531440734863, + "rewards/rejected": -90.3812484741211, + "rewards/weighted_accuracies": 0.5625, + "rewards/weighted_chosen": -0.4108734130859375, + "rewards/weighted_margins": 0.16155776381492615, + "rewards/weighted_rejected": -0.5725204348564148, + "step": 1470 + }, + { + "epoch": 0.7909680005344378, + "grad_norm": 26.816959381103516, + "learning_rate": 1.2846920010716266e-07, + "logits/chosen": 0.014247131533920765, + "logits/rejected": 0.017230987548828125, + "logps/chosen": -199.45938110351562, + "logps/rejected": -212.2609405517578, + "logps/weighted_chosen": -1.156378149986267, + "logps/weighted_rejected": -1.0633026361465454, + "loss": 0.6873, + "rewards/accuracies": 0.659375011920929, + "rewards/chosen": -82.2640609741211, + "rewards/margins": 11.641016006469727, + "rewards/rejected": -93.9296875, + "rewards/weighted_accuracies": 0.5874999761581421, + "rewards/weighted_chosen": -0.43985289335250854, + "rewards/weighted_margins": 0.15669021010398865, + "rewards/weighted_rejected": -0.5962905883789062, + "step": 1480 + }, + { + "epoch": 0.7963123789164273, + "grad_norm": 9.73423957824707, + "learning_rate": 1.2229185417286764e-07, + "logits/chosen": -0.011121368035674095, + "logits/rejected": 0.0305023193359375, + "logps/chosen": -208.5437469482422, + "logps/rejected": -219.5906219482422, + "logps/weighted_chosen": -1.2618286609649658, + "logps/weighted_rejected": -1.607025146484375, + "loss": 0.6811, + "rewards/accuracies": 0.606249988079071, + "rewards/chosen": -81.4375, + "rewards/margins": 10.083984375, + "rewards/rejected": -91.5062484741211, + "rewards/weighted_accuracies": 0.6000000238418579, + "rewards/weighted_chosen": -0.3379165530204773, + "rewards/weighted_margins": 0.1427352875471115, + "rewards/weighted_rejected": -0.480560302734375, + "step": 1490 + }, + { + "epoch": 0.8016567572984167, + "grad_norm": 25.978179931640625, + "learning_rate": 1.1624595785602576e-07, + "logits/chosen": -0.02779388427734375, + "logits/rejected": 0.00923004187643528, + "logps/chosen": -207.6750030517578, + "logps/rejected": -214.234375, + "logps/weighted_chosen": -1.3065948486328125, + "logps/weighted_rejected": -1.779363989830017, + "loss": 0.7042, + "rewards/accuracies": 0.581250011920929, + "rewards/chosen": -82.6429672241211, + "rewards/margins": 7.233593940734863, + "rewards/rejected": -89.8929672241211, + "rewards/weighted_accuracies": 0.5718749761581421, + "rewards/weighted_chosen": -0.4565597474575043, + "rewards/weighted_margins": 0.20427818596363068, + "rewards/weighted_rejected": -0.660656750202179, + "step": 1500 + }, + { + "epoch": 0.8016567572984167, + "eval_logits/chosen": -0.1534786820411682, + "eval_logits/rejected": -0.12998200953006744, + "eval_logps/chosen": -201.64256286621094, + "eval_logps/rejected": -211.74440002441406, + "eval_logps/weighted_chosen": -1.0274064540863037, + "eval_logps/weighted_rejected": -1.3898550271987915, + "eval_loss": 0.66490238904953, + "eval_rewards/accuracies": 0.6135438084602356, + "eval_rewards/chosen": -80.33655548095703, + "eval_rewards/margins": 8.974350929260254, + "eval_rewards/rejected": -89.31975555419922, + "eval_rewards/weighted_accuracies": 0.5651730895042419, + "eval_rewards/weighted_chosen": -0.38825926184654236, + "eval_rewards/weighted_margins": 0.20228415727615356, + "eval_rewards/weighted_rejected": -0.5905433893203735, + "eval_runtime": 1233.4168, + "eval_samples_per_second": 1.59, + "eval_steps_per_second": 0.398, + "step": 1500 + } + ], + "logging_steps": 10, + "max_steps": 1872, + "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..7b414b2 --- /dev/null +++ b/checkpoint-1500/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fa2dcf6fe65fd94a0fde469e207f237b0752c6e558745acec2daa0276a1175ce +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-1872/chat_template.jinja b/checkpoint-1872/chat_template.jinja new file mode 100644 index 0000000..39bd0c9 --- /dev/null +++ b/checkpoint-1872/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-1872/config.json b/checkpoint-1872/config.json new file mode 100644 index 0000000..ceed8ae --- /dev/null +++ b/checkpoint-1872/config.json @@ -0,0 +1,29 @@ +{ + "architectures": [ + "LlamaForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "bos_token_id": 128000, + "eos_token_id": 128009, + "head_dim": 128, + "hidden_act": "silu", + "hidden_size": 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-1872/generation_config.json b/checkpoint-1872/generation_config.json new file mode 100644 index 0000000..38e1b41 --- /dev/null +++ b/checkpoint-1872/generation_config.json @@ -0,0 +1,12 @@ +{ + "bos_token_id": 128000, + "do_sample": true, + "eos_token_id": [ + 128001, + 128009 + ], + "max_length": 4096, + "temperature": 0.6, + "top_p": 0.9, + "transformers_version": "4.54.1" +} diff --git a/checkpoint-1872/latest b/checkpoint-1872/latest new file mode 100644 index 0000000..159bf4b --- /dev/null +++ b/checkpoint-1872/latest @@ -0,0 +1 @@ +global_step1871 \ No newline at end of file diff --git a/checkpoint-1872/model-00001-of-00004.safetensors b/checkpoint-1872/model-00001-of-00004.safetensors new file mode 100644 index 0000000..ccab831 --- /dev/null +++ b/checkpoint-1872/model-00001-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9b1f91ce998738ff854919fcece90440bf284a13e36635adf78d3edc7b0c9583 +size 4976698672 diff --git a/checkpoint-1872/model-00002-of-00004.safetensors b/checkpoint-1872/model-00002-of-00004.safetensors new file mode 100644 index 0000000..fdfe074 --- /dev/null +++ b/checkpoint-1872/model-00002-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e2e889b4e77d1f516c8891784fae73528795e932adc20d7d1f667c2ae85999f3 +size 4999802720 diff --git a/checkpoint-1872/model-00003-of-00004.safetensors b/checkpoint-1872/model-00003-of-00004.safetensors new file mode 100644 index 0000000..d56f7dc --- /dev/null +++ b/checkpoint-1872/model-00003-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:abae5b5c881483db48e9b0e75c4fc41d85da6135c458e0fed88e7134a07b040b +size 4915916176 diff --git a/checkpoint-1872/model-00004-of-00004.safetensors b/checkpoint-1872/model-00004-of-00004.safetensors new file mode 100644 index 0000000..558a162 --- /dev/null +++ b/checkpoint-1872/model-00004-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e1f1af93ae12a50d91c3863142e204beb76f74089a5d36bef78b7d1e6db8a5cc +size 1168138808 diff --git a/checkpoint-1872/model.safetensors.index.json b/checkpoint-1872/model.safetensors.index.json new file mode 100644 index 0000000..58a7ef4 --- /dev/null +++ b/checkpoint-1872/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-1872/rng_state_0.pth b/checkpoint-1872/rng_state_0.pth new file mode 100644 index 0000000..8d84687 --- /dev/null +++ b/checkpoint-1872/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-1872/rng_state_1.pth b/checkpoint-1872/rng_state_1.pth new file mode 100644 index 0000000..54f0119 --- /dev/null +++ b/checkpoint-1872/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-1872/scheduler.pt b/checkpoint-1872/scheduler.pt new file mode 100644 index 0000000..3dda4e0 --- /dev/null +++ b/checkpoint-1872/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e0c045ef5b4d5e7caa97c11866eea0dff56ba3423a3097e77f0a1b0a2ee93c08 +size 1465 diff --git a/checkpoint-1872/special_tokens_map.json b/checkpoint-1872/special_tokens_map.json new file mode 100644 index 0000000..b43be96 --- /dev/null +++ b/checkpoint-1872/special_tokens_map.json @@ -0,0 +1,17 @@ +{ + "bos_token": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "eos_token": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "pad_token": "<|eot_id|>" +} diff --git a/checkpoint-1872/tokenizer.json b/checkpoint-1872/tokenizer.json new file mode 100644 index 0000000..86a3394 --- /dev/null +++ b/checkpoint-1872/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3c5cf44023714fb39b05e71e425f8d7b92805ff73f7988b083b8c87f0bf87393 +size 17209961 diff --git a/checkpoint-1872/tokenizer_config.json b/checkpoint-1872/tokenizer_config.json new file mode 100644 index 0000000..34d134f --- /dev/null +++ b/checkpoint-1872/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": "<|eot_id|>", + "extra_special_tokens": {}, + "model_input_names": [ + "input_ids", + "attention_mask" + ], + "model_max_length": 1000000000000000019884624838656, + "pad_token": "<|eot_id|>", + "tokenizer_class": "PreTrainedTokenizerFast" +} diff --git a/checkpoint-1872/trainer_state.json b/checkpoint-1872/trainer_state.json new file mode 100644 index 0000000..1d4183c --- /dev/null +++ b/checkpoint-1872/trainer_state.json @@ -0,0 +1,4048 @@ +{ + "best_global_step": null, + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 1.0, + "eval_steps": 500, + "global_step": 1872, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.0005344378381989445, + "grad_norm": 63.52920150756836, + "learning_rate": 0.0, + "logits/chosen": -0.144775390625, + "logits/rejected": -0.1036376953125, + "logps/chosen": -110.828125, + "logps/rejected": -116.828125, + "logps/weighted_chosen": -0.2086181640625, + "logps/weighted_rejected": -1.271392822265625, + "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": 0.005344378381989445, + "grad_norm": 102.77810668945312, + "learning_rate": 4.787234042553191e-08, + "logits/chosen": -0.2549913227558136, + "logits/rejected": -0.24070316553115845, + "logps/chosen": -135.32899475097656, + "logps/rejected": -138.99305725097656, + "logps/weighted_chosen": -0.8547499179840088, + "logps/weighted_rejected": -0.9027608036994934, + "loss": 0.6928, + "rewards/accuracies": 0.2951388955116272, + "rewards/chosen": -0.0164930559694767, + "rewards/margins": -0.0349392369389534, + "rewards/rejected": 0.0184461809694767, + "rewards/weighted_accuracies": 0.4027777910232544, + "rewards/weighted_chosen": -0.002239651046693325, + "rewards/weighted_margins": 0.0028160943184047937, + "rewards/weighted_rejected": -0.0050794812850654125, + "step": 10 + }, + { + "epoch": 0.01068875676397889, + "grad_norm": 9.201544761657715, + "learning_rate": 1.0106382978723404e-07, + "logits/chosen": -0.19488906860351562, + "logits/rejected": -0.183399960398674, + "logps/chosen": -116.0374984741211, + "logps/rejected": -115.5054702758789, + "logps/weighted_chosen": -0.6929916143417358, + "logps/weighted_rejected": -0.6838119626045227, + "loss": 0.6917, + "rewards/accuracies": 0.38749998807907104, + "rewards/chosen": 0.01796874962747097, + "rewards/margins": 0.03007812425494194, + "rewards/rejected": -0.01210937462747097, + "rewards/weighted_accuracies": 0.4781250059604645, + "rewards/weighted_chosen": 0.0022293091751635075, + "rewards/weighted_margins": 0.00339679722674191, + "rewards/weighted_rejected": -0.0011629104847088456, + "step": 20 + }, + { + "epoch": 0.016033135145968335, + "grad_norm": 7.109442234039307, + "learning_rate": 1.5425531914893615e-07, + "logits/chosen": -0.1776607483625412, + "logits/rejected": -0.17884215712547302, + "logps/chosen": -123.2867202758789, + "logps/rejected": -129.75, + "logps/weighted_chosen": -0.7567901611328125, + "logps/weighted_rejected": -1.2207733392715454, + "loss": 0.6913, + "rewards/accuracies": 0.3968749940395355, + "rewards/chosen": -0.0074218749068677425, + "rewards/margins": 0.04843749850988388, + "rewards/rejected": -0.05585937574505806, + "rewards/weighted_accuracies": 0.5, + "rewards/weighted_chosen": 0.0007101058727130294, + "rewards/weighted_margins": 0.004384422209113836, + "rewards/weighted_rejected": -0.0036745071411132812, + "step": 30 + }, + { + "epoch": 0.02137751352795778, + "grad_norm": 6.0048956871032715, + "learning_rate": 2.074468085106383e-07, + "logits/chosen": -0.23842163383960724, + "logits/rejected": -0.2298881560564041, + "logps/chosen": -127.34062194824219, + "logps/rejected": -128.1710968017578, + "logps/weighted_chosen": -0.8051223754882812, + "logps/weighted_rejected": -0.5736129879951477, + "loss": 0.6934, + "rewards/accuracies": 0.34062498807907104, + "rewards/chosen": -0.01992187462747097, + "rewards/margins": -0.05351562425494194, + "rewards/rejected": 0.03359375149011612, + "rewards/weighted_accuracies": 0.4593749940395355, + "rewards/weighted_chosen": -7.83920258982107e-05, + "rewards/weighted_margins": -0.0009832382202148438, + "rewards/weighted_rejected": 0.000904083251953125, + "step": 40 + }, + { + "epoch": 0.026721891909947223, + "grad_norm": 32.7233772277832, + "learning_rate": 2.6063829787234044e-07, + "logits/chosen": -0.18835143744945526, + "logits/rejected": -0.170085147023201, + "logps/chosen": -120.88594055175781, + "logps/rejected": -121.3515625, + "logps/weighted_chosen": -0.6867736577987671, + "logps/weighted_rejected": -1.138249158859253, + "loss": 0.6903, + "rewards/accuracies": 0.3656249940395355, + "rewards/chosen": 0.03554687649011612, + "rewards/margins": -0.03164062649011612, + "rewards/rejected": 0.06718750298023224, + "rewards/weighted_accuracies": 0.515625, + "rewards/weighted_chosen": -0.006287765689194202, + "rewards/weighted_margins": 0.0064636231400072575, + "rewards/weighted_rejected": -0.01274185162037611, + "step": 50 + }, + { + "epoch": 0.03206627029193667, + "grad_norm": 55.560707092285156, + "learning_rate": 3.1382978723404253e-07, + "logits/chosen": -0.1308601349592209, + "logits/rejected": -0.1308906525373459, + "logps/chosen": -116.5875015258789, + "logps/rejected": -119.5406265258789, + "logps/weighted_chosen": -0.8019469976425171, + "logps/weighted_rejected": -0.8834930658340454, + "loss": 0.7002, + "rewards/accuracies": 0.3343749940395355, + "rewards/chosen": 0.03476562350988388, + "rewards/margins": -0.06171875074505806, + "rewards/rejected": 0.09648437798023224, + "rewards/weighted_accuracies": 0.4781250059604645, + "rewards/weighted_chosen": -0.01903533935546875, + "rewards/weighted_margins": -2.880096508306451e-05, + "rewards/weighted_rejected": -0.019007300958037376, + "step": 60 + }, + { + "epoch": 0.037410648673926114, + "grad_norm": 20.22484588623047, + "learning_rate": 3.6702127659574467e-07, + "logits/chosen": -0.20572929084300995, + "logits/rejected": -0.19457244873046875, + "logps/chosen": -114.4507827758789, + "logps/rejected": -118.63594055175781, + "logps/weighted_chosen": -0.6070801019668579, + "logps/weighted_rejected": -0.6768280267715454, + "loss": 0.6911, + "rewards/accuracies": 0.42500001192092896, + "rewards/chosen": 0.16484375298023224, + "rewards/margins": 0.07148437201976776, + "rewards/rejected": 0.09335937350988388, + "rewards/weighted_accuracies": 0.4749999940395355, + "rewards/weighted_chosen": -0.010281562805175781, + "rewards/weighted_margins": 0.007259177975356579, + "rewards/weighted_rejected": -0.017521286383271217, + "step": 70 + }, + { + "epoch": 0.04275502705591556, + "grad_norm": 15.305061340332031, + "learning_rate": 4.202127659574468e-07, + "logits/chosen": -0.19508513808250427, + "logits/rejected": -0.12093047797679901, + "logps/chosen": -112.57734680175781, + "logps/rejected": -112.77734375, + "logps/weighted_chosen": -0.4597320556640625, + "logps/weighted_rejected": -0.9711883664131165, + "loss": 0.6952, + "rewards/accuracies": 0.37812501192092896, + "rewards/chosen": 0.04570312425494194, + "rewards/margins": 0.01914062537252903, + "rewards/rejected": 0.02656250074505806, + "rewards/weighted_accuracies": 0.4625000059604645, + "rewards/weighted_chosen": -0.02759857103228569, + "rewards/weighted_margins": 0.021048355847597122, + "rewards/weighted_rejected": -0.04872245714068413, + "step": 80 + }, + { + "epoch": 0.048099405437905, + "grad_norm": 11.991273880004883, + "learning_rate": 4.734042553191489e-07, + "logits/chosen": -0.17391815781593323, + "logits/rejected": -0.11656417697668076, + "logps/chosen": -119.3070297241211, + "logps/rejected": -118.51249694824219, + "logps/weighted_chosen": -1.0416854619979858, + "logps/weighted_rejected": -1.5859466791152954, + "loss": 0.6736, + "rewards/accuracies": 0.4375, + "rewards/chosen": -0.08164062350988388, + "rewards/margins": 0.341796875, + "rewards/rejected": -0.4234375059604645, + "rewards/weighted_accuracies": 0.515625, + "rewards/weighted_chosen": -0.04084587097167969, + "rewards/weighted_margins": 0.10567589104175568, + "rewards/weighted_rejected": -0.14661255478858948, + "step": 90 + }, + { + "epoch": 0.053443783819894446, + "grad_norm": 17.195886611938477, + "learning_rate": 5.26595744680851e-07, + "logits/chosen": -0.204203799366951, + "logits/rejected": -0.14691162109375, + "logps/chosen": -125.05155944824219, + "logps/rejected": -125.3671875, + "logps/weighted_chosen": -0.7374237179756165, + "logps/weighted_rejected": -0.7348953485488892, + "loss": 0.7064, + "rewards/accuracies": 0.4281249940395355, + "rewards/chosen": -0.3832031190395355, + "rewards/margins": 0.15585938096046448, + "rewards/rejected": -0.5390625, + "rewards/weighted_accuracies": 0.46875, + "rewards/weighted_chosen": -0.060018159449100494, + "rewards/weighted_margins": 0.0024120330344885588, + "rewards/weighted_rejected": -0.06242714077234268, + "step": 100 + }, + { + "epoch": 0.058788162201883896, + "grad_norm": 9.103920936584473, + "learning_rate": 5.797872340425531e-07, + "logits/chosen": -0.2138931304216385, + "logits/rejected": -0.19011077284812927, + "logps/chosen": -118.6820297241211, + "logps/rejected": -121.28593444824219, + "logps/weighted_chosen": -0.4821228086948395, + "logps/weighted_rejected": -0.7759445309638977, + "loss": 0.6939, + "rewards/accuracies": 0.4281249940395355, + "rewards/chosen": -0.896484375, + "rewards/margins": 0.09531249850988388, + "rewards/rejected": -0.9917968511581421, + "rewards/weighted_accuracies": 0.528124988079071, + "rewards/weighted_chosen": -0.05413341522216797, + "rewards/weighted_margins": 0.04503212124109268, + "rewards/weighted_rejected": -0.09918193519115448, + "step": 110 + }, + { + "epoch": 0.06413254058387334, + "grad_norm": 8.539896011352539, + "learning_rate": 6.329787234042553e-07, + "logits/chosen": -0.17510680854320526, + "logits/rejected": -0.14726562798023224, + "logps/chosen": -122.58906555175781, + "logps/rejected": -128.4656219482422, + "logps/weighted_chosen": -0.9249496459960938, + "logps/weighted_rejected": -0.762683093547821, + "loss": 0.7325, + "rewards/accuracies": 0.4375, + "rewards/chosen": -1.068359375, + "rewards/margins": 0.28398436307907104, + "rewards/rejected": -1.3523437976837158, + "rewards/weighted_accuracies": 0.4781250059604645, + "rewards/weighted_chosen": -0.098027803003788, + "rewards/weighted_margins": -0.029563140124082565, + "rewards/weighted_rejected": -0.06836700439453125, + "step": 120 + }, + { + "epoch": 0.06947691896586278, + "grad_norm": 15.564044952392578, + "learning_rate": 6.861702127659574e-07, + "logits/chosen": -0.20233002305030823, + "logits/rejected": -0.18396148085594177, + "logps/chosen": -122.2671890258789, + "logps/rejected": -123.8382797241211, + "logps/weighted_chosen": -0.7501281499862671, + "logps/weighted_rejected": -0.821154773235321, + "loss": 0.7078, + "rewards/accuracies": 0.4000000059604645, + "rewards/chosen": -1.45703125, + "rewards/margins": 0.29179686307907104, + "rewards/rejected": -1.7488281726837158, + "rewards/weighted_accuracies": 0.5062500238418579, + "rewards/weighted_chosen": -0.007884788326919079, + "rewards/weighted_margins": 0.006314086727797985, + "rewards/weighted_rejected": -0.014200759120285511, + "step": 130 + }, + { + "epoch": 0.07482129734785223, + "grad_norm": 9.176558494567871, + "learning_rate": 7.393617021276596e-07, + "logits/chosen": -0.16455383598804474, + "logits/rejected": -0.14660033583641052, + "logps/chosen": -116.328125, + "logps/rejected": -119.9664077758789, + "logps/weighted_chosen": -0.9873703122138977, + "logps/weighted_rejected": -0.6730011105537415, + "loss": 0.7174, + "rewards/accuracies": 0.4749999940395355, + "rewards/chosen": -1.6007812023162842, + "rewards/margins": 0.533203125, + "rewards/rejected": -2.133984327316284, + "rewards/weighted_accuracies": 0.515625, + "rewards/weighted_chosen": 0.02084655687212944, + "rewards/weighted_margins": 0.05102996900677681, + "rewards/weighted_rejected": -0.030106354504823685, + "step": 140 + }, + { + "epoch": 0.08016567572984168, + "grad_norm": 10.172534942626953, + "learning_rate": 7.925531914893616e-07, + "logits/chosen": -0.1509498655796051, + "logits/rejected": -0.13156357407569885, + "logps/chosen": -120.0562515258789, + "logps/rejected": -117.85546875, + "logps/weighted_chosen": -0.8198806643486023, + "logps/weighted_rejected": -0.7782562375068665, + "loss": 0.7005, + "rewards/accuracies": 0.46562498807907104, + "rewards/chosen": -2.0234375, + "rewards/margins": 0.607421875, + "rewards/rejected": -2.630859375, + "rewards/weighted_accuracies": 0.543749988079071, + "rewards/weighted_chosen": -0.0781441181898117, + "rewards/weighted_margins": 0.034273337572813034, + "rewards/weighted_rejected": -0.11248989403247833, + "step": 150 + }, + { + "epoch": 0.08551005411183112, + "grad_norm": 25.879701614379883, + "learning_rate": 8.457446808510637e-07, + "logits/chosen": -0.17474937438964844, + "logits/rejected": -0.17009583115577698, + "logps/chosen": -121.2367172241211, + "logps/rejected": -121.62109375, + "logps/weighted_chosen": -0.8304061889648438, + "logps/weighted_rejected": -0.866925060749054, + "loss": 0.7072, + "rewards/accuracies": 0.4781250059604645, + "rewards/chosen": -2.219531297683716, + "rewards/margins": 0.716796875, + "rewards/rejected": -2.936328172683716, + "rewards/weighted_accuracies": 0.47187501192092896, + "rewards/weighted_chosen": -0.07932891696691513, + "rewards/weighted_margins": 0.017035294324159622, + "rewards/weighted_rejected": -0.09644661098718643, + "step": 160 + }, + { + "epoch": 0.09085443249382057, + "grad_norm": 9.220558166503906, + "learning_rate": 8.989361702127659e-07, + "logits/chosen": -0.22574539482593536, + "logits/rejected": -0.19882050156593323, + "logps/chosen": -120.62968444824219, + "logps/rejected": -124.78437805175781, + "logps/weighted_chosen": -0.9401000738143921, + "logps/weighted_rejected": -0.8424423336982727, + "loss": 0.7174, + "rewards/accuracies": 0.4937500059604645, + "rewards/chosen": -2.4033203125, + "rewards/margins": 0.613085925579071, + "rewards/rejected": -3.016406297683716, + "rewards/weighted_accuracies": 0.48750001192092896, + "rewards/weighted_chosen": -0.04837837070226669, + "rewards/weighted_margins": 0.04595222324132919, + "rewards/weighted_rejected": -0.09437103569507599, + "step": 170 + }, + { + "epoch": 0.09619881087581, + "grad_norm": 55.62498092651367, + "learning_rate": 9.52127659574468e-07, + "logits/chosen": -0.2216789275407791, + "logits/rejected": -0.20373383164405823, + "logps/chosen": -120.3851547241211, + "logps/rejected": -121.2328109741211, + "logps/weighted_chosen": -0.435232549905777, + "logps/weighted_rejected": -0.5373809933662415, + "loss": 0.7302, + "rewards/accuracies": 0.47187501192092896, + "rewards/chosen": -1.338281273841858, + "rewards/margins": 0.3658203184604645, + "rewards/rejected": -1.7041015625, + "rewards/weighted_accuracies": 0.53125, + "rewards/weighted_chosen": -0.08163909614086151, + "rewards/weighted_margins": -0.008893203921616077, + "rewards/weighted_rejected": -0.07281456142663956, + "step": 180 + }, + { + "epoch": 0.10154318925779945, + "grad_norm": 39.38788986206055, + "learning_rate": 9.99999129927477e-07, + "logits/chosen": -0.19243164360523224, + "logits/rejected": -0.15758666396141052, + "logps/chosen": -121.8960952758789, + "logps/rejected": -123.01094055175781, + "logps/weighted_chosen": -0.953173816204071, + "logps/weighted_rejected": -1.0853087902069092, + "loss": 0.6702, + "rewards/accuracies": 0.48750001192092896, + "rewards/chosen": -2.100390672683716, + "rewards/margins": 0.662890613079071, + "rewards/rejected": -2.7632813453674316, + "rewards/weighted_accuracies": 0.578125, + "rewards/weighted_chosen": -0.09999503940343857, + "rewards/weighted_margins": 0.12822990119457245, + "rewards/weighted_rejected": -0.22818031907081604, + "step": 190 + }, + { + "epoch": 0.10688756763978889, + "grad_norm": 12.883384704589844, + "learning_rate": 9.99894724888679e-07, + "logits/chosen": -0.17392578721046448, + "logits/rejected": -0.147491455078125, + "logps/chosen": -129.53750610351562, + "logps/rejected": -128.95703125, + "logps/weighted_chosen": -0.7322753667831421, + "logps/weighted_rejected": -1.0728164911270142, + "loss": 0.6777, + "rewards/accuracies": 0.4749999940395355, + "rewards/chosen": -3.215625047683716, + "rewards/margins": 0.9507812261581421, + "rewards/rejected": -4.166406154632568, + "rewards/weighted_accuracies": 0.5375000238418579, + "rewards/weighted_chosen": -0.1104557067155838, + "rewards/weighted_margins": 0.13128623366355896, + "rewards/weighted_rejected": -0.2415626496076584, + "step": 200 + }, + { + "epoch": 0.11223194602177834, + "grad_norm": 17.96660041809082, + "learning_rate": 9.996163469793475e-07, + "logits/chosen": -0.18083801865577698, + "logits/rejected": -0.12766647338867188, + "logps/chosen": -128.1953125, + "logps/rejected": -123.03047180175781, + "logps/weighted_chosen": -0.8699920773506165, + "logps/weighted_rejected": -0.8845611810684204, + "loss": 0.7572, + "rewards/accuracies": 0.5625, + "rewards/chosen": -3.5113282203674316, + "rewards/margins": 1.3683593273162842, + "rewards/rejected": -4.879296779632568, + "rewards/weighted_accuracies": 