初始化项目，由ModelHub XC社区提供模型
Model: cs-552-2026-MMRF/general_knowledge_model Source: Original Platform
This commit is contained in:
ModelHub XC
49
.gitattributes
vendored
Normal file
49
.gitattributes
vendored
Normal file
@@ -0,0 +1,49 @@
|
||||
*.7z filter=lfs diff=lfs merge=lfs -text
|
||||
*.arrow filter=lfs diff=lfs merge=lfs -text
|
||||
*.bin filter=lfs diff=lfs merge=lfs -text
|
||||
*.bz2 filter=lfs diff=lfs merge=lfs -text
|
||||
*.ckpt filter=lfs diff=lfs merge=lfs -text
|
||||
*.ftz filter=lfs diff=lfs merge=lfs -text
|
||||
*.gz filter=lfs diff=lfs merge=lfs -text
|
||||
*.h5 filter=lfs diff=lfs merge=lfs -text
|
||||
*.joblib filter=lfs diff=lfs merge=lfs -text
|
||||
*.lfs.* filter=lfs diff=lfs merge=lfs -text
|
||||
*.mlmodel filter=lfs diff=lfs merge=lfs -text
|
||||
*.model filter=lfs diff=lfs merge=lfs -text
|
||||
*.msgpack filter=lfs diff=lfs merge=lfs -text
|
||||
*.npy filter=lfs diff=lfs merge=lfs -text
|
||||
*.npz filter=lfs diff=lfs merge=lfs -text
|
||||
*.onnx filter=lfs diff=lfs merge=lfs -text
|
||||
*.ot filter=lfs diff=lfs merge=lfs -text
|
||||
*.parquet filter=lfs diff=lfs merge=lfs -text
|
||||
*.pb filter=lfs diff=lfs merge=lfs -text
|
||||
*.pickle filter=lfs diff=lfs merge=lfs -text
|
||||
*.pkl filter=lfs diff=lfs merge=lfs -text
|
||||
*.pt filter=lfs diff=lfs merge=lfs -text
|
||||
*.pth filter=lfs diff=lfs merge=lfs -text
|
||||
*.rar filter=lfs diff=lfs merge=lfs -text
|
||||
*.safetensors filter=lfs diff=lfs merge=lfs -text
|
||||
saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
||||
*.tar.* filter=lfs diff=lfs merge=lfs -text
|
||||
*.tar filter=lfs diff=lfs merge=lfs -text
|
||||
*.tflite filter=lfs diff=lfs merge=lfs -text
|
||||
*.tgz filter=lfs diff=lfs merge=lfs -text
|
||||
*.wasm filter=lfs diff=lfs merge=lfs -text
|
||||
*.xz filter=lfs diff=lfs merge=lfs -text
|
||||
*.zip filter=lfs diff=lfs merge=lfs -text
|
||||
*.zst filter=lfs diff=lfs merge=lfs -text
|
||||
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
||||
checkpoint-1000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
||||
checkpoint-1500/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
||||
checkpoint-2000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
||||
checkpoint-2500/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
||||
checkpoint-3000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
||||
checkpoint-3500/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
||||
checkpoint-4000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
||||
checkpoint-4500/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
||||
checkpoint-500/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
||||
checkpoint-5000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
||||
checkpoint-5500/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
||||
checkpoint-6000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
||||
checkpoint-6241/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
||||
tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
||||
66
README.md
Normal file
66
README.md
Normal file
@@ -0,0 +1,66 @@
|
||||
---
|
||||
library_name: transformers
|
||||
model_name: general_knowledge_model_v8
|
||||
tags:
|
||||
- generated_from_trainer
|
||||
- grpo
|
||||
- trl
|
||||
licence: license
|
||||
---
|
||||
|
||||
# Model Card for general_knowledge_model_v8
|
||||
|
||||
This model is a fine-tuned version of [None](https://huggingface.co/None).
|
||||
It has been trained using [TRL](https://github.com/huggingface/trl).
|
||||
|
||||
## Quick start
|
||||
|
||||
```python
|
||||
from transformers import pipeline
|
||||
|
||||
question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
|
||||
generator = pipeline("text-generation", model="None", device="cuda")
|
||||
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
|
||||
print(output["generated_text"])
|
||||
```
|
||||
|
||||
## Training procedure
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
This model was trained with GRPO, a method introduced in [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://huggingface.co/papers/2402.03300).
|
||||
|
||||
### Framework versions
|
||||
|
||||
- TRL: 1.3.0
|
||||
- Transformers: 5.7.0
|
||||
- Pytorch: 2.10.0+cu128
|
||||
- Datasets: 4.8.5
|
||||
- Tokenizers: 0.22.2
|
||||
|
||||
## Citations
|
||||
|
||||
Cite GRPO as:
|
||||
|
||||
```bibtex
|
||||
@article{shao2024deepseekmath,
|
||||
title = {{DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models}},
|
||||
author = {Zhihong Shao and Peiyi Wang and Qihao Zhu and Runxin Xu and Junxiao Song and Mingchuan Zhang and Y. K. Li and Y. Wu and Daya Guo},
|
||||
year = 2024,
|
||||
eprint = {arXiv:2402.03300},
|
||||
}
|
||||
```
|
||||
|
||||
Cite TRL as:
|
||||
|
||||
```bibtex
|
||||
@software{vonwerra2020trl,
|
||||
title = {{TRL: Transformers Reinforcement Learning}},
|
||||
author = {von Werra, Leandro and Belkada, Younes and Tunstall, Lewis and Beeching, Edward and Thrush, Tristan and Lambert, Nathan and Huang, Shengyi and Rasul, Kashif and Gallouédec, Quentin},
