commit 0f39d4f94dfa3734aa406163d72204a8a631ba2c Author: ModelHub XC Date: Wed May 6 02:55:40 2026 +0800 初始化项目，由ModelHub XC社区提供模型 Model: smirki/UIGEN-FX-4B-Intermediate Source: Original Platform diff --git a/.gitattributes b/.gitattributes new file mode 100644 index 0000000..d2fc868 --- /dev/null +++ b/.gitattributes @@ -0,0 +1,51 @@ +*.7z filter=lfs diff=lfs merge=lfs -text +*.arrow filter=lfs diff=lfs merge=lfs -text +*.bin filter=lfs diff=lfs merge=lfs -text +*.bin.* filter=lfs diff=lfs merge=lfs -text +*.bz2 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 +*.model filter=lfs diff=lfs merge=lfs -text +*.msgpack 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 +*.pt filter=lfs diff=lfs merge=lfs -text +*.pth filter=lfs diff=lfs merge=lfs -text +*.rar filter=lfs diff=lfs merge=lfs -text +saved_model/**/* 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 +*.xz filter=lfs diff=lfs merge=lfs -text +*.zip filter=lfs diff=lfs merge=lfs -text +*.zstandard filter=lfs diff=lfs merge=lfs -text +*.tfevents* filter=lfs diff=lfs merge=lfs -text +*.db* filter=lfs diff=lfs merge=lfs -text +*.ark* filter=lfs diff=lfs merge=lfs -text +**/*ckpt*data* filter=lfs diff=lfs merge=lfs -text +**/*ckpt*.meta filter=lfs diff=lfs merge=lfs -text +**/*ckpt*.index filter=lfs diff=lfs merge=lfs -text +*.safetensors filter=lfs diff=lfs merge=lfs -text +*.ckpt filter=lfs diff=lfs merge=lfs -text +*.gguf* filter=lfs diff=lfs merge=lfs -text +*.ggml filter=lfs diff=lfs merge=lfs -text +*.llamafile* filter=lfs diff=lfs merge=lfs -text +*.pt2 filter=lfs diff=lfs merge=lfs -text +*.mlmodel filter=lfs diff=lfs merge=lfs -text +*.npy filter=lfs diff=lfs merge=lfs -text +*.npz filter=lfs diff=lfs merge=lfs -text +*.pickle filter=lfs diff=lfs merge=lfs -text +*.pkl filter=lfs diff=lfs merge=lfs -text +*.tar filter=lfs diff=lfs merge=lfs -text +*.wasm filter=lfs diff=lfs merge=lfs -text +*.zst filter=lfs diff=lfs merge=lfs -text +*tfevents* filter=lfs diff=lfs merge=lfs -text + +merges.txt filter=lfs diff=lfs merge=lfs -text +tokenizer.json filter=lfs diff=lfs merge=lfs -text +vocab.json filter=lfs diff=lfs merge=lfs -text \ No newline at end of file diff --git a/README.md b/README.md new file mode 100644 index 0000000..e0f95c5 --- /dev/null +++ b/README.md @@ -0,0 +1,23 @@ +--- +base_model: Unsloth/Qwen3-4B-Instruct-2507 +tags: +- text-generation-inference +- transformers +- unsloth +- qwen3 +- trl +- sft +license: apache-2.0 +language: +- en +--- + +# Uploaded model + +- **Developed by:** smirki +- **License:** apache-2.0 +- **Finetuned from model :** Unsloth/Qwen3-4B-Instruct-2507 + +This qwen3 model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library. + +[](https://github.com/unslothai/unsloth) diff --git a/added_tokens.json b/added_tokens.json new file mode 100644 index 0000000..b54f913 --- /dev/null +++ b/added_tokens.json @@ -0,0 +1,28 @@ +{ + "": 151668, + "": 151658, + "": 151666, + "": 151667, + "": 151657, + "": 151665, + "<|box_end|>": 151649, + "<|box_start|>": 151648, + "<|endoftext|>": 151643, + "<|file_sep|>": 151664, + "<|fim_middle|>": 151660, + "<|fim_pad|>": 151662, + "<|fim_prefix|>": 151659, + "<|fim_suffix|>": 151661, + "<|im_end|>": 151645, + "<|im_start|>": 151644, + "<|image_pad|>": 151655, + "<|object_ref_end|>": 151647, + "<|object_ref_start|>": 151646, + "<|quad_end|>": 151651, + "<|quad_start|>": 151650, + "<|repo_name|>": 151663, + "<|video_pad|>": 151656, + "<|vision_end|>": 151653, + "<|vision_pad|>": 151654, + "<|vision_start|>": 151652 +} diff --git a/chat_template.jinja b/chat_template.jinja new file mode 100644 index 0000000..465e19f --- /dev/null +++ b/chat_template.jinja @@ -0,0 +1,289 @@ +{% if 'role' in messages[0] %} +{%- if tools %} + {{- '<|im_start|>system +' }} + {%- if messages[0].role == 'system' %} + {{- messages[0].content + ' + +' }} + {%- endif %} + {{- "# Tools + +You may call one or more functions to assist with the user query. + +You are provided with function signatures within XML tags: +" }} + {%- for tool in tools %} + {{- " +" }} + {{- tool | tojson }} + {%- endfor %} + {{- " + + +For each function call, return a json object with function name and arguments within XML tags: + +{\"name\": , \"arguments\": } +<|im_end|> +" }} +{%- else %} + {%- if messages[0].role == 'system' %} + {{- '<|im_start|>system +' + messages[0].content + '<|im_end|> +' }} + {%- endif %} +{%- endif %} +{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %} +{%- for forward_message in messages %} + {%- set index = (messages|length - 1) - loop.index0 %} + {%- set message = messages[index] %} + {%- set current_content = message.content if message.content is not none else '' %} + {%- set tool_start = '' %} + {%- set tool_start_length = tool_start|length %} + {%- set start_of_message = current_content[:tool_start_length] %} + {%- set tool_end = '' %} + {%- set tool_end_length = tool_end|length %} + {%- set start_pos = (current_content|length) - tool_end_length %} + {%- if start_pos < 0 %} + {%- set start_pos = 0 %} + {%- endif %} + {%- set end_of_message = current_content[start_pos:] %} + {%- if ns.multi_step_tool and message.role == "user" and not(start_of_message == tool_start and end_of_message == tool_end) %} + {%- set ns.multi_step_tool = false %} + {%- set ns.last_query_index = index %} + {%- endif %} +{%- endfor %} +{%- for message in messages %} + {%- if (message.role == "user") or (message.role == "system" and not loop.first) %} + {{- '<|im_start|>' + message.role + ' +' + message.content + '<|im_end|>' + ' +' }} + {%- elif message.role == "assistant" %} + {%- set content = message.content %} + {%- set reasoning_content = '' %} + {%- if message.reasoning_content is defined and message.reasoning_content is not none %} + {%- set reasoning_content = message.reasoning_content %} + {%- else %} + {%- if '' in message.content %} + {%- set content = (message.content.split('')|last).lstrip(' +') %} + {%- set reasoning_content = (message.content.split('')|first).rstrip(' +') %} + {%- set reasoning_content = (reasoning_content.split('')|last).lstrip(' +') %} + {%- endif %} + {%- endif %} + {%- if loop.index0 > ns.last_query_index %} + {%- if loop.last or (not loop.last and reasoning_content) %} + {{- '<|im_start|>' + message.role + ' + +' + reasoning_content.strip(' +') + ' + + +' + content.lstrip(' +') }} + {%- else %} + {{- '<|im_start|>' + message.role + ' +' + content }} + {%- endif %} + {%- else %} + {{- '<|im_start|>' + message.role + ' +' + content }} + {%- endif %} + {%- if message.tool_calls %} + {%- for tool_call in message.tool_calls %} + {%- if (loop.first and content) or (not loop.first) %} + {{- ' +' }} + {%- endif %} + {%- if tool_call.function %} + {%- set tool_call = tool_call.function %} + {%- endif %} + {{- ' +{"name": "' }} + {{- tool_call.name }} + {{- '", "arguments": ' }} + {%- if tool_call.arguments is string %} + {{- tool_call.arguments }} + {%- else %} + {{- tool_call.arguments | tojson }} + {%- endif %} + {{- '} +' }} + {%- endfor %} + {%- endif %} + {{- '<|im_end|> +' }} + {%- elif message.role == "tool" %} + {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %} + {{- '<|im_start|>user' }} + {%- endif %} + {{- ' + +' }} + {{- message.content }} + {{- ' +' }} + {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %} + {{- '<|im_end|> +' }} + {%- endif %} + {%- endif %} +{%- endfor %} +{%- if add_generation_prompt %} + {{- '<|im_start|>assistant +' }} + {%- if enable_thinking is defined and enable_thinking is false %} + {{- ' + + + +' }} + {%- endif %} +{%- endif %} +{% else %} +{%- if tools %} + {{- '<|im_start|>system +' }} + {%- if messages[0].role == 'system' %} + {{- messages[0].content + ' + +' }} + {%- endif %} + {{- "# Tools + +You may call one or more functions to assist with the user query. + +You are provided with function signatures within XML tags: +" }} + {%- for tool in tools %} + {{- " +" }} + {{- tool | tojson }} + {%- endfor %} + {{- " + + +For each function call, return a json object with function name and arguments within XML tags: + +{\"name\": , \"arguments\": } +<|im_end|> +" }} +{%- else %} + {%- if messages[0].role == 'system' %} + {{- '<|im_start|>system +' + messages[0].content + '<|im_end|> +' }} + {%- endif %} +{%- endif %} +{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %} +{%- for forward_message in messages %} + {%- set index = (messages|length - 1) - loop.index0 %} + {%- set message = messages[index] %} + {%- set current_content = message.content if message.content is not none else '' %} + {%- set tool_start = '' %} + {%- set tool_start_length = tool_start|length %} + {%- set start_of_message = current_content[:tool_start_length] %} + {%- set tool_end = '' %} + {%- set tool_end_length = tool_end|length %} + {%- set start_pos = (current_content|length) - tool_end_length %} + {%- if start_pos < 0 %} + {%- set start_pos = 0 %} + {%- endif %} + {%- set end_of_message = current_content[start_pos:] %} + {%- if ns.multi_step_tool and message.role == "user" and not(start_of_message == tool_start and end_of_message == tool_end) %} + {%- set ns.multi_step_tool = false %} + {%- set ns.last_query_index = index %} + {%- endif %} +{%- endfor %} +{%- for message in messages %} + {%- if (message.role == "user") or (message.role == "system" and not loop.first) %} + {{- '<|im_start|>' + message.role + ' +' + message.content + '<|im_end|>' + ' +' }} + {%- elif message.role == "assistant" %} + {%- set content = message.content %} + {%- set reasoning_content = '' %} + {%- if message.reasoning_content is defined and message.reasoning_content is not none %} + {%- set reasoning_content = message.reasoning_content %} + {%- else %} + {%- if '' in message.content %} + {%- set content = (message.content.split('')|last).lstrip(' +') %} + {%- set reasoning_content = (message.content.split('')|first).rstrip(' +') %} + {%- set reasoning_content = (reasoning_content.split('')|last).lstrip(' +') %} + {%- endif %} + {%- endif %} + {%- if loop.index0 > ns.last_query_index %} + {%- if loop.last or (not loop.last and reasoning_content) %} + {{- '<|im_start|>' + message.role + ' + +' + reasoning_content.strip(' +') + ' + + +' + content.lstrip(' +') }} + {%- else %} + {{- '<|im_start|>' + message.role + ' +' + content }} + {%- endif %} + {%- else %} + {{- '<|im_start|>' + message.role + ' +' + content }} + {%- endif %} + {%- if message.tool_calls %} + {%- for tool_call in message.tool_calls %} + {%- if (loop.first and content) or (not loop.first) %} + {{- ' +' }} + {%- endif %} + {%- if tool_call.function %} + {%- set tool_call = tool_call.function %} + {%- endif %} + {{- ' +{"name": "' }} + {{- tool_call.name }} + {{- '", "arguments": ' }} + {%- if tool_call.arguments is string %} + {{- tool_call.arguments }} + {%- else %} + {{- tool_call.arguments | tojson }} + {%- endif %} + {{- '} +' }} + {%- endfor %} + {%- endif %} + {{- '<|im_end|> +' }} + {%- elif message.role == "tool" %} + {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %} + {{- '<|im_start|>user' }} + {%- endif %} + {{- ' + +' }} + {{- message.content }} + {{- ' +' }} + {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %} + {{- '<|im_end|> +' }} + {%- endif %} + {%- endif %} +{%- endfor %} +{%- if add_generation_prompt %} + {{- '<|im_start|>assistant +' }} + {%- if enable_thinking is defined and enable_thinking is false %} + {{- ' + + + +' }} + {%- endif %} +{%- endif %} +{% endif %} \ No newline at end of file diff --git a/config.json b/config.json new file mode 100644 index 0000000..18c0d87 --- /dev/null +++ b/config.json @@ -0,0 +1,70 @@ +{ + "architectures": [ + "Qwen3ForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "eos_token_id": 151645, + "head_dim": 128, + "hidden_act": "silu", + "hidden_size": 2560, + "initializer_range": 0.02, + "intermediate_size": 9728, + "layer_types": [ + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention" + ], + "max_position_embeddings": 262144, + "max_window_layers": 36, + "model_type": "qwen3", + "num_attention_heads": 32, + "num_hidden_layers": 36, + "num_key_value_heads": 8, + "pad_token_id": 151654, + "rms_norm_eps": 1e-06, + "rope_scaling": null, + "rope_theta": 5000000, + "sliding_window": null, + "tie_word_embeddings": true, + "torch_dtype": "bfloat16", + "transformers_version": "4.55.4", + "unsloth_fixed": true, + "unsloth_version": "2025.8.10", + "use_cache": true, + "use_sliding_window": false, + "vocab_size": 151936 +} diff --git a/configuration.json b/configuration.json new file mode 100644 index 0000000..bbeeda1 --- /dev/null +++ b/configuration.json @@ -0,0 +1 @@ +{"framework": "pytorch", "task": "text-generation", "allow_remote": true} \ No newline at end of file diff --git a/generation_config.json b/generation_config.json new file mode 100644 index 0000000..e8c0b71 --- /dev/null +++ b/generation_config.json @@ -0,0 +1,14 @@ +{ + "bos_token_id": 151643, + "do_sample": true, + "eos_token_id": [ + 151645, + 151643 + ], + "max_length": 262144, + "pad_token_id": 151654, + "temperature": 0.7, + "top_k": 20, + "top_p": 0.8, + "transformers_version": "4.55.4" +} diff --git a/merges.txt b/merges.txt new file mode 100644 index 0000000..80c1a19 --- /dev/null +++ b/merges.txt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8831e4f1a044471340f7c0a83d7bd71306a5b867e95fd870f74d0c5308a904d5 +size 1671853 diff --git a/model-00001-of-00002.safetensors b/model-00001-of-00002.safetensors new file mode 100644 index 0000000..f580fa5 --- /dev/null +++ b/model-00001-of-00002.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:20288b037fbd662363ca99169fcbaee5e5423dfc4e03740b2da74b6011bf57c9 +size 4967215360 diff --git a/model-00002-of-00002.safetensors b/model-00002-of-00002.safetensors new file mode 100644 index 0000000..fb6c749 --- /dev/null +++ b/model-00002-of-00002.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7224f43d61b5485bfd2bfe4bcb2cae5b0995c3ea007b8b2296a30b0851e2accf +size 3077766632 diff --git a/model.safetensors.index.json b/model.safetensors.index.json new file mode 100644 index 0000000..b65d806 --- /dev/null +++ b/model.safetensors.index.json @@ -0,0 +1,406 @@ +{ + "metadata": { + "total_parameters": 4022468096, + "total_size": 8044936192 + }, + "weight_map": { + "model.embed_tokens.weight": "model-00001-of-00002.safetensors", + "model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.20.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.20.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.20.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.21.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.28.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.28.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.28.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.28.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.28.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.28.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", + "model.layers.28.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.28.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.28.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", + "model.layers.28.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.28.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.29.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.29.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.29.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.29.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.29.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.29.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", + "model.layers.29.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.29.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.29.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", + "model.layers.29.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.29.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.30.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.30.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.30.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.30.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.30.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.30.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", + "model.layers.30.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.30.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.30.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", + "model.layers.30.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.30.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.31.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.31.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.31.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.31.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.31.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.31.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", + "model.layers.31.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.31.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.31.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", + "model.layers.31.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.31.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.32.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.32.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.32.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.32.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.32.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.32.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", + "model.layers.32.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.32.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.32.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", + "model.layers.32.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.32.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.33.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.33.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.33.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.33.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.33.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.33.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", + "model.layers.33.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.33.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.33.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", + "model.layers.33.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.33.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.34.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.34.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.34.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.34.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.34.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.34.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", + "model.layers.34.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.34.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.34.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", + "model.layers.34.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.34.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.35.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.35.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.35.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.35.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.35.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.35.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", + "model.layers.35.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.35.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.35.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", + "model.layers.35.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.35.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.norm.weight": "model-00002-of-00002.safetensors" + } +} diff --git a/optimizer.pt b/optimizer.pt new file mode 100644 index 0000000..dddcb92 --- /dev/null +++ b/optimizer.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fd1a582ee6213adc5acdb97544e0af330fbff79d2237484c73c4fa0e0a402f13 +size 10493788917 diff --git a/rng_state.pth b/rng_state.pth new file mode 100644 index 0000000..ecd5163 --- /dev/null +++ b/rng_state.