0.5531250238418579, + "rewards/weighted_chosen": -0.07919616997241974, + "rewards/weighted_margins": 0.04895629733800888, + "rewards/weighted_rejected": -0.12815475463867188, + "step": 210 + }, + { + "epoch": 0.11757632440376779, + "grad_norm": 63.790077209472656, + "learning_rate": 9.991640930802883e-07, + "logits/chosen": -0.14619751274585724, + "logits/rejected": -0.13860473036766052, + "logps/chosen": -126.61015319824219, + "logps/rejected": -129.9773406982422, + "logps/weighted_chosen": -1.0776779651641846, + "logps/weighted_rejected": -0.9309722781181335, + "loss": 0.7107, + "rewards/accuracies": 0.5531250238418579, + "rewards/chosen": -4.296484470367432, + "rewards/margins": 1.10546875, + "rewards/rejected": -5.40234375, + "rewards/weighted_accuracies": 0.5406249761581421, + "rewards/weighted_chosen": -0.013729858212172985, + "rewards/weighted_margins": 0.14467468857765198, + "rewards/weighted_rejected": -0.1582542359828949, + "step": 220 + }, + { + "epoch": 0.12292070278575723, + "grad_norm": 9.634794235229492, + "learning_rate": 9.98538120584459e-07, + "logits/chosen": -0.1792152374982834, + "logits/rejected": -0.1362098753452301, + "logps/chosen": -133.7140655517578, + "logps/rejected": -137.7546844482422, + "logps/weighted_chosen": -0.6676223874092102, + "logps/weighted_rejected": -0.9518234133720398, + "loss": 0.7614, + "rewards/accuracies": 0.515625, + "rewards/chosen": -4.493554592132568, + "rewards/margins": 1.0623047351837158, + "rewards/rejected": -5.555859565734863, + "rewards/weighted_accuracies": 0.565625011920929, + "rewards/weighted_chosen": -0.1120578795671463, + "rewards/weighted_margins": 0.031700897961854935, + "rewards/weighted_rejected": -0.14376182854175568, + "step": 230 + }, + { + "epoch": 0.12826508116774668, + "grad_norm": 8.717682838439941, + "learning_rate": 9.977386473421917e-07, + "logits/chosen": -0.09295501559972763, + "logits/rejected": -0.07836151123046875, + "logps/chosen": -121.78202819824219, + "logps/rejected": -126.39453125, + "logps/weighted_chosen": -0.8116515874862671, + "logps/weighted_rejected": -1.2165496349334717, + "loss": 0.6731, + "rewards/accuracies": 0.5874999761581421, + "rewards/chosen": -7.553515434265137, + "rewards/margins": 1.669921875, + "rewards/rejected": -9.224218368530273, + "rewards/weighted_accuracies": 0.534375011920929, + "rewards/weighted_chosen": -0.1755084991455078, + "rewards/weighted_margins": 0.15414047241210938, + "rewards/weighted_rejected": -0.32989272475242615, + "step": 240 + }, + { + "epoch": 0.13360945954973613, + "grad_norm": 35.197269439697266, + "learning_rate": 9.96765951585378e-07, + "logits/chosen": -0.021549224853515625, + "logits/rejected": 0.00345611572265625, + "logps/chosen": -123.0296859741211, + "logps/rejected": -127.9671859741211, + "logps/weighted_chosen": -1.186151146888733, + "logps/weighted_rejected": -1.386474609375, + "loss": 0.7026, + "rewards/accuracies": 0.5625, + "rewards/chosen": -9.133593559265137, + "rewards/margins": 1.150781273841858, + "rewards/rejected": -10.283984184265137, + "rewards/weighted_accuracies": 0.59375, + "rewards/weighted_chosen": -0.3282794952392578, + "rewards/weighted_margins": 0.1091407760977745, + "rewards/weighted_rejected": -0.4371795654296875, + "step": 250 + }, + { + "epoch": 0.13895383793172555, + "grad_norm": 84.2197494506836, + "learning_rate": 9.956203718306388e-07, + "logits/chosen": 0.025267411023378372, + "logits/rejected": 0.06335143744945526, + "logps/chosen": -128.8429718017578, + "logps/rejected": -133.50625610351562, + "logps/weighted_chosen": -0.9628280401229858, + "logps/weighted_rejected": -1.2008453607559204, + "loss": 0.7443, + "rewards/accuracies": 0.5, + "rewards/chosen": -10.861719131469727, + "rewards/margins": 0.871874988079071, + "rewards/rejected": -11.733593940734863, + "rewards/weighted_accuracies": 0.49687498807907104, + "rewards/weighted_chosen": -0.26259079575538635, + "rewards/weighted_margins": 0.06770706176757812, + "rewards/weighted_rejected": -0.3305191099643707, + "step": 260 + }, + { + "epoch": 0.144298216313715, + "grad_norm": 57.09628677368164, + "learning_rate": 9.943023067615136e-07, + "logits/chosen": -0.017973709851503372, + "logits/rejected": -0.0057998658157885075, + "logps/chosen": -137.60311889648438, + "logps/rejected": -140.91561889648438, + "logps/weighted_chosen": -0.8694183230400085, + "logps/weighted_rejected": -0.8910156488418579, + "loss": 0.7462, + "rewards/accuracies": 0.5625, + "rewards/chosen": -10.946093559265137, + "rewards/margins": 1.662109375, + "rewards/rejected": -12.608202934265137, + "rewards/weighted_accuracies": 0.565625011920929, + "rewards/weighted_chosen": -0.06702041625976562, + "rewards/weighted_margins": 0.13949890434741974, + "rewards/weighted_rejected": -0.20627327263355255, + "step": 270 + }, + { + "epoch": 0.14964259469570446, + "grad_norm": 45.0294303894043, + "learning_rate": 9.928122150897112e-07, + "logits/chosen": -0.045014191418886185, + "logits/rejected": 0.02437591552734375, + "logps/chosen": -130.1125030517578, + "logps/rejected": -131.6171875, + "logps/weighted_chosen": -0.7659690976142883, + "logps/weighted_rejected": -0.8499618768692017, + "loss": 0.7079, + "rewards/accuracies": 0.5406249761581421, + "rewards/chosen": -12.664453506469727, + "rewards/margins": 1.584375023841858, + "rewards/rejected": -14.249608993530273, + "rewards/weighted_accuracies": 0.543749988079071, + "rewards/weighted_chosen": -0.22265930473804474, + "rewards/weighted_margins": 0.07556991279125214, + "rewards/weighted_rejected": -0.298013299703598, + "step": 280 + }, + { + "epoch": 0.1549869730776939, + "grad_norm": 9.370684623718262, + "learning_rate": 9.9115061539547e-07, + "logits/chosen": -0.02069244347512722, + "logits/rejected": 0.01569976843893528, + "logps/chosen": -139.4453125, + "logps/rejected": -141.2578125, + "logps/weighted_chosen": -0.8775177001953125, + "logps/weighted_rejected": -1.2375701665878296, + "loss": 0.6949, + "rewards/accuracies": 0.5375000238418579, + "rewards/chosen": -13.07421875, + "rewards/margins": 1.177734375, + "rewards/rejected": -14.249608993530273, + "rewards/weighted_accuracies": 0.546875, + "rewards/weighted_chosen": -0.18484345078468323, + "rewards/weighted_margins": 0.12048187106847763, + "rewards/weighted_rejected": -0.30528029799461365, + "step": 290 + }, + { + "epoch": 0.16033135145968336, + "grad_norm": 30.68979263305664, + "learning_rate": 9.893180859470818e-07, + "logits/chosen": -0.0137786865234375, + "logits/rejected": 0.041573334485292435, + "logps/chosen": -130.22265625, + "logps/rejected": -132.06094360351562, + "logps/weighted_chosen": -1.1769134998321533, + "logps/weighted_rejected": -1.2438628673553467, + "loss": 0.7261, + "rewards/accuracies": 0.590624988079071, + "rewards/chosen": -13.093358993530273, + "rewards/margins": 2.108203172683716, + "rewards/rejected": -15.200780868530273, + "rewards/weighted_accuracies": 0.5843750238418579, + "rewards/weighted_chosen": -0.2848831117153168, + "rewards/weighted_margins": 0.07841148227453232, + "rewards/weighted_rejected": -0.36317139863967896, + "step": 300 + }, + { + "epoch": 0.16567572984167278, + "grad_norm": 71.56196594238281, + "learning_rate": 9.873152644996424e-07, + "logits/chosen": -0.11114044487476349, + "logits/rejected": -0.11599807441234589, + "logps/chosen": -131.01718139648438, + "logps/rejected": -132.28125, + "logps/weighted_chosen": -1.1032531261444092, + "logps/weighted_rejected": -0.8420807123184204, + "loss": 0.7872, + "rewards/accuracies": 0.5718749761581421, + "rewards/chosen": -12.979687690734863, + "rewards/margins": 2.407421827316284, + "rewards/rejected": -15.388280868530273, + "rewards/weighted_accuracies": 0.5218750238418579, + "rewards/weighted_chosen": 0.05532989650964737, + "rewards/weighted_margins": 0.2024696320295334, + "rewards/weighted_rejected": -0.1469375640153885, + "step": 310 + }, + { + "epoch": 0.17102010822366223, + "grad_norm": 12.055798530578613, + "learning_rate": 9.85142848073103e-07, + "logits/chosen": -0.10557708889245987, + "logits/rejected": -0.06234588474035263, + "logps/chosen": -133.39218139648438, + "logps/rejected": -134.2109375, + "logps/weighted_chosen": -0.700244128704071, + "logps/weighted_rejected": -1.4902985095977783, + "loss": 0.7558, + "rewards/accuracies": 0.534375011920929, + "rewards/chosen": -14.2265625, + "rewards/margins": 0.6207031011581421, + "rewards/rejected": -14.846094131469727, + "rewards/weighted_accuracies": 0.515625, + "rewards/weighted_chosen": -0.2170669585466385, + "rewards/weighted_margins": 0.03446922451257706, + "rewards/weighted_rejected": -0.25187987089157104, + "step": 320 + }, + { + "epoch": 0.17636448660565168, + "grad_norm": 41.381507873535156, + "learning_rate": 9.828015927096914e-07, + "logits/chosen": -0.085169218480587, + "logits/rejected": -0.05568351596593857, + "logps/chosen": -143.8234405517578, + "logps/rejected": -142.39608764648438, + "logps/weighted_chosen": -0.977459728717804, + "logps/weighted_rejected": -0.7958831787109375, + "loss": 0.6935, + "rewards/accuracies": 0.550000011920929, + "rewards/chosen": -14.643750190734863, + "rewards/margins": 1.578125, + "rewards/rejected": -16.222265243530273, + "rewards/weighted_accuracies": 0.581250011920929, + "rewards/weighted_chosen": -0.16206398606300354, + "rewards/weighted_margins": 0.07153511047363281, + "rewards/weighted_rejected": -0.23361511528491974, + "step": 330 + }, + { + "epoch": 0.18170886498764113, + "grad_norm": 7.313599586486816, + "learning_rate": 9.802923132107968e-07, + "logits/chosen": -0.10784301906824112, + "logits/rejected": -0.052645109593868256, + "logps/chosen": -143.91561889648438, + "logps/rejected": -147.7234344482422, + "logps/weighted_chosen": -0.7267853021621704, + "logps/weighted_rejected": -1.24786376953125, + "loss": 0.7102, + "rewards/accuracies": 0.5562499761581421, + "rewards/chosen": -16.782812118530273, + "rewards/margins": 1.23046875, + "rewards/rejected": -18.012500762939453, + "rewards/weighted_accuracies": 0.5687500238418579, + "rewards/weighted_chosen": -0.1999809294939041, + "rewards/weighted_margins": 0.056002043187618256, + "rewards/weighted_rejected": -0.2561977505683899, + "step": 340 + }, + { + "epoch": 0.18705324336963056, + "grad_norm": 66.15936279296875, + "learning_rate": 9.776158828534024e-07, + "logits/chosen": -0.12608642876148224, + "logits/rejected": -0.09711913764476776, + "logps/chosen": -143.48281860351562, + "logps/rejected": -188.25936889648438, + "logps/weighted_chosen": -1.260833740234375, + "logps/weighted_rejected": -0.9638397097587585, + "loss": 0.7338, + "rewards/accuracies": 0.612500011920929, + "rewards/chosen": -16.727733612060547, + "rewards/margins": 3.159374952316284, + "rewards/rejected": -19.883594512939453, + "rewards/weighted_accuracies": 0.559374988079071, + "rewards/weighted_chosen": -0.26622849702835083, + "rewards/weighted_margins": 0.05975799635052681, + "rewards/weighted_rejected": -0.3257743716239929, + "step": 350 + }, + { + "epoch": 0.19239762175162, + "grad_norm": 10.4098539352417, + "learning_rate": 9.747732330861695e-07, + "logits/chosen": -0.05247039720416069, + "logits/rejected": -0.024548912420868874, + "logps/chosen": -135.78750610351562, + "logps/rejected": -137.03671264648438, + "logps/weighted_chosen": -0.8440704345703125, + "logps/weighted_rejected": -0.8307411074638367, + "loss": 0.7415, + "rewards/accuracies": 0.578125, + "rewards/chosen": -16.864452362060547, + "rewards/margins": 2.5316405296325684, + "rewards/rejected": -19.396093368530273, + "rewards/weighted_accuracies": 0.5249999761581421, + "rewards/weighted_chosen": 0.005719375796616077, + "rewards/weighted_margins": 0.17883948981761932, + "rewards/weighted_rejected": -0.17323684692382812, + "step": 360 + }, + { + "epoch": 0.19774200013360946, + "grad_norm": 8.8024263381958, + "learning_rate": 9.717653532052742e-07, + "logits/chosen": -0.0410335548222065, + "logits/rejected": -0.049485016614198685, + "logps/chosen": -130.2781219482422, + "logps/rejected": -137.9343719482422, + "logps/weighted_chosen": -1.1705901622772217, + "logps/weighted_rejected": -0.7656036615371704, + "loss": 0.7402, + "rewards/accuracies": 0.6031249761581421, + "rewards/chosen": -17.155858993530273, + "rewards/margins": 2.863085985183716, + "rewards/rejected": -20.016016006469727, + "rewards/weighted_accuracies": 0.53125, + "rewards/weighted_chosen": -0.07308082282543182, + "rewards/weighted_margins": 0.04105224460363388, + "rewards/weighted_rejected": -0.114354707300663, + "step": 370 + }, + { + "epoch": 0.2030863785155989, + "grad_norm": 26.81167221069336, + "learning_rate": 9.685932900101146e-07, + "logits/chosen": -0.07604064792394638, + "logits/rejected": -0.06302948296070099, + "logps/chosen": -134.7859344482422, + "logps/rejected": -137.11093139648438, + "logps/weighted_chosen": -0.98638916015625, + "logps/weighted_rejected": -0.6243560910224915, + "loss": 0.7116, + "rewards/accuracies": 0.5625, + "rewards/chosen": -17.729297637939453, + "rewards/margins": 1.8025391101837158, + "rewards/rejected": -19.528514862060547, + "rewards/weighted_accuracies": 0.546875, + "rewards/weighted_chosen": -0.11945419013500214, + "rewards/weighted_margins": 0.07185058295726776, + "rewards/weighted_rejected": -0.1911766082048416, + "step": 380 + }, + { + "epoch": 0.20843075689758836, + "grad_norm": 40.89613342285156, + "learning_rate": 9.652581474390043e-07, + "logits/chosen": -0.09274902194738388, + "logits/rejected": -0.05452308803796768, + "logps/chosen": -139.3249969482422, + "logps/rejected": -141.53750610351562, + "logps/weighted_chosen": -0.9199386835098267, + "logps/weighted_rejected": -0.9217513799667358, + "loss": 0.7323, + "rewards/accuracies": 0.550000011920929, + "rewards/chosen": -19.379688262939453, + "rewards/margins": 2.321093797683716, + "rewards/rejected": -21.702733993530273, + "rewards/weighted_accuracies": 0.543749988079071, + "rewards/weighted_chosen": -0.05511016771197319, + "rewards/weighted_margins": 0.04057302325963974, + "rewards/weighted_rejected": -0.09564266353845596, + "step": 390 + }, + { + "epoch": 0.21377513527957778, + "grad_norm": 51.860347747802734, + "learning_rate": 9.61761086184981e-07, + "logits/chosen": -0.10268402099609375, + "logits/rejected": -0.05923309177160263, + "logps/chosen": -139.50936889648438, + "logps/rejected": -141.4015655517578, + "logps/weighted_chosen": -0.9449096918106079, + "logps/weighted_rejected": -1.177301049232483, + "loss": 0.7387, + "rewards/accuracies": 0.5625, + "rewards/chosen": -20.024608612060547, + "rewards/margins": 2.9205079078674316, + "rewards/rejected": -22.942577362060547, + "rewards/weighted_accuracies": 0.5625, + "rewards/weighted_chosen": -0.26060789823532104, + "rewards/weighted_margins": 0.017241667956113815, + "rewards/weighted_rejected": -0.2779647707939148, + "step": 400 + }, + { + "epoch": 0.21911951366156723, + "grad_norm": 7.539150238037109, + "learning_rate": 9.581033232918629e-07, + "logits/chosen": -0.015906523913145065, + "logits/rejected": 0.02000865899026394, + "logps/chosen": -138.39999389648438, + "logps/rejected": -141.68905639648438, + "logps/weighted_chosen": -1.0313812494277954, + "logps/weighted_rejected": -1.1196472644805908, + "loss": 0.7283, + "rewards/accuracies": 0.5874999761581421, + "rewards/chosen": -21.702342987060547, + "rewards/margins": 2.826953172683716, + "rewards/rejected": -24.53125, + "rewards/weighted_accuracies": 0.5249999761581421, + "rewards/weighted_chosen": -0.2783859372138977, + "rewards/weighted_margins": 0.06903991848230362, + "rewards/weighted_rejected": -0.3477470278739929, + "step": 410 + }, + { + "epoch": 0.22446389204355668, + "grad_norm": 6.638446807861328, + "learning_rate": 9.542861317306952e-07, + "logits/chosen": -0.05767669528722763, + "logits/rejected": -0.03759918361902237, + "logps/chosen": -140.64688110351562, + "logps/rejected": -141.9718780517578, + "logps/weighted_chosen": -1.5940430164337158, + "logps/weighted_rejected": -1.458154320716858, + "loss": 0.7371, + "rewards/accuracies": 0.5406249761581421, + "rewards/chosen": -22.33203125, + "rewards/margins": 1.1359374523162842, + "rewards/rejected": -23.46484375, + "rewards/weighted_accuracies": 0.515625, + "rewards/weighted_chosen": -0.299835205078125, + "rewards/weighted_margins": 0.032170869410037994, + "rewards/weighted_rejected": -0.3319748044013977, + "step": 420 + }, + { + "epoch": 0.22980827042554614, + "grad_norm": 24.621816635131836, + "learning_rate": 9.503108399567308e-07, + "logits/chosen": -0.15150222182273865, + "logits/rejected": -0.09339599311351776, + "logps/chosen": -149.8671875, + "logps/rejected": -153.6296844482422, + "logps/weighted_chosen": -0.7455719113349915, + "logps/weighted_rejected": -0.826556384563446, + "loss": 0.7125, + "rewards/accuracies": 0.550000011920929, + "rewards/chosen": -22.685546875, + "rewards/margins": 1.7355468273162842, + "rewards/rejected": -24.420312881469727, + "rewards/weighted_accuracies": 0.5406249761581421, + "rewards/weighted_chosen": -0.19403991103172302, + "rewards/weighted_margins": 0.02692871168255806, + "rewards/weighted_rejected": -0.2209148406982422, + "step": 430 + }, + { + "epoch": 0.23515264880753559, + "grad_norm": 10.726137161254883, + "learning_rate": 9.461788314471034e-07, + "logits/chosen": -0.08421440422534943, + "logits/rejected": -0.031351469457149506, + "logps/chosen": -143.1171875, + "logps/rejected": -146.890625, + "logps/weighted_chosen": -0.7524811029434204, + "logps/weighted_rejected": -0.8818389773368835, + "loss": 0.7062, + "rewards/accuracies": 0.550000011920929, + "rewards/chosen": -24.516796112060547, + "rewards/margins": 2.0640625953674316, + "rewards/rejected": -26.577733993530273, + "rewards/weighted_accuracies": 0.528124988079071, + "rewards/weighted_chosen": -0.22655639052391052, + "rewards/weighted_margins": 0.07337188720703125, + "rewards/weighted_rejected": -0.30005186796188354, + "step": 440 + }, + { + "epoch": 0.240497027189525, + "grad_norm": 8.586684226989746, + "learning_rate": 9.418915442193509e-07, + "logits/chosen": -0.03446502611041069, + "logits/rejected": 0.007968139834702015, + "logps/chosen": -144.7843780517578, + "logps/rejected": -155.96249389648438, + "logps/weighted_chosen": -1.1978485584259033, + "logps/weighted_rejected": -1.2447845935821533, + "loss": 0.6841, + "rewards/accuracies": 0.6312500238418579, + "rewards/chosen": -26.516016006469727, + "rewards/margins": 4.714453220367432, + "rewards/rejected": -31.236328125, + "rewards/weighted_accuracies": 0.5218750238418579, + "rewards/weighted_chosen": -0.25520628690719604, + "rewards/weighted_margins": 0.09325332939624786, + "rewards/weighted_rejected": -0.3483436703681946, + "step": 450 + }, + { + "epoch": 0.24584140557151446, + "grad_norm": 10.014894485473633, + "learning_rate": 9.374504703309579e-07, + "logits/chosen": -0.1379852294921875, + "logits/rejected": -0.0861412063241005, + "logps/chosen": -141.83828735351562, + "logps/rejected": -143.34530639648438, + "logps/weighted_chosen": -1.1420501470565796, + "logps/weighted_rejected": -1.055877685546875, + "loss": 0.6804, + "rewards/accuracies": 0.534375011920929, + "rewards/chosen": -25.53515625, + "rewards/margins": 2.038281202316284, + "rewards/rejected": -27.571874618530273, + "rewards/weighted_accuracies": 0.5218750238418579, + "rewards/weighted_chosen": -0.21983489394187927, + "rewards/weighted_margins": 0.17051848769187927, + "rewards/weighted_rejected": -0.3904357850551605, + "step": 460 + }, + { + "epoch": 0.2511857839535039, + "grad_norm": 10.024358749389648, + "learning_rate": 9.328571553600915e-07, + "logits/chosen": -0.1318313628435135, + "logits/rejected": -0.0828731507062912, + "logps/chosen": -144.39688110351562, + "logps/rejected": -147.41796875, + "logps/weighted_chosen": -0.9446808099746704, + "logps/weighted_rejected": -1.138859510421753, + "loss": 0.7137, + "rewards/accuracies": 0.5718749761581421, + "rewards/chosen": -25.697265625, + "rewards/margins": 1.9656250476837158, + "rewards/rejected": -27.665233612060547, + "rewards/weighted_accuracies": 0.4906249940395355, + "rewards/weighted_chosen": -0.2078697234392166, + "rewards/weighted_margins": 0.052256010472774506, + "rewards/weighted_rejected": -0.260000616312027, + "step": 470 + }, + { + "epoch": 0.25653016233549336, + "grad_norm": 13.618422508239746, + "learning_rate": 9.281131978677106e-07, + "logits/chosen": -0.16686096787452698, + "logits/rejected": -0.1246364563703537, + "logps/chosen": -149.2546844482422, + "logps/rejected": -152.47811889648438, + "logps/weighted_chosen": -0.686419665813446, + "logps/weighted_rejected": -1.0338714122772217, + "loss": 0.6805, + "rewards/accuracies": 0.59375, + "rewards/chosen": -25.559375762939453, + "rewards/margins": 3.1332030296325684, + "rewards/rejected": -28.689844131469727, + "rewards/weighted_accuracies": 0.534375011920929, + "rewards/weighted_chosen": -0.22538375854492188, + "rewards/weighted_margins": 0.13199158012866974, + "rewards/weighted_rejected": -0.35721856355667114, + "step": 480 + }, + { + "epoch": 0.2618745407174828, + "grad_norm": 30.852863311767578, + "learning_rate": 9.232202488412361e-07, + "logits/chosen": -0.10437698662281036, + "logits/rejected": -0.05600614473223686, + "logps/chosen": -143.5671844482422, + "logps/rejected": -149.11874389648438, + "logps/weighted_chosen": -1.236029028892517, + "logps/weighted_rejected": -1.6542327404022217, + "loss": 0.6937, + "rewards/accuracies": 0.5375000238418579, + "rewards/chosen": -28.171483993530273, + "rewards/margins": 2.311328172683716, + "rewards/rejected": -30.477733612060547, + "rewards/weighted_accuracies": 0.534375011920929, + "rewards/weighted_chosen": -0.27268296480178833, + "rewards/weighted_margins": 0.08454742282629013, + "rewards/weighted_rejected": -0.35735321044921875, + "step": 490 + }, + { + "epoch": 0.26721891909947226, + "grad_norm": 9.151970863342285, + "learning_rate": 9.181800111199766e-07, + "logits/chosen": -0.1230342835187912, + "logits/rejected": -0.0801507979631424, + "logps/chosen": -147.54843139648438, + "logps/rejected": -147.5437469482422, + "logps/weighted_chosen": -1.5909912586212158, + "logps/weighted_rejected": -1.286199927330017, + "loss": 0.7005, + "rewards/accuracies": 0.578125, + "rewards/chosen": -29.640233993530273, + "rewards/margins": 3.1507811546325684, + "rewards/rejected": -32.789451599121094, + "rewards/weighted_accuracies": 0.5874999761581421, + "rewards/weighted_chosen": -0.34203338623046875, + "rewards/weighted_margins": 0.12214355170726776, + "rewards/weighted_rejected": -0.4641265869140625, + "step": 500 + }, + { + "epoch": 0.26721891909947226, + "eval_logits/chosen": -0.29476672410964966, + "eval_logits/rejected": -0.26891082525253296, + "eval_logps/chosen": -149.5992889404297, + "eval_logps/rejected": -153.5249481201172, + "eval_logps/weighted_chosen": -0.8761399984359741, + "eval_logps/weighted_rejected": -1.120854377746582, + "eval_loss": 0.6999610066413879, + "eval_rewards/accuracies": 0.5784114003181458, + "eval_rewards/chosen": -28.267057418823242, + "eval_rewards/margins": 2.7842414379119873, + "eval_rewards/rejected": -31.05422592163086, + "eval_rewards/weighted_accuracies": 0.5264765620231628, + "eval_rewards/weighted_chosen": -0.23699289560317993, + "eval_rewards/weighted_margins": 0.08454981446266174, + "eval_rewards/weighted_rejected": -0.3215427100658417, + "eval_runtime": 1248.3443, + "eval_samples_per_second": 1.571, + "eval_steps_per_second": 0.393, + "step": 500 + }, + { + "epoch": 0.2725632974814617, + "grad_norm": 7.777375221252441, + "learning_rate": 9.129942388025066e-07, + "logits/chosen": -0.1614234894514084, + "logits/rejected": -0.12539061903953552, + "logps/chosen": -138.70468139648438, + "logps/rejected": -144.50936889648438, + "logps/weighted_chosen": -1.0583159923553467, + "logps/weighted_rejected": -1.0353851318359375, + "loss": 0.7783, + "rewards/accuracies": 0.543749988079071, + "rewards/chosen": -27.291797637939453, + "rewards/margins": 2.485156297683716, + "rewards/rejected": -29.778514862060547, + "rewards/weighted_accuracies": 0.5093749761581421, + "rewards/weighted_chosen": -0.21576309204101562, + "rewards/weighted_margins": 0.003093719482421875, + "rewards/weighted_rejected": -0.219013974070549, + "step": 510 + }, + { + "epoch": 0.2779076758634511, + "grad_norm": 42.880157470703125, + "learning_rate": 9.076647366362082e-07, + "logits/chosen": -0.0863800048828125, + "logits/rejected": -0.04528322070837021, + "logps/chosen": -148.21875, + "logps/rejected": -155.3046875, + "logps/weighted_chosen": -1.141107201576233, + "logps/weighted_rejected": -1.2995758056640625, + "loss": 0.7748, + "rewards/accuracies": 0.612500011920929, + "rewards/chosen": -33.478515625, + "rewards/margins": 4.057812690734863, + "rewards/rejected": -37.53515625, + "rewards/weighted_accuracies": 0.5531250238418579, + "rewards/weighted_chosen": -0.2628006041049957, + "rewards/weighted_margins": -0.00189971923828125, + "rewards/weighted_rejected": -0.26109999418258667, + "step": 520 + }, + { + "epoch": 0.2832520542454406, + "grad_norm": 24.728120803833008, + "learning_rate": 9.021933593891841e-07, + "logits/chosen": -0.020507430657744408, + "logits/rejected": 0.01460342388600111, + "logps/chosen": -154.15780639648438, + "logps/rejected": -156.328125, + "logps/weighted_chosen": -1.0978972911834717, + "logps/weighted_rejected": -0.9736785888671875, + "loss": 