|
||||
license = {Apache-2.0},
|
||||
url = {https://github.com/huggingface/trl},
|
||||
year = {2020}
|
||||
}
|
||||
```
|
||||
89
chat_template.jinja
Normal file
89
chat_template.jinja
Normal file
@@ -0,0 +1,89 @@
|
||||
{%- if tools %}
|
||||
{{- '<|im_start|>system\n' }}
|
||||
{%- if messages[0].role == 'system' %}
|
||||
{{- messages[0].content + '\n\n' }}
|
||||
{%- endif %}
|
||||
{{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
|
||||
{%- for tool in tools %}
|
||||
{{- "\n" }}
|
||||
{{- tool | tojson }}
|
||||
{%- endfor %}
|
||||
{{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
|
||||
{%- else %}
|
||||
{%- if messages[0].role == 'system' %}
|
||||
{{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
|
||||
{%- endif %}
|
||||
{%- endif %}
|
||||
{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
|
||||
{%- for message in messages[::-1] %}
|
||||
{%- set index = (messages|length - 1) - loop.index0 %}
|
||||
{%- if ns.multi_step_tool and message.role == "user" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
|
||||
{%- set ns.multi_step_tool = false %}
|
||||
{%- set ns.last_query_index = index %}
|
||||
{%- endif %}
|
||||
{%- endfor %}
|
||||
{%- for message in messages %}
|
||||
{%- if message.content is string %}
|
||||
{%- set content = message.content %}
|
||||
{%- else %}
|
||||
{%- set content = '' %}
|
||||
{%- endif %}
|
||||
{%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
|
||||
{{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
|
||||
{%- elif message.role == "assistant" %}
|
||||
{%- set reasoning_content = '' %}
|
||||
{%- if message.reasoning_content is string %}
|
||||
{%- set reasoning_content = message.reasoning_content %}
|
||||
{%- else %}
|
||||
{%- if '</think>' in content %}
|
||||
{%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
|
||||
{%- set content = content.split('</think>')[-1].lstrip('\n') %}
|
||||
{%- endif %}
|
||||
{%- endif %}
|
||||
{%- if loop.index0 > ns.last_query_index %}
|
||||
{%- if loop.last or (not loop.last and reasoning_content) %}
|
||||
{{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
|
||||
{%- else %}
|
||||
{{- '<|im_start|>' + message.role + '\n' + content }}
|
||||
{%- endif %}
|
||||
{%- else %}
|
||||
{{- '<|im_start|>' + message.role + '\n' + content }}
|
||||
{%- endif %}
|
||||
{%- if message.tool_calls %}
|
||||
{%- for tool_call in message.tool_calls %}
|
||||
{%- if (loop.first and content) or (not loop.first) %}
|
||||
{{- '\n' }}
|
||||
{%- endif %}
|
||||
{%- if tool_call.function %}
|
||||
{%- set tool_call = tool_call.function %}
|
||||
{%- endif %}
|
||||
{{- '<tool_call>\n{"name": "' }}
|
||||
{{- tool_call.name }}
|
||||
{{- '", "arguments": ' }}
|
||||
{%- if tool_call.arguments is string %}
|
||||
{{- tool_call.arguments }}
|
||||
{%- else %}
|
||||
{{- tool_call.arguments | tojson }}
|
||||
{%- endif %}
|
||||
{{- '}\n</tool_call>' }}
|
||||
{%- endfor %}
|
||||
{%- endif %}
|
||||
{{- '<|im_end|>\n' }}
|
||||
{%- elif message.role == "tool" %}
|
||||
{%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
|
||||
{{- '<|im_start|>user' }}
|
||||
{%- endif %}
|
||||
{{- '\n<tool_response>\n' }}
|
||||
{{- content }}
|
||||
{{- '\n</tool_response>' }}
|
||||
{%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
|
||||
{{- '<|im_end|>\n' }}
|
||||
{%- endif %}
|
||||
{%- endif %}
|
||||
{%- endfor %}
|
||||
{%- if add_generation_prompt %}
|
||||
{{- '<|im_start|>assistant\n' }}
|
||||
{%- if enable_thinking is defined and enable_thinking is false %}
|
||||
{{- '<think>\n\n</think>\n\n' }}
|
||||
{%- endif %}
|
||||
{%- endif %}
|
||||
89
checkpoint-1000/chat_template.jinja
Normal file
89
checkpoint-1000/chat_template.jinja
Normal file
@@ -0,0 +1,89 @@
|
||||
{%- if tools %}
|
||||
{{- '<|im_start|>system\n' }}
|
||||
{%- if messages[0].role == 'system' %}
|
||||
{{- messages[0].content + '\n\n' }}
|
||||
{%- endif %}
|
||||
{{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
|
||||
{%- for tool in tools %}
|
||||
{{- "\n" }}
|
||||
{{- tool | tojson }}
|
||||
{%- endfor %}
|
||||
{{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
|
||||
{%- else %}
|
||||
{%- if messages[0].role == 'system' %}
|
||||
{{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
|
||||
{%- endif %}
|
||||
{%- endif %}
|
||||
{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
|
||||
{%- for message in messages[::-1] %}
|
||||
{%- set index = (messages|length - 1) - loop.index0 %}
|
||||
{%- if ns.multi_step_tool and message.role == "user" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
|
||||
{%- set ns.multi_step_tool = false %}
|
||||
{%- set ns.last_query_index = index %}
|
||||
{%- endif %}
|
||||
{%- endfor %}
|
||||
{%- for message in messages %}
|
||||
{%- if message.content is string %}
|
||||
{%- set content = message.content %}
|
||||
{%- else %}
|
||||
{%- set content = '' %}
|
||||
{%- endif %}
|
||||
{%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