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:89645f9f6a2a8823b334e0748e4f5ea683a27b3d02632f0a241397db55d82fad +size 14645 diff --git a/scheduler.pt b/scheduler.pt new file mode 100644 index 0000000..03bb251 --- /dev/null +++ b/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:96fbc3d8dd719da20f5f7085173be583f34a0113e2ceff55f0b9b3cfe37c14ee +size 1465 diff --git a/special_tokens_map.json b/special_tokens_map.json new file mode 100644 index 0000000..9b8043f --- /dev/null +++ b/special_tokens_map.json @@ -0,0 +1,31 @@ +{ + "additional_special_tokens": [ + "<|im_start|>", + "<|im_end|>", + "<|object_ref_start|>", + "<|object_ref_end|>", + "<|box_start|>", + "<|box_end|>", + "<|quad_start|>", + "<|quad_end|>", + "<|vision_start|>", + "<|vision_end|>", + "<|vision_pad|>", + "<|image_pad|>", + "<|video_pad|>" + ], + "eos_token": { + "content": "<|im_end|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "pad_token": { + "content": "<|vision_pad|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + } +} diff --git a/tokenizer.json b/tokenizer.json new file mode 100644 index 0000000..cd71f61 --- /dev/null +++ b/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:aeb13307a71acd8fe81861d94ad54ab689df773318809eed3cbe794b4492dae4 +size 11422654 diff --git a/tokenizer_config.json b/tokenizer_config.json new file mode 100644 index 0000000..da9f80f --- /dev/null +++ b/tokenizer_config.json @@ -0,0 +1,240 @@ +{ + "add_bos_token": false, + "add_prefix_space": false, + "added_tokens_decoder": { + "151643": { + "content": "<|endoftext|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "151644": { + "content": "<|im_start|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "151645": { + "content": "<|im_end|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "151646": { + "content": "<|object_ref_start|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "151647": { + "content": "<|object_ref_end|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "151648": { + "content": "<|box_start|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "151649": { + "content": "<|box_end|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "151650": { + "content": "<|quad_start|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "151651": { + "content": "<|quad_end|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "151652": { + "content": "<|vision_start|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "151653": { + "content": "<|vision_end|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "151654": { + "content": "<|vision_pad|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "151655": { + "content": "<|image_pad|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "151656": { + "content": "<|video_pad|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "151657": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "151658": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "151659": { + "content": "<|fim_prefix|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "151660": { + "content": "<|fim_middle|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "151661": { + "content": "<|fim_suffix|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "151662": { + "content": "<|fim_pad|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "151663": { + "content": "<|repo_name|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "151664": { + "content": "<|file_sep|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "151665": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "151666": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "151667": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "151668": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + } + }, + "additional_special_tokens": [ + "<|im_start|>", + "<|im_end|>", + "<|object_ref_start|>", + "<|object_ref_end|>", + "<|box_start|>", + "<|box_end|>", + "<|quad_start|>", + "<|quad_end|>", + "<|vision_start|>", + "<|vision_end|>", + "<|vision_pad|>", + "<|image_pad|>", + "<|video_pad|>" + ], + "bos_token": null, + "clean_up_tokenization_spaces": false, + "eos_token": "<|im_end|>", + "errors": "replace", + "extra_special_tokens": {}, + "model_max_length": 262144, + "pad_token": "<|vision_pad|>", + "padding_side": "right", + "split_special_tokens": false, + "tokenizer_class": "Qwen2Tokenizer", + "unk_token": null +} diff --git a/trainer_state.json b/trainer_state.json new file mode 100644 index 0000000..23f0f44 --- /dev/null +++ b/trainer_state.json @@ -0,0 +1,2834 @@ +{ + "best_global_step": null, + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 4.808177991581479, + "eval_steps": 500, + "global_step": 4000, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.012026458208057728, + "grad_norm": 5.46875, + "learning_rate": 3.6000000000000003e-06, + "loss": 0.6162, + "step": 10 + }, + { + "epoch": 0.024052916416115455, + "grad_norm": 1.6171875, + "learning_rate": 7.600000000000001e-06, + "loss": 0.5185, + "step": 20 + }, + { + "epoch": 0.03607937462417318, + "grad_norm": 0.9140625, + "learning_rate": 1.16e-05, + "loss": 0.4609, + "step": 30 + }, + { + "epoch": 0.04810583283223091, + "grad_norm": 0.84375, + "learning_rate": 1.5600000000000003e-05, + "loss": 0.4184, + "step": 40 + }, + { + "epoch": 0.06013229104028864, + "grad_norm": 0.7890625, + "learning_rate": 1.9600000000000002e-05, + "loss": 0.4258, + "step": 50 + }, + { + "epoch": 0.07215874924834637, + "grad_norm": 0.72265625, + "learning_rate": 1.9999763370213405e-05, + "loss": 0.4325, + "step": 60 + }, + { + "epoch": 0.0841852074564041, + "grad_norm": 0.796875, + "learning_rate": 1.9998945405081654e-05, + "loss": 0.429, + "step": 70 + }, + { + "epoch": 0.09621166566446182, + "grad_norm": 0.7421875, + "learning_rate": 1.9997543231029267e-05, + "loss": 0.405, + "step": 80 + }, + { + "epoch": 0.10823812387251955, + "grad_norm": 0.92578125, + "learning_rate": 1.9995556929981147e-05, + "loss": 0.4061, + "step": 90 + }, + { + "epoch": 0.12026458208057728, + "grad_norm": 0.85546875, + "learning_rate": 1.9992986617991008e-05, + "loss": 0.4029, + "step": 100 + }, + { + "epoch": 0.132291040288635, + "grad_norm": 1.0234375, + "learning_rate": 1.9989832445234612e-05, + "loss": 0.3865, + "step": 110 + }, + { + "epoch": 0.14431749849669273, + "grad_norm": 0.66015625, + "learning_rate": 1.998609459600098e-05, + "loss": 0.4057, + "step": 120 + }, + { + "epoch": 0.15634395670475046, + "grad_norm": 0.7109375, + "learning_rate": 1.9981773288681624e-05, + "loss": 0.4188, + "step": 130 + }, + { + "epoch": 0.1683704149128082, + "grad_norm": 0.78515625, + "learning_rate": 1.99768687757578e-05, + "loss": 0.3985, + "step": 140 + }, + { + "epoch": 0.1803968731208659, + "grad_norm": 0.7265625, + "learning_rate": 1.9971381343785755e-05, + "loss": 0.4024, + "step": 150 + }, + { + "epoch": 0.19242333132892364, + "grad_norm": 0.73046875, + "learning_rate": 1.996531131337997e-05, + "loss": 0.3711, + "step": 160 + }, + { + "epoch": 0.20444978953698137, + "grad_norm": 0.82421875, + "learning_rate": 1.9958659039194434e-05, + "loss": 0.3807, + "step": 170 + }, + { + "epoch": 0.2164762477450391, + "grad_norm": 0.890625, + "learning_rate": 1.9951424909901927e-05, + "loss": 0.416, + "step": 180 + }, + { + "epoch": 0.22850270595309682, + "grad_norm": 0.76953125, + "learning_rate": 1.994360934817132e-05, + "loss": 0.3846, + "step": 190 + }, + { + "epoch": 0.24052916416115455, + "grad_norm": 0.71875, + "learning_rate": 1.9935212810642844e-05, + "loss": 0.3587, + "step": 200 + }, + { + "epoch": 0.25255562236921225, + "grad_norm": 0.71484375, + "learning_rate": 1.9926235787901457e-05, + "loss": 0.3799, + "step": 210 + }, + { + "epoch": 0.26458208057727, + "grad_norm": 0.78515625, + "learning_rate": 1.991667880444815e-05, + "loss": 0.4036, + "step": 220 + }, + { + "epoch": 0.2766085387853277, + "grad_norm": 0.80859375, + "learning_rate": 1.990654241866931e-05, + "loss": 0.4102, + "step": 230 + }, + { + "epoch": 0.28863499699338546, + "grad_norm": 0.81640625, + "learning_rate": 1.9895827222804086e-05, + "loss": 0.391, + "step": 240 + }, + { + "epoch": 0.30066145520144316, + "grad_norm": 0.76171875, + "learning_rate": 1.9884533842909807e-05, + "loss": 0.3655, + "step": 250 + }, + { + "epoch": 0.3126879134095009, + "grad_norm": 0.79296875, + "learning_rate": 1.9872662938825392e-05, + "loss": 0.3754, + "step": 260 + }, + { + "epoch": 0.3247143716175586, + "grad_norm": 0.99609375, + "learning_rate": 1.9860215204132783e-05, + "loss": 0.3552, + "step": 270 + }, + { + "epoch": 0.3367408298256164, + "grad_norm": 2.984375, + "learning_rate": 1.9847191366116453e-05, + "loss": 0.3657, + "step": 280 + }, + { + "epoch": 0.34876728803367407, + "grad_norm": 0.66796875, + "learning_rate": 1.983359218572088e-05, + "loss": 0.3919, + "step": 290 + }, + { + "epoch": 0.3607937462417318, + "grad_norm": 0.73828125, + "learning_rate": 1.9819418457506105e-05, + "loss": 0.3825, + "step": 300 + }, + { + "epoch": 0.3728202044497895, + "grad_norm": 1.0859375, + "learning_rate": 1.9804671009601316e-05, + "loss": 0.3928, + "step": 310 + }, + { + "epoch": 0.3848466626578473, + "grad_norm": 1.28125, + "learning_rate": 1.9789350703656443e-05, + "loss": 0.3772, + "step": 320 + }, + { + "epoch": 0.396873120865905, + "grad_norm": 0.87890625, + "learning_rate": 1.9773458434791822e-05, + "loss": 0.3892, + "step": 330 + }, + { + "epoch": 0.40889957907396274, + "grad_norm": 0.96875, + "learning_rate": 1.9756995131545898e-05, + "loss": 0.3875, + "step": 340 + }, + { + "epoch": 0.42092603728202044, + "grad_norm": 0.7265625, + "learning_rate": 1.973996175582098e-05, + "loss": 0.3758, + "step": 350 + }, + { + "epoch": 0.4329524954900782, + "grad_norm": 0.77734375, + "learning_rate": 1.9722359302827025e-05, + "loss": 0.3584, + "step": 360 + }, + { + "epoch": 0.4449789536981359, + "grad_norm": 0.67578125, + "learning_rate": 1.97041888010235e-05, + "loss": 0.3456, + "step": 370 + }, + { + "epoch": 0.45700541190619365, + "grad_norm": 0.66796875, + "learning_rate": 1.968545131205929e-05, + "loss": 0.3502, + "step": 380 + }, + { + "epoch": 0.46903187011425135, + "grad_norm": 0.74609375, + "learning_rate": 1.966614793071067e-05, + "loss": 0.3296, + "step": 390 + }, + { + "epoch": 0.4810583283223091, + "grad_norm": 1.21875, + "learning_rate": 1.964627978481734e-05, + "loss": 0.3501, + "step": 400 + }, + { + "epoch": 0.4930847865303668, + "grad_norm": 1.046875, + "learning_rate": 1.962584803521653e-05, + "loss": 