0.6998, + "rewards/accuracies": 0.578125, + "rewards/chosen": -40.849220275878906, + "rewards/margins": 2.678515672683716, + "rewards/rejected": -43.53046798706055, + "rewards/weighted_accuracies": 0.512499988079071, + "rewards/weighted_chosen": -0.2724204957485199, + "rewards/weighted_margins": 0.09940185397863388, + "rewards/weighted_rejected": -0.3718253970146179, + "step": 530 + }, + { + "epoch": 0.28859643262743, + "grad_norm": 11.090974807739258, + "learning_rate": 8.965820112047629e-07, + "logits/chosen": -0.07119598239660263, + "logits/rejected": -0.04578246921300888, + "logps/chosen": -163.8781280517578, + "logps/rejected": -172.9187469482422, + "logps/weighted_chosen": -0.9050750732421875, + "logps/weighted_rejected": -1.05401611328125, + "loss": 0.6621, + "rewards/accuracies": 0.559374988079071, + "rewards/chosen": -42.501564025878906, + "rewards/margins": 4.053515434265137, + "rewards/rejected": -46.54999923706055, + "rewards/weighted_accuracies": 0.6000000238418579, + "rewards/weighted_chosen": -0.1999252289533615, + "rewards/weighted_margins": 0.1526111662387848, + "rewards/weighted_rejected": -0.3521057069301605, + "step": 540 + }, + { + "epoch": 0.2939408110094195, + "grad_norm": 13.709502220153809, + "learning_rate": 8.908326449388213e-07, + "logits/chosen": -0.07627105712890625, + "logits/rejected": -0.03014526329934597, + "logps/chosen": -156.328125, + "logps/rejected": -161.06405639648438, + "logps/weighted_chosen": -1.228173851966858, + "logps/weighted_rejected": -1.3901275396347046, + "loss": 0.7368, + "rewards/accuracies": 0.6156250238418579, + "rewards/chosen": -41.07109451293945, + "rewards/margins": 4.337500095367432, + "rewards/rejected": -45.392967224121094, + "rewards/weighted_accuracies": 0.512499988079071, + "rewards/weighted_chosen": -0.172210693359375, + "rewards/weighted_margins": 0.07322921603918076, + "rewards/weighted_rejected": -0.2451316863298416, + "step": 550 + }, + { + "epoch": 0.2992851893914089, + "grad_norm": 10.929818153381348, + "learning_rate": 8.849472614801527e-07, + "logits/chosen": -0.07251434028148651, + "logits/rejected": -0.02443542517721653, + "logps/chosen": -161.86874389648438, + "logps/rejected": -165.0749969482422, + "logps/weighted_chosen": -0.752777099609375, + "logps/weighted_rejected": -1.0892517566680908, + "loss": 0.6642, + "rewards/accuracies": 0.606249988079071, + "rewards/chosen": -44.9609375, + "rewards/margins": 3.3421874046325684, + "rewards/rejected": -48.30546951293945, + "rewards/weighted_accuracies": 0.5874999761581421, + "rewards/weighted_chosen": -0.1927940398454666, + "rewards/weighted_margins": 0.19287796318531036, + "rewards/weighted_rejected": -0.38528138399124146, + "step": 560 + }, + { + "epoch": 0.30462956777339834, + "grad_norm": 11.730274200439453, + "learning_rate": 8.789279090541208e-07, + "logits/chosen": -0.05120544508099556, + "logits/rejected": 0.01478500384837389, + "logps/chosen": -161.8796844482422, + "logps/rejected": -172.51406860351562, + "logps/weighted_chosen": -0.8590911626815796, + "logps/weighted_rejected": -1.1839568614959717, + "loss": 0.6821, + "rewards/accuracies": 0.6187499761581421, + "rewards/chosen": -43.99609375, + "rewards/margins": 4.257031440734863, + "rewards/rejected": -48.24609375, + "rewards/weighted_accuracies": 0.515625, + "rewards/weighted_chosen": -0.29502803087234497, + "rewards/weighted_margins": 0.16868743300437927, + "rewards/weighted_rejected": -0.4639228880405426, + "step": 570 + }, + { + "epoch": 0.3099739461553878, + "grad_norm": 16.967859268188477, + "learning_rate": 8.72776682509837e-07, + "logits/chosen": 0.011944961734116077, + "logits/rejected": 0.08212585747241974, + "logps/chosen": -157.47500610351562, + "logps/rejected": -163.43594360351562, + "logps/weighted_chosen": -0.8815460205078125, + "logps/weighted_rejected": -1.220306396484375, + "loss": 0.6705, + "rewards/accuracies": 0.606249988079071, + "rewards/chosen": -43.369529724121094, + "rewards/margins": 4.614453315734863, + "rewards/rejected": -47.984375, + "rewards/weighted_accuracies": 0.565625011920929, + "rewards/weighted_chosen": -0.3169898986816406, + "rewards/weighted_margins": 0.19199447333812714, + "rewards/weighted_rejected": -0.5094040036201477, + "step": 580 + }, + { + "epoch": 0.31531832453737724, + "grad_norm": 9.289381980895996, + "learning_rate": 8.664957225911138e-07, + "logits/chosen": -0.04825134202837944, + "logits/rejected": -0.01284942589700222, + "logps/chosen": -163.6437530517578, + "logps/rejected": -184.9796905517578, + "logps/weighted_chosen": -0.775604248046875, + "logps/weighted_rejected": -1.1336090564727783, + "loss": 0.6889, + "rewards/accuracies": 0.596875011920929, + "rewards/chosen": -42.566795349121094, + "rewards/margins": 3.8046875, + "rewards/rejected": -46.36328125, + "rewards/weighted_accuracies": 0.565625011920929, + "rewards/weighted_chosen": -0.30945053696632385, + "rewards/weighted_margins": 0.116083525121212, + "rewards/weighted_rejected": -0.4253120422363281, + "step": 590 + }, + { + "epoch": 0.3206627029193667, + "grad_norm": 7.966893672943115, + "learning_rate": 8.600872151914451e-07, + "logits/chosen": 0.06866455078125, + "logits/rejected": 0.09470061957836151, + "logps/chosen": -158.2546844482422, + "logps/rejected": -167.48281860351562, + "logps/weighted_chosen": -1.3719971179962158, + "logps/weighted_rejected": -1.7499603033065796, + "loss": 0.7417, + "rewards/accuracies": 0.5843750238418579, + "rewards/chosen": -44.11249923706055, + "rewards/margins": 3.3988280296325684, + "rewards/rejected": -47.50468826293945, + "rewards/weighted_accuracies": 0.5375000238418579, + "rewards/weighted_chosen": -0.21634292602539062, + "rewards/weighted_margins": 0.19443893432617188, + "rewards/weighted_rejected": -0.41086500883102417, + "step": 600 + }, + { + "epoch": 0.32600708130135614, + "grad_norm": 49.298519134521484, + "learning_rate": 8.535533905932737e-07, + "logits/chosen": 0.06249542161822319, + "logits/rejected": 0.12158355861902237, + "logps/chosen": -173.4250030517578, + "logps/rejected": -183.015625, + "logps/weighted_chosen": -0.9283081293106079, + "logps/weighted_rejected": -1.2264496088027954, + "loss": 0.6778, + "rewards/accuracies": 0.590624988079071, + "rewards/chosen": -48.342185974121094, + "rewards/margins": 5.698437690734863, + "rewards/rejected": -54.049217224121094, + "rewards/weighted_accuracies": 0.578125, + "rewards/weighted_chosen": -0.3028602600097656, + "rewards/weighted_margins": 0.1688835173845291, + "rewards/weighted_rejected": -0.4717727601528168, + "step": 610 + }, + { + "epoch": 0.33135145968334556, + "grad_norm": 19.377832412719727, + "learning_rate": 8.468965226918105e-07, + "logits/chosen": 0.037828825414180756, + "logits/rejected": 0.0580875389277935, + "logps/chosen": -170.1281280517578, + "logps/rejected": -178.9093780517578, + "logps/weighted_chosen": -1.4841537475585938, + "logps/weighted_rejected": -1.118981957435608, + "loss": 0.7542, + "rewards/accuracies": 0.5874999761581421, + "rewards/chosen": -46.46210861206055, + "rewards/margins": 3.794921875, + "rewards/rejected": -50.26640701293945, + "rewards/weighted_accuracies": 0.596875011920929, + "rewards/weighted_chosen": -0.40849533677101135, + "rewards/weighted_margins": 0.0321044921875, + "rewards/weighted_rejected": -0.4405921995639801, + "step": 620 + }, + { + "epoch": 0.33669583806533504, + "grad_norm": 10.768741607666016, + "learning_rate": 8.40118928203676e-07, + "logits/chosen": 0.05730438232421875, + "logits/rejected": 0.07560119777917862, + "logps/chosen": -152.3718719482422, + "logps/rejected": -164.1218719482422, + "logps/weighted_chosen": -1.4566253423690796, + "logps/weighted_rejected": -1.3677704334259033, + "loss": 0.7025, + "rewards/accuracies": 0.643750011920929, + "rewards/chosen": -43.721092224121094, + "rewards/margins": 4.366015434265137, + "rewards/rejected": -48.07500076293945, + "rewards/weighted_accuracies": 0.590624988079071, + "rewards/weighted_chosen": -0.30116957426071167, + "rewards/weighted_margins": 0.09331168979406357, + "rewards/weighted_rejected": -0.39441680908203125, + "step": 630 + }, + { + "epoch": 0.34204021644732446, + "grad_norm": 8.60303020477295, + "learning_rate": 8.332229658606382e-07, + "logits/chosen": -0.055472563952207565, + "logits/rejected": -0.03988037258386612, + "logps/chosen": -175.6671905517578, + "logps/rejected": -180.06875610351562, + "logps/weighted_chosen": -0.8608856201171875, + "logps/weighted_rejected": -0.986541748046875, + "loss": 0.7297, + "rewards/accuracies": 0.596875011920929, + "rewards/chosen": -44.84062576293945, + "rewards/margins": 4.931250095367432, + "rewards/rejected": -49.77031326293945, + "rewards/weighted_accuracies": 0.515625, + "rewards/weighted_chosen": -0.23048630356788635, + "rewards/weighted_margins": 0.10724487155675888, + "rewards/weighted_rejected": -0.33786240220069885, + "step": 640 + }, + { + "epoch": 0.34738459482931394, + "grad_norm": 26.148893356323242, + "learning_rate": 8.262110355887302e-07, + "logits/chosen": -0.055730439722537994, + "logits/rejected": -0.00980529747903347, + "logps/chosen": -166.38125610351562, + "logps/rejected": -169.40469360351562, + "logps/weighted_chosen": -0.7286010980606079, + "logps/weighted_rejected": -0.8990936279296875, + "loss": 0.7103, + "rewards/accuracies": 0.609375, + "rewards/chosen": -46.994529724121094, + "rewards/margins": 3.8031249046325684, + "rewards/rejected": -50.78515625, + "rewards/weighted_accuracies": 0.518750011920929, + "rewards/weighted_chosen": -0.2977249026298523, + "rewards/weighted_margins": 0.08674392849206924, + "rewards/weighted_rejected": -0.3846077024936676, + "step": 650 + }, + { + "epoch": 0.35272897321130336, + "grad_norm": 11.823346138000488, + "learning_rate": 8.190855776730293e-07, + "logits/chosen": -0.07448883354663849, + "logits/rejected": -0.014659118838608265, + "logps/chosen": -171.80313110351562, + "logps/rejected": -182.2687530517578, + "logps/weighted_chosen": -0.9873107671737671, + "logps/weighted_rejected": -1.4951263666152954, + "loss": 0.7347, + "rewards/accuracies": 0.596875011920929, + "rewards/chosen": -50.564064025878906, + "rewards/margins": 4.532812595367432, + "rewards/rejected": -55.109375, + "rewards/weighted_accuracies": 0.5531250238418579, + "rewards/weighted_chosen": -0.28445130586624146, + "rewards/weighted_margins": 0.06389693915843964, + "rewards/weighted_rejected": -0.34831467270851135, + "step": 660 + }, + { + "epoch": 0.3580733515932928, + "grad_norm": 13.33145523071289, + "learning_rate": 8.118490719083917e-07, + "logits/chosen": -0.10270843654870987, + "logits/rejected": -0.062209319323301315, + "logps/chosen": -191.3000030517578, + "logps/rejected": -193.4109344482422, + "logps/weighted_chosen": -1.0586059093475342, + "logps/weighted_rejected": -1.3695876598358154, + "loss": 0.6575, + "rewards/accuracies": 0.546875, + "rewards/chosen": -58.06640625, + "rewards/margins": 4.108984470367432, + "rewards/rejected": -62.18359375, + "rewards/weighted_accuracies": 0.565625011920929, + "rewards/weighted_chosen": -0.2945011258125305, + "rewards/weighted_margins": 0.18829631805419922, + "rewards/weighted_rejected": -0.4826883375644684, + "step": 670 + }, + { + "epoch": 0.36341772997528227, + "grad_norm": 23.127918243408203, + "learning_rate": 8.045040367364368e-07, + "logits/chosen": 0.03829193115234375, + "logits/rejected": 0.055706024169921875, + "logps/chosen": -166.234375, + "logps/rejected": -169.96249389648438, + "logps/weighted_chosen": -1.025201439857483, + "logps/weighted_rejected": -1.302404761314392, + "loss": 0.7166, + "rewards/accuracies": 0.5843750238418579, + "rewards/chosen": -56.759376525878906, + "rewards/margins": 4.711718559265137, + "rewards/rejected": -61.46796798706055, + "rewards/weighted_accuracies": 0.590624988079071, + "rewards/weighted_chosen": -0.4052978456020355, + "rewards/weighted_margins": 0.09704723209142685, + "rewards/weighted_rejected": -0.5025444030761719, + "step": 680 + }, + { + "epoch": 0.3687621083572717, + "grad_norm": 7.137462615966797, + "learning_rate": 7.970530283690819e-07, + "logits/chosen": 0.04682769626379013, + "logits/rejected": 0.066925048828125, + "logps/chosen": -172.50155639648438, + "logps/rejected": -180.54061889648438, + "logps/weighted_chosen": -1.301055908203125, + "logps/weighted_rejected": -1.128961205482483, + "loss": 0.6864, + "rewards/accuracies": 0.6312500238418579, + "rewards/chosen": -58.544532775878906, + "rewards/margins": 5.397265434265137, + "rewards/rejected": -63.9453125, + "rewards/weighted_accuracies": 0.59375, + "rewards/weighted_chosen": -0.3540634214878082, + "rewards/weighted_margins": 0.1536247283220291, + "rewards/weighted_rejected": -0.5076873898506165, + "step": 690 + }, + { + "epoch": 0.3741064867392611, + "grad_norm": 7.699697971343994, + "learning_rate": 7.89498639898931e-07, + "logits/chosen": 0.05636443942785263, + "logits/rejected": 0.09223861992359161, + "logps/chosen": -169.8562469482422, + "logps/rejected": -173.65625, + "logps/weighted_chosen": -0.9024292230606079, + "logps/weighted_rejected": -1.1884033679962158, + "loss": 0.6642, + "rewards/accuracies": 0.6031249761581421, + "rewards/chosen": -59.00078201293945, + "rewards/margins": 4.887109279632568, + "rewards/rejected": -63.889060974121094, + "rewards/weighted_accuracies": 0.5562499761581421, + "rewards/weighted_chosen": -0.3017433285713196, + "rewards/weighted_margins": 0.20005035400390625, + "rewards/weighted_rejected": -0.5016593933105469, + "step": 700 + }, + { + "epoch": 0.3794508651212506, + "grad_norm": 18.41610336303711, + "learning_rate": 7.818435003968305e-07, + "logits/chosen": -0.03687286376953125, + "logits/rejected": -0.00025310515775345266, + "logps/chosen": -181.0812530517578, + "logps/rejected": -189.1828155517578, + "logps/weighted_chosen": -1.0574157238006592, + "logps/weighted_rejected": -1.2661011219024658, + "loss": 0.6918, + "rewards/accuracies": 0.590624988079071, + "rewards/chosen": -58.548439025878906, + "rewards/margins": 5.687890529632568, + "rewards/rejected": -64.234375, + "rewards/weighted_accuracies": 0.550000011920929, + "rewards/weighted_chosen": -0.33240431547164917, + "rewards/weighted_margins": 0.08642272651195526, + "rewards/weighted_rejected": -0.4188430905342102, + "step": 710 + }, + { + "epoch": 0.38479524350324, + "grad_norm": 26.664974212646484, + "learning_rate": 7.740902739969008e-07, + "logits/chosen": -0.08480529487133026, + "logits/rejected": -0.07062835991382599, + "logps/chosen": -171.16561889648438, + "logps/rejected": -175.68905639648438, + "logps/weighted_chosen": -0.828997790813446, + "logps/weighted_rejected": -0.9821502566337585, + "loss": 0.6819, + "rewards/accuracies": 0.6187499761581421, + "rewards/chosen": -49.982810974121094, + "rewards/margins": 4.688672065734863, + "rewards/rejected": -54.672264099121094, + "rewards/weighted_accuracies": 0.5531250238418579, + "rewards/weighted_chosen": -0.18975982069969177, + "rewards/weighted_margins": 0.1927967071533203, + "rewards/weighted_rejected": -0.38214111328125, + "step": 720 + }, + { + "epoch": 0.3901396218852295, + "grad_norm": 8.73801040649414, + "learning_rate": 7.662416589693695e-07, + "logits/chosen": -0.00596694927662611, + "logits/rejected": 0.019218826666474342, + "logps/chosen": -161.8249969482422, + "logps/rejected": -166.8312530517578, + "logps/weighted_chosen": -1.3982207775115967, + "logps/weighted_rejected": -1.7395598888397217, + "loss": 0.7772, + "rewards/accuracies": 0.6625000238418579, + "rewards/chosen": -46.8515625, + "rewards/margins": 5.096484184265137, + "rewards/rejected": -51.95000076293945, + "rewards/weighted_accuracies": 0.590624988079071, + "rewards/weighted_chosen": -0.23512879014015198, + "rewards/weighted_margins": 0.10187530517578125, + "rewards/weighted_rejected": -0.3369201719760895, + "step": 730 + }, + { + "epoch": 0.3954840002672189, + "grad_norm": 19.624563217163086, + "learning_rate": 7.583003867815192e-07, + "logits/chosen": -0.0639801025390625, + "logits/rejected": -0.05755653232336044, + "logps/chosen": -189.7468719482422, + "logps/rejected": -196.7062530517578, + "logps/weighted_chosen": -1.1809478998184204, + "logps/weighted_rejected": -1.3041900396347046, + "loss": 0.698, + "rewards/accuracies": 0.578125, + "rewards/chosen": -58.209373474121094, + "rewards/margins": 4.971875190734863, + "rewards/rejected": -63.1640625, + "rewards/weighted_accuracies": 0.518750011920929, + "rewards/weighted_chosen": -0.39410895109176636, + "rewards/weighted_margins": 0.08784560859203339, + "rewards/weighted_rejected": -0.4818374514579773, + "step": 740 + }, + { + "epoch": 0.40082837864920834, + "grad_norm": 19.04839324951172, + "learning_rate": 7.502692211470869e-07, + "logits/chosen": -0.05941009521484375, + "logits/rejected": -0.02816619910299778, + "logps/chosen": -182.8984375, + "logps/rejected": -190.1203155517578, + "logps/weighted_chosen": -0.853833019733429, + "logps/weighted_rejected": -1.1572723388671875, + "loss": 0.7079, + "rewards/accuracies": 0.574999988079071, + "rewards/chosen": -60.310157775878906, + "rewards/margins": 6.424609184265137, + "rewards/rejected": -66.73905944824219, + "rewards/weighted_accuracies": 0.546875, + "rewards/weighted_chosen": -0.4228218197822571, + "rewards/weighted_margins": 0.10426559299230576, + "rewards/weighted_rejected": -0.5270500183105469, + "step": 750 + }, + { + "epoch": 0.4061727570311978, + "grad_norm": 12.947752952575684, + "learning_rate": 7.421509570644387e-07, + "logits/chosen": -0.05351562425494194, + "logits/rejected": -0.0015235900646075606, + "logps/chosen": -180.66250610351562, + "logps/rejected": -185.6828155517578, + "logps/weighted_chosen": -1.2698882818222046, + "logps/weighted_rejected": -1.5226471424102783, + "loss": 0.6831, + "rewards/accuracies": 0.6343749761581421, + "rewards/chosen": -62.2109375, + "rewards/margins": 5.124218940734863, + "rewards/rejected": -67.33280944824219, + "rewards/weighted_accuracies": 0.59375, + "rewards/weighted_chosen": -0.32732313871383667, + "rewards/weighted_margins": 0.18027648329734802, + "rewards/weighted_rejected": -0.5074516534805298, + "step": 760 + }, + { + "epoch": 0.41151713541318724, + "grad_norm": 10.666881561279297, + "learning_rate": 7.339484198438566e-07, + "logits/chosen": 0.00444793701171875, + "logits/rejected": 0.03400421142578125, + "logps/chosen": -165.93124389648438, + "logps/rejected": -172.2937469482422, + "logps/weighted_chosen": -1.7782714366912842, + "logps/weighted_rejected": -1.320520043373108, + "loss": 0.6969, + "rewards/accuracies": 0.6000000238418579, + "rewards/chosen": -57.14374923706055, + "rewards/margins": 7.307421684265137, + "rewards/rejected": -64.4468765258789, + "rewards/weighted_accuracies": 0.5625, + "rewards/weighted_chosen": -0.3784065246582031, + "rewards/weighted_margins": 0.17845840752124786, + "rewards/weighted_rejected": -0.557085394859314, + "step": 770 + }, + { + "epoch": 0.4168615137951767, + "grad_norm": 11.188801765441895, + "learning_rate": 7.256644641242773e-07, + "logits/chosen": -0.03662109375, + "logits/rejected": 0.02340393140912056, + "logps/chosen": -170.4765625, + "logps/rejected": -175.390625, + "logps/weighted_chosen": -1.057519555091858, + "logps/weighted_rejected": -1.3154876232147217, + "loss": 0.7128, + "rewards/accuracies": 0.5562499761581421, + "rewards/chosen": -56.12578201293945, + "rewards/margins": 3.3226561546325684, + "rewards/rejected": -59.43828201293945, + "rewards/weighted_accuracies": 0.546875, + "rewards/weighted_chosen": -0.37098997831344604, + "rewards/weighted_margins": 0.07502517849206924, + "rewards/weighted_rejected": -0.4461258053779602, + "step": 780 + }, + { + "epoch": 0.42220589217716614, + "grad_norm": 9.07581615447998, + "learning_rate": 7.173019728798234e-07, + "logits/chosen": -0.03949751704931259, + "logits/rejected": -0.02464141882956028, + "logps/chosen": -169.55313110351562, + "logps/rejected": -177.18124389648438, + "logps/weighted_chosen": -0.8541290163993835, + "logps/weighted_rejected": -1.1333099603652954, + "loss": 0.6788, + "rewards/accuracies": 0.6625000238418579, + "rewards/chosen": -55.47734451293945, + "rewards/margins": 6.255859375, + "rewards/rejected": -61.71406173706055, + "rewards/weighted_accuracies": 0.574999988079071, + "rewards/weighted_chosen": -0.307028204202652, + "rewards/weighted_margins": 0.12036208808422089, + "rewards/weighted_rejected": -0.4274864196777344, + "step": 790 + }, + { + "epoch": 0.42755027055915557, + "grad_norm": 19.591575622558594, + "learning_rate": 7.088638564164738e-07, + "logits/chosen": -0.08669586479663849, + "logits/rejected": -0.05663032457232475, + "logps/chosen": -185.1593780517578, + "logps/rejected": -195.95938110351562, + "logps/weighted_chosen": -1.067468285560608, + "logps/weighted_rejected": -1.238824486732483, + "loss": 0.693, + "rewards/accuracies": 0.628125011920929, + "rewards/chosen": -63.07421875, + "rewards/margins": 6.429296970367432, + "rewards/rejected": -69.49296569824219, + "rewards/weighted_accuracies": 0.581250011920929, + "rewards/weighted_chosen": -0.3178604245185852, + "rewards/weighted_margins": 0.1114906296133995, + "rewards/weighted_rejected": -0.4294891357421875, + "step": 800 + }, + { + "epoch": 0.43289464894114504, + "grad_norm": 13.074841499328613, + "learning_rate": 7.003530513592215e-07, + "logits/chosen": -0.05083770677447319, + "logits/rejected": -0.0205841064453125, + "logps/chosen": -177.95156860351562, + "logps/rejected": -189.63436889648438, + "logps/weighted_chosen": -1.188201904296875, + "logps/weighted_rejected": -1.1756408214569092, + "loss": 0.6926, + "rewards/accuracies": 0.6187499761581421, + "rewards/chosen": -62.192970275878906, + "rewards/margins": 7.530859470367432, + "rewards/rejected": -69.73905944824219, + "rewards/weighted_accuracies": 0.543749988079071, + "rewards/weighted_chosen": -0.3717803955078125, + "rewards/weighted_margins": 0.12430572509765625, + "rewards/weighted_rejected": -0.49610137939453125, + "step": 810 + }, + { + "epoch": 0.43823902732313447, + "grad_norm": 51.0267448425293, + "learning_rate": 6.917725196300726e-07, + "logits/chosen": -0.1438644379377365, + "logits/rejected": -0.15563049912452698, + "logps/chosen": -191.125, + "logps/rejected": -203.52188110351562, + "logps/weighted_chosen": -1.097863793373108, + "logps/weighted_rejected": -0.971697986125946, + "loss": 0.6974, + "rewards/accuracies": 0.596875011920929, + "rewards/chosen": -63.764060974121094, + "rewards/margins": 7.415234565734863, + "rewards/rejected": -71.171875, + "rewards/weighted_accuracies": 0.5531250238418579, + "rewards/weighted_chosen": -0.3276786804199219, + "rewards/weighted_margins": 0.08215713500976562, + "rewards/weighted_rejected": -0.4100082516670227, + "step": 820 + }, + { + "epoch": 0.44358340570512395, + "grad_norm": 20.933467864990234, + "learning_rate": 6.831252474172411e-07, + "logits/chosen": -0.05628051608800888, + "logits/rejected": -0.03361053392291069, + "logps/chosen": -176.0031280517578, + "logps/rejected": -186.4343719482422, + "logps/weighted_chosen": -1.6239410638809204, + "logps/weighted_rejected": -1.3564453125, + "loss": 0.6817, + "rewards/accuracies": 0.609375, + "rewards/chosen": -60.45234298706055, + "rewards/margins": 6.214453220367432, + "rewards/rejected": -66.66015625, + "rewards/weighted_accuracies": 0.574999988079071, + "rewards/weighted_chosen": -0.2794052064418793, + "rewards/weighted_margins": 0.22185592353343964, + "rewards/weighted_rejected": -0.500841498374939, + "step": 830 + }, + { + "epoch": 0.44892778408711337, + "grad_norm": 11.393433570861816, + "learning_rate": 6.74414244135898e-07, + "logits/chosen": -0.15757599472999573, + "logits/rejected": -0.116943359375, + "logps/chosen": -177.0984344482422, + "logps/rejected": -185.13436889648438, + "logps/weighted_chosen": -0.666271984577179, + "logps/weighted_rejected": -0.7827666997909546, + "loss": 0.6919, + "rewards/accuracies": 0.625, + "rewards/chosen": -60.349998474121094, + "rewards/margins": 6.200390815734863, + "rewards/rejected": -66.5289077758789, + "rewards/weighted_accuracies": 0.581250011920929, + "rewards/weighted_chosen": -0.33733367919921875, + "rewards/weighted_margins": 0.08320007473230362, + "rewards/weighted_rejected": -0.42046278715133667, + "step": 840 + }, + { + "epoch": 0.4542721624691028, + "grad_norm": 9.733312606811523, + "learning_rate": 6.656425413808388e-07, + "logits/chosen": -0.12353897094726562, + "logits/rejected": -0.08713607490062714, + "logps/chosen": -179.66561889648438, + "logps/rejected": -183.109375, + "logps/weighted_chosen": -1.0826873779296875, + "logps/weighted_rejected": -1.106787085533142, + "loss": 0.6875, + "rewards/accuracies": 0.5562499761581421, + "rewards/chosen": -60.806251525878906, + "rewards/margins": 2.733203172683716, + "rewards/rejected": -63.51250076293945, + "rewards/weighted_accuracies": 0.5843750238418579, + "rewards/weighted_chosen": -0.22345809638500214, + "rewards/weighted_margins": 0.1886741667985916, + "rewards/weighted_rejected": -0.4119575619697571, + "step": 850 + }, + { + "epoch": 0.45961654085109227, + "grad_norm": 9.04237174987793, + "learning_rate": 