|
||||
{{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
|
||||
{%- elif message.role == "assistant" %}
|
||||
{%- set reasoning_content = '' %}
|
||||
{%- if message.reasoning_content is string %}
|
||||
{%- set reasoning_content = message.reasoning_content %}
|
||||
{%- else %}
|
||||
{%- if '</think>' in content %}
|
||||
{%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
|
||||
{%- set content = content.split('</think>')[-1].lstrip('\n') %}
|
||||
{%- endif %}
|
||||
{%- endif %}
|
||||
{%- if loop.index0 > ns.last_query_index %}
|
||||
{%- if loop.last or (not loop.last and reasoning_content) %}
|
||||
{{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
|
||||
{%- else %}
|
||||
{{- '<|im_start|>' + message.role + '\n' + content }}
|
||||
{%- endif %}
|
||||
{%- else %}
|
||||
{{- '<|im_start|>' + message.role + '\n' + content }}
|
||||
{%- endif %}
|
||||
{%- if message.tool_calls %}
|
||||
{%- for tool_call in message.tool_calls %}
|
||||
{%- if (loop.first and content) or (not loop.first) %}
|
||||
{{- '\n' }}
|
||||
{%- endif %}
|
||||
{%- if tool_call.function %}
|
||||
{%- set tool_call = tool_call.function %}
|
||||
{%- endif %}
|
||||
{{- '<tool_call>\n{"name": "' }}
|
||||
{{- tool_call.name }}
|
||||
{{- '", "arguments": ' }}
|
||||
{%- if tool_call.arguments is string %}
|
||||
{{- tool_call.arguments }}
|
||||
{%- else %}
|
||||
{{- tool_call.arguments | tojson }}
|
||||
{%- endif %}
|
||||
{{- '}\n</tool_call>' }}
|
||||
{%- endfor %}
|
||||
{%- endif %}
|
||||
{{- '<|im_end|>\n' }}
|
||||
{%- elif message.role == "tool" %}
|
||||
{%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
|
||||
{{- '<|im_start|>user' }}
|
||||
{%- endif %}
|
||||
{{- '\n<tool_response>\n' }}
|
||||
{{- content }}
|
||||
{{- '\n</tool_response>' }}
|
||||
{%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
|
||||
{{- '<|im_end|>\n' }}
|
||||
{%- endif %}
|
||||
{%- endif %}
|
||||
{%- endfor %}
|
||||
{%- if add_generation_prompt %}
|
||||
{{- '<|im_start|>assistant\n' }}
|
||||
{%- if enable_thinking is defined and enable_thinking is false %}
|
||||
{{- '<think>\n\n</think>\n\n' }}
|
||||
{%- endif %}
|
||||
{%- endif %}
|
||||
63
checkpoint-1000/config.json
Normal file
63
checkpoint-1000/config.json
Normal file
@@ -0,0 +1,63 @@
|
||||
{
|
||||
"architectures": [
|
||||
"Qwen3ForCausalLM"
|
||||
],
|
||||
"attention_bias": false,
|
||||
"attention_dropout": 0.0,
|
||||
"bos_token_id": null,
|
||||
"dtype": "bfloat16",
|
||||
"eos_token_id": 151645,
|
||||
"head_dim": 128,
|
||||
"hidden_act": "silu",
|
||||
"hidden_size": 2048,
|
||||
"initializer_range": 0.02,
|
||||
"intermediate_size": 6144,
|
||||
"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"
|
||||
],
|
||||
"max_position_embeddings": 40960,
|
||||
"max_window_layers": 28,
|
||||
"model_type": "qwen3",
|
||||
"num_attention_heads": 16,
|
||||
"num_hidden_layers": 28,
|
||||
"num_key_value_heads": 8,
|
||||
"pad_token_id": 151643,
|
||||
"rms_norm_eps": 1e-06,
|
||||
"rope_parameters": {
|
||||
"rope_theta": 1000000,
|
||||
"rope_type": "default"
|
||||
},
|
||||
"sliding_window": null,
|
||||
"tie_word_embeddings": true,
|
||||
"transformers_version": "5.7.0",
|
||||
"use_cache": false,
|
||||
"use_sliding_window": false,
|
||||
"vocab_size": 151936
|
||||
}
|
||||
12
checkpoint-1000/generation_config.json
Normal file
12
checkpoint-1000/generation_config.json
Normal file
@@ -0,0 +1,12 @@
|
||||
{
|
||||
"do_sample": true,
|
||||
"eos_token_id": [
|
||||
151645,
|
||||
151643
|
||||
],
|
||||
"pad_token_id": 151643,
|
||||
"temperature": 0.6,
|
||||
"top_k": 20,
|
||||
"top_p": 0.95,
|
||||
"transformers_version": "5.7.0"
|
||||
}
|
||||
3
checkpoint-1000/model.safetensors
Normal file
3
checkpoint-1000/model.safetensors
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:fa6a5ffa325a42521faee189eed67b3922773291e206e209e90b3c770b4d4539
|
||||
size 3441185608
|
||||
3
checkpoint-1000/optimizer.pt
Normal file
3
checkpoint-1000/optimizer.pt
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:2c51b9e4e783c1596ac7d185b6f7589123a938675998d0d32a5f077dfa2dfefd
|
||||
size 6882572207
|
||||
3
checkpoint-1000/rng_state.pth
Normal file
3
checkpoint-1000/rng_state.pth
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:61c19bab1174704a4a4441475683bf1270277af15d2e2c95e964789128e482c4
|
||||
size 14645
|
||||
3
checkpoint-1000/scheduler.pt
Normal file
3
checkpoint-1000/scheduler.pt
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:03766410d0ee20761a88c2327709b696e608f6be1dc35ba4d82dcb2eea2325f7
|
||||
size 1465
|
||||
3
checkpoint-1000/tokenizer.json
Normal file
3
checkpoint-1000/tokenizer.json
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:be75606093db2094d7cd20f3c2f385c212750648bd6ea4fb2bf507a6a4c55506
|
||||
size 11422650
|
||||
30
checkpoint-1000/tokenizer_config.json
Normal file
30
checkpoint-1000/tokenizer_config.json
Normal file
@@ -0,0 +1,30 @@
|
||||
{
|
||||
"add_prefix_space": false,
|
||||
"backend": "tokenizers",
|
||||
"bos_token": null,