0.3404, + "step": 410 + }, + { + "epoch": 0.5051112447384245, + "grad_norm": 0.9765625, + "learning_rate": 1.960485387567517e-05, + "loss": 0.3427, + "step": 420 + }, + { + "epoch": 0.5171377029464823, + "grad_norm": 0.8046875, + "learning_rate": 1.9583298532820147e-05, + "loss": 0.3723, + "step": 430 + }, + { + "epoch": 0.52916416115454, + "grad_norm": 0.796875, + "learning_rate": 1.956118326606664e-05, + "loss": 0.3835, + "step": 440 + }, + { + "epoch": 0.5411906193625977, + "grad_norm": 0.65625, + "learning_rate": 1.9538509367544524e-05, + "loss": 0.3488, + "step": 450 + }, + { + "epoch": 0.5532170775706554, + "grad_norm": 0.91796875, + "learning_rate": 1.951527816202289e-05, + "loss": 0.3653, + "step": 460 + }, + { + "epoch": 0.5652435357787132, + "grad_norm": 0.69921875, + "learning_rate": 1.949149100683263e-05, + "loss": 0.3542, + "step": 470 + }, + { + "epoch": 0.5772699939867709, + "grad_norm": 0.81640625, + "learning_rate": 1.946714929178714e-05, + "loss": 0.3543, + "step": 480 + }, + { + "epoch": 0.5892964521948286, + "grad_norm": 0.91796875, + "learning_rate": 1.944225443910111e-05, + "loss": 0.3619, + "step": 490 + }, + { + "epoch": 0.6013229104028863, + "grad_norm": 0.9609375, + "learning_rate": 1.941680790330744e-05, + "loss": 0.3641, + "step": 500 + }, + { + "epoch": 0.613349368610944, + "grad_norm": 0.765625, + "learning_rate": 1.9390811171172244e-05, + "loss": 0.3643, + "step": 510 + }, + { + "epoch": 0.6253758268190018, + "grad_norm": 0.78515625, + "learning_rate": 1.936426576160799e-05, + "loss": 0.3449, + "step": 520 + }, + { + "epoch": 0.6374022850270595, + "grad_norm": 0.79296875, + "learning_rate": 1.9337173225584743e-05, + "loss": 0.3096, + "step": 530 + }, + { + "epoch": 0.6494287432351172, + "grad_norm": 0.8515625, + "learning_rate": 1.9309535146039567e-05, + "loss": 0.3355, + "step": 540 + }, + { + "epoch": 0.6614552014431749, + "grad_norm": 0.84375, + "learning_rate": 1.9281353137784017e-05, + "loss": 0.366, + "step": 550 + }, + { + "epoch": 0.6734816596512327, + "grad_norm": 0.66015625, + "learning_rate": 1.9252628847409805e-05, + "loss": 0.3321, + "step": 560 + }, + { + "epoch": 0.6855081178592904, + "grad_norm": 0.78125, + "learning_rate": 1.9223363953192595e-05, + "loss": 0.37, + "step": 570 + }, + { + "epoch": 0.6975345760673481, + "grad_norm": 0.79296875, + "learning_rate": 1.9193560164993927e-05, + "loss": 0.3492, + "step": 580 + }, + { + "epoch": 0.7095610342754058, + "grad_norm": 0.89453125, + "learning_rate": 1.916321922416134e-05, + "loss": 0.3085, + "step": 590 + }, + { + "epoch": 0.7215874924834637, + "grad_norm": 0.87109375, + "learning_rate": 1.913234290342661e-05, + "loss": 0.3165, + "step": 600 + }, + { + "epoch": 0.7336139506915214, + "grad_norm": 5.15625, + "learning_rate": 1.9100933006802195e-05, + "loss": 0.3482, + "step": 610 + }, + { + "epoch": 0.745640408899579, + "grad_norm": 0.78125, + "learning_rate": 1.9068991369475807e-05, + "loss": 0.3407, + "step": 620 + }, + { + "epoch": 0.7576668671076368, + "grad_norm": 0.70703125, + "learning_rate": 1.903651985770321e-05, + "loss": 0.3611, + "step": 630 + }, + { + "epoch": 0.7696933253156946, + "grad_norm": 0.88671875, + "learning_rate": 1.900352036869917e-05, + "loss": 0.3781, + "step": 640 + }, + { + "epoch": 0.7817197835237523, + "grad_norm": 0.84765625, + "learning_rate": 1.8969994830526603e-05, + "loss": 0.3468, + "step": 650 + }, + { + "epoch": 0.79374624173181, + "grad_norm": 0.96875, + "learning_rate": 1.8935945201983926e-05, + "loss": 0.3564, + "step": 660 + }, + { + "epoch": 0.8057726999398677, + "grad_norm": 1.1015625, + "learning_rate": 1.8901373472490622e-05, + "loss": 0.3796, + "step": 670 + }, + { + "epoch": 0.8177991581479255, + "grad_norm": 0.68359375, + "learning_rate": 1.8866281661970984e-05, + "loss": 0.332, + "step": 680 + }, + { + "epoch": 0.8298256163559832, + "grad_norm": 0.73828125, + "learning_rate": 1.8830671820736115e-05, + "loss": 0.3278, + "step": 690 + }, + { + "epoch": 0.8418520745640409, + "grad_norm": 0.78125, + "learning_rate": 1.8794546029364122e-05, + "loss": 0.3482, + "step": 700 + }, + { + "epoch": 0.8538785327720986, + "grad_norm": 0.76953125, + "learning_rate": 1.8757906398578555e-05, + "loss": 0.3811, + "step": 710 + }, + { + "epoch": 0.8659049909801564, + "grad_norm": 0.8515625, + "learning_rate": 1.8720755069125097e-05, + "loss": 0.3303, + "step": 720 + }, + { + "epoch": 0.8779314491882141, + "grad_norm": 0.7421875, + "learning_rate": 1.8683094211646463e-05, + "loss": 0.3525, + "step": 730 + }, + { + "epoch": 0.8899579073962718, + "grad_norm": 0.66796875, + "learning_rate": 1.864492602655561e-05, + "loss": 0.3209, + "step": 740 + }, + { + "epoch": 0.9019843656043295, + "grad_norm": 0.78125, + "learning_rate": 1.8606252743907127e-05, + "loss": 0.3121, + "step": 750 + }, + { + "epoch": 0.9140108238123873, + "grad_norm": 0.85546875, + "learning_rate": 1.8567076623266987e-05, + "loss": 0.3621, + "step": 760 + }, + { + "epoch": 0.926037282020445, + "grad_norm": 1.578125, + "learning_rate": 1.8527399953580492e-05, + "loss": 0.3416, + "step": 770 + }, + { + "epoch": 0.9380637402285027, + "grad_norm": 0.92578125, + "learning_rate": 1.8487225053038557e-05, + "loss": 0.3373, + "step": 780 + }, + { + "epoch": 0.9500901984365604, + "grad_norm": 0.71484375, + "learning_rate": 1.8446554268942252e-05, + "loss": 0.3214, + "step": 790 + }, + { + "epoch": 0.9621166566446182, + "grad_norm": 0.71484375, + "learning_rate": 1.8405389977565666e-05, + "loss": 0.3552, + "step": 800 + }, + { + "epoch": 0.9741431148526759, + "grad_norm": 0.84765625, + "learning_rate": 1.8363734584017065e-05, + "loss": 0.3538, + "step": 810 + }, + { + "epoch": 0.9861695730607336, + "grad_norm": 1.0390625, + "learning_rate": 1.832159052209837e-05, + "loss": 0.353, + "step": 820 + }, + { + "epoch": 0.9981960312687913, + "grad_norm": 0.8671875, + "learning_rate": 1.827896025416295e-05, + "loss": 0.332, + "step": 830 + }, + { + "epoch": 1.009621166566446, + "grad_norm": 0.80859375, + "learning_rate": 1.8235846270971754e-05, + "loss": 0.3417, + "step": 840 + }, + { + "epoch": 1.0216476247745039, + "grad_norm": 0.92578125, + "learning_rate": 1.8192251091547807e-05, + "loss": 0.3196, + "step": 850 + }, + { + "epoch": 1.0336740829825617, + "grad_norm": 0.7890625, + "learning_rate": 1.814817726302899e-05, + "loss": 0.3178, + "step": 860 + }, + { + "epoch": 1.0457005411906193, + "grad_norm": 0.91796875, + "learning_rate": 1.8103627360519257e-05, + "loss": 0.3172, + "step": 870 + }, + { + "epoch": 1.057726999398677, + "grad_norm": 0.74609375, + "learning_rate": 1.805860398693816e-05, + "loss": 0.3258, + "step": 880 + }, + { + "epoch": 1.069753457606735, + "grad_norm": 0.74609375, + "learning_rate": 1.801310977286876e-05, + "loss": 0.3157, + "step": 890 + }, + { + "epoch": 1.0817799158147925, + "grad_norm": 0.76171875, + "learning_rate": 1.7967147376403947e-05, + "loss": 0.3331, + "step": 900 + }, + { + "epoch": 1.0938063740228503, + "grad_norm": 0.83203125, + "learning_rate": 1.7920719482991137e-05, + "loss": 0.3189, + "step": 910 + }, + { + "epoch": 1.1058328322309081, + "grad_norm": 0.78125, + "learning_rate": 1.7873828805275346e-05, + "loss": 0.3234, + "step": 920 + }, + { + "epoch": 1.1178592904389657, + "grad_norm": 0.67578125, + "learning_rate": 1.7826478082940727e-05, + "loss": 0.3046, + "step": 930 + }, + { + "epoch": 1.1298857486470235, + "grad_norm": 0.7265625, + "learning_rate": 1.777867008255048e-05, + "loss": 0.3151, + "step": 940 + }, + { + "epoch": 1.141912206855081, + "grad_norm": 0.8671875, + "learning_rate": 1.7730407597385207e-05, + "loss": 0.2885, + "step": 950 + }, + { + "epoch": 1.153938665063139, + "grad_norm": 0.81640625, + "learning_rate": 1.768169344727973e-05, + "loss": 0.3145, + "step": 960 + }, + { + "epoch": 1.1659651232711967, + "grad_norm": 0.80078125, + "learning_rate": 1.7632530478458313e-05, + "loss": 0.3297, + "step": 970 + }, + { + "epoch": 1.1779915814792543, + "grad_norm": 0.75, + "learning_rate": 1.758292156336837e-05, + "loss": 0.3108, + "step": 980 + }, + { + "epoch": 1.1900180396873121, + "grad_norm": 0.72265625, + "learning_rate": 1.7532869600512663e-05, + "loss": 0.3091, + "step": 990 + }, + { + "epoch": 1.2020444978953697, + "grad_norm": 0.86328125, + "learning_rate": 1.7482377514279907e-05, + "loss": 0.328, + "step": 1000 + }, + { + "epoch": 1.2140709561034275, + "grad_norm": 0.9609375, + "learning_rate": 1.7431448254773943e-05, + "loss": 0.3206, + "step": 1010 + }, + { + "epoch": 1.2260974143114853, + "grad_norm": 0.93359375, + "learning_rate": 1.738008479764135e-05, + "loss": 0.3399, + "step": 1020 + }, + { + "epoch": 1.238123872519543, + "grad_norm": 0.8515625, + "learning_rate": 1.7328290143897603e-05, + "loss": 0.3099, + "step": 1030 + }, + { + "epoch": 1.2501503307276007, + "grad_norm": 0.7734375, + "learning_rate": 1.7276067319751717e-05, + "loss": 0.3041, + "step": 1040 + }, + { + "epoch": 1.2621767889356583, + "grad_norm": 0.74609375, + "learning_rate": 1.7223419376429445e-05, + "loss": 0.3217, + "step": 1050 + }, + { + "epoch": 1.2742032471437161, + "grad_norm": 0.78125, + "learning_rate": 1.7170349389995004e-05, + "loss": 0.3122, + "step": 1060 + }, + { + "epoch": 1.286229705351774, + "grad_norm": 0.8046875, + "learning_rate": 1.711686046117134e-05, + "loss": 0.3154, + "step": 1070 + }, + { + "epoch": 1.2982561635598318, + "grad_norm": 0.87109375, + "learning_rate": 1.7062955715158977e-05, + "loss": 0.3146, + "step": 1080 + }, + { + "epoch": 1.3102826217678893, + "grad_norm": 0.8046875, + "learning_rate": 1.7008638301453404e-05, + "loss": 0.3434, + "step": 1090 + }, + { + "epoch": 1.3223090799759472, + "grad_norm": 0.8515625, + "learning_rate": 1.6953911393661073e-05, + "loss": 0.3083, + "step": 1100 + }, + { + "epoch": 1.3343355381840047, + "grad_norm": 0.90234375, + "learning_rate": 1.6898778189313966e-05, + "loss": 0.3274, + "step": 1110 + }, + { + "epoch": 1.3463619963920626, + "grad_norm": 0.8671875, + "learning_rate": 1.684324190968278e-05, + "loss": 0.3244, + "step": 1120 + }, + { + "epoch": 1.3583884546001204, + "grad_norm": 0.8984375, + "learning_rate": 1.6787305799588705e-05, + "loss": 0.3357, + "step": 1130 + }, + { + "epoch": 1.370414912808178, + "grad_norm": 0.828125, + "learning_rate": 