6.568131918714294e-07, + "logits/chosen": -0.09643097221851349, + "logits/rejected": -0.08455581963062286, + "logps/chosen": -178.4453125, + "logps/rejected": -187.99063110351562, + "logps/weighted_chosen": -0.840087890625, + "logps/weighted_rejected": -1.158935546875, + "loss": 0.7138, + "rewards/accuracies": 0.596875011920929, + "rewards/chosen": -61.162498474121094, + "rewards/margins": 5.386328220367432, + "rewards/rejected": -66.54765319824219, + "rewards/weighted_accuracies": 0.550000011920929, + "rewards/weighted_chosen": -0.2982444763183594, + "rewards/weighted_margins": 0.08280716091394424, + "rewards/weighted_rejected": -0.38090819120407104, + "step": 860 + }, + { + "epoch": 0.4649609192330817, + "grad_norm": 7.623125076293945, + "learning_rate": 6.47929268389203e-07, + "logits/chosen": -0.06428680568933487, + "logits/rejected": -0.04297828674316406, + "logps/chosen": -190.86093139648438, + "logps/rejected": -205.9656219482422, + "logps/weighted_chosen": -0.9338012933731079, + "logps/weighted_rejected": -0.741748034954071, + "loss": 0.6808, + "rewards/accuracies": 0.6000000238418579, + "rewards/chosen": -65.76640319824219, + "rewards/margins": 7.994921684265137, + "rewards/rejected": -73.75703430175781, + "rewards/weighted_accuracies": 0.5, + "rewards/weighted_chosen": -0.28187790513038635, + "rewards/weighted_margins": 0.11656570434570312, + "rewards/weighted_rejected": -0.39813536405563354, + "step": 870 + }, + { + "epoch": 0.47030529761507117, + "grad_norm": 36.31256103515625, + "learning_rate": 6.389938627084732e-07, + "logits/chosen": -0.04478149488568306, + "logits/rejected": -0.054534912109375, + "logps/chosen": -181.5046844482422, + "logps/rejected": -192.84375, + "logps/weighted_chosen": -1.712792992591858, + "logps/weighted_rejected": -1.4381287097930908, + "loss": 0.7264, + "rewards/accuracies": 0.581250011920929, + "rewards/chosen": -64.04609680175781, + "rewards/margins": 6.507421970367432, + "rewards/rejected": -70.56953430175781, + "rewards/weighted_accuracies": 0.515625, + "rewards/weighted_chosen": -0.3980148434638977, + "rewards/weighted_margins": 0.10754700005054474, + "rewards/weighted_rejected": -0.5054702758789062, + "step": 880 + }, + { + "epoch": 0.4756496759970606, + "grad_norm": 6.0423126220703125, + "learning_rate": 6.300100845203373e-07, + "logits/chosen": -0.07687301933765411, + "logits/rejected": -0.04276275634765625, + "logps/chosen": -194.1609344482422, + "logps/rejected": -200.828125, + "logps/weighted_chosen": -0.6031554937362671, + "logps/weighted_rejected": -0.86151123046875, + "loss": 0.6793, + "rewards/accuracies": 0.59375, + "rewards/chosen": -63.5546875, + "rewards/margins": 5.604296684265137, + "rewards/rejected": -69.13984680175781, + "rewards/weighted_accuracies": 0.559374988079071, + "rewards/weighted_chosen": -0.27074965834617615, + "rewards/weighted_margins": 0.11608581244945526, + "rewards/weighted_rejected": -0.38665771484375, + "step": 890 + }, + { + "epoch": 0.48099405437905, + "grad_norm": 21.55549430847168, + "learning_rate": 6.20981060350445e-07, + "logits/chosen": -0.12602996826171875, + "logits/rejected": -0.08534087985754013, + "logps/chosen": -186.3312530517578, + "logps/rejected": -190.0625, + "logps/weighted_chosen": -0.8782455325126648, + "logps/weighted_rejected": -1.0832672119140625, + "loss": 0.6442, + "rewards/accuracies": 0.574999988079071, + "rewards/chosen": -65.2984390258789, + "rewards/margins": 6.321875095367432, + "rewards/rejected": -71.61796569824219, + "rewards/weighted_accuracies": 0.612500011920929, + "rewards/weighted_chosen": -0.3018638491630554, + "rewards/weighted_margins": 0.19889220595359802, + "rewards/weighted_rejected": -0.5003150701522827, + "step": 900 + }, + { + "epoch": 0.4863384327610395, + "grad_norm": 15.255428314208984, + "learning_rate": 6.119099324709087e-07, + "logits/chosen": -0.08440399169921875, + "logits/rejected": -0.03687591478228569, + "logps/chosen": -183.1640625, + "logps/rejected": -192.125, + "logps/weighted_chosen": -0.7069732546806335, + "logps/weighted_rejected": -1.348388671875, + "loss": 0.7239, + "rewards/accuracies": 0.609375, + "rewards/chosen": -65.2796859741211, + "rewards/margins": 7.408593654632568, + "rewards/rejected": -72.6890640258789, + "rewards/weighted_accuracies": 0.5093749761581421, + "rewards/weighted_chosen": -0.3756294250488281, + "rewards/weighted_margins": 0.05287017673254013, + "rewards/weighted_rejected": -0.4286758303642273, + "step": 910 + }, + { + "epoch": 0.4916828111430289, + "grad_norm": 10.355165481567383, + "learning_rate": 6.027998578067316e-07, + "logits/chosen": -0.06318511813879013, + "logits/rejected": -0.02981414832174778, + "logps/chosen": -199.36563110351562, + "logps/rejected": -208.70938110351562, + "logps/weighted_chosen": -0.6963897943496704, + "logps/weighted_rejected": -0.969342052936554, + "loss": 0.676, + "rewards/accuracies": 0.6031249761581421, + "rewards/chosen": -71.73905944824219, + "rewards/margins": 6.481640815734863, + "rewards/rejected": -78.2171859741211, + "rewards/weighted_accuracies": 0.550000011920929, + "rewards/weighted_chosen": -0.2880393862724304, + "rewards/weighted_margins": 0.1389778107404709, + "rewards/weighted_rejected": -0.426931768655777, + "step": 920 + }, + { + "epoch": 0.4970271895250184, + "grad_norm": 5.530470371246338, + "learning_rate": 5.936540068371394e-07, + "logits/chosen": 0.006648254580795765, + "logits/rejected": 0.005427551455795765, + "logps/chosen": -193.99374389648438, + "logps/rejected": -202.4406280517578, + "logps/weighted_chosen": -1.107049584388733, + "logps/weighted_rejected": -1.3529266119003296, + "loss": 0.6926, + "rewards/accuracies": 0.612500011920929, + "rewards/chosen": -74.79765319824219, + "rewards/margins": 5.872656345367432, + "rewards/rejected": -80.6624984741211, + "rewards/weighted_accuracies": 0.559374988079071, + "rewards/weighted_chosen": -0.43673935532569885, + "rewards/weighted_margins": 0.12621307373046875, + "rewards/weighted_rejected": -0.5626624822616577, + "step": 930 + }, + { + "epoch": 0.5023715679070078, + "grad_norm": 5.006593704223633, + "learning_rate": 5.844755624921918e-07, + "logits/chosen": -0.005947875790297985, + "logits/rejected": 0.02864990197122097, + "logps/chosen": -187.453125, + "logps/rejected": -192.7375030517578, + "logps/weighted_chosen": -1.132598876953125, + "logps/weighted_rejected": -1.0424163341522217, + "loss": 0.7378, + "rewards/accuracies": 0.609375, + "rewards/chosen": -70.77812194824219, + "rewards/margins": 5.782812595367432, + "rewards/rejected": -76.56562805175781, + "rewards/weighted_accuracies": 0.512499988079071, + "rewards/weighted_chosen": -0.44052886962890625, + "rewards/weighted_margins": 0.07312469184398651, + "rewards/weighted_rejected": -0.5137832760810852, + "step": 940 + }, + { + "epoch": 0.5077159462889973, + "grad_norm": 16.502933502197266, + "learning_rate": 5.752677190450634e-07, + "logits/chosen": -0.021286392584443092, + "logits/rejected": -0.00705032330006361, + "logps/chosen": -186.49374389648438, + "logps/rejected": -196.24844360351562, + "logps/weighted_chosen": -1.068359375, + "logps/weighted_rejected": -1.3232848644256592, + "loss": 0.7251, + "rewards/accuracies": 0.625, + "rewards/chosen": -67.70234680175781, + "rewards/margins": 7.561327934265137, + "rewards/rejected": -75.28047180175781, + "rewards/weighted_accuracies": 0.528124988079071, + "rewards/weighted_chosen": -0.390481561422348, + "rewards/weighted_margins": 0.058994293212890625, + "rewards/weighted_rejected": -0.44965285062789917, + "step": 950 + }, + { + "epoch": 0.5130603246709867, + "grad_norm": 12.000792503356934, + "learning_rate": 5.66033681000375e-07, + "logits/chosen": -0.03893890231847763, + "logits/rejected": 0.01745758019387722, + "logps/chosen": -189.5234375, + "logps/rejected": -201.2624969482422, + "logps/weighted_chosen": -1.0573517084121704, + "logps/weighted_rejected": -1.4250671863555908, + "loss": 0.6656, + "rewards/accuracies": 0.637499988079071, + "rewards/chosen": -69.40625, + "rewards/margins": 8.408203125, + "rewards/rejected": -77.78593444824219, + "rewards/weighted_accuracies": 0.5718749761581421, + "rewards/weighted_chosen": -0.2941024899482727, + "rewards/weighted_margins": 0.1710365265607834, + "rewards/weighted_rejected": -0.4653762876987457, + "step": 960 + }, + { + "epoch": 0.5184047030529761, + "grad_norm": 8.829608917236328, + "learning_rate": 5.567766619789658e-07, + "logits/chosen": -0.03634033352136612, + "logits/rejected": 0.0003803253057412803, + "logps/chosen": -190.27969360351562, + "logps/rejected": -199.0500030517578, + "logps/weighted_chosen": -1.1300811767578125, + "logps/weighted_rejected": -1.3106689453125, + "loss": 0.6931, + "rewards/accuracies": 0.565625011920929, + "rewards/chosen": -67.55078125, + "rewards/margins": 7.454297065734863, + "rewards/rejected": -74.9703140258789, + "rewards/weighted_accuracies": 0.606249988079071, + "rewards/weighted_chosen": -0.3190757632255554, + "rewards/weighted_margins": 0.11787338554859161, + "rewards/weighted_rejected": -0.4370986819267273, + "step": 970 + }, + { + "epoch": 0.5237490814349656, + "grad_norm": 9.326337814331055, + "learning_rate": 5.474998835994916e-07, + "logits/chosen": 0.009844208136200905, + "logits/rejected": 0.051239013671875, + "logps/chosen": -195.78125, + "logps/rejected": -208.5343780517578, + "logps/weighted_chosen": -1.179107666015625, + "logps/weighted_rejected": -1.546759009361267, + "loss": 0.6677, + "rewards/accuracies": 0.6343749761581421, + "rewards/chosen": -70.6781234741211, + "rewards/margins": 6.912109375, + "rewards/rejected": -77.5609359741211, + "rewards/weighted_accuracies": 0.5625, + "rewards/weighted_chosen": -0.28845518827438354, + "rewards/weighted_margins": 0.1378173828125, + "rewards/weighted_rejected": -0.42629241943359375, + "step": 980 + }, + { + "epoch": 0.529093459816955, + "grad_norm": 10.308549880981445, + "learning_rate": 5.382065743572411e-07, + "logits/chosen": -0.025909423828125, + "logits/rejected": 0.01560287456959486, + "logps/chosen": -192.8093719482422, + "logps/rejected": -205.2843780517578, + "logps/weighted_chosen": -0.8877533078193665, + "logps/weighted_rejected": -1.2557891607284546, + "loss": 0.6811, + "rewards/accuracies": 0.671875, + "rewards/chosen": -70.57890319824219, + "rewards/margins": 8.194921493530273, + "rewards/rejected": -78.80000305175781, + "rewards/weighted_accuracies": 0.534375011920929, + "rewards/weighted_chosen": -0.3638969361782074, + "rewards/weighted_margins": 0.13902893662452698, + "rewards/weighted_rejected": -0.5028846859931946, + "step": 990 + }, + { + "epoch": 0.5344378381989445, + "grad_norm": 6.612018585205078, + "learning_rate": 5.288999685005571e-07, + "logits/chosen": -0.03016967698931694, + "logits/rejected": -0.0050643919967114925, + "logps/chosen": -192.640625, + "logps/rejected": -201.4656219482422, + "logps/weighted_chosen": -0.626147449016571, + "logps/weighted_rejected": -0.8550964593887329, + "loss": 0.672, + "rewards/accuracies": 0.5843750238418579, + "rewards/chosen": -72.01640319824219, + "rewards/margins": 6.489062309265137, + "rewards/rejected": -78.5062484741211, + "rewards/weighted_accuracies": 0.581250011920929, + "rewards/weighted_chosen": -0.2895965576171875, + "rewards/weighted_margins": 0.1136070266366005, + "rewards/weighted_rejected": -0.4031356871128082, + "step": 1000 + }, + { + "epoch": 0.5344378381989445, + "eval_logits/chosen": -0.1623995155096054, + "eval_logits/rejected": -0.13676175475120544, + "eval_logps/chosen": -193.06211853027344, + "eval_logps/rejected": -200.78919982910156, + "eval_logps/weighted_chosen": -1.0030580759048462, + "eval_logps/weighted_rejected": -1.3398514986038208, + "eval_loss": 0.6716485619544983, + "eval_rewards/accuracies": 0.5982688665390015, + "eval_rewards/chosen": -71.73777770996094, + "eval_rewards/margins": 6.556771755218506, + "eval_rewards/rejected": -78.29378509521484, + "eval_rewards/weighted_accuracies": 0.5570265054702759, + "eval_rewards/weighted_chosen": -0.36391088366508484, + "eval_rewards/weighted_margins": 0.17662912607192993, + "eval_rewards/weighted_rejected": -0.5405399203300476, + "eval_runtime": 1144.2369, + "eval_samples_per_second": 1.714, + "eval_steps_per_second": 0.429, + "step": 1000 + }, + { + "epoch": 0.539782216580934, + "grad_norm": 6.273863315582275, + "learning_rate": 5.19583304905257e-07, + "logits/chosen": -0.04050292819738388, + "logits/rejected": -0.010775757022202015, + "logps/chosen": -187.6999969482422, + "logps/rejected": -195.6999969482422, + "logps/weighted_chosen": -0.898364245891571, + "logps/weighted_rejected": -1.051977515220642, + "loss": 0.699, + "rewards/accuracies": 0.606249988079071, + "rewards/chosen": -72.2406234741211, + "rewards/margins": 6.910546779632568, + "rewards/rejected": -79.1578140258789, + "rewards/weighted_accuracies": 0.574999988079071, + "rewards/weighted_chosen": -0.3900245726108551, + "rewards/weighted_margins": 0.1439811736345291, + "rewards/weighted_rejected": -0.5338424444198608, + "step": 1010 + }, + { + "epoch": 0.5451265949629234, + "grad_norm": 8.971517562866211, + "learning_rate": 5.102598259474434e-07, + "logits/chosen": -0.005509376525878906, + "logits/rejected": 0.02565765380859375, + "logps/chosen": -184.39999389648438, + "logps/rejected": -187.3781280517578, + "logps/weighted_chosen": -0.8080657720565796, + "logps/weighted_rejected": -1.043609619140625, + "loss": 0.7008, + "rewards/accuracies": 0.621874988079071, + "rewards/chosen": -72.9703140258789, + "rewards/margins": 5.461718559265137, + "rewards/rejected": -78.44844055175781, + "rewards/weighted_accuracies": 0.5562499761581421, + "rewards/weighted_chosen": -0.3826606869697571, + "rewards/weighted_margins": 0.062392424792051315, + "rewards/weighted_rejected": -0.44501036405563354, + "step": 1020 + }, + { + "epoch": 0.5504709733449128, + "grad_norm": 12.744133949279785, + "learning_rate": 5.009327763750932e-07, + "logits/chosen": -0.049317169934511185, + "logits/rejected": -0.0027618408203125, + "logps/chosen": -192.65625, + "logps/rejected": -202.55624389648438, + "logps/weighted_chosen": -0.8524872064590454, + "logps/weighted_rejected": -1.3228943347930908, + "loss": 0.6779, + "rewards/accuracies": 0.6156250238418579, + "rewards/chosen": -74.0992202758789, + "rewards/margins": 7.507031440734863, + "rewards/rejected": -81.62187194824219, + "rewards/weighted_accuracies": 0.5531250238418579, + "rewards/weighted_chosen": -0.3904098570346832, + "rewards/weighted_margins": 0.14227446913719177, + "rewards/weighted_rejected": -0.5326492190361023, + "step": 1030 + }, + { + "epoch": 0.5558153517269022, + "grad_norm": 6.275015354156494, + "learning_rate": 4.916054021788265e-07, + "logits/chosen": -0.04349822923541069, + "logits/rejected": -0.005797195248305798, + "logps/chosen": -192.04061889648438, + "logps/rejected": -196.96875, + "logps/weighted_chosen": -1.060455322265625, + "logps/weighted_rejected": -1.1041381359100342, + "loss": 0.676, + "rewards/accuracies": 0.621874988079071, + "rewards/chosen": -73.9984359741211, + "rewards/margins": 6.270703315734863, + "rewards/rejected": -80.27656555175781, + "rewards/weighted_accuracies": 0.5687500238418579, + "rewards/weighted_chosen": -0.3287033140659332, + "rewards/weighted_margins": 0.17382964491844177, + "rewards/weighted_rejected": -0.5024040341377258, + "step": 1040 + }, + { + "epoch": 0.5611597301088918, + "grad_norm": 7.003974437713623, + "learning_rate": 4.822809494622379e-07, + "logits/chosen": -0.07877655327320099, + "logits/rejected": -0.06013183668255806, + "logps/chosen": -189.9453125, + "logps/rejected": -201.99374389648438, + "logps/weighted_chosen": -1.28973388671875, + "logps/weighted_rejected": -1.1621277332305908, + "loss": 0.7014, + "rewards/accuracies": 0.6000000238418579, + "rewards/chosen": -69.9781265258789, + "rewards/margins": 8.531641006469727, + "rewards/rejected": -78.4828109741211, + "rewards/weighted_accuracies": 0.5218750238418579, + "rewards/weighted_chosen": -0.38208693265914917, + "rewards/weighted_margins": 0.09144439548254013, + "rewards/weighted_rejected": -0.4734359681606293, + "step": 1050 + }, + { + "epoch": 0.5665041084908812, + "grad_norm": 7.428668975830078, + "learning_rate": 4.729626633121914e-07, + "logits/chosen": -0.065679170191288, + "logits/rejected": -0.02408447302877903, + "logps/chosen": -191.80313110351562, + "logps/rejected": -202.32186889648438, + "logps/weighted_chosen": -1.075415015220642, + "logps/weighted_rejected": -1.3513977527618408, + "loss": 0.6657, + "rewards/accuracies": 0.5718749761581421, + "rewards/chosen": -73.8218765258789, + "rewards/margins": 7.512499809265137, + "rewards/rejected": -81.32499694824219, + "rewards/weighted_accuracies": 0.590624988079071, + "rewards/weighted_chosen": -0.32668304443359375, + "rewards/weighted_margins": 0.21370239555835724, + "rewards/weighted_rejected": -0.5402389764785767, + "step": 1060 + }, + { + "epoch": 0.5718484868728706, + "grad_norm": 10.470946311950684, + "learning_rate": 4.636537866694692e-07, + "logits/chosen": -0.023516083136200905, + "logits/rejected": 0.0013282776344567537, + "logps/chosen": -193.6531219482422, + "logps/rejected": -200.671875, + "logps/weighted_chosen": -1.336614966392517, + "logps/weighted_rejected": -1.549737572669983, + "loss": 0.6482, + "rewards/accuracies": 0.6312500238418579, + "rewards/chosen": -75.12890625, + "rewards/margins": 6.398828029632568, + "rewards/rejected": -81.51094055175781, + "rewards/weighted_accuracies": 0.612500011920929, + "rewards/weighted_chosen": -0.3157196044921875, + "rewards/weighted_margins": 0.2204277068376541, + "rewards/weighted_rejected": -0.5359970331192017, + "step": 1070 + }, + { + "epoch": 0.57719286525486, + "grad_norm": 5.486422061920166, + "learning_rate": 4.5435755920016557e-07, + "logits/chosen": -0.15928420424461365, + "logits/rejected": -0.16242370009422302, + "logps/chosen": -213.3874969482422, + "logps/rejected": -219.72500610351562, + "logps/weighted_chosen": -0.9821227788925171, + "logps/weighted_rejected": -0.865551769733429, + "loss": 0.6778, + "rewards/accuracies": 0.6000000238418579, + "rewards/chosen": -76.74687194824219, + "rewards/margins": 8.090624809265137, + "rewards/rejected": -84.828125, + "rewards/weighted_accuracies": 0.5375000238418579, + "rewards/weighted_chosen": -0.32687264680862427, + "rewards/weighted_margins": 0.08376693725585938, + "rewards/weighted_rejected": -0.4107864499092102, + "step": 1080 + }, + { + "epoch": 0.5825372436368494, + "grad_norm": 8.151789665222168, + "learning_rate": 4.450772161682221e-07, + "logits/chosen": -0.068756103515625, + "logits/rejected": -0.015399170108139515, + "logps/chosen": -199.9187469482422, + "logps/rejected": -204.18124389648438, + "logps/weighted_chosen": -1.588891625404358, + "logps/weighted_rejected": -1.757287621498108, + "loss": 0.6788, + "rewards/accuracies": 0.628125011920929, + "rewards/chosen": -74.18281555175781, + "rewards/margins": 8.891406059265137, + "rewards/rejected": -83.09687805175781, + "rewards/weighted_accuracies": 0.5218750238418579, + "rewards/weighted_chosen": -0.30588531494140625, + "rewards/weighted_margins": 0.1938468962907791, + "rewards/weighted_rejected": -0.49979859590530396, + "step": 1090 + }, + { + "epoch": 0.587881622018839, + "grad_norm": 6.956281661987305, + "learning_rate": 4.3581598730949376e-07, + "logits/chosen": -0.064672090113163, + "logits/rejected": -0.038168907165527344, + "logps/chosen": -203.828125, + "logps/rejected": -218.46249389648438, + "logps/weighted_chosen": -1.4421265125274658, + "logps/weighted_rejected": -1.3359496593475342, + "loss": 0.6969, + "rewards/accuracies": 0.659375011920929, + "rewards/chosen": -78.1781234741211, + "rewards/margins": 10.614843368530273, + "rewards/rejected": -88.7796859741211, + "rewards/weighted_accuracies": 0.5625, + "rewards/weighted_chosen": -0.38915711641311646, + "rewards/weighted_margins": 0.0950775146484375, + "rewards/weighted_rejected": -0.4842666685581207, + "step": 1100 + }, + { + "epoch": 0.5932260004008284, + "grad_norm": 9.076947212219238, + "learning_rate": 4.2657709570774007e-07, + "logits/chosen": -0.06801605224609375, + "logits/rejected": -0.05637931823730469, + "logps/chosen": -195.67813110351562, + "logps/rejected": -204.66250610351562, + "logps/weighted_chosen": -1.285638451576233, + "logps/weighted_rejected": -1.3720200061798096, + "loss": 0.6571, + "rewards/accuracies": 0.625, + "rewards/chosen": -76.2890625, + "rewards/margins": 8.073827743530273, + "rewards/rejected": -84.35624694824219, + "rewards/weighted_accuracies": 0.5687500238418579, + "rewards/weighted_chosen": -0.4137374758720398, + "rewards/weighted_margins": 0.19571837782859802, + "rewards/weighted_rejected": -0.6095176935195923, + "step": 1110 + }, + { + "epoch": 0.5985703787828178, + "grad_norm": 7.4237751960754395, + "learning_rate": 4.173637566729304e-07, + "logits/chosen": -0.09175796806812286, + "logits/rejected": -0.07670821994543076, + "logps/chosen": -196.52188110351562, + "logps/rejected": -210.00625610351562, + "logps/weighted_chosen": -1.169805884361267, + "logps/weighted_rejected": -1.150994896888733, + "loss": 0.6803, + "rewards/accuracies": 0.6499999761581421, + "rewards/chosen": -76.2906265258789, + "rewards/margins": 10.744921684265137, + "rewards/rejected": -87.03125, + "rewards/weighted_accuracies": 0.59375, + "rewards/weighted_chosen": -0.39191436767578125, + "rewards/weighted_margins": 0.15792007744312286, + "rewards/weighted_rejected": -0.549652099609375, + "step": 1120 + }, + { + "epoch": 0.6039147571648072, + "grad_norm": 6.803237438201904, + "learning_rate": 4.0817917662225467e-07, + "logits/chosen": -0.14934463798999786, + "logits/rejected": -0.13103675842285156, + "logps/chosen": -204.88436889648438, + "logps/rejected": -215.25155639648438, + "logps/weighted_chosen": -1.14501953125, + "logps/weighted_rejected": -1.147515892982483, + "loss": 0.6706, + "rewards/accuracies": 0.659375011920929, + "rewards/chosen": -80.02812194824219, + "rewards/margins": 9.226953506469727, + "rewards/rejected": -89.2734375, + "rewards/weighted_accuracies": 0.565625011920929, + "rewards/weighted_chosen": -0.3231521546840668, + "rewards/weighted_margins": 0.16764679551124573, + "rewards/weighted_rejected": -0.49060362577438354, + "step": 1130 + }, + { + "epoch": 0.6092591355467967, + "grad_norm": 7.308501720428467, + "learning_rate": 3.9902655196422957e-07, + "logits/chosen": -0.17415618896484375, + "logits/rejected": -0.13456878066062927, + "logps/chosen": -204.93594360351562, + "logps/rejected": -214.28750610351562, + "logps/weighted_chosen": -0.799731433391571, + "logps/weighted_rejected": -1.1109497547149658, + "loss": 0.691, + "rewards/accuracies": 0.596875011920929, + "rewards/chosen": -82.1039047241211, + "rewards/margins": 7.87109375, + "rewards/rejected": -89.9781265258789, + "rewards/weighted_accuracies": 0.5625, + "rewards/weighted_chosen": -0.4164085388183594, + "rewards/weighted_margins": 0.125599667429924, + "rewards/weighted_rejected": -0.5417296886444092, + "step": 1140 + }, + { + "epoch": 0.6146035139287862, + "grad_norm": 13.811458587646484, + "learning_rate": 3.8990906798628763e-07, + "logits/chosen": -0.08937682956457138, + "logits/rejected": -0.06486892700195312, + "logps/chosen": -210.97500610351562, + "logps/rejected": -214.6125030517578, + "logps/weighted_chosen": -1.324133276939392, + "logps/weighted_rejected": -1.3257598876953125, + "loss": 0.7048, + "rewards/accuracies": 0.528124988079071, + "rewards/chosen": -80.96875, + "rewards/margins": 6.654687404632568, + "rewards/rejected": -87.6578140258789, + "rewards/weighted_accuracies": 0.543749988079071, + "rewards/weighted_chosen": -0.3775787353515625, + "rewards/weighted_margins": 0.11986999213695526, + "rewards/weighted_rejected": -0.4972671568393707, + "step": 1150 + }, + { + "epoch": 0.6199478923107756, + "grad_norm": 10.00770378112793, + "learning_rate": 3.8082989774623587e-07, + "logits/chosen": -0.01934986189007759, + "logits/rejected": -0.015110015869140625, + "logps/chosen": -188.74063110351562, + "logps/rejected": -204.4187469482422, + "logps/weighted_chosen": -1.20220947265625, + "logps/weighted_rejected": -1.109490990638733, + "loss": 0.6717, + "rewards/accuracies": 0.637499988079071, + "rewards/chosen": -77.26094055175781, + "rewards/margins": 9.564844131469727, + "rewards/rejected": -86.8343734741211, + "rewards/weighted_accuracies": 0.578125, + "rewards/weighted_chosen": -0.3926025331020355, + "rewards/weighted_margins": 0.13916321098804474, + "rewards/weighted_rejected": -0.5318969488143921, + "step": 1160 + }, + { + "epoch": 0.625292270692765, + "grad_norm": 28.483694076538086, + "learning_rate": 3.717922009679725e-07, + "logits/chosen": -0.053440093994140625, + "logits/rejected": -0.0072647095657885075, + "logps/chosen": -201.890625, + "logps/rejected": -210.5187530517578, + "logps/weighted_chosen": -1.0479736328125, + "logps/weighted_rejected": -1.5367584228515625, + "loss": 0.6686, + "rewards/accuracies": 0.621874988079071, + "rewards/chosen": -78.7554702758789, + "rewards/margins": 8.385546684265137, + "rewards/rejected": -87.1624984741211, + "rewards/weighted_accuracies": 0.5687500238418579, + "rewards/weighted_chosen": -0.352639764547348, + "rewards/weighted_margins": 