|
||||
"clean_up_tokenization_spaces": false,
|
||||
"eos_token": "<|im_end|>",
|
||||
"errors": "replace",
|
||||
"extra_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|>"
|
||||
],
|
||||
"is_local": false,
|
||||
"local_files_only": false,
|
||||
"model_max_length": 131072,
|
||||
"pad_token": "<|endoftext|>",
|
||||
"split_special_tokens": false,
|
||||
"tokenizer_class": "Qwen2Tokenizer",
|
||||
"unk_token": null
|
||||
}
|
||||
234
checkpoint-1000/trainer_state.json
Normal file
234
checkpoint-1000/trainer_state.json
Normal file
@@ -0,0 +1,234 @@
|
||||
{
|
||||
"best_global_step": null,
|
||||
"best_metric": null,
|
||||
"best_model_checkpoint": null,
|
||||
"epoch": 0.16024998998437562,
|
||||
"eval_steps": 500,
|
||||
"global_step": 1000,
|
||||
"is_hyper_param_search": false,
|
||||
"is_local_process_zero": true,
|
||||
"is_world_process_zero": true,
|
||||
"log_history": [
|
||||
{
|
||||
"entropy": 2.223189059495926,
|
||||
"epoch": 0.008012499499218782,
|
||||
"grad_norm": 2.875,
|
||||
"learning_rate": 1.9842973882390643e-05,
|
||||
"loss": 2.3428109741210936,
|
||||
"mean_token_accuracy": 0.4963963569700718,
|
||||
"num_tokens": 305366.0,
|
||||
"step": 50
|
||||
},
|
||||
{
|
||||
"entropy": 2.188130117058754,
|
||||
"epoch": 0.016024998998437564,
|
||||
"grad_norm": 3.328125,
|
||||
"learning_rate": 1.96827431501362e-05,
|
||||
"loss": 2.225358734130859,
|
||||
"mean_token_accuracy": 0.5110921229422093,
|
||||
"num_tokens": 605042.0,
|
||||
"step": 100
|
||||
},
|
||||
{
|
||||
"entropy": 2.197347580194473,
|
||||
"epoch": 0.024037498497656344,
|
||||
"grad_norm": 3.453125,
|
||||
"learning_rate": 1.952251241788175e-05,
|
||||
"loss": 2.2299420166015627,
|
||||
"mean_token_accuracy": 0.5093490639328957,
|
||||
"num_tokens": 906888.0,
|
||||
"step": 150
|
||||
},
|
||||
{
|
||||
"entropy": 2.1913950395584108,
|
||||
"epoch": 0.03204999799687513,
|
||||
"grad_norm": 2.90625,
|
||||
"learning_rate": 1.9362281685627303e-05,
|
||||
"loss": 2.210995941162109,
|
||||
"mean_token_accuracy": 0.5097909501194954,
|
||||
"num_tokens": 1213125.0,
|
||||
"step": 200
|
||||
},
|
||||
{
|
||||
"entropy": 2.1611998677253723,
|
||||
"epoch": 0.040062497496093905,
|
||||
"grad_norm": 3.25,
|
||||
"learning_rate": 1.920205095337286e-05,
|
||||
"loss": 2.1867474365234374,
|
||||
"mean_token_accuracy": 0.5154410660266876,
|
||||
"num_tokens": 1509026.0,
|
||||
"step": 250
|
||||
},
|
||||
{
|
||||
"entropy": 2.168051454424858,
|
||||
"epoch": 0.04807499699531269,
|
||||
"grad_norm": 2.9375,
|
||||
"learning_rate": 1.904182022111841e-05,
|
||||
"loss": 2.190071716308594,
|
||||
"mean_token_accuracy": 0.5156884534657001,
|
||||
"num_tokens": 1806850.0,
|
||||
"step": 300
|
||||
},
|
||||
{
|
||||
"entropy": 2.1737154543399813,
|
||||
"epoch": 0.05608749649453147,
|
||||
"grad_norm": 3.421875,
|
||||
"learning_rate": 1.8881589488863967e-05,
|
||||
"loss": 2.199147186279297,
|
||||
"mean_token_accuracy": 0.514045044630766,
|
||||
"num_tokens": 2104505.0,
|
||||
"step": 350
|
||||
},
|
||||
{
|
||||
"entropy": 2.148039159178734,
|
||||
"epoch": 0.06409999599375026,
|
||||
"grad_norm": 3.109375,
|
||||
"learning_rate": 1.872135875660952e-05,
|
||||
"loss": 2.176946105957031,
|
||||
"mean_token_accuracy": 0.5177112428843975,
|
||||
"num_tokens": 2391088.0,
|
||||
"step": 400
|
||||
},
|
||||
{
|
||||
"entropy": 2.194890711903572,
|
||||
"epoch": 0.07211249549296903,
|
||||
"grad_norm": 3.28125,
|
||||
"learning_rate": 1.856112802435507e-05,
|
||||
"loss": 2.216288757324219,
|
||||
"mean_token_accuracy": 0.5110335703194141,
|
||||
"num_tokens": 2691272.0,
|
||||
"step": 450
|
||||
},
|
||||
{
|
||||
"entropy": 2.1557786524295808,
|
||||
"epoch": 0.08012499499218781,
|
||||
"grad_norm": 4.15625,
|
||||
"learning_rate": 1.8400897292100627e-05,
|
||||
"loss": 2.1812680053710936,
|
||||
"mean_token_accuracy": 0.5176697050035,
|
||||
"num_tokens": 2991855.0,
|
||||
"step": 500
|
||||
},
|
||||
{
|
||||
"entropy": 2.1708167713880537,
|
||||
"epoch": 0.08813749449140659,
|
||||
"grad_norm": 4.34375,
|
||||
"learning_rate": 1.824066655984618e-05,
|
||||
"loss": 2.1921002197265627,
|
||||
"mean_token_accuracy": 0.5156507430970669,
|
||||
"num_tokens": 3295049.0,
|
||||
"step": 550
|
||||
},
|
||||
{
|
||||
"entropy": 2.1686314886808393,
|
||||
"epoch": 0.09614999399062538,
|
||||
"grad_norm": 2.875,
|
||||
"learning_rate": 1.8080435827591735e-05,
|
||||
"loss": 2.1914617919921877,
|
||||
"mean_token_accuracy": 0.5139613692462445,
|
||||
"num_tokens": 3597206.0,
|
||||
"step": 600
|
||||
},
|
||||
{
|
||||
"entropy": 2.15742906332016,
|
||||
"epoch": 0.10416249348984416,
|
||||
"grad_norm": 3.109375,
|
||||
"learning_rate": 1.7920205095337287e-05,
|
||||
"loss": 2.174853973388672,
|
||||
"mean_token_accuracy": 0.5159518288075924,
|
||||
"num_tokens": 3905268.0,
|
||||
"step": 650
|
||||
},
|
||||
{
|
||||
"entropy": 2.162363375425339,
|
||||
"epoch": 0.11217499298906294,
|
||||
"grad_norm": 3.171875,
|
||||
"learning_rate": 1.7759974363082843e-05,