1.6730973127213858e-05, + "loss": 0.3209, + "step": 1140 + }, + { + "epoch": 1.3824413710162358, + "grad_norm": 0.6796875, + "learning_rate": 1.6674247183910313e-05, + "loss": 0.2987, + "step": 1150 + }, + { + "epoch": 1.3944678292242934, + "grad_norm": 0.6171875, + "learning_rate": 1.6617131284007805e-05, + "loss": 0.3268, + "step": 1160 + }, + { + "epoch": 1.4064942874323512, + "grad_norm": 0.734375, + "learning_rate": 1.6559628764620094e-05, + "loss": 0.3087, + "step": 1170 + }, + { + "epoch": 1.418520745640409, + "grad_norm": 0.85546875, + "learning_rate": 1.6501742985449965e-05, + "loss": 0.3381, + "step": 1180 + }, + { + "epoch": 1.4305472038484666, + "grad_norm": 0.94921875, + "learning_rate": 1.6443477328592954e-05, + "loss": 0.3134, + "step": 1190 + }, + { + "epoch": 1.4425736620565244, + "grad_norm": 0.87109375, + "learning_rate": 1.6384835198339722e-05, + "loss": 0.3038, + "step": 1200 + }, + { + "epoch": 1.454600120264582, + "grad_norm": 0.6953125, + "learning_rate": 1.6325820020977168e-05, + "loss": 0.3153, + "step": 1210 + }, + { + "epoch": 1.4666265784726398, + "grad_norm": 0.75390625, + "learning_rate": 1.626643524458823e-05, + "loss": 0.3144, + "step": 1220 + }, + { + "epoch": 1.4786530366806976, + "grad_norm": 0.75390625, + "learning_rate": 1.6206684338850433e-05, + "loss": 0.3114, + "step": 1230 + }, + { + "epoch": 1.4906794948887554, + "grad_norm": 0.828125, + "learning_rate": 1.6146570794833156e-05, + "loss": 0.3207, + "step": 1240 + }, + { + "epoch": 1.502705953096813, + "grad_norm": 0.80078125, + "learning_rate": 1.6086098124793665e-05, + "loss": 0.309, + "step": 1250 + }, + { + "epoch": 1.5147324113048706, + "grad_norm": 1.0, + "learning_rate": 1.6025269861971907e-05, + "loss": 0.3054, + "step": 1260 + }, + { + "epoch": 1.5267588695129284, + "grad_norm": 0.8984375, + "learning_rate": 1.596408956038406e-05, + "loss": 0.3089, + "step": 1270 + }, + { + "epoch": 1.5387853277209862, + "grad_norm": 1.015625, + "learning_rate": 1.5902560794614896e-05, + "loss": 0.3185, + "step": 1280 + }, + { + "epoch": 1.550811785929044, + "grad_norm": 0.87109375, + "learning_rate": 1.5840687159608915e-05, + "loss": 0.3255, + "step": 1290 + }, + { + "epoch": 1.5628382441371016, + "grad_norm": 1.0078125, + "learning_rate": 1.5778472270460322e-05, + "loss": 0.3099, + "step": 1300 + }, + { + "epoch": 1.5748647023451592, + "grad_norm": 0.82421875, + "learning_rate": 1.5715919762201783e-05, + "loss": 0.3386, + "step": 1310 + }, + { + "epoch": 1.586891160553217, + "grad_norm": 0.921875, + "learning_rate": 1.5653033289592065e-05, + "loss": 0.3068, + "step": 1320 + }, + { + "epoch": 1.5989176187612748, + "grad_norm": 0.82421875, + "learning_rate": 1.558981652690248e-05, + "loss": 0.2844, + "step": 1330 + }, + { + "epoch": 1.6109440769693326, + "grad_norm": 1.0, + "learning_rate": 1.552627316770222e-05, + "loss": 0.3259, + "step": 1340 + }, + { + "epoch": 1.6229705351773902, + "grad_norm": 1.1015625, + "learning_rate": 1.546240692464255e-05, + "loss": 0.3173, + "step": 1350 + }, + { + "epoch": 1.634996993385448, + "grad_norm": 1.03125, + "learning_rate": 1.539822152923989e-05, + "loss": 0.3349, + "step": 1360 + }, + { + "epoch": 1.6470234515935056, + "grad_norm": 0.7578125, + "learning_rate": 1.5333720731657776e-05, + "loss": 0.3076, + "step": 1370 + }, + { + "epoch": 1.6590499098015634, + "grad_norm": 0.828125, + "learning_rate": 1.5268908300487785e-05, + "loss": 0.3062, + "step": 1380 + }, + { + "epoch": 1.6710763680096212, + "grad_norm": 0.8984375, + "learning_rate": 1.5203788022529315e-05, + "loss": 0.3092, + "step": 1390 + }, + { + "epoch": 1.683102826217679, + "grad_norm": 0.7890625, + "learning_rate": 1.5138363702568347e-05, + "loss": 0.2981, + "step": 1400 + }, + { + "epoch": 1.6951292844257366, + "grad_norm": 0.7265625, + "learning_rate": 1.5072639163155136e-05, + "loss": 0.3007, + "step": 1410 + }, + { + "epoch": 1.7071557426337942, + "grad_norm": 0.84765625, + "learning_rate": 1.5006618244380881e-05, + "loss": 0.3241, + "step": 1420 + }, + { + "epoch": 1.719182200841852, + "grad_norm": 0.96484375, + "learning_rate": 1.4940304803653353e-05, + "loss": 0.3244, + "step": 1430 + }, + { + "epoch": 1.7312086590499098, + "grad_norm": 0.83203125, + "learning_rate": 1.4873702715471519e-05, + "loss": 0.2866, + "step": 1440 + }, + { + "epoch": 1.7432351172579676, + "grad_norm": 0.9296875, + "learning_rate": 1.4806815871199161e-05, + "loss": 0.3228, + "step": 1450 + }, + { + "epoch": 1.7552615754660252, + "grad_norm": 0.84375, + "learning_rate": 1.4739648178837525e-05, + "loss": 0.3114, + "step": 1460 + }, + { + "epoch": 1.7672880336740828, + "grad_norm": 0.9453125, + "learning_rate": 1.467220356279699e-05, + "loss": 0.2948, + "step": 1470 + }, + { + "epoch": 1.7793144918821406, + "grad_norm": 1.0078125, + "learning_rate": 1.4604485963667752e-05, + "loss": 0.2876, + "step": 1480 + }, + { + "epoch": 1.7913409500901984, + "grad_norm": 0.71875, + "learning_rate": 1.453649933798962e-05, + "loss": 0.3029, + "step": 1490 + }, + { + "epoch": 1.8033674082982563, + "grad_norm": 0.9140625, + "learning_rate": 1.4468247658020828e-05, + "loss": 0.2988, + "step": 1500 + }, + { + "epoch": 1.8153938665063138, + "grad_norm": 1.0078125, + "learning_rate": 1.4399734911505951e-05, + "loss": 0.3203, + "step": 1510 + }, + { + "epoch": 1.8274203247143717, + "grad_norm": 0.79296875, + "learning_rate": 1.433096510144291e-05, + "loss": 0.311, + "step": 1520 + }, + { + "epoch": 1.8394467829224292, + "grad_norm": 0.93359375, + "learning_rate": 1.4261942245849097e-05, + "loss": 0.3002, + "step": 1530 + }, + { + "epoch": 1.851473241130487, + "grad_norm": 0.8359375, + "learning_rate": 1.4192670377526601e-05, + "loss": 0.3104, + "step": 1540 + }, + { + "epoch": 1.8634996993385449, + "grad_norm": 1.5, + "learning_rate": 1.4123153543826598e-05, + "loss": 0.3138, + "step": 1550 + }, + { + "epoch": 1.8755261575466027, + "grad_norm": 0.87109375, + "learning_rate": 1.4053395806412869e-05, + "loss": 0.3063, + "step": 1560 + }, + { + "epoch": 1.8875526157546603, + "grad_norm": 0.6484375, + "learning_rate": 1.3983401241024484e-05, + "loss": 0.2997, + "step": 1570 + }, + { + "epoch": 1.8995790739627179, + "grad_norm": 0.765625, + "learning_rate": 1.3913173937237679e-05, + "loss": 0.2849, + "step": 1580 + }, + { + "epoch": 1.9116055321707757, + "grad_norm": 0.83984375, + "learning_rate": 1.3842717998226909e-05, + "loss": 0.2895, + "step": 1590 + }, + { + "epoch": 1.9236319903788335, + "grad_norm": 1.0703125, + "learning_rate": 1.3772037540525108e-05, + "loss": 0.3287, + "step": 1600 + }, + { + "epoch": 1.9356584485868913, + "grad_norm": 0.9921875, + "learning_rate": 1.3701136693783174e-05, + "loss": 0.3209, + "step": 1610 + }, + { + "epoch": 1.9476849067949489, + "grad_norm": 0.83984375, + "learning_rate": 1.3630019600528696e-05, + "loss": 0.3362, + "step": 1620 + }, + { + "epoch": 1.9597113650030065, + "grad_norm": 0.91796875, + "learning_rate": 1.3558690415923901e-05, + "loss": 0.3068, + "step": 1630 + }, + { + "epoch": 1.9717378232110643, + "grad_norm": 0.79296875, + "learning_rate": 1.348715330752289e-05, + "loss": 0.2909, + "step": 1640 + }, + { + "epoch": 1.983764281419122, + "grad_norm": 0.75390625, + "learning_rate": 1.3415412455028149e-05, + "loss": 0.2934, + "step": 1650 + }, + { + "epoch": 1.99579073962718, + "grad_norm": 0.90234375, + "learning_rate": 1.3343472050046313e-05, + "loss": 0.3228, + "step": 1660 + }, + { + "epoch": 2.0072158749248348, + "grad_norm": 0.77734375, + "learning_rate": 1.32713362958433e-05, + "loss": 0.3191, + "step": 1670 + }, + { + "epoch": 2.019242333132892, + "grad_norm": 0.87109375, + "learning_rate": 1.3199009407098686e-05, + "loss": 0.3069, + "step": 1680 + }, + { + "epoch": 2.03126879134095, + "grad_norm": 0.8203125, + "learning_rate": 1.3126495609659497e-05, + "loss": 0.3005, + "step": 1690 + }, + { + "epoch": 2.0432952495490078, + "grad_norm": 0.80078125, + "learning_rate": 1.3053799140293273e-05, + "loss": 0.2981, + "step": 1700 + }, + { + "epoch": 2.0553217077570656, + "grad_norm": 0.92578125, + "learning_rate": 1.2980924246440533e-05, + "loss": 0.2818, + "step": 1710 + }, + { + "epoch": 2.0673481659651234, + "grad_norm": 0.92578125, + "learning_rate": 1.2907875185966623e-05, + "loss": 0.3, + "step": 1720 + }, + { + "epoch": 2.079374624173181, + "grad_norm": 0.9921875, + "learning_rate": 1.283465622691292e-05, + "loss": 0.3039, + "step": 1730 + }, + { + "epoch": 2.0914010823812386, + "grad_norm": 1.15625, + "learning_rate": 1.2761271647247482e-05, + "loss": 0.262, + "step": 1740 + }, + { + "epoch": 2.1034275405892964, + "grad_norm": 0.9453125, + "learning_rate": 1.2687725734615089e-05, + "loss": 0.289, + "step": 1750 + }, + { + "epoch": 2.115453998797354, + "grad_norm": 0.80859375, + "learning_rate": 1.2614022786086735e-05, + "loss": 0.2881, + "step": 1760 + }, + { + "epoch": 2.127480457005412, + "grad_norm": 0.94140625, + "learning_rate": 1.254016710790855e-05, + "loss": 0.2775, + "step": 1770 + }, + { + "epoch": 2.13950691521347, + "grad_norm": 0.7421875, + "learning_rate": 1.2466163015250216e-05, + "loss": 0.2843, + "step": 1780 + }, + { + "epoch": 2.151533373421527, + "grad_norm": 0.80859375, + "learning_rate": 1.2392014831952834e-05, + "loss": 0.289, + "step": 1790 + }, + { + "epoch": 2.163559831629585, + "grad_norm": 0.8203125, + "learning_rate": 1.2317726890276292e-05, + "loss": 0.2979, + "step": 1800 + }, + { + "epoch": 2.175586289837643, + "grad_norm": 0.96875, + "learning_rate": 1.2243303530646153e-05, + "loss": 0.2703, + "step": 1810 + }, + { + "epoch": 2.1876127480457006, + "grad_norm": 0.91796875, + "learning_rate": 1.2168749101400049e-05, + "loss": 0.2687, + "step": 1820 + }, + { + "epoch": 2.1996392062537584, + "grad_norm": 0.734375, + "learning_rate": 1.2094067958533625e-05, + "loss": 0.2906, + "step": 1830 + }, + { + "epoch": 2.2116656644618162, + "grad_norm": 0.83984375, + "learning_rate": 1.201926446544603e-05, + "loss": 0.2868, + "step": 1840 + }, + { + "epoch": 2.2236921226698736, + "grad_norm": 1.4921875, + "learning_rate": 1.1944342992684977e-05, + "loss": 0.3086, + "step": 1850 + }, + { + "epoch": 2.2357185808779314, + "grad_norm": 0.94921875, + "learning_rate": 1.1869307917691381e-05, + "loss": 0.305, + "step": 1860 + }, + { + "epoch": 