0.159870907664299, + "rewards/weighted_rejected": -0.5122512578964233, + "step": 1170 + }, + { + "epoch": 0.6306366490747545, + "grad_norm": 7.887178897857666, + "learning_rate": 3.6279912294184207e-07, + "logits/chosen": -0.06793518364429474, + "logits/rejected": -0.03975982591509819, + "logps/chosen": -208.45938110351562, + "logps/rejected": -216.97500610351562, + "logps/weighted_chosen": -1.1531188488006592, + "logps/weighted_rejected": -1.3199646472930908, + "loss": 0.664, + "rewards/accuracies": 0.625, + "rewards/chosen": -83.76875305175781, + "rewards/margins": 8.121874809265137, + "rewards/rejected": -91.8890609741211, + "rewards/weighted_accuracies": 0.6000000238418579, + "rewards/weighted_chosen": -0.3757583498954773, + "rewards/weighted_margins": 0.18969421088695526, + "rewards/weighted_rejected": -0.5654312372207642, + "step": 1180 + }, + { + "epoch": 0.6359810274567439, + "grad_norm": 21.66887664794922, + "learning_rate": 3.5385379343001507e-07, + "logits/chosen": -0.041289519518613815, + "logits/rejected": -0.01831817626953125, + "logps/chosen": -198.00936889648438, + "logps/rejected": -204.0437469482422, + "logps/weighted_chosen": -1.02520751953125, + "logps/weighted_rejected": -1.4494140148162842, + "loss": 0.6816, + "rewards/accuracies": 0.596875011920929, + "rewards/chosen": -79.15156555175781, + "rewards/margins": 7.596484184265137, + "rewards/rejected": -86.73515319824219, + "rewards/weighted_accuracies": 0.6031249761581421, + "rewards/weighted_chosen": -0.3985244631767273, + "rewards/weighted_margins": 0.12162628024816513, + "rewards/weighted_rejected": -0.5204399228096008, + "step": 1190 + }, + { + "epoch": 0.6413254058387334, + "grad_norm": 11.715490341186523, + "learning_rate": 3.4495932557727114e-07, + "logits/chosen": -0.0670623779296875, + "logits/rejected": -0.04256897047162056, + "logps/chosen": -200.75936889648438, + "logps/rejected": -214.21875, + "logps/weighted_chosen": -1.1181519031524658, + "logps/weighted_rejected": -1.4523468017578125, + "loss": 0.6876, + "rewards/accuracies": 0.6468750238418579, + "rewards/chosen": -81.3304672241211, + "rewards/margins": 10.844531059265137, + "rewards/rejected": -92.1703109741211, + "rewards/weighted_accuracies": 0.550000011920929, + "rewards/weighted_chosen": -0.4529586732387543, + "rewards/weighted_margins": 0.178059384226799, + "rewards/weighted_rejected": -0.6310364007949829, + "step": 1200 + }, + { + "epoch": 0.6466697842207229, + "grad_norm": 5.799643516540527, + "learning_rate": 3.3611881482756464e-07, + "logits/chosen": -0.02250823937356472, + "logits/rejected": -0.02404480054974556, + "logps/chosen": -209.1593780517578, + "logps/rejected": -213.9921875, + "logps/weighted_chosen": -1.55322265625, + "logps/weighted_rejected": -1.4918396472930908, + "loss": 0.6919, + "rewards/accuracies": 0.574999988079071, + "rewards/chosen": -84.61796569824219, + "rewards/margins": 6.630078315734863, + "rewards/rejected": -91.2593765258789, + "rewards/weighted_accuracies": 0.528124988079071, + "rewards/weighted_chosen": -0.4175216555595398, + "rewards/weighted_margins": 0.15537413954734802, + "rewards/weighted_rejected": -0.5732803344726562, + "step": 1210 + }, + { + "epoch": 0.6520141626027123, + "grad_norm": 10.090937614440918, + "learning_rate": 3.2733533784675273e-07, + "logits/chosen": 0.013554382137954235, + "logits/rejected": 0.04766502231359482, + "logps/chosen": -197.359375, + "logps/rejected": -209.89688110351562, + "logps/weighted_chosen": -1.3024734258651733, + "logps/weighted_rejected": -1.784857153892517, + "loss": 0.69, + "rewards/accuracies": 0.621874988079071, + "rewards/chosen": -79.140625, + "rewards/margins": 10.651171684265137, + "rewards/rejected": -89.7906265258789, + "rewards/weighted_accuracies": 0.5531250238418579, + "rewards/weighted_chosen": -0.4760070741176605, + "rewards/weighted_margins": 0.15030670166015625, + "rewards/weighted_rejected": -0.6262580752372742, + "step": 1220 + }, + { + "epoch": 0.6573585409847017, + "grad_norm": 7.221394062042236, + "learning_rate": 3.1861195145185603e-07, + "logits/chosen": -0.03308219835162163, + "logits/rejected": 0.00255584716796875, + "logps/chosen": -206.4031219482422, + "logps/rejected": -215.953125, + "logps/weighted_chosen": -1.181341528892517, + "logps/weighted_rejected": -1.657598853111267, + "loss": 0.6826, + "rewards/accuracies": 0.6000000238418579, + "rewards/chosen": -83.6734390258789, + "rewards/margins": 7.490234375, + "rewards/rejected": -91.1578140258789, + "rewards/weighted_accuracies": 0.5531250238418579, + "rewards/weighted_chosen": -0.3481903076171875, + "rewards/weighted_margins": 0.16850128769874573, + "rewards/weighted_rejected": -0.5166076421737671, + "step": 1230 + }, + { + "epoch": 0.6627029193666911, + "grad_norm": 50.77132797241211, + "learning_rate": 3.099516915472289e-07, + "logits/chosen": 0.006022644229233265, + "logits/rejected": 0.08475188910961151, + "logps/chosen": -200.8328094482422, + "logps/rejected": -205.4609375, + "logps/weighted_chosen": -0.924468994140625, + "logps/weighted_rejected": -1.4624541997909546, + "loss": 0.6753, + "rewards/accuracies": 0.6031249761581421, + "rewards/chosen": -80.05937194824219, + "rewards/margins": 5.724609375, + "rewards/rejected": -85.7828140258789, + "rewards/weighted_accuracies": 0.5874999761581421, + "rewards/weighted_chosen": -0.3424621522426605, + "rewards/weighted_margins": 0.1885635405778885, + "rewards/weighted_rejected": -0.5309646725654602, + "step": 1240 + }, + { + "epoch": 0.6680472977486807, + "grad_norm": 7.760387897491455, + "learning_rate": 3.013575720680062e-07, + "logits/chosen": 0.020127106457948685, + "logits/rejected": 0.042375944554805756, + "logps/chosen": -202.19375610351562, + "logps/rejected": -208.9968719482422, + "logps/weighted_chosen": -1.1934722661972046, + "logps/weighted_rejected": -1.6453979015350342, + "loss": 0.665, + "rewards/accuracies": 0.6187499761581421, + "rewards/chosen": -81.3656234741211, + "rewards/margins": 8.227343559265137, + "rewards/rejected": -89.56718444824219, + "rewards/weighted_accuracies": 0.550000011920929, + "rewards/weighted_chosen": -0.3932647705078125, + "rewards/weighted_margins": 0.21837767958641052, + "rewards/weighted_rejected": -0.6111129522323608, + "step": 1250 + }, + { + "epoch": 0.6733916761306701, + "grad_norm": 16.958906173706055, + "learning_rate": 2.928325839311974e-07, + "logits/chosen": -0.02686614915728569, + "logits/rejected": -0.016692351549863815, + "logps/chosen": -194.46719360351562, + "logps/rejected": -203.3093719482422, + "logps/weighted_chosen": -1.3333160877227783, + "logps/weighted_rejected": -1.2732360363006592, + "loss": 0.7271, + "rewards/accuracies": 0.6031249761581421, + "rewards/chosen": -79.31562805175781, + "rewards/margins": 8.610547065734863, + "rewards/rejected": -87.9281234741211, + "rewards/weighted_accuracies": 0.5531250238418579, + "rewards/weighted_chosen": -0.4805351197719574, + "rewards/weighted_margins": 0.173136904835701, + "rewards/weighted_rejected": -0.6535751223564148, + "step": 1260 + }, + { + "epoch": 0.6787360545126595, + "grad_norm": 19.331541061401367, + "learning_rate": 2.843796939947896e-07, + "logits/chosen": -0.015142058953642845, + "logits/rejected": 0.009578704833984375, + "logps/chosen": -199.8125, + "logps/rejected": -213.27499389648438, + "logps/weighted_chosen": -1.1416137218475342, + "logps/weighted_rejected": -1.472558617591858, + "loss": 0.6635, + "rewards/accuracies": 0.6187499761581421, + "rewards/chosen": -82.5484390258789, + "rewards/margins": 10.644140243530273, + "rewards/rejected": -93.1968765258789, + "rewards/weighted_accuracies": 0.550000011920929, + "rewards/weighted_chosen": -0.4067550599575043, + "rewards/weighted_margins": 0.1748603880405426, + "rewards/weighted_rejected": -0.5817291140556335, + "step": 1270 + }, + { + "epoch": 0.6840804328946489, + "grad_norm": 15.563817977905273, + "learning_rate": 2.7600184402522575e-07, + "logits/chosen": 0.02512359619140625, + "logits/rejected": 0.06499175727367401, + "logps/chosen": -196.04061889648438, + "logps/rejected": -209.2218780517578, + "logps/weighted_chosen": -1.2925841808319092, + "logps/weighted_rejected": -1.1955718994140625, + "loss": 0.6787, + "rewards/accuracies": 0.668749988079071, + "rewards/chosen": -79.7046890258789, + "rewards/margins": 11.271875381469727, + "rewards/rejected": -90.94921875, + "rewards/weighted_accuracies": 0.596875011920929, + "rewards/weighted_chosen": -0.4067886471748352, + "rewards/weighted_margins": 0.17460021376609802, + "rewards/weighted_rejected": -0.5814231634140015, + "step": 1280 + }, + { + "epoch": 0.6894248112766383, + "grad_norm": 13.546245574951172, + "learning_rate": 2.6770194967361303e-07, + "logits/chosen": -0.0179290771484375, + "logits/rejected": 0.0178680419921875, + "logps/chosen": -213.72500610351562, + "logps/rejected": -227.05313110351562, + "logps/weighted_chosen": -1.001013159751892, + "logps/weighted_rejected": -1.686248779296875, + "loss": 0.6712, + "rewards/accuracies": 0.6031249761581421, + "rewards/chosen": -86.60624694824219, + "rewards/margins": 8.127344131469727, + "rewards/rejected": -94.73124694824219, + "rewards/weighted_accuracies": 0.5406249761581421, + "rewards/weighted_chosen": -0.3357299864292145, + "rewards/weighted_margins": 0.20237883925437927, + "rewards/weighted_rejected": -0.5380722284317017, + "step": 1290 + }, + { + "epoch": 0.6947691896586279, + "grad_norm": 14.806809425354004, + "learning_rate": 2.594828994610211e-07, + "logits/chosen": 0.027711868286132812, + "logits/rejected": 0.05919818952679634, + "logps/chosen": -193.265625, + "logps/rejected": -206.9031219482422, + "logps/weighted_chosen": -1.0292694568634033, + "logps/weighted_rejected": -1.295262098312378, + "loss": 0.6787, + "rewards/accuracies": 0.6000000238418579, + "rewards/chosen": -80.3687515258789, + "rewards/margins": 8.969141006469727, + "rewards/rejected": -89.3671875, + "rewards/weighted_accuracies": 0.5562499761581421, + "rewards/weighted_chosen": -0.3460655212402344, + "rewards/weighted_margins": 0.17020568251609802, + "rewards/weighted_rejected": -0.5160125494003296, + "step": 1300 + }, + { + "epoch": 0.7001135680406173, + "grad_norm": 12.198771476745605, + "learning_rate": 2.513475537732219e-07, + "logits/chosen": 0.0022941590286791325, + "logits/rejected": 0.025787353515625, + "logps/chosen": -204.703125, + "logps/rejected": -217.66250610351562, + "logps/weighted_chosen": -1.215307593345642, + "logps/weighted_rejected": -1.44464111328125, + "loss": 0.665, + "rewards/accuracies": 0.609375, + "rewards/chosen": -84.4281234741211, + "rewards/margins": 8.751562118530273, + "rewards/rejected": -93.18281555175781, + "rewards/weighted_accuracies": 0.559374988079071, + "rewards/weighted_chosen": -0.3498634397983551, + "rewards/weighted_margins": 0.19211502373218536, + "rewards/weighted_rejected": -0.541577160358429, + "step": 1310 + }, + { + "epoch": 0.7054579464226067, + "grad_norm": 18.498014450073242, + "learning_rate": 2.4329874386522113e-07, + "logits/chosen": 0.0073295594193041325, + "logits/rejected": 0.055358123034238815, + "logps/chosen": -199.6437530517578, + "logps/rejected": -208.70156860351562, + "logps/weighted_chosen": -1.0841186046600342, + "logps/weighted_rejected": -1.4295532703399658, + "loss": 0.6551, + "rewards/accuracies": 0.59375, + "rewards/chosen": -83.765625, + "rewards/margins": 8.8515625, + "rewards/rejected": -92.6343765258789, + "rewards/weighted_accuracies": 0.596875011920929, + "rewards/weighted_chosen": -0.4359336793422699, + "rewards/weighted_margins": 0.21386031806468964, + "rewards/weighted_rejected": -0.6501365900039673, + "step": 1320 + }, + { + "epoch": 0.7108023248045962, + "grad_norm": 15.962686538696289, + "learning_rate": 2.3533927087592713e-07, + "logits/chosen": -0.000156402587890625, + "logits/rejected": 0.01469268836081028, + "logps/chosen": -206.29061889648438, + "logps/rejected": -216.75, + "logps/weighted_chosen": -1.1605316400527954, + "logps/weighted_rejected": -1.276606798171997, + "loss": 0.7058, + "rewards/accuracies": 0.59375, + "rewards/chosen": -83.1468734741211, + "rewards/margins": 7.533203125, + "rewards/rejected": -90.6796875, + "rewards/weighted_accuracies": 0.528124988079071, + "rewards/weighted_chosen": -0.422750860452652, + "rewards/weighted_margins": 0.102025605738163, + "rewards/weighted_rejected": -0.5250183343887329, + "step": 1330 + }, + { + "epoch": 0.7161467031865856, + "grad_norm": 10.239470481872559, + "learning_rate": 2.2747190485330193e-07, + "logits/chosen": 0.02880706824362278, + "logits/rejected": 0.07036437839269638, + "logps/chosen": -199.8859405517578, + "logps/rejected": -209.57186889648438, + "logps/weighted_chosen": -1.2549285888671875, + "logps/weighted_rejected": -1.352288842201233, + "loss": 0.6718, + "rewards/accuracies": 0.637499988079071, + "rewards/chosen": -80.2109375, + "rewards/margins": 10.519922256469727, + "rewards/rejected": -90.71875, + "rewards/weighted_accuracies": 0.612500011920929, + "rewards/weighted_chosen": -0.3369941711425781, + "rewards/weighted_margins": 0.2133018523454666, + "rewards/weighted_rejected": -0.5501846075057983, + "step": 1340 + }, + { + "epoch": 0.7214910815685751, + "grad_norm": 12.942761421203613, + "learning_rate": 2.1969938379033166e-07, + "logits/chosen": -0.0024871826171875, + "logits/rejected": -0.01717834547162056, + "logps/chosen": -206.4250030517578, + "logps/rejected": -220.47030639648438, + "logps/weighted_chosen": -1.2035309076309204, + "logps/weighted_rejected": -1.3001312017440796, + "loss": 0.7016, + "rewards/accuracies": 0.596875011920929, + "rewards/chosen": -82.8765640258789, + "rewards/margins": 8.98828125, + "rewards/rejected": -91.8671875, + "rewards/weighted_accuracies": 0.5718749761581421, + "rewards/weighted_chosen": -0.37746506929397583, + "rewards/weighted_margins": 0.11672210693359375, + "rewards/weighted_rejected": -0.4940811097621918, + "step": 1350 + }, + { + "epoch": 0.7268354599505645, + "grad_norm": 16.300827026367188, + "learning_rate": 2.1202441267215265e-07, + "logits/chosen": 0.04172973707318306, + "logits/rejected": 0.06365661323070526, + "logps/chosen": -183.5437469482422, + "logps/rejected": -198.49063110351562, + "logps/weighted_chosen": -1.542272925376892, + "logps/weighted_rejected": -1.678741455078125, + "loss": 0.7155, + "rewards/accuracies": 0.643750011920929, + "rewards/chosen": -78.47343444824219, + "rewards/margins": 10.018359184265137, + "rewards/rejected": -88.47343444824219, + "rewards/weighted_accuracies": 0.5625, + "rewards/weighted_chosen": -0.4637313783168793, + "rewards/weighted_margins": 0.1779930144548416, + "rewards/weighted_rejected": -0.6415374875068665, + "step": 1360 + }, + { + "epoch": 0.732179838332554, + "grad_norm": 17.129104614257812, + "learning_rate": 2.0444966253466572e-07, + "logits/chosen": -0.01758270338177681, + "logits/rejected": 0.049257658421993256, + "logps/chosen": -198.3562469482422, + "logps/rejected": -212.31875610351562, + "logps/weighted_chosen": -1.0854461193084717, + "logps/weighted_rejected": -1.6400238275527954, + "loss": 0.6718, + "rewards/accuracies": 0.606249988079071, + "rewards/chosen": -81.81718444824219, + "rewards/margins": 8.70703125, + "rewards/rejected": -90.52812194824219, + "rewards/weighted_accuracies": 0.5874999761581421, + "rewards/weighted_chosen": -0.3319801390171051, + "rewards/weighted_margins": 0.17666855454444885, + "rewards/weighted_rejected": -0.5086410641670227, + "step": 1370 + }, + { + "epoch": 0.7375242167145434, + "grad_norm": 15.24010181427002, + "learning_rate": 1.9697776953496397e-07, + "logits/chosen": -0.0036972046364098787, + "logits/rejected": 0.04438323900103569, + "logps/chosen": -204.8406219482422, + "logps/rejected": -215.5343780517578, + "logps/weighted_chosen": -1.2466247081756592, + "logps/weighted_rejected": -1.588470458984375, + "loss": 0.6659, + "rewards/accuracies": 0.6187499761581421, + "rewards/chosen": -82.30078125, + "rewards/margins": 10.021875381469727, + "rewards/rejected": -92.33515930175781, + "rewards/weighted_accuracies": 0.574999988079071, + "rewards/weighted_chosen": -0.3886352479457855, + "rewards/weighted_margins": 0.18134918808937073, + "rewards/weighted_rejected": -0.5699905157089233, + "step": 1380 + }, + { + "epoch": 0.7428685950965328, + "grad_norm": 13.349143028259277, + "learning_rate": 1.8961133403390113e-07, + "logits/chosen": -0.03137054294347763, + "logits/rejected": 0.01921997033059597, + "logps/chosen": -205.91250610351562, + "logps/rejected": -215.74374389648438, + "logps/weighted_chosen": -0.754425048828125, + "logps/weighted_rejected": -1.830108642578125, + "loss": 0.6648, + "rewards/accuracies": 0.596875011920929, + "rewards/chosen": -82.6187515258789, + "rewards/margins": 9.654296875, + "rewards/rejected": -92.2750015258789, + "rewards/weighted_accuracies": 0.590624988079071, + "rewards/weighted_chosen": -0.38862380385398865, + "rewards/weighted_margins": 0.23016509413719177, + "rewards/weighted_rejected": -0.618792712688446, + "step": 1390 + }, + { + "epoch": 0.7482129734785222, + "grad_norm": 12.553123474121094, + "learning_rate": 1.823529196911156e-07, + "logits/chosen": -0.07255706936120987, + "logits/rejected": -0.03428497165441513, + "logps/chosen": -207.0109405517578, + "logps/rejected": -213.7937469482422, + "logps/weighted_chosen": -1.1017853021621704, + "logps/weighted_rejected": -1.335351586341858, + "loss": 0.6721, + "rewards/accuracies": 0.609375, + "rewards/chosen": -83.64375305175781, + "rewards/margins": 7.791796684265137, + "rewards/rejected": -91.4296875, + "rewards/weighted_accuracies": 0.59375, + "rewards/weighted_chosen": -0.35135191679000854, + "rewards/weighted_margins": 0.21256867051124573, + "rewards/weighted_rejected": -0.5637038946151733, + "step": 1400 + }, + { + "epoch": 0.7535573518605118, + "grad_norm": 10.907644271850586, + "learning_rate": 1.7520505257282886e-07, + "logits/chosen": 0.02479248121380806, + "logits/rejected": 0.05216522142291069, + "logps/chosen": -198.2531280517578, + "logps/rejected": -207.4499969482422, + "logps/weighted_chosen": -1.016326904296875, + "logps/weighted_rejected": -1.4452117681503296, + "loss": 0.6724, + "rewards/accuracies": 0.6312500238418579, + "rewards/chosen": -80.75312805175781, + "rewards/margins": 9.733593940734863, + "rewards/rejected": -90.4671859741211, + "rewards/weighted_accuracies": 0.578125, + "rewards/weighted_chosen": -0.3796447813510895, + "rewards/weighted_margins": 0.1787368804216385, + "rewards/weighted_rejected": -0.5583381652832031, + "step": 1410 + }, + { + "epoch": 0.7589017302425012, + "grad_norm": 9.989439964294434, + "learning_rate": 1.6817022027272572e-07, + "logits/chosen": 0.03133545070886612, + "logits/rejected": 0.03775329515337944, + "logps/chosen": -207.06405639648438, + "logps/rejected": -224.08749389648438, + "logps/weighted_chosen": -1.253021240234375, + "logps/weighted_rejected": -1.788061499595642, + "loss": 0.703, + "rewards/accuracies": 0.640625, + "rewards/chosen": -82.796875, + "rewards/margins": 11.674609184265137, + "rewards/rejected": -94.50468444824219, + "rewards/weighted_accuracies": 0.5687500238418579, + "rewards/weighted_chosen": -0.3869224488735199, + "rewards/weighted_margins": 0.14797210693359375, + "rewards/weighted_rejected": -0.5350815057754517, + "step": 1420 + }, + { + "epoch": 0.7642461086244906, + "grad_norm": 12.953351974487305, + "learning_rate": 1.6125087104622502e-07, + "logits/chosen": -0.03018951416015625, + "logits/rejected": 0.010332489386200905, + "logps/chosen": -204.82186889648438, + "logps/rejected": -211.2375030517578, + "logps/weighted_chosen": -1.2213348150253296, + "logps/weighted_rejected": -1.4504730701446533, + "loss": 0.679, + "rewards/accuracies": 0.6187499761581421, + "rewards/chosen": -81.8109359741211, + "rewards/margins": 8.600781440734863, + "rewards/rejected": -90.39375305175781, + "rewards/weighted_accuracies": 0.6000000238418579, + "rewards/weighted_chosen": -0.43824997544288635, + "rewards/weighted_margins": 0.12977370619773865, + "rewards/weighted_rejected": -0.56793212890625, + "step": 1430 + }, + { + "epoch": 0.76959048700648, + "grad_norm": 51.06248474121094, + "learning_rate": 1.5444941295843945e-07, + "logits/chosen": -0.0010368346702307463, + "logits/rejected": -0.0032970428001135588, + "logps/chosen": -206.72500610351562, + "logps/rejected": -225.6125030517578, + "logps/weighted_chosen": -0.7966338992118835, + "logps/weighted_rejected": -0.8868865966796875, + "loss": 0.6564, + "rewards/accuracies": 0.640625, + "rewards/chosen": -83.40937805175781, + "rewards/margins": 12.608983993530273, + "rewards/rejected": -96.03125, + "rewards/weighted_accuracies": 0.5625, + "rewards/weighted_chosen": -0.2895980775356293, + "rewards/weighted_margins": 0.20346298813819885, + "rewards/weighted_rejected": -0.49294739961624146, + "step": 1440 + }, + { + "epoch": 0.7749348653884695, + "grad_norm": 7.5457658767700195, + "learning_rate": 1.4776821304612392e-07, + "logits/chosen": -0.01905670203268528, + "logits/rejected": 0.01137466449290514, + "logps/chosen": -196.77969360351562, + "logps/rejected": -210.0859375, + "logps/weighted_chosen": -0.9277313351631165, + "logps/weighted_rejected": -1.508001685142517, + "loss": 0.6648, + "rewards/accuracies": 0.625, + "rewards/chosen": -79.80155944824219, + "rewards/margins": 9.823827743530273, + "rewards/rejected": -89.63984680175781, + "rewards/weighted_accuracies": 0.578125, + "rewards/weighted_chosen": -0.3523406982421875, + "rewards/weighted_margins": 0.22584763169288635, + "rewards/weighted_rejected": -0.5784717798233032, + "step": 1450 + }, + { + "epoch": 0.780279243770459, + "grad_norm": 22.205232620239258, + "learning_rate": 1.4120959649390136e-07, + "logits/chosen": -0.06232147291302681, + "logits/rejected": -0.025969695299863815, + "logps/chosen": -205.7578125, + "logps/rejected": -216.8156280517578, + "logps/weighted_chosen": -1.046392798423767, + "logps/weighted_rejected": -1.1339843273162842, + "loss": 0.6717, + "rewards/accuracies": 0.643750011920929, + "rewards/chosen": -82.03125, + "rewards/margins": 10.355077743530273, + "rewards/rejected": -92.41093444824219, + "rewards/weighted_accuracies": 0.6000000238418579, + "rewards/weighted_chosen": -0.39169007539749146, + "rewards/weighted_margins": 0.185791015625, + "rewards/weighted_rejected": -0.5777603387832642, + "step": 1460 + }, + { + "epoch": 0.7856236221524484, + "grad_norm": 15.942556381225586, + "learning_rate": 1.347758458250541e-07, + "logits/chosen": 0.012713623233139515, + "logits/rejected": 0.04683647304773331, + "logps/chosen": -203.40469360351562, + "logps/rejected": -219.640625, + "logps/weighted_chosen": -1.0474426746368408, + "logps/weighted_rejected": -1.3139832019805908, + "loss": 0.6875, + "rewards/accuracies": 0.5843750238418579, + "rewards/chosen": -82.4453125, + "rewards/margins": 7.944531440734863, + "rewards/rejected": -90.3812484741211, + "rewards/weighted_accuracies": 0.5625, + "rewards/weighted_chosen": -0.4108734130859375, + "rewards/weighted_margins": 0.16155776381492615, + "rewards/weighted_rejected": -0.5725204348564148, + "step": 1470 + }, + { + "epoch": 0.7909680005344378, + "grad_norm": 26.816959381103516, + "learning_rate": 1.2846920010716266e-07, + "logits/chosen": 0.014247131533920765, + "logits/rejected": 0.017230987548828125, + "logps/chosen": -199.45938110351562, + "logps/rejected": -212.2609405517578, + "logps/weighted_chosen": -1.156378149986267, + "logps/weighted_rejected": -1.0633026361465454, + "loss": 0.6873, + "rewards/accuracies": 0.659375011920929, + "rewards/chosen": -82.2640609741211, + "rewards/margins": 11.641016006469727, + "rewards/rejected": -93.9296875, + "rewards/weighted_accuracies": 0.5874999761581421, + "rewards/weighted_chosen": -0.43985289335250854, + "rewards/weighted_margins": 0.15669021010398865, + "rewards/weighted_rejected": -0.5962905883789062, + "step": 1480 + }, + { + "epoch": 0.7963123789164273, + "grad_norm": 9.73423957824707, + "learning_rate": 1.2229185417286764e-07, + "logits/chosen": -0.011121368035674095, + "logits/rejected": 0.0305023193359375, + "logps/chosen": -208.5437469482422, + "logps/rejected": -219.5906219482422, + "logps/weighted_chosen": -1.2618286609649658, + "logps/weighted_rejected": -1.607025146484375, + "loss": 0.6811, + "rewards/accuracies": 0.606249988079071, + "rewards/chosen": -81.4375, + "rewards/margins": 10.083984375, + "rewards/rejected": -91.5062484741211, + "rewards/weighted_accuracies": 0.6000000238418579, + "rewards/weighted_chosen": -0.3379165530204773, + "rewards/weighted_margins": 0.1427352875471115, + "rewards/weighted_rejected": -0.480560302734375, + "step": 1490 + }, + { + "epoch": 0.8016567572984167, + "grad_norm": 25.978179931640625, + "learning_rate": 1.1624595785602576e-07, + "logits/chosen": -0.02779388427734375, + "logits/rejected": 0.00923004187643528, + "logps/chosen": -207.6750030517578, + "logps/rejected": -214.234375, + "logps/weighted_chosen": -1.3065948486328125, + "logps/weighted_rejected": -1.779363989830017, + "loss": 0.7042, + "rewards/accuracies": 0.581250011920929, + "rewards/chosen": -82.6429672241211, + "rewards/margins": 7.233593940734863, + "rewards/rejected": -89.8929672241211, + "rewards/weighted_accuracies": 0.5718749761581421, + "rewards/weighted_chosen": -0.4565597474575043, + "rewards/weighted_margins": 