|
||||
"loss": 2.184231872558594,
|
||||
"mean_token_accuracy": 0.516631413847208,
|
||||
"num_tokens": 4207934.0,
|
||||
"step": 700
|
||||
},
|
||||
{
|
||||
"entropy": 2.1516289287805557,
|
||||
"epoch": 0.12018749248828171,
|
||||
"grad_norm": 3.609375,
|
||||
"learning_rate": 1.7599743630828395e-05,
|
||||
"loss": 2.167020263671875,
|
||||
"mean_token_accuracy": 0.516358617246151,
|
||||
"num_tokens": 4515774.0,
|
||||
"step": 750
|
||||
},
|
||||
{
|
||||
"entropy": 2.1549544781446457,
|
||||
"epoch": 0.1281999919875005,
|
||||
"grad_norm": 3.28125,
|
||||
"learning_rate": 1.7439512898573947e-05,
|
||||
"loss": 2.181917877197266,
|
||||
"mean_token_accuracy": 0.5182220858335495,
|
||||
"num_tokens": 4814709.0,
|
||||
"step": 800
|
||||
},
|
||||
{
|
||||
"entropy": 2.1532615756988527,
|
||||
"epoch": 0.1362124914867193,
|
||||
"grad_norm": 3.078125,
|
||||
"learning_rate": 1.7279282166319503e-05,
|
||||
"loss": 2.1659007263183594,
|
||||
"mean_token_accuracy": 0.5180425970256328,
|
||||
"num_tokens": 5112585.0,
|
||||
"step": 850
|
||||
},
|
||||
{
|
||||
"entropy": 2.1704017293453215,
|
||||
"epoch": 0.14422499098593805,
|
||||
"grad_norm": 2.921875,
|
||||
"learning_rate": 1.7119051434065055e-05,
|
||||
"loss": 2.1934996032714844,
|
||||
"mean_token_accuracy": 0.5154964691400528,
|
||||
"num_tokens": 5411369.0,
|
||||
"step": 900
|
||||
},
|
||||
{
|
||||
"entropy": 2.1349107217788696,
|
||||
"epoch": 0.15223749048515683,
|
||||
"grad_norm": 2.984375,
|
||||
"learning_rate": 1.6958820701810608e-05,
|
||||
"loss": 2.1598545837402345,
|
||||
"mean_token_accuracy": 0.5208920259773732,
|
||||
"num_tokens": 5707060.0,
|
||||
"step": 950
|
||||
},
|
||||
{
|
||||
"entropy": 2.1541705918312073,
|
||||
"epoch": 0.16024998998437562,
|
||||
"grad_norm": 3.203125,
|
||||
"learning_rate": 1.6798589969556163e-05,
|
||||
"loss": 2.170956573486328,
|
||||
"mean_token_accuracy": 0.5204938448965549,
|
||||
"num_tokens": 6004079.0,
|
||||
"step": 1000
|
||||
}
|
||||
],
|
||||
"logging_steps": 50,
|
||||
"max_steps": 6241,
|
||||
"num_input_tokens_seen": 0,
|
||||
"num_train_epochs": 1,
|
||||
"save_steps": 500,
|
||||
"stateful_callbacks": {
|
||||
"TrainerControl": {
|
||||
"args": {
|
||||
"should_epoch_stop": false,
|
||||
"should_evaluate": false,
|
||||
"should_log": false,
|
||||
"should_save": true,
|
||||
"should_training_stop": false
|
||||
},
|
||||
"attributes": {}
|
||||
}
|
||||
},
|
||||
"total_flos": 7.095551986099814e+16,
|
||||
"train_batch_size": 4,
|
||||
"trial_name": null,
|
||||
"trial_params": null
|
||||
}
|
||||
3
checkpoint-1000/training_args.bin
Normal file
3
checkpoint-1000/training_args.bin
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:78622f63304ec0913ed0b4ea7a1b26c08a5621b3661a380d798c4ea426235671
|
||||
size 5713
|
||||
89
checkpoint-1500/chat_template.jinja
Normal file
89
checkpoint-1500/chat_template.jinja
Normal file
@@ -0,0 +1,89 @@
|
||||
{%- if tools %}
|
||||
{{- '<|im_start|>system\n' }}
|
||||
{%- if messages[0].role == 'system' %}
|
||||
{{- messages[0].content + '\n\n' }}
|
||||
{%- endif %}
|
||||
{{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
|
||||
{%- for tool in tools %}
|
||||
{{- "\n" }}
|
||||
{{- tool | tojson }}
|
||||
{%- endfor %}
|
||||
{{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
|
||||
{%- else %}
|
||||
{%- if messages[0].role == 'system' %}
|
||||
{{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
|
||||
{%- endif %}
|
||||
{%- endif %}
|
||||
{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
|
||||
{%- for message in messages[::-1] %}
|
||||
{%- set index = (messages|length - 1) - loop.index0 %}
|
||||
{%- if ns.multi_step_tool and message.role == "user" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
|
||||
{%- set ns.multi_step_tool = false %}
|
||||
{%- set ns.last_query_index = index %}
|
||||
{%- endif %}
|
||||
{%- endfor %}
|
||||
{%- for message in messages %}
|
||||
{%- if message.content is string %}
|
||||
{%- set content = message.content %}
|
||||
{%- else %}
|
||||
{%- set content = '' %}
|
||||
{%- endif %}
|
||||
{%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
|
||||
{{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
|
||||
{%- elif message.role == "assistant" %}
|
||||
{%- set reasoning_content = '' %}
|
||||
{%- if message.reasoning_content is string %}
|
||||
{%- set reasoning_content = message.reasoning_content %}
|
||||
{%- else %}
|
||||
{%- if '</think>' in content %}
|
||||
{%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
|
||||
{%- set content = content.split('</think>')[-1].lstrip('\n') %}
|
||||
{%- endif %}
|
||||
{%- endif %}
|
||||
{%- if loop.index0 > ns.last_query_index %}
|
||||
{%- if loop.last or (not loop.last and reasoning_content) %}
|
||||
{{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
|
||||
{%- else %}
|
||||
{{- '<|im_start|>' + message.role + '\n' + content }}
|
||||
{%- endif %}
|
||||
{%- else %}
|
||||
{{- '<|im_start|>' + message.role + '\n' + content }}
|
||||