2.247745039085989, + "grad_norm": 1.1640625, + "learning_rate": 1.1794163624543598e-05, + "loss": 0.2888, + "step": 1870 + }, + { + "epoch": 2.259771497294047, + "grad_norm": 0.765625, + "learning_rate": 1.1718914503701289e-05, + "loss": 0.2839, + "step": 1880 + }, + { + "epoch": 2.2717979555021044, + "grad_norm": 0.93359375, + "learning_rate": 1.1643564951748877e-05, + "loss": 0.3057, + "step": 1890 + }, + { + "epoch": 2.283824413710162, + "grad_norm": 0.84765625, + "learning_rate": 1.156811937113869e-05, + "loss": 0.282, + "step": 1900 + }, + { + "epoch": 2.29585087191822, + "grad_norm": 0.859375, + "learning_rate": 1.1492582169933715e-05, + "loss": 0.3051, + "step": 1910 + }, + { + "epoch": 2.307877330126278, + "grad_norm": 1.0859375, + "learning_rate": 1.1416957761550072e-05, + "loss": 0.2816, + "step": 1920 + }, + { + "epoch": 2.3199037883343356, + "grad_norm": 0.9296875, + "learning_rate": 1.1341250564499132e-05, + "loss": 0.2889, + "step": 1930 + }, + { + "epoch": 2.3319302465423934, + "grad_norm": 0.72265625, + "learning_rate": 1.1265465002129365e-05, + "loss": 0.3014, + "step": 1940 + }, + { + "epoch": 2.343956704750451, + "grad_norm": 0.6875, + "learning_rate": 1.1189605502367901e-05, + "loss": 0.2762, + "step": 1950 + }, + { + "epoch": 2.3559831629585086, + "grad_norm": 0.98046875, + "learning_rate": 1.1113676497461807e-05, + "loss": 0.2937, + "step": 1960 + }, + { + "epoch": 2.3680096211665664, + "grad_norm": 0.9375, + "learning_rate": 1.1037682423719137e-05, + "loss": 0.3122, + "step": 1970 + }, + { + "epoch": 2.3800360793746242, + "grad_norm": 0.828125, + "learning_rate": 1.0961627721249717e-05, + "loss": 0.299, + "step": 1980 + }, + { + "epoch": 2.392062537582682, + "grad_norm": 0.84765625, + "learning_rate": 1.0885516833705743e-05, + "loss": 0.2899, + "step": 1990 + }, + { + "epoch": 2.4040889957907394, + "grad_norm": 0.796875, + "learning_rate": 1.0809354208022125e-05, + "loss": 0.3103, + "step": 2000 + }, + { + "epoch": 2.4161154539987972, + "grad_norm": 0.9140625, + "learning_rate": 1.0733144294156695e-05, + "loss": 0.2846, + "step": 2010 + }, + { + "epoch": 2.428141912206855, + "grad_norm": 0.90234375, + "learning_rate": 1.0656891544830179e-05, + "loss": 0.287, + "step": 2020 + }, + { + "epoch": 2.440168370414913, + "grad_norm": 0.94921875, + "learning_rate": 1.058060041526607e-05, + "loss": 0.2998, + "step": 2030 + }, + { + "epoch": 2.4521948286229707, + "grad_norm": 0.82421875, + "learning_rate": 1.0504275362930286e-05, + "loss": 0.2828, + "step": 2040 + }, + { + "epoch": 2.4642212868310285, + "grad_norm": 0.84375, + "learning_rate": 1.0427920847270769e-05, + "loss": 0.2806, + "step": 2050 + }, + { + "epoch": 2.476247745039086, + "grad_norm": 0.8984375, + "learning_rate": 1.0351541329456907e-05, + "loss": 0.3148, + "step": 2060 + }, + { + "epoch": 2.4882742032471437, + "grad_norm": 0.796875, + "learning_rate": 1.0275141272118894e-05, + "loss": 0.28, + "step": 2070 + }, + { + "epoch": 2.5003006614552015, + "grad_norm": 0.77734375, + "learning_rate": 1.0198725139086985e-05, + "loss": 0.2655, + "step": 2080 + }, + { + "epoch": 2.5123271196632593, + "grad_norm": 0.76171875, + "learning_rate": 1.0122297395130692e-05, + "loss": 0.2716, + "step": 2090 + }, + { + "epoch": 2.5243535778713166, + "grad_norm": 1.2265625, + "learning_rate": 1.004586250569792e-05, + "loss": 0.2692, + "step": 2100 + }, + { + "epoch": 2.5363800360793745, + "grad_norm": 0.90234375, + "learning_rate": 9.969424936654067e-06, + "loss": 0.2487, + "step": 2110 + }, + { + "epoch": 2.5484064942874323, + "grad_norm": 0.94921875, + "learning_rate": 9.89298915402108e-06, + "loss": 0.2857, + "step": 2120 + }, + { + "epoch": 2.56043295249549, + "grad_norm": 0.78125, + "learning_rate": 9.816559623716546e-06, + "loss": 0.2729, + "step": 2130 + }, + { + "epoch": 2.572459410703548, + "grad_norm": 1.0546875, + "learning_rate": 9.740140811292734e-06, + "loss": 0.309, + "step": 2140 + }, + { + "epoch": 2.5844858689116057, + "grad_norm": 0.79296875, + "learning_rate": 9.663737181675704e-06, + "loss": 0.2575, + "step": 2150 + }, + { + "epoch": 2.5965123271196635, + "grad_norm": 0.734375, + "learning_rate": 9.58735319890444e-06, + "loss": 0.2889, + "step": 2160 + }, + { + "epoch": 2.608538785327721, + "grad_norm": 0.96875, + "learning_rate": 9.510993325870004e-06, + "loss": 0.2594, + "step": 2170 + }, + { + "epoch": 2.6205652435357787, + "grad_norm": 0.76171875, + "learning_rate": 9.4346620240548e-06, + "loss": 0.2949, + "step": 2180 + }, + { + "epoch": 2.6325917017438365, + "grad_norm": 0.77734375, + "learning_rate": 9.358363753271905e-06, + "loss": 0.2885, + "step": 2190 + }, + { + "epoch": 2.6446181599518943, + "grad_norm": 1.1015625, + "learning_rate": 9.282102971404485e-06, + "loss": 0.3021, + "step": 2200 + }, + { + "epoch": 2.6566446181599517, + "grad_norm": 1.015625, + "learning_rate": 9.205884134145344e-06, + "loss": 0.2882, + "step": 2210 + }, + { + "epoch": 2.6686710763680095, + "grad_norm": 0.9453125, + "learning_rate": 9.12971169473658e-06, + "loss": 0.3009, + "step": 2220 + }, + { + "epoch": 2.6806975345760673, + "grad_norm": 0.95703125, + "learning_rate": 9.053590103709406e-06, + "loss": 0.3009, + "step": 2230 + }, + { + "epoch": 2.692723992784125, + "grad_norm": 0.8125, + "learning_rate": 8.977523808624116e-06, + "loss": 0.261, + "step": 2240 + }, + { + "epoch": 2.704750450992183, + "grad_norm": 0.828125, + "learning_rate": 8.901517253810215e-06, + "loss": 0.2604, + "step": 2250 + }, + { + "epoch": 2.7167769092002407, + "grad_norm": 0.828125, + "learning_rate": 8.825574880106771e-06, + "loss": 0.2923, + "step": 2260 + }, + { + "epoch": 2.728803367408298, + "grad_norm": 0.90625, + "learning_rate": 8.749701124602932e-06, + "loss": 0.2661, + "step": 2270 + }, + { + "epoch": 2.740829825616356, + "grad_norm": 1.8125, + "learning_rate": 8.673900420378691e-06, + "loss": 0.2751, + "step": 2280 + }, + { + "epoch": 2.7528562838244137, + "grad_norm": 2.578125, + "learning_rate": 8.598177196245867e-06, + "loss": 0.2703, + "step": 2290 + }, + { + "epoch": 2.7648827420324715, + "grad_norm": 0.9296875, + "learning_rate": 8.522535876489349e-06, + "loss": 0.2773, + "step": 2300 + }, + { + "epoch": 2.776909200240529, + "grad_norm": 0.79296875, + "learning_rate": 8.446980880608589e-06, + "loss": 0.276, + "step": 2310 + }, + { + "epoch": 2.7889356584485867, + "grad_norm": 0.8828125, + "learning_rate": 8.371516623059394e-06, + "loss": 0.261, + "step": 2320 + }, + { + "epoch": 2.8009621166566445, + "grad_norm": 0.82421875, + "learning_rate": 8.296147512995993e-06, + "loss": 0.2894, + "step": 2330 + }, + { + "epoch": 2.8129885748647023, + "grad_norm": 4.09375, + "learning_rate": 8.220877954013435e-06, + "loss": 0.2721, + "step": 2340 + }, + { + "epoch": 2.82501503307276, + "grad_norm": 0.92578125, + "learning_rate": 8.145712343890288e-06, + "loss": 0.3016, + "step": 2350 + }, + { + "epoch": 2.837041491280818, + "grad_norm": 1.0625, + "learning_rate": 8.070655074331692e-06, + "loss": 0.3162, + "step": 2360 + }, + { + "epoch": 2.8490679494888758, + "grad_norm": 0.890625, + "learning_rate": 7.995710530712772e-06, + "loss": 0.3188, + "step": 2370 + }, + { + "epoch": 2.861094407696933, + "grad_norm": 0.81640625, + "learning_rate": 7.92088309182241e-06, + "loss": 0.2633, + "step": 2380 + }, + { + "epoch": 2.873120865904991, + "grad_norm": 0.82421875, + "learning_rate": 7.846177129607391e-06, + "loss": 0.2797, + "step": 2390 + }, + { + "epoch": 2.8851473241130488, + "grad_norm": 0.74609375, + "learning_rate": 7.771597008916992e-06, + "loss": 0.293, + "step": 2400 + }, + { + "epoch": 2.8971737823211066, + "grad_norm": 0.91796875, + "learning_rate": 7.697147087247932e-06, + "loss": 0.3114, + "step": 2410 + }, + { + "epoch": 2.909200240529164, + "grad_norm": 2.21875, + "learning_rate": 7.622831714489782e-06, + "loss": 0.3038, + "step": 2420 + }, + { + "epoch": 2.9212266987372217, + "grad_norm": 0.87890625, + "learning_rate": 7.548655232670828e-06, + "loss": 0.3096, + "step": 2430 + }, + { + "epoch": 2.9332531569452795, + "grad_norm": 1.125, + "learning_rate": 7.474621975704353e-06, + "loss": 0.2976, + "step": 2440 + }, + { + "epoch": 2.9452796151533374, + "grad_norm": 0.8515625, + "learning_rate": 7.400736269135448e-06, + "loss": 0.2838, + "step": 2450 + }, + { + "epoch": 2.957306073361395, + "grad_norm": 0.9453125, + "learning_rate": 7.327002429888257e-06, + "loss": 0.2764, + "step": 2460 + }, + { + "epoch": 2.969332531569453, + "grad_norm": 0.94921875, + "learning_rate": 7.253424766013768e-06, + "loss": 0.3026, + "step": 2470 + }, + { + "epoch": 2.981358989777511, + "grad_norm": 0.87109375, + "learning_rate": 7.1800075764381086e-06, + "loss": 0.3092, + "step": 2480 + }, + { + "epoch": 2.993385447985568, + "grad_norm": 0.9375, + "learning_rate": 7.106755150711356e-06, + "loss": 0.2882, + "step": 2490 + }, + { + "epoch": 3.004810583283223, + "grad_norm": 0.87109375, + "learning_rate": 7.033671768756924e-06, + "loss": 0.2851, + "step": 2500 + }, + { + "epoch": 3.016837041491281, + "grad_norm": 0.81640625, + "learning_rate": 6.960761700621504e-06, + "loss": 0.2679, + "step": 2510 + }, + { + "epoch": 3.0288634996993387, + "grad_norm": 0.73828125, + "learning_rate": 6.888029206225562e-06, + "loss": 0.2953, + "step": 2520 + }, + { + "epoch": 3.0408899579073965, + "grad_norm": 0.9609375, + "learning_rate": 6.815478535114458e-06, + "loss": 0.2658, + "step": 2530 + }, + { + "epoch": 3.052916416115454, + "grad_norm": 1.09375, + "learning_rate": 6.743113926210154e-06, + "loss": 0.2695, + "step": 2540 + }, + { + "epoch": 3.0649428743235116, + "grad_norm": 0.71875, + "learning_rate": 6.670939607563544e-06, + "loss": 0.2796, + "step": 2550 + }, + { + "epoch": 3.0769693325315695, + "grad_norm": 0.87109375, + "learning_rate": 6.598959796107423e-06, + "loss": 0.2839, + "step": 2560 + }, + { + "epoch": 3.0889957907396273, + "grad_norm": 0.8203125, + "learning_rate": 6.527178697410101e-06, + "loss": 0.2702, + "step": 2570 + }, + { + "epoch": 3.101022248947685, + "grad_norm": 0.91015625, + "learning_rate": 6.455600505429693e-06, + "loss": 0.2815, + "step": 2580 + }, + { + "epoch": 3.1130487071557424, + "grad_norm": 0.88671875, + "learning_rate": 6.384229402269069e-06, + "loss": 0.272, + "step": 2590 + }, + { + "epoch": 