0.20427818596363068, + "rewards/weighted_rejected": -0.660656750202179, + "step": 1500 + }, + { + "epoch": 0.8016567572984167, + "eval_logits/chosen": -0.1534786820411682, + "eval_logits/rejected": -0.12998200953006744, + "eval_logps/chosen": -201.64256286621094, + "eval_logps/rejected": -211.74440002441406, + "eval_logps/weighted_chosen": -1.0274064540863037, + "eval_logps/weighted_rejected": -1.3898550271987915, + "eval_loss": 0.66490238904953, + "eval_rewards/accuracies": 0.6135438084602356, + "eval_rewards/chosen": -80.33655548095703, + "eval_rewards/margins": 8.974350929260254, + "eval_rewards/rejected": -89.31975555419922, + "eval_rewards/weighted_accuracies": 0.5651730895042419, + "eval_rewards/weighted_chosen": -0.38825926184654236, + "eval_rewards/weighted_margins": 0.20228415727615356, + "eval_rewards/weighted_rejected": -0.5905433893203735, + "eval_runtime": 1233.4168, + "eval_samples_per_second": 1.59, + "eval_steps_per_second": 0.398, + "step": 1500 + }, + { + "epoch": 0.8070011356804062, + "grad_norm": 8.884445190429688, + "learning_rate": 1.103336152435274e-07, + "logits/chosen": 0.0062271119095385075, + "logits/rejected": 0.03198547288775444, + "logps/chosen": -200.34375, + "logps/rejected": -211.6671905517578, + "logps/weighted_chosen": -0.8016601800918579, + "logps/weighted_rejected": -1.186651587486267, + "loss": 0.6617, + "rewards/accuracies": 0.6000000238418579, + "rewards/chosen": -83.3515625, + "rewards/margins": 8.618749618530273, + "rewards/rejected": -91.9468765258789, + "rewards/weighted_accuracies": 0.565625011920929, + "rewards/weighted_chosen": -0.372201532125473, + "rewards/weighted_margins": 0.16056060791015625, + "rewards/weighted_rejected": -0.5327514410018921, + "step": 1510 + }, + { + "epoch": 0.8123455140623956, + "grad_norm": 7.671918869018555, + "learning_rate": 1.0455688394303419e-07, + "logits/chosen": -0.004675674252212048, + "logits/rejected": 0.05005035549402237, + "logps/chosen": -204.24063110351562, + "logps/rejected": -209.0500030517578, + "logps/weighted_chosen": -1.006256103515625, + "logps/weighted_rejected": -1.2681457996368408, + "loss": 0.6866, + "rewards/accuracies": 0.578125, + "rewards/chosen": -84.80937194824219, + "rewards/margins": 5.722265720367432, + "rewards/rejected": -90.5546875, + "rewards/weighted_accuracies": 0.574999988079071, + "rewards/weighted_chosen": -0.41632080078125, + "rewards/weighted_margins": 0.1294807493686676, + "rewards/weighted_rejected": -0.5454360842704773, + "step": 1520 + }, + { + "epoch": 0.8176898924443851, + "grad_norm": 7.2523908615112305, + "learning_rate": 9.891777436689208e-08, + "logits/chosen": -0.008026123046875, + "logits/rejected": 0.03105773963034153, + "logps/chosen": -201.49844360351562, + "logps/rejected": -214.48593139648438, + "logps/weighted_chosen": -0.7988342046737671, + "logps/weighted_rejected": -1.346063256263733, + "loss": 0.6675, + "rewards/accuracies": 0.5874999761581421, + "rewards/chosen": -81.26875305175781, + "rewards/margins": 7.553906440734863, + "rewards/rejected": -88.828125, + "rewards/weighted_accuracies": 0.6031249761581421, + "rewards/weighted_chosen": -0.3720565736293793, + "rewards/weighted_margins": 0.18174858391284943, + "rewards/weighted_rejected": -0.5538619756698608, + "step": 1530 + }, + { + "epoch": 0.8230342708263745, + "grad_norm": 7.528415679931641, + "learning_rate": 9.341824903247003e-08, + "logits/chosen": 0.0003150939883198589, + "logits/rejected": 0.0193023681640625, + "logps/chosen": -200.9968719482422, + "logps/rejected": -206.00936889648438, + "logps/weighted_chosen": -1.262664794921875, + "logps/weighted_rejected": -1.222955346107483, + "loss": 0.7044, + "rewards/accuracies": 0.606249988079071, + "rewards/chosen": -81.1015625, + "rewards/margins": 7.956250190734863, + "rewards/rejected": -89.03437805175781, + "rewards/weighted_accuracies": 0.574999988079071, + "rewards/weighted_chosen": -0.3925842344760895, + "rewards/weighted_margins": 0.1385902464389801, + "rewards/weighted_rejected": -0.5313628911972046, + "step": 1540 + }, + { + "epoch": 0.8283786492083639, + "grad_norm": 12.314373970031738, + "learning_rate": 8.806022187916634e-08, + "logits/chosen": -0.01698608323931694, + "logits/rejected": -0.0005397796630859375, + "logps/chosen": -204.90469360351562, + "logps/rejected": -220.0968780517578, + "logps/weighted_chosen": -0.826275646686554, + "logps/weighted_rejected": -0.7519165277481079, + "loss": 0.6882, + "rewards/accuracies": 0.612500011920929, + "rewards/chosen": -81.35234069824219, + "rewards/margins": 9.787890434265137, + "rewards/rejected": -91.16328430175781, + "rewards/weighted_accuracies": 0.5562499761581421, + "rewards/weighted_chosen": -0.3715835511684418, + "rewards/weighted_margins": 0.13009873032569885, + "rewards/weighted_rejected": -0.5020004510879517, + "step": 1550 + }, + { + "epoch": 0.8337230275903534, + "grad_norm": 7.998411178588867, + "learning_rate": 8.284555760232176e-08, + "logits/chosen": -0.03365020826458931, + "logits/rejected": -0.011144256219267845, + "logps/chosen": -187.03125, + "logps/rejected": -198.0343780517578, + "logps/weighted_chosen": -0.8406280279159546, + "logps/weighted_rejected": -0.948577880859375, + "loss": 0.6634, + "rewards/accuracies": 0.625, + "rewards/chosen": -76.2421875, + "rewards/margins": 10.604687690734863, + "rewards/rejected": -86.85624694824219, + "rewards/weighted_accuracies": 0.574999988079071, + "rewards/weighted_chosen": -0.3922592103481293, + "rewards/weighted_margins": 0.20080260932445526, + "rewards/weighted_rejected": -0.5934112668037415, + "step": 1560 + }, + { + "epoch": 0.8390674059723429, + "grad_norm": 8.670356750488281, + "learning_rate": 7.777607100426925e-08, + "logits/chosen": -0.054158784449100494, + "logits/rejected": -0.03361816331744194, + "logps/chosen": -207.2312469482422, + "logps/rejected": -216.125, + "logps/weighted_chosen": -1.2455352544784546, + "logps/weighted_rejected": -1.077966332435608, + "loss": 0.6691, + "rewards/accuracies": 0.59375, + "rewards/chosen": -82.80937194824219, + "rewards/margins": 9.444531440734863, + "rewards/rejected": -92.26094055175781, + "rewards/weighted_accuracies": 0.606249988079071, + "rewards/weighted_chosen": -0.35657501220703125, + "rewards/weighted_margins": 0.17081375420093536, + "rewards/weighted_rejected": -0.5276878476142883, + "step": 1570 + }, + { + "epoch": 0.8444117843543323, + "grad_norm": 12.28963565826416, + "learning_rate": 7.285352636274922e-08, + "logits/chosen": 0.01006469689309597, + "logits/rejected": 0.03425750881433487, + "logps/chosen": -195.58749389648438, + "logps/rejected": -205.8781280517578, + "logps/weighted_chosen": -1.3043944835662842, + "logps/weighted_rejected": -1.3767731189727783, + "loss": 0.7349, + "rewards/accuracies": 0.621874988079071, + "rewards/chosen": -77.5609359741211, + "rewards/margins": 10.790624618530273, + "rewards/rejected": -88.34062194824219, + "rewards/weighted_accuracies": 0.543749988079071, + "rewards/weighted_chosen": -0.4487808346748352, + "rewards/weighted_margins": 0.135456845164299, + "rewards/weighted_rejected": -0.5850158929824829, + "step": 1580 + }, + { + "epoch": 0.8497561627363217, + "grad_norm": 12.43053913116455, + "learning_rate": 6.80796368169067e-08, + "logits/chosen": -0.03640174865722656, + "logits/rejected": 0.024863053113222122, + "logps/chosen": -209.8828125, + "logps/rejected": -222.59219360351562, + "logps/weighted_chosen": -0.916961669921875, + "logps/weighted_rejected": -1.4554595947265625, + "loss": 0.6726, + "rewards/accuracies": 0.637499988079071, + "rewards/chosen": -80.7679672241211, + "rewards/margins": 12.613672256469727, + "rewards/rejected": -93.3812484741211, + "rewards/weighted_accuracies": 0.6031249761581421, + "rewards/weighted_chosen": -0.3256774842739105, + "rewards/weighted_margins": 0.18812866508960724, + "rewards/weighted_rejected": -0.5140304565429688, + "step": 1590 + }, + { + "epoch": 0.8551005411183111, + "grad_norm": 7.3841352462768555, + "learning_rate": 6.345606377108658e-08, + "logits/chosen": 0.04971466213464737, + "logits/rejected": 0.09067229926586151, + "logps/chosen": -199.7687530517578, + "logps/rejected": -215.95938110351562, + "logps/weighted_chosen": -0.9194701910018921, + "logps/weighted_rejected": -1.5334380865097046, + "loss": 0.6658, + "rewards/accuracies": 0.6468750238418579, + "rewards/chosen": -81.9398422241211, + "rewards/margins": 11.660547256469727, + "rewards/rejected": -93.58906555175781, + "rewards/weighted_accuracies": 0.546875, + "rewards/weighted_chosen": -0.3771987855434418, + "rewards/weighted_margins": 0.18507537245750427, + "rewards/weighted_rejected": -0.5623397827148438, + "step": 1600 + }, + { + "epoch": 0.8604449195003007, + "grad_norm": 6.630573749542236, + "learning_rate": 5.89844163166332e-08, + "logits/chosen": -0.0077033997513353825, + "logits/rejected": 0.0152740478515625, + "logps/chosen": -206.56094360351562, + "logps/rejected": -213.8093719482422, + "logps/weighted_chosen": -1.148492455482483, + "logps/weighted_rejected": -1.014794945716858, + "loss": 0.6821, + "rewards/accuracies": 0.5874999761581421, + "rewards/chosen": -82.2054672241211, + "rewards/margins": 8.006640434265137, + "rewards/rejected": -90.23515319824219, + "rewards/weighted_accuracies": 0.5249999761581421, + "rewards/weighted_chosen": -0.3819526731967926, + "rewards/weighted_margins": 0.1473899781703949, + "rewards/weighted_rejected": -0.529217541217804, + "step": 1610 + }, + { + "epoch": 0.8657892978822901, + "grad_norm": 9.12763500213623, + "learning_rate": 5.466625067189601e-08, + "logits/chosen": 0.0029403686057776213, + "logits/rejected": 0.038446806371212006, + "logps/chosen": -195.69686889648438, + "logps/rejected": -205.2781219482422, + "logps/weighted_chosen": -0.8219360113143921, + "logps/weighted_rejected": -1.164337158203125, + "loss": 0.6719, + "rewards/accuracies": 0.6312500238418579, + "rewards/chosen": -79.7906265258789, + "rewards/margins": 9.067968368530273, + "rewards/rejected": -88.85469055175781, + "rewards/weighted_accuracies": 0.5625, + "rewards/weighted_chosen": -0.3714141845703125, + "rewards/weighted_margins": 0.172444149851799, + "rewards/weighted_rejected": -0.5441833734512329, + "step": 1620 + }, + { + "epoch": 0.8711336762642795, + "grad_norm": 31.594406127929688, + "learning_rate": 5.0503069640635944e-08, + "logits/chosen": -0.013871383853256702, + "logits/rejected": 0.03087768517434597, + "logps/chosen": -203.55313110351562, + "logps/rejected": -214.140625, + "logps/weighted_chosen": -1.257110595703125, + "logps/weighted_rejected": -1.71246337890625, + "loss": 0.6744, + "rewards/accuracies": 0.6187499761581421, + "rewards/chosen": -82.45625305175781, + "rewards/margins": 9.573046684265137, + "rewards/rejected": -92.05781555175781, + "rewards/weighted_accuracies": 0.578125, + "rewards/weighted_chosen": -0.33105164766311646, + "rewards/weighted_margins": 0.23786774277687073, + "rewards/weighted_rejected": -0.5689346194267273, + "step": 1630 + }, + { + "epoch": 0.8764780546462689, + "grad_norm": 6.86949348449707, + "learning_rate": 4.649632208902127e-08, + "logits/chosen": -0.06173248216509819, + "logits/rejected": -0.0015998840099200606, + "logps/chosen": -193.95938110351562, + "logps/rejected": -206.36563110351562, + "logps/weighted_chosen": -0.985150158405304, + "logps/weighted_rejected": -1.1464660167694092, + "loss": 0.6605, + "rewards/accuracies": 0.596875011920929, + "rewards/chosen": -76.65937805175781, + "rewards/margins": 9.346484184265137, + "rewards/rejected": -86.00468444824219, + "rewards/weighted_accuracies": 0.606249988079071, + "rewards/weighted_chosen": -0.3394966125488281, + "rewards/weighted_margins": 0.15953750908374786, + "rewards/weighted_rejected": -0.4988296627998352, + "step": 1640 + }, + { + "epoch": 0.8818224330282584, + "grad_norm": 8.658909797668457, + "learning_rate": 4.2647402441394534e-08, + "logits/chosen": -0.00011138916306663305, + "logits/rejected": 0.032086946070194244, + "logps/chosen": -195.4421844482422, + "logps/rejected": -208.1671905517578, + "logps/weighted_chosen": -1.0746338367462158, + "logps/weighted_rejected": -1.328485131263733, + "loss": 0.6769, + "rewards/accuracies": 0.578125, + "rewards/chosen": -81.19999694824219, + "rewards/margins": 7.428124904632568, + "rewards/rejected": -88.65937805175781, + "rewards/weighted_accuracies": 0.5218750238418579, + "rewards/weighted_chosen": -0.42961883544921875, + "rewards/weighted_margins": 0.11412658542394638, + "rewards/weighted_rejected": -0.5438331365585327, + "step": 1650 + }, + { + "epoch": 0.8871668114102479, + "grad_norm": 8.569114685058594, + "learning_rate": 3.895765019498609e-08, + "logits/chosen": 0.002704620361328125, + "logits/rejected": 0.06543274223804474, + "logps/chosen": -200.77499389648438, + "logps/rejected": -209.3781280517578, + "logps/weighted_chosen": -1.202539086341858, + "logps/weighted_rejected": -1.7360382080078125, + "loss": 0.6903, + "rewards/accuracies": 0.578125, + "rewards/chosen": -81.5859375, + "rewards/margins": 7.512890815734863, + "rewards/rejected": -89.10624694824219, + "rewards/weighted_accuracies": 0.5562499761581421, + "rewards/weighted_chosen": -0.38261109590530396, + "rewards/weighted_margins": 0.17353057861328125, + "rewards/weighted_rejected": -0.5557746887207031, + "step": 1660 + }, + { + "epoch": 0.8925111897922373, + "grad_norm": 8.67790412902832, + "learning_rate": 3.542834945374412e-08, + "logits/chosen": 0.02303619310259819, + "logits/rejected": 0.04090576246380806, + "logps/chosen": -198.5749969482422, + "logps/rejected": -209.38436889648438, + "logps/weighted_chosen": -0.764483630657196, + "logps/weighted_rejected": -1.1029876470565796, + "loss": 0.6775, + "rewards/accuracies": 0.609375, + "rewards/chosen": -79.39921569824219, + "rewards/margins": 7.441015720367432, + "rewards/rejected": -86.8218765258789, + "rewards/weighted_accuracies": 0.5093749761581421, + "rewards/weighted_chosen": -0.3830154538154602, + "rewards/weighted_margins": 0.15334168076515198, + "rewards/weighted_rejected": -0.5364456176757812, + "step": 1670 + }, + { + "epoch": 0.8978555681742267, + "grad_norm": 21.397300720214844, + "learning_rate": 3.206072848144176e-08, + "logits/chosen": -0.00791015662252903, + "logits/rejected": 0.04454975202679634, + "logps/chosen": -196.8625030517578, + "logps/rejected": -209.47030639648438, + "logps/weighted_chosen": -1.026220679283142, + "logps/weighted_rejected": -1.4850037097930908, + "loss": 0.6715, + "rewards/accuracies": 0.606249988079071, + "rewards/chosen": -79.52656555175781, + "rewards/margins": 9.549219131469727, + "rewards/rejected": -89.0843734741211, + "rewards/weighted_accuracies": 0.5843750238418579, + "rewards/weighted_chosen": -0.3757896423339844, + "rewards/weighted_margins": 0.19904938340187073, + "rewards/weighted_rejected": -0.5747489929199219, + "step": 1680 + }, + { + "epoch": 0.9031999465562162, + "grad_norm": 7.6560378074646, + "learning_rate": 2.8855959274217802e-08, + "logits/chosen": 0.011219787411391735, + "logits/rejected": 0.015446471981704235, + "logps/chosen": -192.77188110351562, + "logps/rejected": -202.75, + "logps/weighted_chosen": -1.6045806407928467, + "logps/weighted_rejected": -1.1945679187774658, + "loss": 0.6618, + "rewards/accuracies": 0.6343749761581421, + "rewards/chosen": -78.68046569824219, + "rewards/margins": 10.335546493530273, + "rewards/rejected": -89.0546875, + "rewards/weighted_accuracies": 0.543749988079071, + "rewards/weighted_chosen": -0.3234237730503082, + "rewards/weighted_margins": 0.20303955674171448, + "rewards/weighted_rejected": -0.5267227292060852, + "step": 1690 + }, + { + "epoch": 0.9085443249382056, + "grad_norm": 7.867788314819336, + "learning_rate": 2.5815157152700106e-08, + "logits/chosen": -0.0039764405228197575, + "logits/rejected": 0.052965544164180756, + "logps/chosen": -198.703125, + "logps/rejected": -227.3562469482422, + "logps/weighted_chosen": -1.1858246326446533, + "logps/weighted_rejected": -1.7776000499725342, + "loss": 0.6681, + "rewards/accuracies": 0.628125011920929, + "rewards/chosen": -81.5078125, + "rewards/margins": 25.093358993530273, + "rewards/rejected": -106.625, + "rewards/weighted_accuracies": 0.5625, + "rewards/weighted_chosen": -0.3344528079032898, + "rewards/weighted_margins": 0.23516464233398438, + "rewards/weighted_rejected": -0.5694076418876648, + "step": 1700 + }, + { + "epoch": 0.9138887033201951, + "grad_norm": 31.351016998291016, + "learning_rate": 2.2939380373852513e-08, + "logits/chosen": 0.04047546535730362, + "logits/rejected": 0.057331085205078125, + "logps/chosen": -192.50155639648438, + "logps/rejected": -199.328125, + "logps/weighted_chosen": -1.187780737876892, + "logps/weighted_rejected": -1.1523864269256592, + "loss": 0.7191, + "rewards/accuracies": 0.628125011920929, + "rewards/chosen": -80.76875305175781, + "rewards/margins": 6.9296875, + "rewards/rejected": -87.68281555175781, + "rewards/weighted_accuracies": 0.565625011920929, + "rewards/weighted_chosen": -0.4314765930175781, + "rewards/weighted_margins": 0.061667632311582565, + "rewards/weighted_rejected": -0.4930679202079773, + "step": 1710 + }, + { + "epoch": 0.9192330817021845, + "grad_norm": 12.911133766174316, + "learning_rate": 2.0229629762681334e-08, + "logits/chosen": -0.0029060363303869963, + "logits/rejected": 0.054146576672792435, + "logps/chosen": -202.1999969482422, + "logps/rejected": -203.1953125, + "logps/weighted_chosen": -1.305535912513733, + "logps/weighted_rejected": -1.5592772960662842, + "loss": 0.6688, + "rewards/accuracies": 0.6468750238418579, + "rewards/chosen": -80.04765319824219, + "rewards/margins": 6.839453220367432, + "rewards/rejected": -86.9046859741211, + "rewards/weighted_accuracies": 0.565625011920929, + "rewards/weighted_chosen": -0.3944839537143707, + "rewards/weighted_margins": 0.15517273545265198, + "rewards/weighted_rejected": -0.5497649908065796, + "step": 1720 + }, + { + "epoch": 0.924577460084174, + "grad_norm": 7.706750869750977, + "learning_rate": 1.7686848363929363e-08, + "logits/chosen": -0.020279694348573685, + "logits/rejected": 0.029271889477968216, + "logps/chosen": -200.734375, + "logps/rejected": -208.2390594482422, + "logps/weighted_chosen": -0.884796142578125, + "logps/weighted_rejected": -1.5206177234649658, + "loss": 0.6686, + "rewards/accuracies": 0.5562499761581421, + "rewards/chosen": -79.7640609741211, + "rewards/margins": 7.583203315734863, + "rewards/rejected": -87.37187194824219, + "rewards/weighted_accuracies": 0.5625, + "rewards/weighted_chosen": -0.4030723571777344, + "rewards/weighted_margins": 0.17832335829734802, + "rewards/weighted_rejected": -0.5815140008926392, + "step": 1730 + }, + { + "epoch": 0.9299218384661634, + "grad_norm": 8.226058006286621, + "learning_rate": 1.5311921113878355e-08, + "logits/chosen": -0.0008687973022460938, + "logits/rejected": 0.028296660631895065, + "logps/chosen": -194.00155639648438, + "logps/rejected": -203.703125, + "logps/weighted_chosen": -1.365270972251892, + "logps/weighted_rejected": -1.539056420326233, + "loss": 0.6705, + "rewards/accuracies": 0.621874988079071, + "rewards/chosen": -77.375, + "rewards/margins": 8.365625381469727, + "rewards/rejected": -85.73750305175781, + "rewards/weighted_accuracies": 0.5562499761581421, + "rewards/weighted_chosen": -0.45653611421585083, + "rewards/weighted_margins": 0.19713135063648224, + "rewards/weighted_rejected": -0.6534255743026733, + "step": 1740 + }, + { + "epoch": 0.9352662168481528, + "grad_norm": 21.11993408203125, + "learning_rate": 1.3105674532374212e-08, + "logits/chosen": 0.022278213873505592, + "logits/rejected": 0.06085510179400444, + "logps/chosen": -189.58438110351562, + "logps/rejected": -204.9343719482422, + "logps/weighted_chosen": -0.7065078616142273, + "logps/weighted_rejected": -1.562200903892517, + "loss": 0.6816, + "rewards/accuracies": 0.6000000238418579, + "rewards/chosen": -78.2125015258789, + "rewards/margins": 9.004687309265137, + "rewards/rejected": -87.2289047241211, + "rewards/weighted_accuracies": 0.5375000238418579, + "rewards/weighted_chosen": -0.3986854553222656, + "rewards/weighted_margins": 0.12375183403491974, + "rewards/weighted_rejected": -0.5223197937011719, + "step": 1750 + }, + { + "epoch": 0.9406105952301423, + "grad_norm": 7.333647727966309, + "learning_rate": 1.106887643518256e-08, + "logits/chosen": -0.011242675594985485, + "logits/rejected": -0.014645385555922985, + "logps/chosen": -207.7468719482422, + "logps/rejected": -222.8937530517578, + "logps/weighted_chosen": -1.1957489252090454, + "logps/weighted_rejected": -1.2758667469024658, + "loss": 0.6751, + "rewards/accuracies": 0.621874988079071, + "rewards/chosen": -82.3929672241211, + "rewards/margins": 12.168749809265137, + "rewards/rejected": -94.54219055175781, + "rewards/weighted_accuracies": 0.5687500238418579, + "rewards/weighted_chosen": -0.3890426754951477, + "rewards/weighted_margins": 0.12043914943933487, + "rewards/weighted_rejected": -0.5094535946846008, + "step": 1760 + }, + { + "epoch": 0.9459549736121318, + "grad_norm": 6.959185600280762, + "learning_rate": 9.202235666774027e-09, + "logits/chosen": -0.015836333855986595, + "logits/rejected": -0.0019592284224927425, + "logps/chosen": -208.76406860351562, + "logps/rejected": -219.86874389648438, + "logps/weighted_chosen": -1.0631835460662842, + "logps/weighted_rejected": -1.0057373046875, + "loss": 0.704, + "rewards/accuracies": 0.6312500238418579, + "rewards/chosen": -85.24140930175781, + "rewards/margins": 8.126562118530273, + "rewards/rejected": -93.36405944824219, + "rewards/weighted_accuracies": 0.565625011920929, + "rewards/weighted_chosen": -0.43046340346336365, + "rewards/weighted_margins": 0.13985595107078552, + "rewards/weighted_rejected": -0.5703773498535156, + "step": 1770 + }, + { + "epoch": 0.9512993519941212, + "grad_norm": 18.106613159179688, + "learning_rate": 7.50640185363305e-09, + "logits/chosen": 0.01944122277200222, + "logits/rejected": 0.0510406494140625, + "logps/chosen": -201.4640655517578, + "logps/rejected": -213.0968780517578, + "logps/weighted_chosen": -1.5302276611328125, + "logps/weighted_rejected": -1.740319848060608, + "loss": 0.699, + "rewards/accuracies": 0.596875011920929, + "rewards/chosen": -82.03047180175781, + "rewards/margins": 7.597265720367432, + "rewards/rejected": -89.6078109741211, + "rewards/weighted_accuracies": 0.534375011920929, + "rewards/weighted_chosen": -0.37673646211624146, + "rewards/weighted_margins": 0.13721999526023865, + "rewards/weighted_rejected": -0.5135101079940796, + "step": 1780 + }, + { + "epoch": 0.9566437303761106, + "grad_norm": 11.966142654418945, + "learning_rate": 5.981965178175563e-09, + "logits/chosen": -0.07247161865234375, + "logits/rejected": -0.02783966064453125, + "logps/chosen": -205.18124389648438, + "logps/rejected": -219.86874389648438, + "logps/weighted_chosen": -0.6668335199356079, + "logps/weighted_rejected": -1.101470947265625, + "loss": 0.6685, + "rewards/accuracies": 0.621874988079071, + "rewards/chosen": -85.0562515258789, + "rewards/margins": 10.532031059265137, + "rewards/rejected": -95.59062194824219, + "rewards/weighted_accuracies": 0.590624988079071, + "rewards/weighted_chosen": -0.32790833711624146, + "rewards/weighted_margins": 0.19003143906593323, + "rewards/weighted_rejected": -0.517895519733429, + "step": 1790 + }, + { + "epoch": 0.9619881087581, + "grad_norm": 9.166258811950684, + "learning_rate": 4.629456173354196e-09, + "logits/chosen": 0.03516960144042969, + "logits/rejected": 0.053380586206912994, + "logps/chosen": -197.13436889648438, + "logps/rejected": -208.0656280517578, + "logps/weighted_chosen": -1.8832824230194092, + "logps/weighted_rejected": -1.6900818347930908, + "loss": 0.6711, + "rewards/accuracies": 0.625, + "rewards/chosen": -79.5218734741211, + "rewards/margins": 8.657422065734863, + "rewards/rejected": -88.18281555175781, + "rewards/weighted_accuracies": 0.609375, + "rewards/weighted_chosen": -0.3943443298339844, + "rewards/weighted_margins": 0.23465195298194885, + "rewards/weighted_rejected": -0.6290283203125, + "step": 1800 + }, + { + "epoch": 0.9673324871400896, + "grad_norm": 7.5731706619262695, + "learning_rate": 3.4493455380225634e-09, + "logits/chosen": -0.02142333984375, + "logits/rejected": 0.01344223041087389, + "logps/chosen": -208.6281280517578, + "logps/rejected": -217.1765594482422, + "logps/weighted_chosen": -0.970935046672821, + "logps/weighted_rejected": -1.060217261314392, + "loss": 0.6414, + "rewards/accuracies": 0.6312500238418579, + "rewards/chosen": -80.3062515258789, + "rewards/margins": 11.758593559265137, + "rewards/rejected": -92.08125305175781, + "rewards/weighted_accuracies": 0.596875011920929, + "rewards/weighted_chosen": -0.3379959166049957, + "rewards/weighted_margins": 0.2506675720214844, + "rewards/weighted_rejected": -0.5888374447822571, + "step": 1810 + }, + { + "epoch": 0.972676865522079, + "grad_norm": 10.392830848693848, + "learning_rate": 2.4420439731234754e-09, + "logits/chosen": -0.02555542066693306, + "logits/rejected": 0.033538818359375, + "logps/chosen": -211.2375030517578, + "logps/rejected": -217.2218780517578, + "logps/weighted_chosen": -1.076715111732483, + "logps/weighted_rejected": -1.1117675304412842, + "loss": 