{%- endif %}
|
||||
{%- if message.tool_calls %}
|
||||
{%- for tool_call in message.tool_calls %}
|
||||
{%- if (loop.first and content) or (not loop.first) %}
|
||||
{{- '\n' }}
|
||||
{%- endif %}
|
||||
{%- if tool_call.function %}
|
||||
{%- set tool_call = tool_call.function %}
|
||||
{%- endif %}
|
||||
{{- '<tool_call>\n{"name": "' }}
|
||||
{{- tool_call.name }}
|
||||
{{- '", "arguments": ' }}
|
||||
{%- if tool_call.arguments is string %}
|
||||
{{- tool_call.arguments }}
|
||||
{%- else %}
|
||||
{{- tool_call.arguments | tojson }}
|
||||
{%- endif %}
|
||||
{{- '}\n</tool_call>' }}
|
||||
{%- endfor %}
|
||||
{%- endif %}
|
||||
{{- '<|im_end|>\n' }}
|
||||
{%- elif message.role == "tool" %}
|
||||
{%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
|
||||
{{- '<|im_start|>user' }}
|
||||
{%- endif %}
|
||||
{{- '\n<tool_response>\n' }}
|
||||
{{- content }}
|
||||
{{- '\n</tool_response>' }}
|
||||
{%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
|
||||
{{- '<|im_end|>\n' }}
|
||||
{%- endif %}
|
||||
{%- endif %}
|
||||
{%- endfor %}
|
||||
{%- if add_generation_prompt %}
|
||||
{{- '<|im_start|>assistant\n' }}
|
||||
{%- if enable_thinking is defined and enable_thinking is false %}
|
||||
{{- '<think>\n\n</think>\n\n' }}
|
||||
{%- endif %}
|
||||
{%- endif %}
|
||||
63
checkpoint-1500/config.json
Normal file
63
checkpoint-1500/config.json
Normal file
@@ -0,0 +1,63 @@
|
||||
{
|
||||
"architectures": [
|
||||
"Qwen3ForCausalLM"
|
||||
],
|
||||
"attention_bias": false,
|
||||
"attention_dropout": 0.0,
|
||||
"bos_token_id": null,
|
||||
"dtype": "bfloat16",
|
||||
"eos_token_id": 151645,
|
||||
"head_dim": 128,
|
||||
"hidden_act": "silu",
|
||||
"hidden_size": 2048,
|
||||
"initializer_range": 0.02,
|
||||
"intermediate_size": 6144,
|
||||
"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"
|
||||
],
|
||||
"max_position_embeddings": 40960,
|
||||
"max_window_layers": 28,
|
||||
"model_type": "qwen3",
|
||||
"num_attention_heads": 16,
|
||||
"num_hidden_layers": 28,
|
||||
"num_key_value_heads": 8,
|
||||
"pad_token_id": 151643,
|
||||
"rms_norm_eps": 1e-06,
|
||||
"rope_parameters": {
|
||||
"rope_theta": 1000000,
|
||||
"rope_type": "default"
|
||||
},
|
||||
"sliding_window": null,
|
||||
"tie_word_embeddings": true,
|
||||
"transformers_version": "5.7.0",
|
||||
"use_cache": false,
|
||||
"use_sliding_window": false,
|
||||
"vocab_size": 151936
|
||||
}
|
||||
12
checkpoint-1500/generation_config.json
Normal file
12
checkpoint-1500/generation_config.json
Normal file
@@ -0,0 +1,12 @@
|
||||
{
|
||||
"do_sample": true,
|
||||
"eos_token_id": [
|
||||
151645,
|
||||
151643
|
||||
],
|
||||
"pad_token_id": 151643,
|
||||
"temperature": 0.6,
|
||||
"top_k": 20,
|
||||
"top_p": 0.95,
|
||||
"transformers_version": "5.7.0"
|
||||
}
|
||||
3
checkpoint-1500/model.safetensors
Normal file
3
checkpoint-1500/model.safetensors
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:438e34183f5e94f608c7669e0b2cb247069e6a47ebcded96167ec3afb7af8101
|
||||
size 3441185608
|
||||
3
checkpoint-1500/optimizer.pt
Normal file
3
checkpoint-1500/optimizer.pt
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:11e561d4c48540c45fb0712b82d5ffdd92c408f08ae4ed12ecbced4e409b1208
|
||||
size 6882572207
|
||||
3
checkpoint-1500/rng_state.pth
Normal file
3
checkpoint-1500/rng_state.pth
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:61c19bab1174704a4a4441475683bf1270277af15d2e2c95e964789128e482c4
|
||||
size 14645
|
||||
3
checkpoint-1500/scheduler.pt
Normal file
3
checkpoint-1500/scheduler.pt
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:b666bb023db840fb022cfa0357e113ba9d208c4303ad969fafd6098567ba3038
|
||||
size 1465
|
||||
3
checkpoint-1500/tokenizer.json
Normal file
3
checkpoint-1500/tokenizer.json
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:be75606093db2094d7cd20f3c2f385c212750648bd6ea4fb2bf507a6a4c55506
|
||||
size 11422650
|
||||
30
checkpoint-1500/tokenizer_config.json
Normal file
30
checkpoint-1500/tokenizer_config.json
Normal file
@@ -0,0 +1,30 @@
|
||||
{
|
||||
"add_prefix_space": false,
|
||||
"backend": "tokenizers",
|
||||
"bos_token": null,
|
||||
"clean_up_tokenization_spaces": false,
|
||||
"eos_token": "<|im_end|>",
|
||||
"errors": "replace",
|
||||
"extra_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|>"
|
||||
],
|
||||
"is_local": false,
|
||||
"local_files_only": false,
|
||||
"model_max_length": 131072,
|
||||
"pad_token": "<|endoftext|>",
|
||||
"split_special_tokens": false,
|
||||
"tokenizer_class": "Qwen2Tokenizer",
|
||||
"unk_token": null
|
||||
}
|
||||
334
checkpoint-1500/trainer_state.json
Normal file
334
checkpoint-1500/trainer_state.json
Normal file
@@ -0,0 +1,334 @@
|
||||
{
|
||||
"best_global_step": null,
|
||||
"best_metric": null,
|
||||
"best_model_checkpoint": null,
|
||||
"epoch": 0.24037498497656343,
|
||||
"eval_steps": 500,
|
||||
"global_step": 1500,
|
||||
"is_hyper_param_search": false,
|
||||
"is_local_process_zero": true,
|
||||
"is_world_process_zero": true,