3.1250751653638003, + "grad_norm": 0.87890625, + "learning_rate": 6.3130695579315086e-06, + "loss": 0.2801, + "step": 2600 + }, + { + "epoch": 3.137101623571858, + "grad_norm": 0.8359375, + "learning_rate": 6.242125130077062e-06, + "loss": 0.2813, + "step": 2610 + }, + { + "epoch": 3.149128081779916, + "grad_norm": 0.8046875, + "learning_rate": 6.1714002637796274e-06, + "loss": 0.2775, + "step": 2620 + }, + { + "epoch": 3.1611545399879737, + "grad_norm": 1.4765625, + "learning_rate": 6.100899091284773e-06, + "loss": 0.2937, + "step": 2630 + }, + { + "epoch": 3.173180998196031, + "grad_norm": 0.84765625, + "learning_rate": 6.030625731768292e-06, + "loss": 0.2828, + "step": 2640 + }, + { + "epoch": 3.185207456404089, + "grad_norm": 0.80859375, + "learning_rate": 5.960584291095534e-06, + "loss": 0.2655, + "step": 2650 + }, + { + "epoch": 3.1972339146121467, + "grad_norm": 0.78125, + "learning_rate": 5.890778861581524e-06, + "loss": 0.2907, + "step": 2660 + }, + { + "epoch": 3.2092603728202045, + "grad_norm": 0.77734375, + "learning_rate": 5.821213521751843e-06, + "loss": 0.2921, + "step": 2670 + }, + { + "epoch": 3.2212868310282623, + "grad_norm": 0.91015625, + "learning_rate": 5.751892336104337e-06, + "loss": 0.3095, + "step": 2680 + }, + { + "epoch": 3.23331328923632, + "grad_norm": 0.98828125, + "learning_rate": 5.68281935487165e-06, + "loss": 0.2533, + "step": 2690 + }, + { + "epoch": 3.2453397474443775, + "grad_norm": 1.1171875, + "learning_rate": 5.613998613784565e-06, + "loss": 0.2944, + "step": 2700 + }, + { + "epoch": 3.2573662056524353, + "grad_norm": 1.484375, + "learning_rate": 5.545434133836226e-06, + "loss": 0.307, + "step": 2710 + }, + { + "epoch": 3.269392663860493, + "grad_norm": 0.94140625, + "learning_rate": 5.4771299210471815e-06, + "loss": 0.2876, + "step": 2720 + }, + { + "epoch": 3.281419122068551, + "grad_norm": 0.8515625, + "learning_rate": 5.409089966231353e-06, + "loss": 0.2812, + "step": 2730 + }, + { + "epoch": 3.2934455802766087, + "grad_norm": 0.82421875, + "learning_rate": 5.341318244762831e-06, + "loss": 0.2517, + "step": 2740 + }, + { + "epoch": 3.305472038484666, + "grad_norm": 0.73046875, + "learning_rate": 5.273818716343626e-06, + "loss": 0.2396, + "step": 2750 + }, + { + "epoch": 3.317498496692724, + "grad_norm": 0.9296875, + "learning_rate": 5.20659532477232e-06, + "loss": 0.29, + "step": 2760 + }, + { + "epoch": 3.3295249549007817, + "grad_norm": 0.9375, + "learning_rate": 5.13965199771362e-06, + "loss": 0.3046, + "step": 2770 + }, + { + "epoch": 3.3415514131088395, + "grad_norm": 0.80859375, + "learning_rate": 5.0729926464689e-06, + "loss": 0.2467, + "step": 2780 + }, + { + "epoch": 3.3535778713168973, + "grad_norm": 1.296875, + "learning_rate": 5.006621165747649e-06, + "loss": 0.2675, + "step": 2790 + }, + { + "epoch": 3.365604329524955, + "grad_norm": 0.921875, + "learning_rate": 4.94054143343995e-06, + "loss": 0.2609, + "step": 2800 + }, + { + "epoch": 3.3776307877330125, + "grad_norm": 0.796875, + "learning_rate": 4.87475731038987e-06, + "loss": 0.249, + "step": 2810 + }, + { + "epoch": 3.3896572459410703, + "grad_norm": 0.94921875, + "learning_rate": 4.809272640169895e-06, + "loss": 0.2446, + "step": 2820 + }, + { + "epoch": 3.401683704149128, + "grad_norm": 2.140625, + "learning_rate": 4.744091248856384e-06, + "loss": 0.2829, + "step": 2830 + }, + { + "epoch": 3.413710162357186, + "grad_norm": 0.9921875, + "learning_rate": 4.6792169448059795e-06, + "loss": 0.2811, + "step": 2840 + }, + { + "epoch": 3.4257366205652433, + "grad_norm": 0.94140625, + "learning_rate": 4.614653518433142e-06, + "loss": 0.2659, + "step": 2850 + }, + { + "epoch": 3.437763078773301, + "grad_norm": 0.7421875, + "learning_rate": 4.550404741988644e-06, + "loss": 0.2771, + "step": 2860 + }, + { + "epoch": 3.449789536981359, + "grad_norm": 0.7890625, + "learning_rate": 4.486474369339211e-06, + "loss": 0.2765, + "step": 2870 + }, + { + "epoch": 3.4618159951894167, + "grad_norm": 0.72265625, + "learning_rate": 4.422866135748155e-06, + "loss": 0.2769, + "step": 2880 + }, + { + "epoch": 3.4738424533974746, + "grad_norm": 0.75, + "learning_rate": 4.359583757657158e-06, + "loss": 0.2525, + "step": 2890 + }, + { + "epoch": 3.4858689116055324, + "grad_norm": 0.90625, + "learning_rate": 4.296630932469127e-06, + "loss": 0.2806, + "step": 2900 + }, + { + "epoch": 3.4978953698135897, + "grad_norm": 1.21875, + "learning_rate": 4.2340113383321655e-06, + "loss": 0.2706, + "step": 2910 + }, + { + "epoch": 3.5099218280216475, + "grad_norm": 1.1796875, + "learning_rate": 4.17172863392467e-06, + "loss": 0.2772, + "step": 2920 + }, + { + "epoch": 3.5219482862297053, + "grad_norm": 0.8046875, + "learning_rate": 4.1097864582415524e-06, + "loss": 0.3001, + "step": 2930 + }, + { + "epoch": 3.533974744437763, + "grad_norm": 0.84765625, + "learning_rate": 4.04818843038164e-06, + "loss": 0.2898, + "step": 2940 + }, + { + "epoch": 3.546001202645821, + "grad_norm": 0.875, + "learning_rate": 3.986938149336222e-06, + "loss": 0.2692, + "step": 2950 + }, + { + "epoch": 3.5580276608538783, + "grad_norm": 0.86328125, + "learning_rate": 3.9260391937787565e-06, + "loss": 0.2922, + "step": 2960 + }, + { + "epoch": 3.570054119061936, + "grad_norm": 0.84375, + "learning_rate": 3.865495121855799e-06, + "loss": 0.2824, + "step": 2970 + }, + { + "epoch": 3.582080577269994, + "grad_norm": 0.7578125, + "learning_rate": 3.805309470979087e-06, + "loss": 0.2504, + "step": 2980 + }, + { + "epoch": 3.5941070354780518, + "grad_norm": 1.0078125, + "learning_rate": 3.7454857576188875e-06, + "loss": 0.2892, + "step": 2990 + }, + { + "epoch": 3.6061334936861096, + "grad_norm": 0.90625, + "learning_rate": 3.6860274770985116e-06, + "loss": 0.2858, + "step": 3000 + }, + { + "epoch": 3.6181599518941674, + "grad_norm": 0.890625, + "learning_rate": 3.6269381033901084e-06, + "loss": 0.2898, + "step": 3010 + }, + { + "epoch": 3.6301864101022248, + "grad_norm": 1.6328125, + "learning_rate": 3.5682210889116974e-06, + "loss": 0.3009, + "step": 3020 + }, + { + "epoch": 3.6422128683102826, + "grad_norm": 0.921875, + "learning_rate": 3.509879864325435e-06, + "loss": 0.258, + "step": 3030 + }, + { + "epoch": 3.6542393265183404, + "grad_norm": 0.84375, + "learning_rate": 3.451917838337192e-06, + "loss": 0.2618, + "step": 3040 + }, + { + "epoch": 3.666265784726398, + "grad_norm": 0.80078125, + "learning_rate": 3.3943383974973685e-06, + "loss": 0.2775, + "step": 3050 + }, + { + "epoch": 3.6782922429344556, + "grad_norm": 1.125, + "learning_rate": 3.3371449060030568e-06, + "loss": 0.2701, + "step": 3060 + }, + { + "epoch": 3.6903187011425134, + "grad_norm": 0.7265625, + "learning_rate": 3.280340705501455e-06, + "loss": 0.2815, + "step": 3070 + }, + { + "epoch": 3.702345159350571, + "grad_norm": 0.90234375, + "learning_rate": 3.2239291148946363e-06, + "loss": 0.2832, + "step": 3080 + }, + { + "epoch": 3.714371617558629, + "grad_norm": 0.76171875, + "learning_rate": 3.167913430145644e-06, + "loss": 0.2863, + "step": 3090 + }, + { + "epoch": 3.726398075766687, + "grad_norm": 0.796875, + "learning_rate": 3.1122969240858935e-06, + "loss": 0.2486, + "step": 3100 + }, + { + "epoch": 3.7384245339747446, + "grad_norm": 0.9296875, + "learning_rate": 3.057082846223982e-06, + "loss": 0.2483, + "step": 3110 + }, + { + "epoch": 3.7504509921828024, + "grad_norm": 0.796875, + "learning_rate": 3.0022744225557987e-06, + "loss": 0.2812, + "step": 3120 + }, + { + "epoch": 3.76247745039086, + "grad_norm": 0.7734375, + "learning_rate": 2.9478748553760637e-06, + "loss": 0.249, + "step": 3130 + }, + { + "epoch": 3.7745039085989176, + "grad_norm": 0.6796875, + "learning_rate": 2.89388732309121e-06, + "loss": 0.2609, + "step": 3140 + }, + { + "epoch": 3.7865303668069754, + "grad_norm": 0.98828125, + "learning_rate": 2.840314980033687e-06, + "loss": 0.2657, + "step": 3150 + }, + { + "epoch": 3.7985568250150332, + "grad_norm": 0.68359375, + "learning_rate": 2.787160956277666e-06, + "loss": 0.2657, + "step": 3160 + }, + { + "epoch": 3.8105832832230906, + "grad_norm": 0.91015625, + "learning_rate": 2.734428357456145e-06, + "loss": 0.2867, + "step": 3170 + }, + { + "epoch": 3.8226097414311484, + "grad_norm": 0.7265625, + "learning_rate": 2.682120264579514e-06, + "loss": 0.257, + "step": 3180 + }, + { + "epoch": 3.834636199639206, + "grad_norm": 0.890625, + "learning_rate": 2.6302397338555243e-06, + "loss": 0.2805, + "step": 3190 + }, + { + "epoch": 3.846662657847264, + "grad_norm": 1.0, + "learning_rate": 2.5787897965107368e-06, + "loss": 0.2527, + "step": 3200 + }, + { + "epoch": 3.858689116055322, + "grad_norm": 0.76953125, + "learning_rate": 2.527773458613404e-06, + "loss": 0.2908, + "step": 3210 + }, + { + "epoch": 3.8707155742633796, + "grad_norm": 1.1796875, + "learning_rate": 2.47719370089784e-06, + "loss": 0.255, + "step": 3220 + }, + { + "epoch": 3.882742032471437, + "grad_norm": 0.9921875, + "learning_rate": 2.4270534785902723e-06, + "loss": 0.3079, + "step": 3230 + }, + { + "epoch": 3.894768490679495, + "grad_norm": 1.0390625, + "learning_rate": 2.377355721236159e-06, + "loss": 0.2571, + "step": 3240 + }, + { + "epoch": 3.9067949488875526, + "grad_norm": 0.85546875, + "learning_rate": 2.3281033325290458e-06, + "loss": 0.3038, + "step": 3250 + }, + { + "epoch": 3.9188214070956104, + "grad_norm": 0.8359375, + "learning_rate": 2.2792991901408914e-06, + "loss": 0.2857, + "step": 3260 + }, + { + "epoch": 3.930847865303668, + "grad_norm": 0.8515625, + "learning_rate": 2.2309461455539474e-06, + "loss": 0.2911, + "step": 3270 + }, + { + "epoch": 3.9428743235117256, + "grad_norm": 0.921875, + "learning_rate": 2.1830470238941514e-06, + "loss": 0.2809, + "step": 3280 + }, + { + "epoch": 3.9549007817197834, + "grad_norm": 0.8828125, + "learning_rate": 2.1356046237660543e-06, + "loss": 0.2843, + "step": 3290 + }, + { + "epoch": 3.9669272399278412, + "grad_norm": 0.91796875, + "learning_rate": 2.0886217170893252e-06, + "loss": 0.274, + "step": 3300 + }, + { + "epoch": 3.978953698135899, + "grad_norm": 0.90625, + "learning_rate": 2.042101048936773e-06, + "loss": 0.2725, + "step": 3310 + }, + { + "epoch": 3.990980156343957, + "grad_norm": 0.8828125, + "learning_rate": 1.996045337373974e-06, + "loss": 0.2818, + "step": 3320 + }, + { + "epoch": 