0.658, + "rewards/accuracies": 0.581250011920929, + "rewards/chosen": -86.1390609741211, + "rewards/margins": 7.975390434265137, + "rewards/rejected": -94.12968444824219, + "rewards/weighted_accuracies": 0.6031249761581421, + "rewards/weighted_chosen": -0.387350469827652, + "rewards/weighted_margins": 0.1678314208984375, + "rewards/weighted_rejected": -0.5549789667129517, + "step": 1820 + }, + { + "epoch": 0.9780212439040684, + "grad_norm": 14.731633186340332, + "learning_rate": 1.607902038756881e-09, + "logits/chosen": -0.03051147423684597, + "logits/rejected": 0.03038787841796875, + "logps/chosen": -201.2140655517578, + "logps/rejected": -211.74063110351562, + "logps/weighted_chosen": -1.221441626548767, + "logps/weighted_rejected": -1.850244164466858, + "loss": 0.698, + "rewards/accuracies": 0.6156250238418579, + "rewards/chosen": -82.11015319824219, + "rewards/margins": 7.809765815734863, + "rewards/rejected": -89.9296875, + "rewards/weighted_accuracies": 0.5218750238418579, + "rewards/weighted_chosen": -0.4083351194858551, + "rewards/weighted_margins": 0.17291259765625, + "rewards/weighted_rejected": -0.581036388874054, + "step": 1830 + }, + { + "epoch": 0.9833656222860578, + "grad_norm": 16.210765838623047, + "learning_rate": 9.472100321785137e-10, + "logits/chosen": 0.01699371263384819, + "logits/rejected": 0.042603302747011185, + "logps/chosen": -193.3234405517578, + "logps/rejected": -205.1609344482422, + "logps/weighted_chosen": -1.5597137212753296, + "logps/weighted_rejected": -1.877203345298767, + "loss": 0.6571, + "rewards/accuracies": 0.6031249761581421, + "rewards/chosen": -80.43437194824219, + "rewards/margins": 10.099609375, + "rewards/rejected": -90.53593444824219, + "rewards/weighted_accuracies": 0.5531250238418579, + "rewards/weighted_chosen": -0.4782699644565582, + "rewards/weighted_margins": 0.19840240478515625, + "rewards/weighted_rejected": -0.6767608523368835, + "step": 1840 + }, + { + "epoch": 0.9887100006680473, + "grad_norm": 15.271677017211914, + "learning_rate": 4.601978867705969e-10, + "logits/chosen": 0.00860519427806139, + "logits/rejected": 0.059200286865234375, + "logps/chosen": -193.88436889648438, + "logps/rejected": -201.98593139648438, + "logps/weighted_chosen": -1.474829077720642, + "logps/weighted_rejected": -1.616754174232483, + "loss": 0.6632, + "rewards/accuracies": 0.596875011920929, + "rewards/chosen": -78.921875, + "rewards/margins": 7.835546970367432, + "rewards/rejected": -86.7671890258789, + "rewards/weighted_accuracies": 0.6000000238418579, + "rewards/weighted_chosen": -0.33439406752586365, + "rewards/weighted_margins": 0.28738290071487427, + "rewards/weighted_rejected": -0.6216728091239929, + "step": 1850 + }, + { + "epoch": 0.9940543790500368, + "grad_norm": 37.25553512573242, + "learning_rate": 1.4703509202118604e-10, + "logits/chosen": -0.05092964321374893, + "logits/rejected": -0.014559936709702015, + "logps/chosen": -205.4250030517578, + "logps/rejected": -215.70938110351562, + "logps/weighted_chosen": -0.953631579875946, + "logps/weighted_rejected": -1.1918243169784546, + "loss": 0.697, + "rewards/accuracies": 0.609375, + "rewards/chosen": -83.56562805175781, + "rewards/margins": 7.594531059265137, + "rewards/rejected": -91.1812515258789, + "rewards/weighted_accuracies": 0.5375000238418579, + "rewards/weighted_chosen": -0.3130737245082855, + "rewards/weighted_margins": 0.1581375151872635, + "rewards/weighted_rejected": -0.47164687514305115, + "step": 1860 + }, + { + "epoch": 0.9993987574320262, + "grad_norm": 7.526350021362305, + "learning_rate": 7.830634538019686e-12, + "logits/chosen": -0.017978286370635033, + "logits/rejected": 0.02026519738137722, + "logps/chosen": -205.7078094482422, + "logps/rejected": -213.390625, + "logps/weighted_chosen": -1.1760375499725342, + "logps/weighted_rejected": -1.36285400390625, + "loss": 0.6788, + "rewards/accuracies": 0.612500011920929, + "rewards/chosen": -81.91874694824219, + "rewards/margins": 8.227734565734863, + "rewards/rejected": -90.15156555175781, + "rewards/weighted_accuracies": 0.596875011920929, + "rewards/weighted_chosen": -0.3904052674770355, + "rewards/weighted_margins": 0.1767173707485199, + "rewards/weighted_rejected": -0.5669296383857727, + "step": 1870 + } + ], + "logging_steps": 10, + "max_steps": 1872, + "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-1872/training_args.bin b/checkpoint-1872/training_args.bin new file mode 100644 index 0000000..7b414b2 --- /dev/null +++ b/checkpoint-1872/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fa2dcf6fe65fd94a0fde469e207f237b0752c6e558745acec2daa0276a1175ce +size 8785 diff --git a/checkpoint-1872/zero_to_fp32.py b/checkpoint-1872/zero_to_fp32.py new file mode 100644 index 0000000..0e75914 --- /dev/null +++ b/checkpoint-1872/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..ceed8ae --- /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": 128009, + "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..38e1b41 --- /dev/null +++ b/checkpoint-500/generation_config.json @@ -0,0 +1,12 @@ +{ + "bos_token_id": 128000, + "do_sample": true, + "eos_token_id": [ + 128001, + 128009 + ], + "max_length": 4096, + "temperature": 0.6, + "top_p": 0.9, + "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..e81786f --- /dev/null +++ b/checkpoint-500/model-00001-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dfc821f62ba385c986b784fb763fe22ca0d7effd194474c8fd4a3199ae86ae35 +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..70055b9 --- /dev/null +++ b/checkpoint-500/model-00002-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d1f7ef6be4563bf511b2af3b783edbd94200100fa1753a6ccbdaaa53cea6a6b3 +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..7c3160c --- /dev/null +++ b/checkpoint-500/model-00003-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d15ae769d4e28adbdda67812c2b9cb888f12623d9a1c18c12ae4d9abccf20cd9 +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..63e90dd --- /dev/null +++ b/checkpoint-500/model-00004-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:19e72126c98b268cb7d5c3e26a1c0c3075e6e144d1bad12bf8110ac1e2741541 +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..07a2dab --- /dev/null +++ b/checkpoint-500/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d9b2b8de739e57a2f4df669edfa7e2fe815a983332153866972a4acf1d8d50ab +size 1465 diff --git a/checkpoint-500/special_tokens_map.json b/checkpoint-500/special_tokens_map.json new file mode 100644 index 0000000..b43be96 --- /dev/null +++ b/checkpoint-500/special_tokens_map.json @@ -0,0 +1,17 @@ +{ + "bos_token": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "eos_token": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "pad_token": "<|eot_id|>" +} diff --git a/checkpoint-500/tokenizer.json b/checkpoint-500/tokenizer.json new file mode 100644 index 0000000..86a3394 --- /dev/null +++ b/checkpoint-500/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3c5cf44023714fb39b05e71e425f8d7b92805ff73f7988b083b8c87f0bf87393 +size 17209961 diff --git a/checkpoint-500/tokenizer_config.json b/checkpoint-500/tokenizer_config.json new file mode 100644 index 0000000..34d134f --- /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": "<|eot_id|>", + "extra_special_tokens": {}, + "model_input_names": [ + "input_ids", + "attention_mask" + ], + "model_max_length": 1000000000000000019884624838656, + "pad_token": "<|eot_id|>", + "tokenizer_class": "PreTrainedTokenizerFast" +} diff --git a/checkpoint-500/trainer_state.json b/checkpoint-500/trainer_state.json new file mode 100644 index 0000000..345207b --- /dev/null +++ b/checkpoint-500/trainer_state.json @@ -0,0 +1,1127 @@ +{ + "best_global_step": null, + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 0.26721891909947226, + "eval_steps": 500, + "global_step": 500, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.0005344378381989445, + "grad_norm": 63.52920150756836, + "learning_rate": 0.0, + "logits/chosen": -0.144775390625, + "logits/rejected": -0.1036376953125, + "logps/chosen": -110.828125, + "logps/rejected": -116.828125, + "logps/weighted_chosen": -0.2086181640625, + "logps/weighted_rejected": -1.271392822265625, + "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": 0.005344378381989445, + "grad_norm": 102.77810668945312, + "learning_rate": 4.787234042553191e-08, + "logits/chosen": -0.2549913227558136, + "logits/rejected": -0.24070316553115845, + "logps/chosen": -135.32899475097656, + "logps/rejected": -138.99305725097656, + "logps/weighted_chosen": -0.8547499179840088, + "logps/weighted_rejected": -0.9027608036994934, + "loss": 0.6928, + "rewards/accuracies": 0.2951388955116272, + "rewards/chosen": -0.0164930559694767, + "rewards/margins": -0.0349392369389534, + "rewards/rejected": 0.0184461809694767, + "rewards/weighted_accuracies": 0.4027777910232544, + "rewards/weighted_chosen": -0.002239651046693325, + "rewards/weighted_margins": 0.0028160943184047937, + "rewards/weighted_rejected": -0.0050794812850654125, + "step": 10 + }, + { + "epoch": 0.01068875676397889, + "grad_norm": 9.201544761657715, + "learning_rate": 1.0106382978723404e-07, + "logits/chosen": -0.19488906860351562, + "logits/rejected": -0.183399960398674, + "logps/chosen": -116.0374984741211, + "logps/rejected": -115.5054702758789, + "logps/weighted_chosen": -0.6929916143417358, + "logps/weighted_rejected": -0.6838119626045227, + "loss": 0.6917, + "rewards/accuracies": 0.38749998807907104, + "rewards/chosen": 0.01796874962747097, + "rewards/margins": 0.03007812425494194, + "rewards/rejected": -0.01210937462747097, + "rewards/weighted_accuracies": 0.4781250059604645, + "rewards/weighted_chosen": 0.0022293091751635075, + "rewards/weighted_margins": 0.00339679722674191, + "rewards/weighted_rejected": -0.0011629104847088456, + "step": 20 + }, + { + "epoch": 0.016033135145968335, + "grad_norm": 7.109442234039307, + "learning_rate": 1.5425531914893615e-07, + "logits/chosen": -0.1776607483625412, + "logits/rejected": -0.17884215712547302, + "logps/chosen": -123.2867202758789, + "logps/rejected": -129.75, + "logps/weighted_chosen": -0.7567901611328125, + "logps/weighted_rejected": -1.2207733392715454, + "loss": 0.6913, + "rewards/accuracies": 0.3968749940395355, + "rewards/chosen": -0.0074218749068677425, + "rewards/margins": 0.04843749850988388, + "rewards/rejected": -0.05585937574505806, + "rewards/weighted_accuracies": 0.5, + "rewards/weighted_chosen": 0.0007101058727130294, + "rewards/weighted_margins": 0.004384422209113836, + "rewards/weighted_rejected": -0.0036745071411132812, + "step": 30 + }, + { + "epoch": 0.02137751352795778, + "grad_norm": 6.0048956871032715, + "learning_rate": 2.074468085106383e-07, + "logits/chosen": -0.23842163383960724, + "logits/rejected": -0.2298881560564041, + "logps/chosen": -127.34062194824219, + "logps/rejected": -128.1710968017578, + "logps/weighted_chosen": -0.8051223754882812, + "logps/weighted_rejected": -0.5736129879951477, + "loss": 0.6934, + "rewards/accuracies": 0.34062498807907104, + "rewards/chosen": -0.01992187462747097, + "rewards/margins": -0.05351562425494194, + "rewards/rejected": 0.03359375149011612, + "rewards/weighted_accuracies": 0.4593749940395355, + "rewards/weighted_chosen": -7.83920258982107e-05, + "rewards/weighted_margins": -0.0009832382202148438, + "rewards/weighted_rejected": 0.000904083251953125, + "step": 40 + }, + { + "epoch": 0.026721891909947223, + "grad_norm": 32.7233772277832, + "learning_rate": 2.6063829787234044e-07, + "logits/chosen": -0.18835143744945526, + "logits/rejected": -0.170085147023201, + "logps/chosen": -120.88594055175781, + "logps/rejected": -121.3515625, + "logps/weighted_chosen": -0.6867736577987671, + "logps/weighted_rejected": -1.138249158859253, + "loss": 0.6903, + "rewards/accuracies": 0.3656249940395355, + "rewards/chosen": 0.03554687649011612, + "rewards/margins": -0.03164062649011612, + "rewards/rejected": 0.06718750298023224, + "rewards/weighted_accuracies": 0.515625, + "rewards/weighted_chosen": -0.006287765689194202, + "rewards/weighted_margins": 0.0064636231400072575, + "rewards/weighted_rejected": -0.01274185162037611, + "step": 50 + }, + { + "epoch": 0.03206627029193667, + "grad_norm": 55.560707092285156, + "learning_rate": 3.1382978723404253e-07, + "logits/chosen": -0.1308601349592209, + "logits/rejected": -0.1308906525373459, + "logps/chosen": -116.5875015258789, + "logps/rejected": -119.5406265258789, + "logps/weighted_chosen": -0.8019469976425171, + "logps/weighted_rejected": -0.8834930658340454, + "loss": 0.7002, + "rewards/accuracies": 0.3343749940395355, + "rewards/chosen": 0.03476562350988388, + "rewards/margins": -0.06171875074505806, + "rewards/rejected": 0.09648437798023224, + "rewards/weighted_accuracies": 0.4781250059604645, + "rewards/weighted_chosen": -0.01903533935546875, + "rewards/weighted_margins": -2.880096508306451e-05, + "rewards/weighted_rejected": -0.019007300958037376, + "step": 60 + }, + { + "epoch": 0.037410648673926114, + "grad_norm": 20.22484588623047, + "learning_rate": 3.6702127659574467e-07, + "logits/chosen": -0.20572929084300995, + "logits/rejected": -0.19457244873046875, + "logps/chosen": -114.4507827758789, + "logps/rejected": -118.63594055175781, + "logps/weighted_chosen": -0.6070801019668579, + "logps/weighted_rejected": -0.6768280267715454, + "loss": 0.6911, + "rewards/accuracies": 0.42500001192092896, + "rewards/chosen": 0.16484375298023224, + "rewards/margins": 0.07148437201976776, + "rewards/rejected": 0.09335937350988388, + "rewards/weighted_accuracies": 0.4749999940395355, + "rewards/weighted_chosen": -0.010281562805175781, + "rewards/weighted_margins": 0.007259177975356579, + "rewards/weighted_rejected": -0.017521286383271217, + "step": 70 + }, + { + "epoch": 0.04275502705591556, + "grad_norm": 15.305061340332031, + "learning_rate": 4.202127659574468e-07, + "logits/chosen": -0.19508513808250427, + "logits/rejected": -0.12093047797679901, + "logps/chosen": -112.57734680175781, + "logps/rejected": -112.77734375, + "logps/weighted_chosen": -0.4597320556640625, + "logps/weighted_rejected": -0.9711883664131165, + "loss": 0.6952, + "rewards/accuracies": 0.37812501192092896, + "rewards/chosen": 0.04570312425494194, + "rewards/margins": 0.01914062537252903, + "rewards/rejected": 0.02656250074505806, + "rewards/weighted_accuracies": 0.4625000059604645, + "rewards/weighted_chosen": -0.02759857103228569, + "rewards/weighted_margins": 0.021048355847597122, + "rewards/weighted_rejected": -0.04872245714068413, + "step": 80 + }, + { + "epoch": 0.048099405437905, + "grad_norm": 11.991273880004883, + "learning_rate": 4.734042553191489e-07, + "logits/chosen": -0.17391815781593323, + "logits/rejected": -0.11656417697668076, + "logps/chosen": -119.3070297241211, + "logps/rejected": -118.51249694824219, + "logps/weighted_chosen": -1.0416854619979858, + "logps/weighted_rejected": -1.5859466791152954, + "loss": 0.6736, + "rewards/accuracies": 0.4375, + "rewards/chosen": -0.08164062350988388, + "rewards/margins": 0.341796875, + "rewards/rejected": -0.4234375059604645, + "rewards/weighted_accuracies": 0.515625, + "rewards/weighted_chosen": -0.04084587097167969, + "rewards/weighted_margins": 0.10567589104175568, + "rewards/weighted_rejected": -0.14661255478858948, + "step": 90 + }, + { + "epoch": 0.053443783819894446, + "grad_norm": 17.195886611938477, + "learning_rate": 5.26595744680851e-07, + "logits/chosen": -0.204203799366951, + "logits/rejected": -0.14691162109375, + "logps/chosen": -125.05155944824219, + "logps/rejected": -125.3671875, + "logps/weighted_chosen": -0.7374237179756165, + "logps/weighted_rejected": -0.7348953485488892, + "loss": 0.7064, + "rewards/accuracies": 0.4281249940395355, + "rewards/chosen": -0.3832031190395355, + "rewards/margins": 0.15585938096046448, + "rewards/rejected": -0.5390625, + "rewards/weighted_accuracies": 0.46875, + "rewards/weighted_chosen": -0.060018159449100494, + "rewards/weighted_margins": 0.0024120330344885588, + "rewards/weighted_rejected": -0.06242714077234268, + "step": 100 + }, + { + "epoch": 0.058788162201883896, + "grad_norm": 9.103920936584473, + "learning_rate": 5.797872340425531e-07, + "logits/chosen": -0.2138931304216385, + "logits/rejected": -0.19011077284812927, + "logps/chosen": -118.6820297241211, + "logps/rejected": -121.28593444824219, + "logps/weighted_chosen": -0.4821228086948395, + "logps/weighted_rejected": -0.7759445309638977, + "loss": 0.6939, + "rewards/accuracies": 0.4281249940395355, + "rewards/chosen": -0.896484375, + "rewards/margins": 0.09531249850988388, + "rewards/rejected": -0.9917968511581421, + "rewards/weighted_accuracies": 0.528124988079071, + "rewards/weighted_chosen": -0.05413341522216797, + "rewards/weighted_margins": 0.04503212124109268, + "rewards/weighted_rejected": -0.09918193519115448, + "step": 110 + }, + { + "epoch": 0.06413254058387334, + "grad_norm": 8.539896011352539, + "learning_rate": 6.329787234042553e-07, + "logits/chosen": -0.17510680854320526, + "logits/rejected": -0.14726562798023224, + "logps/chosen": -122.58906555175781, + "logps/rejected": -128.4656219482422, + "logps/weighted_chosen": -0.9249496459960938, + "logps/weighted_rejected": -0.762683093547821, + "loss": 0.7325, + "rewards/accuracies": 0.4375, + "rewards/chosen": -1.068359375, + "rewards/margins": 0.28398436307907104, + "rewards/rejected": -1.3523437976837158, + "rewards/weighted_accuracies": 0.4781250059604645, + "rewards/weighted_chosen": -0.098027803003788, + "rewards/weighted_margins": -0.029563140124082565, + "rewards/weighted_rejected": -0.06836700439453125, + "step": 120 + }, + { + "epoch": 0.06947691896586278, + "grad_norm": 15.564044952392578, + "learning_rate": 6.861702127659574e-07, + "logits/chosen": -0.20233002305030823, + "logits/rejected": -0.18396148085594177, + "logps/chosen": -122.2671890258789, + "logps/rejected": -123.8382797241211, + "logps/weighted_chosen": -0.7501281499862671, + "logps/weighted_rejected": -0.821154773235321, + "loss": 0.7078, + "rewards/accuracies": 0.4000000059604645, + "rewards/chosen": -1.45703125, + "rewards/margins": 0.29179686307907104, + "rewards/rejected": -1.7488281726837158, + "rewards/weighted_accuracies": 0.5062500238418579, + "rewards/weighted_chosen": -0.007884788326919079, + "rewards/weighted_margins": 0.006314086727797985, + "rewards/weighted_rejected": -0.014200759120285511, + "step": 130 + }, + { + "epoch": 0.07482129734785223, + "grad_norm": 9.176558494567871, + "learning_rate": 7.393617021276596e-07, + "logits/chosen": -0.16455383598804474, + "logits/rejected": -0.14660033583641052, + "logps/chosen": -116.328125, + "logps/rejected": -119.9664077758789, + "logps/weighted_chosen": -0.9873703122138977, + "logps/weighted_rejected": -0.6730011105537415, + "loss": 0.7174, + "rewards/accuracies": 0.4749999940395355, + "rewards/chosen": -1.6007812023162842, + "rewards/margins": 0.533203125, + "rewards/rejected": -2.133984327316284, + "rewards/weighted_accuracies": 0.515625, + "rewards/weighted_chosen": 0.02084655687212944, + "rewards/weighted_margins": 0.05102996900677681, + "rewards/weighted_rejected": -0.030106354504823685, + "step": 140 + }, + { + "epoch": 0.08016567572984168, + "grad_norm": 10.172534942626953, + "learning_rate": 7.925531914893616e-07, + "logits/chosen": -0.1509498655796051, + "logits/rejected": -0.13156357407569885, + "logps/chosen": -120.0562515258789, + "logps/rejected": -117.85546875, + "logps/weighted_chosen": -0.8198806643486023, + "logps/weighted_rejected": -0.7782562375068665, + "loss": 0.7005, + "rewards/accuracies": 0.46562498807907104, + "rewards/chosen": -2.0234375, + "rewards/margins": 0.607421875, + "rewards/rejected": -2.630859375, + "rewards/weighted_accuracies": 0.543749988079071, + "rewards/weighted_chosen": -0.0781441181898117, + "rewards/weighted_margins": 0.034273337572813034, + "rewards/weighted_rejected": -0.11248989403247833, + "step": 150 + }, + { + "epoch": 0.08551005411183112, + "grad_norm": 25.879701614379883, + "learning_rate": 8.457446808510637e-07, + "logits/chosen": -0.17474937438964844, + "logits/rejected": -0.17009583115577698, + "logps/chosen": -121.2367172241211, + "logps/rejected": -121.62109375, + "logps/weighted_chosen": -0.8304061889648438, + "logps/weighted_rejected": -0.866925060749054, + "loss": 0.7072, + "rewards/accuracies": 0.4781250059604645, + "rewards/chosen": -2.219531297683716, + "rewards/margins": 0.716796875, + "rewards/rejected": -2.936328172683716, + "rewards/weighted_accuracies": 0.47187501192092896, + "rewards/weighted_chosen": -0.07932891696691513, + "rewards/weighted_margins": 0.017035294324159622, + "rewards/weighted_rejected": -0.09644661098718643, + "step": 160 + }, + { + "epoch": 0.09085443249382057, + "grad_norm": 9.220558166503906, + "learning_rate": 8.989361702127659e-07, + "logits/chosen": -0.22574539482593536, + "logits/rejected": -0.19882050156593323, + "logps/chosen": -120.62968444824219, + "logps/rejected": -124.78437805175781, + "logps/weighted_chosen": -0.9401000738143921, + "logps/weighted_rejected": -0.8424423336982727, + "loss": 0.7174, + "rewards/accuracies": 0.4937500059604645, + "rewards/chosen": -2.4033203125, + "rewards/margins": 0.613085925579071, + "rewards/rejected": -3.016406297683716, + "rewards/weighted_accuracies": 0.48750001192092896, + "rewards/weighted_chosen": -0.04837837070226669, + "rewards/weighted_margins": 0.04595222324132919, + "rewards/weighted_rejected": -0.09437103569507599, + "step": 170 + }, + { + "epoch": 0.09619881087581, + "grad_norm": 55.62498092651367, + "learning_rate": 9.52127659574468e-07, + "logits/chosen": -0.2216789275407791, + "logits/rejected": -0.20373383164405823, + "logps/chosen": -120.3851547241211, + "logps/rejected": -121.2328109741211, + "logps/weighted_chosen": -0.435232549905777, + "logps/weighted_rejected": -0.5373809933662415, + "loss": 0.7302, + "rewards/accuracies": 0.47187501192092896, + "rewards/chosen": -1.338281273841858, + "rewards/margins": 0.3658203184604645, + "rewards/rejected": -1.7041015625, + "rewards/weighted_accuracies": 0.53125, + "rewards/weighted_chosen": -0.08163909614086151, + "rewards/weighted_margins": -0.008893203921616077, + "rewards/weighted_rejected": -0.07281456142663956, + "step": 180 + }, + { + "epoch": 0.10154318925779945, + "grad_norm": 39.38788986206055, + "learning_rate": 9.99999129927477e-07, + "logits/chosen": -0.19243164360523224, + "logits/rejected": -0.15758666396141052, + "logps/chosen": -121.8960952758789, + "logps/rejected": -123.01094055175781, + "logps/weighted_chosen": -0.953173816204071, + "logps/weighted_rejected": -1.0853087902069092, + "loss": 0.6702, + "rewards/accuracies": 0.48750001192092896, + "rewards/chosen": -2.100390672683716, + "rewards/margins": 0.662890613079071, + "rewards/rejected": -2.7632813453674316, + "rewards/weighted_accuracies": 0.578125, + "rewards/weighted_chosen": -0.09999503940343857, + "rewards/weighted_margins": 0.12822990119457245, + "rewards/weighted_rejected": -0.22818031907081604, + "step": 190 + }, + { + "epoch": 0.10688756763978889, + "grad_norm": 12.883384704589844, + "learning_rate": 9.99894724888679e-07, + "logits/chosen": -0.17392578721046448, + "logits/rejected": -0.147491455078125, + "logps/chosen": -129.53750610351562, + "logps/rejected": -128.95703125, + "logps/weighted_chosen": -0.7322753667831421, + "logps/weighted_rejected": -1.0728164911270142, + "loss": 0.6777, + "rewards/accuracies": 0.4749999940395355, + "rewards/chosen": -3.215625047683716, + "rewards/margins": 0.9507812261581421, + "rewards/rejected": -4.166406154632568, + "rewards/weighted_accuracies": 0.5375000238418579, + "rewards/weighted_chosen": -0.1104557067155838, + "rewards/weighted_margins": 0.13128623366355896, + "rewards/weighted_rejected": -0.2415626496076584, + "step": 200 + }, + { + "epoch": 0.11223194602177834, + "grad_norm": 17.96660041809082, + "learning_rate": 9.996163469793475e-07, + "logits/chosen": -0.18083801865577698, + "logits/rejected": -0.12766647338867188, + "logps/chosen": -128.1953125, + "logps/rejected": -123.03047180175781, + "logps/weighted_chosen": -0.8699920773506165, + "logps/weighted_rejected": -0.8845611810684204, + "loss": 0.7572, + "rewards/accuracies": 0.5625, + "rewards/chosen": -3.5113282203674316, + "rewards/margins": 1.3683593273162842, + "rewards/rejected": -4.879296779632568, + "rewards/weighted_accuracies": 0.5531250238418579, + "rewards/weighted_chosen": -0.07919616997241974, + "rewards/weighted_margins": 0.04895629733800888, + "rewards/weighted_rejected": -0.12815475463867188, + "step": 210 + }, + { + "epoch": 0.11757632440376779, + "grad_norm": 63.790077209472656, + "learning_rate": 9.991640930802883e-07, + "logits/chosen": -0.14619751274585724, + "logits/rejected": -0.13860473036766052, + "logps/chosen": -126.61015319824219, + "logps/rejected": -129.9773406982422, + "logps/weighted_chosen": -1.0776779651641846, + "logps/weighted_rejected": -0.9309722781181335, + "loss": 0.7107, + "rewards/accuracies": 0.5531250238418579, + "rewards/chosen": -4.296484470367432, + "rewards/margins": 1.10546875, + "rewards/rejected": -5.40234375, + "rewards/weighted_accuracies": 0.5406249761581421, + "rewards/weighted_chosen": -0.013729858212172985, + "rewards/weighted_margins": 