|
||||
"log_history": [
|
||||
{
|
||||
"entropy": 2.223189059495926,
|
||||
"epoch": 0.008012499499218782,
|
||||
"grad_norm": 2.875,
|
||||
"learning_rate": 1.9842973882390643e-05,
|
||||
"loss": 2.3428109741210936,
|
||||
"mean_token_accuracy": 0.4963963569700718,
|
||||
"num_tokens": 305366.0,
|
||||
"step": 50
|
||||
},
|
||||
{
|
||||
"entropy": 2.188130117058754,
|
||||
"epoch": 0.016024998998437564,
|
||||
"grad_norm": 3.328125,
|
||||
"learning_rate": 1.96827431501362e-05,
|
||||
"loss": 2.225358734130859,
|
||||
"mean_token_accuracy": 0.5110921229422093,
|
||||
"num_tokens": 605042.0,
|
||||
"step": 100
|
||||
},
|
||||
{
|
||||
"entropy": 2.197347580194473,
|
||||
"epoch": 0.024037498497656344,
|
||||
"grad_norm": 3.453125,
|
||||
"learning_rate": 1.952251241788175e-05,
|
||||
"loss": 2.2299420166015627,
|
||||
"mean_token_accuracy": 0.5093490639328957,
|
||||
"num_tokens": 906888.0,
|
||||
"step": 150
|
||||
},
|
||||
{
|
||||
"entropy": 2.1913950395584108,
|
||||
"epoch": 0.03204999799687513,
|
||||
"grad_norm": 2.90625,
|
||||
"learning_rate": 1.9362281685627303e-05,
|
||||
"loss": 2.210995941162109,
|
||||
"mean_token_accuracy": 0.5097909501194954,
|
||||
"num_tokens": 1213125.0,
|
||||
"step": 200
|
||||
},
|
||||
{
|
||||
"entropy": 2.1611998677253723,
|
||||
"epoch": 0.040062497496093905,
|
||||
"grad_norm": 3.25,
|
||||
"learning_rate": 1.920205095337286e-05,
|
||||
"loss": 2.1867474365234374,
|
||||
"mean_token_accuracy": 0.5154410660266876,
|
||||
"num_tokens": 1509026.0,
|
||||
"step": 250
|
||||
},
|
||||
{
|
||||
"entropy": 2.168051454424858,
|
||||
"epoch": 0.04807499699531269,
|
||||
"grad_norm": 2.9375,
|
||||
"learning_rate": 1.904182022111841e-05,
|
||||
"loss": 2.190071716308594,
|
||||
"mean_token_accuracy": 0.5156884534657001,
|
||||
"num_tokens": 1806850.0,
|
||||
"step": 300
|
||||
},
|
||||
{
|
||||
"entropy": 2.1737154543399813,
|
||||
"epoch": 0.05608749649453147,
|
||||
"grad_norm": 3.421875,
|
||||
"learning_rate": 1.8881589488863967e-05,
|
||||
"loss": 2.199147186279297,
|
||||
"mean_token_accuracy": 0.514045044630766,
|
||||
"num_tokens": 2104505.0,
|
||||
"step": 350
|
||||
},
|
||||
{
|
||||
"entropy": 2.148039159178734,
|
||||
"epoch": 0.06409999599375026,
|
||||
"grad_norm": 3.109375,
|
||||
"learning_rate": 1.872135875660952e-05,
|
||||
"loss": 2.176946105957031,
|
||||
"mean_token_accuracy": 0.5177112428843975,
|
||||
"num_tokens": 2391088.0,
|
||||
"step": 400
|
||||
},
|
||||
{
|
||||
"entropy": 2.194890711903572,
|
||||
"epoch": 0.07211249549296903,
|
||||
"grad_norm": 3.28125,
|
||||
"learning_rate": 1.856112802435507e-05,
|
||||
"loss": 2.216288757324219,
|
||||
"mean_token_accuracy": 0.5110335703194141,
|
||||
"num_tokens": 2691272.0,
|
||||
"step": 450
|
||||
},
|
||||
{
|
||||
"entropy": 2.1557786524295808,
|
||||
"epoch": 0.08012499499218781,
|
||||
"grad_norm": 4.15625,
|
||||
"learning_rate": 1.8400897292100627e-05,
|
||||
"loss": 2.1812680053710936,
|
||||
"mean_token_accuracy": 0.5176697050035,
|
||||
"num_tokens": 2991855.0,
|
||||
"step": 500
|
||||
},
|
||||
{
|
||||
"entropy": 2.1708167713880537,
|
||||
"epoch": 0.08813749449140659,
|
||||
"grad_norm": 4.34375,
|
||||
"learning_rate": 1.824066655984618e-05,
|
||||
"loss": 2.1921002197265627,
|
||||
"mean_token_accuracy": 0.5156507430970669,
|
||||
"num_tokens": 3295049.0,
|
||||
"step": 550
|
||||
},
|
||||
{
|
||||
"entropy": 2.1686314886808393,
|
||||
"epoch": 0.09614999399062538,
|
||||
"grad_norm": 2.875,
|
||||
"learning_rate": 1.8080435827591735e-05,
|
||||
"loss": 2.1914617919921877,
|
||||
"mean_token_accuracy": 0.5139613692462445,
|
||||
"num_tokens": 3597206.0,
|
||||
"step": 600
|
||||
},
|
||||
{
|
||||
"entropy": 2.15742906332016,
|
||||
"epoch": 0.10416249348984416,
|
||||
"grad_norm": 3.109375,
|
||||
"learning_rate": 1.7920205095337287e-05,
|
||||
"loss": 2.174853973388672,
|
||||
"mean_token_accuracy": 0.5159518288075924,
|
||||
"num_tokens": 3905268.0,
|
||||
"step": 650
|
||||
},
|
||||
{
|
||||
"entropy": 2.162363375425339,
|
||||
"epoch": 0.11217499298906294,
|
||||
"grad_norm": 3.171875,
|
||||
"learning_rate": 1.7759974363082843e-05,
|
||||
"loss": 2.184231872558594,
|
||||
"mean_token_accuracy": 0.516631413847208,
|
||||
"num_tokens": 4207934.0,
|
||||
"step": 700
|
||||
},
|
||||
{
|
||||
"entropy": 2.1516289287805557,
|
||||
"epoch": 0.12018749248828171,
|
||||
"grad_norm": 3.609375,
|
||||
"learning_rate": 1.7599743630828395e-05,
|
||||
"loss": 2.167020263671875,
|
||||
"mean_token_accuracy": 0.516358617246151,
|
||||
"num_tokens": 4515774.0,
|
||||
"step": 750
|
||||
},
|
||||
{
|
||||
"entropy": 2.1549544781446457,
|
||||
"epoch": 0.1281999919875005,
|
||||
"grad_norm": 3.28125,
|
||||
"learning_rate": 1.7439512898573947e-05,
|
||||
"loss": 2.181917877197266,
|
||||
"mean_token_accuracy": 0.5182220858335495,
|
||||
"num_tokens": 4814709.0,
|
||||
"step": 800
|
||||
},
|
||||
{
|
||||
"entropy": 2.1532615756988527,
|
||||
"epoch": 0.1362124914867193,
|
||||
"grad_norm": 3.078125,