4.002405291641612, + "grad_norm": 0.80859375, + "learning_rate": 1.9504572733004702e-06, + "loss": 0.2622, + "step": 3330 + }, + { + "epoch": 4.0144317498496696, + "grad_norm": 0.93359375, + "learning_rate": 1.9053395202925252e-06, + "loss": 0.2648, + "step": 3340 + }, + { + "epoch": 4.026458208057727, + "grad_norm": 0.9296875, + "learning_rate": 1.860694714447524e-06, + "loss": 0.2998, + "step": 3350 + }, + { + "epoch": 4.038484666265784, + "grad_norm": 0.8515625, + "learning_rate": 1.8165254642299323e-06, + "loss": 0.2615, + "step": 3360 + }, + { + "epoch": 4.050511124473842, + "grad_norm": 1.0625, + "learning_rate": 1.7728343503189126e-06, + "loss": 0.2773, + "step": 3370 + }, + { + "epoch": 4.0625375826819, + "grad_norm": 0.84375, + "learning_rate": 1.7296239254575231e-06, + "loss": 0.266, + "step": 3380 + }, + { + "epoch": 4.074564040889958, + "grad_norm": 0.82421875, + "learning_rate": 1.6868967143035764e-06, + "loss": 0.2465, + "step": 3390 + }, + { + "epoch": 4.0865904990980155, + "grad_norm": 0.83984375, + "learning_rate": 1.6446552132821413e-06, + "loss": 0.2516, + "step": 3400 + }, + { + "epoch": 4.098616957306073, + "grad_norm": 1.0703125, + "learning_rate": 1.602901890439663e-06, + "loss": 0.295, + "step": 3410 + }, + { + "epoch": 4.110643415514131, + "grad_norm": 0.70703125, + "learning_rate": 1.5616391852997835e-06, + "loss": 0.2872, + "step": 3420 + }, + { + "epoch": 4.122669873722189, + "grad_norm": 0.76171875, + "learning_rate": 1.5208695087207914e-06, + "loss": 0.2714, + "step": 3430 + }, + { + "epoch": 4.134696331930247, + "grad_norm": 0.77734375, + "learning_rate": 1.4805952427547732e-06, + "loss": 0.2662, + "step": 3440 + }, + { + "epoch": 4.146722790138305, + "grad_norm": 0.86328125, + "learning_rate": 1.4408187405084318e-06, + "loss": 0.2765, + "step": 3450 + }, + { + "epoch": 4.158749248346362, + "grad_norm": 0.9921875, + "learning_rate": 1.4015423260055983e-06, + "loss": 0.2724, + "step": 3460 + }, + { + "epoch": 4.170775706554419, + "grad_norm": 0.97265625, + "learning_rate": 1.3627682940514586e-06, + "loss": 0.2828, + "step": 3470 + }, + { + "epoch": 4.182802164762477, + "grad_norm": 0.8046875, + "learning_rate": 1.3244989100984573e-06, + "loss": 0.2956, + "step": 3480 + }, + { + "epoch": 4.194828622970535, + "grad_norm": 0.91796875, + "learning_rate": 1.2867364101139522e-06, + "loss": 0.2706, + "step": 3490 + }, + { + "epoch": 4.206855081178593, + "grad_norm": 1.03125, + "learning_rate": 1.2494830004495551e-06, + "loss": 0.2728, + "step": 3500 + }, + { + "epoch": 4.218881539386651, + "grad_norm": 0.91015625, + "learning_rate": 1.21274085771224e-06, + "loss": 0.2736, + "step": 3510 + }, + { + "epoch": 4.230907997594708, + "grad_norm": 0.9765625, + "learning_rate": 1.1765121286371505e-06, + "loss": 0.2695, + "step": 3520 + }, + { + "epoch": 4.242934455802766, + "grad_norm": 0.8515625, + "learning_rate": 1.1407989299621913e-06, + "loss": 0.2659, + "step": 3530 + }, + { + "epoch": 4.254960914010824, + "grad_norm": 0.984375, + "learning_rate": 1.1056033483043405e-06, + "loss": 0.2649, + "step": 3540 + }, + { + "epoch": 4.266987372218882, + "grad_norm": 0.99609375, + "learning_rate": 1.0709274400377355e-06, + "loss": 0.2771, + "step": 3550 + }, + { + "epoch": 4.27901383042694, + "grad_norm": 1.046875, + "learning_rate": 1.036773231173538e-06, + "loss": 0.263, + "step": 3560 + }, + { + "epoch": 4.291040288634997, + "grad_norm": 0.859375, + "learning_rate": 1.0031427172415397e-06, + "loss": 0.2676, + "step": 3570 + }, + { + "epoch": 4.303066746843054, + "grad_norm": 0.78125, + "learning_rate": 9.700378631735918e-07, + "loss": 0.2565, + "step": 3580 + }, + { + "epoch": 4.315093205051112, + "grad_norm": 0.85546875, + "learning_rate": 9.374606031887767e-07, + "loss": 0.2778, + "step": 3590 + }, + { + "epoch": 4.32711966325917, + "grad_norm": 0.7578125, + "learning_rate": 9.054128406804185e-07, + "loss": 0.2817, + "step": 3600 + }, + { + "epoch": 4.339146121467228, + "grad_norm": 0.96875, + "learning_rate": 8.738964481048584e-07, + "loss": 0.2666, + "step": 3610 + }, + { + "epoch": 4.351172579675286, + "grad_norm": 0.9921875, + "learning_rate": 8.429132668720574e-07, + "loss": 0.2394, + "step": 3620 + }, + { + "epoch": 4.363199037883343, + "grad_norm": 0.875, + "learning_rate": 8.124651072380152e-07, + "loss": 0.2643, + "step": 3630 + }, + { + "epoch": 4.375225496091401, + "grad_norm": 0.88671875, + "learning_rate": 7.825537481989864e-07, + "loss": 0.2693, + "step": 3640 + }, + { + "epoch": 4.387251954299459, + "grad_norm": 0.9296875, + "learning_rate": 7.531809373875565e-07, + "loss": 0.2834, + "step": 3650 + }, + { + "epoch": 4.399278412507517, + "grad_norm": 0.7421875, + "learning_rate": 7.243483909705229e-07, + "loss": 0.263, + "step": 3660 + }, + { + "epoch": 4.411304870715575, + "grad_norm": 0.88671875, + "learning_rate": 6.960577935486212e-07, + "loss": 0.2734, + "step": 3670 + }, + { + "epoch": 4.4233313289236325, + "grad_norm": 1.3046875, + "learning_rate": 6.683107980581105e-07, + "loss": 0.2961, + "step": 3680 + }, + { + "epoch": 4.435357787131689, + "grad_norm": 0.765625, + "learning_rate": 6.411090256741848e-07, + "loss": 0.2575, + "step": 3690 + }, + { + "epoch": 4.447384245339747, + "grad_norm": 1.0, + "learning_rate": 6.144540657162645e-07, + "loss": 0.2965, + "step": 3700 + }, + { + "epoch": 4.459410703547805, + "grad_norm": 1.0390625, + "learning_rate": 5.883474755551266e-07, + "loss": 0.28, + "step": 3710 + }, + { + "epoch": 4.471437161755863, + "grad_norm": 0.78125, + "learning_rate": 5.627907805219169e-07, + "loss": 0.2565, + "step": 3720 + }, + { + "epoch": 4.483463619963921, + "grad_norm": 0.7265625, + "learning_rate": 5.377854738190302e-07, + "loss": 0.266, + "step": 3730 + }, + { + "epoch": 4.495490078171978, + "grad_norm": 1.2890625, + "learning_rate": 5.133330164328653e-07, + "loss": 0.273, + "step": 3740 + }, + { + "epoch": 4.507516536380036, + "grad_norm": 0.765625, + "learning_rate": 4.894348370484648e-07, + "loss": 0.2763, + "step": 3750 + }, + { + "epoch": 4.519542994588094, + "grad_norm": 1.078125, + "learning_rate": 4.660923319660382e-07, + "loss": 0.2882, + "step": 3760 + }, + { + "epoch": 4.531569452796152, + "grad_norm": 0.83203125, + "learning_rate": 4.4330686501938835e-07, + "loss": 0.2963, + "step": 3770 + }, + { + "epoch": 4.543595911004209, + "grad_norm": 0.828125, + "learning_rate": 4.2107976749621415e-07, + "loss": 0.2585, + "step": 3780 + }, + { + "epoch": 4.5556223692122675, + "grad_norm": 1.5078125, + "learning_rate": 3.994123380603376e-07, + "loss": 0.271, + "step": 3790 + }, + { + "epoch": 4.567648827420324, + "grad_norm": 0.8828125, + "learning_rate": 3.783058426758246e-07, + "loss": 0.281, + "step": 3800 + }, + { + "epoch": 4.579675285628382, + "grad_norm": 1.0078125, + "learning_rate": 3.5776151453301267e-07, + "loss": 0.2838, + "step": 3810 + }, + { + "epoch": 4.59170174383644, + "grad_norm": 0.875, + "learning_rate": 3.3778055397647e-07, + "loss": 0.2859, + "step": 3820 + }, + { + "epoch": 4.603728202044498, + "grad_norm": 0.828125, + "learning_rate": 3.183641284348504e-07, + "loss": 0.2604, + "step": 3830 + }, + { + "epoch": 4.615754660252556, + "grad_norm": 0.921875, + "learning_rate": 2.995133723526944e-07, + "loss": 0.2832, + "step": 3840 + }, + { + "epoch": 4.6277811184606135, + "grad_norm": 4.53125, + "learning_rate": 2.812293871241401e-07, + "loss": 0.2856, + "step": 3850 + }, + { + "epoch": 4.639807576668671, + "grad_norm": 0.86328125, + "learning_rate": 2.6351324102857345e-07, + "loss": 0.2634, + "step": 3860 + }, + { + "epoch": 4.651834034876729, + "grad_norm": 0.7734375, + "learning_rate": 2.463659691682185e-07, + "loss": 0.2626, + "step": 3870 + }, + { + "epoch": 4.663860493084787, + "grad_norm": 0.7421875, + "learning_rate": 2.297885734076455e-07, + "loss": 0.2823, + "step": 3880 + }, + { + "epoch": 4.675886951292844, + "grad_norm": 1.046875, + "learning_rate": 2.1378202231525092e-07, + "loss": 0.2659, + "step": 3890 + }, + { + "epoch": 4.687913409500902, + "grad_norm": 0.76953125, + "learning_rate": 1.983472511066542e-07, + "loss": 0.253, + "step": 3900 + }, + { + "epoch": 4.699939867708959, + "grad_norm": 0.92578125, + "learning_rate": 1.8348516159006346e-07, + "loss": 0.2716, + "step": 3910 + }, + { + "epoch": 4.711966325917017, + "grad_norm": 0.95703125, + "learning_rate": 1.6919662211358212e-07, + "loss": 0.2812, + "step": 3920 + }, + { + "epoch": 4.723992784125075, + "grad_norm": 0.8984375, + "learning_rate": 1.554824675144706e-07, + "loss": 0.2459, + "step": 3930 + }, + { + "epoch": 4.736019242333133, + "grad_norm": 0.7890625, + "learning_rate": 1.4234349907038204e-07, + "loss": 0.2626, + "step": 3940 + }, + { + "epoch": 4.748045700541191, + "grad_norm": 0.86328125, + "learning_rate": 1.2978048445252745e-07, + "loss": 0.2714, + "step": 3950 + }, + { + "epoch": 4.7600721587492485, + "grad_norm": 0.79296875, + "learning_rate": 1.1779415768083946e-07, + "loss": 0.2722, + "step": 3960 + }, + { + "epoch": 4.772098616957306, + "grad_norm": 1.03125, + "learning_rate": 1.0638521908107657e-07, + "loss": 0.2544, + "step": 3970 + }, + { + "epoch": 4.784125075165364, + "grad_norm": 0.8515625, + "learning_rate": 9.555433524390701e-08, + "loss": 0.269, + "step": 3980 + }, + { + "epoch": 4.796151533373422, + "grad_norm": 0.76171875, + "learning_rate": 8.530213898596317e-08, + "loss": 0.2516, + "step": 3990 + }, + { + "epoch": 4.808177991581479, + "grad_norm": 0.96875, + "learning_rate": 7.562922931286687e-08, + "loss": 0.3124, + "step": 4000 + } + ], + "logging_steps": 10, + "max_steps": 4160, + "num_input_tokens_seen": 0, + "num_train_epochs": 5, + "save_steps": 200, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 6.305687538737086e+18, + "train_batch_size": 4, + "trial_name": null, + "trial_params": null +} diff --git a/training_args.bin b/training_args.bin new file mode 100644 index 0000000..b59388c --- /dev/null +++ b/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fe348d236181d593a53f346e5d1bd6ab9c926047f41ada724f0882071f24c332 +size 6161 diff --git a/vocab.json b/vocab.json new file mode 100644 index 0000000..6c49fc6 --- /dev/null +++ b/vocab.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ca10d7e9fb3ed18575dd1e277a2579c16d108e32f27439684afa0e10b1440910 +size 2776833