0.14467468857765198, + "rewards/weighted_rejected": -0.1582542359828949, + "step": 220 + }, + { + "epoch": 0.12292070278575723, + "grad_norm": 9.634794235229492, + "learning_rate": 9.98538120584459e-07, + "logits/chosen": -0.1792152374982834, + "logits/rejected": -0.1362098753452301, + "logps/chosen": -133.7140655517578, + "logps/rejected": -137.7546844482422, + "logps/weighted_chosen": -0.6676223874092102, + "logps/weighted_rejected": -0.9518234133720398, + "loss": 0.7614, + "rewards/accuracies": 0.515625, + "rewards/chosen": -4.493554592132568, + "rewards/margins": 1.0623047351837158, + "rewards/rejected": -5.555859565734863, + "rewards/weighted_accuracies": 0.565625011920929, + "rewards/weighted_chosen": -0.1120578795671463, + "rewards/weighted_margins": 0.031700897961854935, + "rewards/weighted_rejected": -0.14376182854175568, + "step": 230 + }, + { + "epoch": 0.12826508116774668, + "grad_norm": 8.717682838439941, + "learning_rate": 9.977386473421917e-07, + "logits/chosen": -0.09295501559972763, + "logits/rejected": -0.07836151123046875, + "logps/chosen": -121.78202819824219, + "logps/rejected": -126.39453125, + "logps/weighted_chosen": -0.8116515874862671, + "logps/weighted_rejected": -1.2165496349334717, + "loss": 0.6731, + "rewards/accuracies": 0.5874999761581421, + "rewards/chosen": -7.553515434265137, + "rewards/margins": 1.669921875, + "rewards/rejected": -9.224218368530273, + "rewards/weighted_accuracies": 0.534375011920929, + "rewards/weighted_chosen": -0.1755084991455078, + "rewards/weighted_margins": 0.15414047241210938, + "rewards/weighted_rejected": -0.32989272475242615, + "step": 240 + }, + { + "epoch": 0.13360945954973613, + "grad_norm": 35.197269439697266, + "learning_rate": 9.96765951585378e-07, + "logits/chosen": -0.021549224853515625, + "logits/rejected": 0.00345611572265625, + "logps/chosen": -123.0296859741211, + "logps/rejected": -127.9671859741211, + "logps/weighted_chosen": -1.186151146888733, + "logps/weighted_rejected": -1.386474609375, + "loss": 0.7026, + "rewards/accuracies": 0.5625, + "rewards/chosen": -9.133593559265137, + "rewards/margins": 1.150781273841858, + "rewards/rejected": -10.283984184265137, + "rewards/weighted_accuracies": 0.59375, + "rewards/weighted_chosen": -0.3282794952392578, + "rewards/weighted_margins": 0.1091407760977745, + "rewards/weighted_rejected": -0.4371795654296875, + "step": 250 + }, + { + "epoch": 0.13895383793172555, + "grad_norm": 84.2197494506836, + "learning_rate": 9.956203718306388e-07, + "logits/chosen": 0.025267411023378372, + "logits/rejected": 0.06335143744945526, + "logps/chosen": -128.8429718017578, + "logps/rejected": -133.50625610351562, + "logps/weighted_chosen": -0.9628280401229858, + "logps/weighted_rejected": -1.2008453607559204, + "loss": 0.7443, + "rewards/accuracies": 0.5, + "rewards/chosen": -10.861719131469727, + "rewards/margins": 0.871874988079071, + "rewards/rejected": -11.733593940734863, + "rewards/weighted_accuracies": 0.49687498807907104, + "rewards/weighted_chosen": -0.26259079575538635, + "rewards/weighted_margins": 0.06770706176757812, + "rewards/weighted_rejected": -0.3305191099643707, + "step": 260 + }, + { + "epoch": 0.144298216313715, + "grad_norm": 57.09628677368164, + "learning_rate": 9.943023067615136e-07, + "logits/chosen": -0.017973709851503372, + "logits/rejected": -0.0057998658157885075, + "logps/chosen": -137.60311889648438, + "logps/rejected": -140.91561889648438, + "logps/weighted_chosen": -0.8694183230400085, + "logps/weighted_rejected": -0.8910156488418579, + "loss": 0.7462, + "rewards/accuracies": 0.5625, + "rewards/chosen": -10.946093559265137, + "rewards/margins": 1.662109375, + "rewards/rejected": -12.608202934265137, + "rewards/weighted_accuracies": 0.565625011920929, + "rewards/weighted_chosen": -0.06702041625976562, + "rewards/weighted_margins": 0.13949890434741974, + "rewards/weighted_rejected": -0.20627327263355255, + "step": 270 + }, + { + "epoch": 0.14964259469570446, + "grad_norm": 45.0294303894043, + "learning_rate": 9.928122150897112e-07, + "logits/chosen": -0.045014191418886185, + "logits/rejected": 0.02437591552734375, + "logps/chosen": -130.1125030517578, + "logps/rejected": -131.6171875, + "logps/weighted_chosen": -0.7659690976142883, + "logps/weighted_rejected": -0.8499618768692017, + "loss": 0.7079, + "rewards/accuracies": 0.5406249761581421, + "rewards/chosen": -12.664453506469727, + "rewards/margins": 1.584375023841858, + "rewards/rejected": -14.249608993530273, + "rewards/weighted_accuracies": 0.543749988079071, + "rewards/weighted_chosen": -0.22265930473804474, + "rewards/weighted_margins": 0.07556991279125214, + "rewards/weighted_rejected": -0.298013299703598, + "step": 280 + }, + { + "epoch": 0.1549869730776939, + "grad_norm": 9.370684623718262, + "learning_rate": 9.9115061539547e-07, + "logits/chosen": -0.02069244347512722, + "logits/rejected": 0.01569976843893528, + "logps/chosen": -139.4453125, + "logps/rejected": -141.2578125, + "logps/weighted_chosen": -0.8775177001953125, + "logps/weighted_rejected": -1.2375701665878296, + "loss": 0.6949, + "rewards/accuracies": 0.5375000238418579, + "rewards/chosen": -13.07421875, + "rewards/margins": 1.177734375, + "rewards/rejected": -14.249608993530273, + "rewards/weighted_accuracies": 0.546875, + "rewards/weighted_chosen": -0.18484345078468323, + "rewards/weighted_margins": 0.12048187106847763, + "rewards/weighted_rejected": -0.30528029799461365, + "step": 290 + }, + { + "epoch": 0.16033135145968336, + "grad_norm": 30.68979263305664, + "learning_rate": 9.893180859470818e-07, + "logits/chosen": -0.0137786865234375, + "logits/rejected": 0.041573334485292435, + "logps/chosen": -130.22265625, + "logps/rejected": -132.06094360351562, + "logps/weighted_chosen": -1.1769134998321533, + "logps/weighted_rejected": -1.2438628673553467, + "loss": 0.7261, + "rewards/accuracies": 0.590624988079071, + "rewards/chosen": -13.093358993530273, + "rewards/margins": 2.108203172683716, + "rewards/rejected": -15.200780868530273, + "rewards/weighted_accuracies": 0.5843750238418579, + "rewards/weighted_chosen": -0.2848831117153168, + "rewards/weighted_margins": 0.07841148227453232, + "rewards/weighted_rejected": -0.36317139863967896, + "step": 300 + }, + { + "epoch": 0.16567572984167278, + "grad_norm": 71.56196594238281, + "learning_rate": 9.873152644996424e-07, + "logits/chosen": -0.11114044487476349, + "logits/rejected": -0.11599807441234589, + "logps/chosen": -131.01718139648438, + "logps/rejected": -132.28125, + "logps/weighted_chosen": -1.1032531261444092, + "logps/weighted_rejected": -0.8420807123184204, + "loss": 0.7872, + "rewards/accuracies": 0.5718749761581421, + "rewards/chosen": -12.979687690734863, + "rewards/margins": 2.407421827316284, + "rewards/rejected": -15.388280868530273, + "rewards/weighted_accuracies": 0.5218750238418579, + "rewards/weighted_chosen": 0.05532989650964737, + "rewards/weighted_margins": 0.2024696320295334, + "rewards/weighted_rejected": -0.1469375640153885, + "step": 310 + }, + { + "epoch": 0.17102010822366223, + "grad_norm": 12.055798530578613, + "learning_rate": 9.85142848073103e-07, + "logits/chosen": -0.10557708889245987, + "logits/rejected": -0.06234588474035263, + "logps/chosen": -133.39218139648438, + "logps/rejected": -134.2109375, + "logps/weighted_chosen": -0.700244128704071, + "logps/weighted_rejected": -1.4902985095977783, + "loss": 0.7558, + "rewards/accuracies": 0.534375011920929, + "rewards/chosen": -14.2265625, + "rewards/margins": 0.6207031011581421, + "rewards/rejected": -14.846094131469727, + "rewards/weighted_accuracies": 0.515625, + "rewards/weighted_chosen": -0.2170669585466385, + "rewards/weighted_margins": 0.03446922451257706, + "rewards/weighted_rejected": -0.25187987089157104, + "step": 320 + }, + { + "epoch": 0.17636448660565168, + "grad_norm": 41.381507873535156, + "learning_rate": 9.828015927096914e-07, + "logits/chosen": -0.085169218480587, + "logits/rejected": -0.05568351596593857, + "logps/chosen": -143.8234405517578, + "logps/rejected": -142.39608764648438, + "logps/weighted_chosen": -0.977459728717804, + "logps/weighted_rejected": -0.7958831787109375, + "loss": 0.6935, + "rewards/accuracies": 0.550000011920929, + "rewards/chosen": -14.643750190734863, + "rewards/margins": 1.578125, + "rewards/rejected": -16.222265243530273, + "rewards/weighted_accuracies": 0.581250011920929, + "rewards/weighted_chosen": -0.16206398606300354, + "rewards/weighted_margins": 0.07153511047363281, + "rewards/weighted_rejected": -0.23361511528491974, + "step": 330 + }, + { + "epoch": 0.18170886498764113, + "grad_norm": 7.313599586486816, + "learning_rate": 9.802923132107968e-07, + "logits/chosen": -0.10784301906824112, + "logits/rejected": -0.052645109593868256, + "logps/chosen": -143.91561889648438, + "logps/rejected": -147.7234344482422, + "logps/weighted_chosen": -0.7267853021621704, + "logps/weighted_rejected": -1.24786376953125, + "loss": 0.7102, + "rewards/accuracies": 0.5562499761581421, + "rewards/chosen": -16.782812118530273, + "rewards/margins": 1.23046875, + "rewards/rejected": -18.012500762939453, + "rewards/weighted_accuracies": 0.5687500238418579, + "rewards/weighted_chosen": -0.1999809294939041, + "rewards/weighted_margins": 0.056002043187618256, + "rewards/weighted_rejected": -0.2561977505683899, + "step": 340 + }, + { + "epoch": 0.18705324336963056, + "grad_norm": 66.15936279296875, + "learning_rate": 9.776158828534024e-07, + "logits/chosen": -0.12608642876148224, + "logits/rejected": -0.09711913764476776, + "logps/chosen": -143.48281860351562, + "logps/rejected": -188.25936889648438, + "logps/weighted_chosen": -1.260833740234375, + "logps/weighted_rejected": -0.9638397097587585, + "loss": 0.7338, + "rewards/accuracies": 0.612500011920929, + "rewards/chosen": -16.727733612060547, + "rewards/margins": 3.159374952316284, + "rewards/rejected": -19.883594512939453, + "rewards/weighted_accuracies": 0.559374988079071, + "rewards/weighted_chosen": -0.26622849702835083, + "rewards/weighted_margins": 0.05975799635052681, + "rewards/weighted_rejected": -0.3257743716239929, + "step": 350 + }, + { + "epoch": 0.19239762175162, + "grad_norm": 10.4098539352417, + "learning_rate": 9.747732330861695e-07, + "logits/chosen": -0.05247039720416069, + "logits/rejected": -0.024548912420868874, + "logps/chosen": -135.78750610351562, + "logps/rejected": -137.03671264648438, + "logps/weighted_chosen": -0.8440704345703125, + "logps/weighted_rejected": -0.8307411074638367, + "loss": 0.7415, + "rewards/accuracies": 0.578125, + "rewards/chosen": -16.864452362060547, + "rewards/margins": 2.5316405296325684, + "rewards/rejected": -19.396093368530273, + "rewards/weighted_accuracies": 0.5249999761581421, + "rewards/weighted_chosen": 0.005719375796616077, + "rewards/weighted_margins": 0.17883948981761932, + "rewards/weighted_rejected": -0.17323684692382812, + "step": 360 + }, + { + "epoch": 0.19774200013360946, + "grad_norm": 8.8024263381958, + "learning_rate": 9.717653532052742e-07, + "logits/chosen": -0.0410335548222065, + "logits/rejected": -0.049485016614198685, + "logps/chosen": -130.2781219482422, + "logps/rejected": -137.9343719482422, + "logps/weighted_chosen": -1.1705901622772217, + "logps/weighted_rejected": -0.7656036615371704, + "loss": 0.7402, + "rewards/accuracies": 0.6031249761581421, + "rewards/chosen": -17.155858993530273, + "rewards/margins": 2.863085985183716, + "rewards/rejected": -20.016016006469727, + "rewards/weighted_accuracies": 0.53125, + "rewards/weighted_chosen": -0.07308082282543182, + "rewards/weighted_margins": 0.04105224460363388, + "rewards/weighted_rejected": -0.114354707300663, + "step": 370 + }, + { + "epoch": 0.2030863785155989, + "grad_norm": 26.81167221069336, + "learning_rate": 9.685932900101146e-07, + "logits/chosen": -0.07604064792394638, + "logits/rejected": -0.06302948296070099, + "logps/chosen": -134.7859344482422, + "logps/rejected": -137.11093139648438, + "logps/weighted_chosen": -0.98638916015625, + "logps/weighted_rejected": -0.6243560910224915, + "loss": 0.7116, + "rewards/accuracies": 0.5625, + "rewards/chosen": -17.729297637939453, + "rewards/margins": 1.8025391101837158, + "rewards/rejected": -19.528514862060547, + "rewards/weighted_accuracies": 0.546875, + "rewards/weighted_chosen": -0.11945419013500214, + "rewards/weighted_margins": 0.07185058295726776, + "rewards/weighted_rejected": -0.1911766082048416, + "step": 380 + }, + { + "epoch": 0.20843075689758836, + "grad_norm": 40.89613342285156, + "learning_rate": 9.652581474390043e-07, + "logits/chosen": -0.09274902194738388, + "logits/rejected": -0.05452308803796768, + "logps/chosen": -139.3249969482422, + "logps/rejected": -141.53750610351562, + "logps/weighted_chosen": -0.9199386835098267, + "logps/weighted_rejected": -0.9217513799667358, + "loss": 0.7323, + "rewards/accuracies": 0.550000011920929, + "rewards/chosen": -19.379688262939453, + "rewards/margins": 2.321093797683716, + "rewards/rejected": -21.702733993530273, + "rewards/weighted_accuracies": 0.543749988079071, + "rewards/weighted_chosen": -0.05511016771197319, + "rewards/weighted_margins": 0.04057302325963974, + "rewards/weighted_rejected": -0.09564266353845596, + "step": 390 + }, + { + "epoch": 0.21377513527957778, + "grad_norm": 51.860347747802734, + "learning_rate": 9.61761086184981e-07, + "logits/chosen": -0.10268402099609375, + "logits/rejected": -0.05923309177160263, + "logps/chosen": -139.50936889648438, + "logps/rejected": -141.4015655517578, + "logps/weighted_chosen": -0.9449096918106079, + "logps/weighted_rejected": -1.177301049232483, + "loss": 0.7387, + "rewards/accuracies": 0.5625, + "rewards/chosen": -20.024608612060547, + "rewards/margins": 2.9205079078674316, + "rewards/rejected": -22.942577362060547, + "rewards/weighted_accuracies": 0.5625, + "rewards/weighted_chosen": -0.26060789823532104, + "rewards/weighted_margins": 0.017241667956113815, + "rewards/weighted_rejected": -0.2779647707939148, + "step": 400 + }, + { + "epoch": 0.21911951366156723, + "grad_norm": 7.539150238037109, + "learning_rate": 9.581033232918629e-07, + "logits/chosen": -0.015906523913145065, + "logits/rejected": 0.02000865899026394, + "logps/chosen": -138.39999389648438, + "logps/rejected": -141.68905639648438, + "logps/weighted_chosen": -1.0313812494277954, + "logps/weighted_rejected": -1.1196472644805908, + "loss": 0.7283, + "rewards/accuracies": 0.5874999761581421, + "rewards/chosen": -21.702342987060547, + "rewards/margins": 2.826953172683716, + "rewards/rejected": -24.53125, + "rewards/weighted_accuracies": 0.5249999761581421, + "rewards/weighted_chosen": -0.2783859372138977, + "rewards/weighted_margins": 0.06903991848230362, + "rewards/weighted_rejected": -0.3477470278739929, + "step": 410 + }, + { + "epoch": 0.22446389204355668, + "grad_norm": 6.638446807861328, + "learning_rate": 9.542861317306952e-07, + "logits/chosen": -0.05767669528722763, + "logits/rejected": -0.03759918361902237, + "logps/chosen": -140.64688110351562, + "logps/rejected": -141.9718780517578, + "logps/weighted_chosen": -1.5940430164337158, + "logps/weighted_rejected": -1.458154320716858, + "loss": 0.7371, + "rewards/accuracies": 0.5406249761581421, + "rewards/chosen": -22.33203125, + "rewards/margins": 1.1359374523162842, + "rewards/rejected": -23.46484375, + "rewards/weighted_accuracies": 0.515625, + "rewards/weighted_chosen": -0.299835205078125, + "rewards/weighted_margins": 0.032170869410037994, + "rewards/weighted_rejected": -0.3319748044013977, + "step": 420 + }, + { + "epoch": 0.22980827042554614, + "grad_norm": 24.621816635131836, + "learning_rate": 9.503108399567308e-07, + "logits/chosen": -0.15150222182273865, + "logits/rejected": -0.09339599311351776, + "logps/chosen": -149.8671875, + "logps/rejected": -153.6296844482422, + "logps/weighted_chosen": -0.7455719113349915, + "logps/weighted_rejected": -0.826556384563446, + "loss": 0.7125, + "rewards/accuracies": 0.550000011920929, + "rewards/chosen": -22.685546875, + "rewards/margins": 1.7355468273162842, + "rewards/rejected": -24.420312881469727, + "rewards/weighted_accuracies": 0.5406249761581421, + "rewards/weighted_chosen": -0.19403991103172302, + "rewards/weighted_margins": 0.02692871168255806, + "rewards/weighted_rejected": -0.2209148406982422, + "step": 430 + }, + { + "epoch": 0.23515264880753559, + "grad_norm": 10.726137161254883, + "learning_rate": 9.461788314471034e-07, + "logits/chosen": -0.08421440422534943, + "logits/rejected": -0.031351469457149506, + "logps/chosen": -143.1171875, + "logps/rejected": -146.890625, + "logps/weighted_chosen": -0.7524811029434204, + "logps/weighted_rejected": -0.8818389773368835, + "loss": 0.7062, + "rewards/accuracies": 0.550000011920929, + "rewards/chosen": -24.516796112060547, + "rewards/margins": 2.0640625953674316, + "rewards/rejected": -26.577733993530273, + "rewards/weighted_accuracies": 0.528124988079071, + "rewards/weighted_chosen": -0.22655639052391052, + "rewards/weighted_margins": 0.07337188720703125, + "rewards/weighted_rejected": -0.30005186796188354, + "step": 440 + }, + { + "epoch": 0.240497027189525, + "grad_norm": 8.586684226989746, + "learning_rate": 9.418915442193509e-07, + "logits/chosen": -0.03446502611041069, + "logits/rejected": 0.007968139834702015, + "logps/chosen": -144.7843780517578, + "logps/rejected": -155.96249389648438, + "logps/weighted_chosen": -1.1978485584259033, + "logps/weighted_rejected": -1.2447845935821533, + "loss": 0.6841, + "rewards/accuracies": 0.6312500238418579, + "rewards/chosen": -26.516016006469727, + "rewards/margins": 4.714453220367432, + "rewards/rejected": -31.236328125, + "rewards/weighted_accuracies": 0.5218750238418579, + "rewards/weighted_chosen": -0.25520628690719604, + "rewards/weighted_margins": 0.09325332939624786, + "rewards/weighted_rejected": -0.3483436703681946, + "step": 450 + }, + { + "epoch": 0.24584140557151446, + "grad_norm": 10.014894485473633, + "learning_rate": 9.374504703309579e-07, + "logits/chosen": -0.1379852294921875, + "logits/rejected": -0.0861412063241005, + "logps/chosen": -141.83828735351562, + "logps/rejected": -143.34530639648438, + "logps/weighted_chosen": -1.1420501470565796, + "logps/weighted_rejected": -1.055877685546875, + "loss": 0.6804, + "rewards/accuracies": 0.534375011920929, + "rewards/chosen": -25.53515625, + "rewards/margins": 2.038281202316284, + "rewards/rejected": -27.571874618530273, + "rewards/weighted_accuracies": 0.5218750238418579, + "rewards/weighted_chosen": -0.21983489394187927, + "rewards/weighted_margins": 0.17051848769187927, + "rewards/weighted_rejected": -0.3904357850551605, + "step": 460 + }, + { + "epoch": 0.2511857839535039, + "grad_norm": 10.024358749389648, + "learning_rate": 9.328571553600915e-07, + "logits/chosen": -0.1318313628435135, + "logits/rejected": -0.0828731507062912, + "logps/chosen": -144.39688110351562, + "logps/rejected": -147.41796875, + "logps/weighted_chosen": -0.9446808099746704, + "logps/weighted_rejected": -1.138859510421753, + "loss": 0.7137, + "rewards/accuracies": 0.5718749761581421, + "rewards/chosen": -25.697265625, + "rewards/margins": 1.9656250476837158, + "rewards/rejected": -27.665233612060547, + "rewards/weighted_accuracies": 0.4906249940395355, + "rewards/weighted_chosen": -0.2078697234392166, + "rewards/weighted_margins": 0.052256010472774506, + "rewards/weighted_rejected": -0.260000616312027, + "step": 470 + }, + { + "epoch": 0.25653016233549336, + "grad_norm": 13.618422508239746, + "learning_rate": 9.281131978677106e-07, + "logits/chosen": -0.16686096787452698, + "logits/rejected": -0.1246364563703537, + "logps/chosen": -149.2546844482422, + "logps/rejected": -152.47811889648438, + "logps/weighted_chosen": -0.686419665813446, + "logps/weighted_rejected": -1.0338714122772217, + "loss": 0.6805, + "rewards/accuracies": 0.59375, + "rewards/chosen": -25.559375762939453, + "rewards/margins": 3.1332030296325684, + "rewards/rejected": -28.689844131469727, + "rewards/weighted_accuracies": 0.534375011920929, + "rewards/weighted_chosen": -0.22538375854492188, + "rewards/weighted_margins": 0.13199158012866974, + "rewards/weighted_rejected": -0.35721856355667114, + "step": 480 + }, + { + "epoch": 0.2618745407174828, + "grad_norm": 30.852863311767578, + "learning_rate": 9.232202488412361e-07, + "logits/chosen": -0.10437698662281036, + "logits/rejected": -0.05600614473223686, + "logps/chosen": -143.5671844482422, + "logps/rejected": -149.11874389648438, + "logps/weighted_chosen": -1.236029028892517, + "logps/weighted_rejected": -1.6542327404022217, + "loss": 0.6937, + "rewards/accuracies": 0.5375000238418579, + "rewards/chosen": -28.171483993530273, + "rewards/margins": 2.311328172683716, + "rewards/rejected": -30.477733612060547, + "rewards/weighted_accuracies": 0.534375011920929, + "rewards/weighted_chosen": -0.27268296480178833, + "rewards/weighted_margins": 0.08454742282629013, + "rewards/weighted_rejected": -0.35735321044921875, + "step": 490 + }, + { + "epoch": 0.26721891909947226, + "grad_norm": 9.151970863342285, + "learning_rate": 9.181800111199766e-07, + "logits/chosen": -0.1230342835187912, + "logits/rejected": -0.0801507979631424, + "logps/chosen": -147.54843139648438, + "logps/rejected": -147.5437469482422, + "logps/weighted_chosen": -1.5909912586212158, + "logps/weighted_rejected": -1.286199927330017, + "loss": 0.7005, + "rewards/accuracies": 0.578125, + "rewards/chosen": -29.640233993530273, + "rewards/margins": 3.1507811546325684, + "rewards/rejected": -32.789451599121094, + "rewards/weighted_accuracies": 0.5874999761581421, + "rewards/weighted_chosen": -0.34203338623046875, + "rewards/weighted_margins": 0.12214355170726776, + "rewards/weighted_rejected": -0.4641265869140625, + "step": 500 + }, + { + "epoch": 0.26721891909947226, + "eval_logits/chosen": -0.29476672410964966, + "eval_logits/rejected": -0.26891082525253296, + "eval_logps/chosen": -149.5992889404297, + "eval_logps/rejected": -153.5249481201172, + "eval_logps/weighted_chosen": -0.8761399984359741, + "eval_logps/weighted_rejected": -1.120854377746582, + "eval_loss": 0.6999610066413879, + "eval_rewards/accuracies": 0.5784114003181458, + "eval_rewards/chosen": -28.267057418823242, + "eval_rewards/margins": 2.7842414379119873, + "eval_rewards/rejected": -31.05422592163086, + "eval_rewards/weighted_accuracies": 0.5264765620231628, + "eval_rewards/weighted_chosen": -0.23699289560317993, + "eval_rewards/weighted_margins": 0.08454981446266174, + "eval_rewards/weighted_rejected": -0.3215427100658417, + "eval_runtime": 1248.3443, + "eval_samples_per_second": 1.571, + "eval_steps_per_second": 0.393, + "step": 500 + } + ], + "logging_steps": 10, + "max_steps": 1872, + "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..7b414b2 --- /dev/null +++ b/checkpoint-500/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fa2dcf6fe65fd94a0fde469e207f237b0752c6e558745acec2daa0276a1175ce +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..ceed8ae --- /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": 128009, + "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..ccab831 --- /dev/null +++ b/model-00001-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9b1f91ce998738ff854919fcece90440bf284a13e36635adf78d3edc7b0c9583 +size 4976698672 diff --git a/model-00002-of-00004.safetensors b/model-00002-of-00004.safetensors new file mode 100644 index 0000000..fdfe074 --- /dev/null +++ b/model-00002-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e2e889b4e77d1f516c8891784fae73528795e932adc20d7d1f667c2ae85999f3 +size 4999802720 diff --git a/model-00003-of-00004.safetensors b/model-00003-of-00004.safetensors new file mode 100644 index 0000000..d56f7dc --- /dev/null +++ b/model-00003-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:abae5b5c881483db48e9b0e75c4fc41d85da6135c458e0fed88e7134a07b040b +size 4915916176 diff --git a/model-00004-of-00004.safetensors b/model-00004-of-00004.safetensors new file mode 100644 index 0000000..558a162 --- /dev/null +++ b/model-00004-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e1f1af93ae12a50d91c3863142e204beb76f74089a5d36bef78b7d1e6db8a5cc +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..b43be96 --- /dev/null +++ b/special_tokens_map.json @@ -0,0 +1,17 @@ +{ + "bos_token": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "eos_token": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "pad_token": "<|eot_id|>" +} diff --git a/tokenizer.json b/tokenizer.json new file mode 100644 index 0000000..86a3394 --- /dev/null +++ b/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3c5cf44023714fb39b05e71e425f8d7b92805ff73f7988b083b8c87f0bf87393 +size 17209961 diff --git a/tokenizer_config.json b/tokenizer_config.json new file mode 100644 index 0000000..34d134f --- /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": "<|eot_id|>", + "extra_special_tokens": {}, + "model_input_names": [ + "input_ids", + "attention_mask" + ], + "model_max_length": 1000000000000000019884624838656, + "pad_token": "<|eot_id|>", + "tokenizer_class": "PreTrainedTokenizerFast" +} diff --git a/training_args.bin b/training_args.bin new file mode 100644 index 0000000..7b414b2 --- /dev/null +++ b/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fa2dcf6fe65fd94a0fde469e207f237b0752c6e558745acec2daa0276a1175ce +size 8785