|
||||
"learning_rate": 1.7279282166319503e-05,
|
||||
"loss": 2.1659007263183594,
|
||||
"mean_token_accuracy": 0.5180425970256328,
|
||||
"num_tokens": 5112585.0,
|
||||
"step": 850
|
||||
},
|
||||
{
|
||||
"entropy": 2.1704017293453215,
|
||||
"epoch": 0.14422499098593805,
|
||||
"grad_norm": 2.921875,
|
||||
"learning_rate": 1.7119051434065055e-05,
|
||||
"loss": 2.1934996032714844,
|
||||
"mean_token_accuracy": 0.5154964691400528,
|
||||
"num_tokens": 5411369.0,
|
||||
"step": 900
|
||||
},
|
||||
{
|
||||
"entropy": 2.1349107217788696,
|
||||
"epoch": 0.15223749048515683,
|
||||
"grad_norm": 2.984375,
|
||||
"learning_rate": 1.6958820701810608e-05,
|
||||
"loss": 2.1598545837402345,
|
||||
"mean_token_accuracy": 0.5208920259773732,
|
||||
"num_tokens": 5707060.0,
|
||||
"step": 950
|
||||
},
|
||||
{
|
||||
"entropy": 2.1541705918312073,
|
||||
"epoch": 0.16024998998437562,
|
||||
"grad_norm": 3.203125,
|
||||
"learning_rate": 1.6798589969556163e-05,
|
||||
"loss": 2.170956573486328,
|
||||
"mean_token_accuracy": 0.5204938448965549,
|
||||
"num_tokens": 6004079.0,
|
||||
"step": 1000
|
||||
},
|
||||
{
|
||||
"entropy": 2.1523867017030716,
|
||||
"epoch": 0.1682624894835944,
|
||||
"grad_norm": 3.09375,
|
||||
"learning_rate": 1.6638359237301716e-05,
|
||||
"loss": 2.1695118713378907,
|
||||
"mean_token_accuracy": 0.5177376234531402,
|
||||
"num_tokens": 6315462.0,
|
||||
"step": 1050
|
||||
},
|
||||
{
|
||||
"entropy": 2.1356022965908052,
|
||||
"epoch": 0.17627498898281319,
|
||||
"grad_norm": 3.21875,
|
||||
"learning_rate": 1.6478128505047268e-05,
|
||||
"loss": 2.1513107299804686,
|
||||
"mean_token_accuracy": 0.5217180585861206,
|
||||
"num_tokens": 6613060.0,
|
||||
"step": 1100
|
||||
},
|
||||
{
|
||||
"entropy": 2.1483460652828215,
|
||||
"epoch": 0.18428748848203197,
|
||||
"grad_norm": 3.21875,
|
||||
"learning_rate": 1.6317897772792824e-05,
|
||||
"loss": 2.1671595764160156,
|
||||
"mean_token_accuracy": 0.5199222292006016,
|
||||
"num_tokens": 6909655.0,
|
||||
"step": 1150
|
||||
},
|
||||
{
|
||||
"entropy": 2.1074040442705155,
|
||||
"epoch": 0.19229998798125075,
|
||||
"grad_norm": 3.828125,
|
||||
"learning_rate": 1.6157667040538376e-05,
|
||||
"loss": 2.12770751953125,
|
||||
"mean_token_accuracy": 0.5252153894305229,
|
||||
"num_tokens": 7203263.0,
|
||||
"step": 1200
|
||||
},
|
||||
{
|
||||
"entropy": 2.1354027593135836,
|
||||
"epoch": 0.20031248748046954,
|
||||
"grad_norm": 2.921875,
|
||||
"learning_rate": 1.5997436308283932e-05,
|
||||
"loss": 2.156183166503906,
|
||||
"mean_token_accuracy": 0.518907299786806,
|
||||
"num_tokens": 7505411.0,
|
||||
"step": 1250
|
||||
},
|
||||
{
|
||||
"entropy": 2.142924894094467,
|
||||
"epoch": 0.20832498697968832,
|
||||
"grad_norm": 3.5625,
|
||||
"learning_rate": 1.5837205576029484e-05,
|
||||
"loss": 2.158486633300781,
|
||||
"mean_token_accuracy": 0.5205909128487111,
|
||||
"num_tokens": 7803562.0,
|
||||
"step": 1300
|
||||
},
|
||||
{
|
||||
"entropy": 2.1305291467905043,
|
||||
"epoch": 0.2163374864789071,
|
||||
"grad_norm": 3.34375,
|
||||
"learning_rate": 1.5676974843775036e-05,
|
||||
"loss": 2.151043701171875,
|
||||
"mean_token_accuracy": 0.5211617235839366,
|
||||
"num_tokens": 8108037.0,
|
||||
"step": 1350
|
||||
},
|
||||
{
|
||||
"entropy": 2.160947166085243,
|
||||
"epoch": 0.2243499859781259,
|
||||
"grad_norm": 3.234375,
|
||||
"learning_rate": 1.5516744111520592e-05,
|
||||
"loss": 2.177792663574219,
|
||||
"mean_token_accuracy": 0.5167004990577698,
|
||||
"num_tokens": 8408674.0,
|
||||
"step": 1400
|
||||
},
|
||||
{
|
||||
"entropy": 2.120403967499733,
|
||||
"epoch": 0.23236248547734467,
|
||||
"grad_norm": 3.375,
|
||||
"learning_rate": 1.5356513379266144e-05,
|
||||
"loss": 2.1462562561035154,
|
||||
"mean_token_accuracy": 0.5225985661149025,
|
||||
"num_tokens": 8707650.0,
|
||||
"step": 1450
|
||||
},
|
||||
{
|
||||
"entropy": 2.138882914185524,
|
||||
"epoch": 0.24037498497656343,
|
||||
"grad_norm": 3.171875,
|
||||
"learning_rate": 1.5196282647011698e-05,
|
||||
"loss": 2.1499905395507812,
|
||||
"mean_token_accuracy": 0.5204668046534061,
|
||||
"num_tokens": 9006296.0,
|
||||
"step": 1500
|
||||
}
|
||||
],
|
||||
"logging_steps": 50,
|
||||
"max_steps": 6241,
|
||||
"num_input_tokens_seen": 0,
|
||||
"num_train_epochs": 1,
|
||||
"save_steps": 500,
|
||||
"stateful_callbacks": {
|
||||
"TrainerControl": {
|
||||
"args": {
|
||||
"should_epoch_stop": false,
|
||||
"should_evaluate": false,
|
||||
"should_log": false,
|
||||
"should_save": true,
|
||||
"should_training_stop": false
|
||||
},
|
||||
"attributes": {}
|
||||
}
|
||||
},
|
||||
"total_flos": 1.064051705174999e+17,
|
||||
"train_batch_size": 4,
|
||||
"trial_name": null,
|
||||
"trial_params": null
|
||||
}
|
||||
3
checkpoint-1500/training_args.bin
Normal file
3
checkpoint-1500/training_args.bin
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:78622f63304ec0913ed0b4ea7a1b26c08a5621b3661a380d798c4ea426235671
|
||||
size 5713
|
||||
89
checkpoint-2000/chat_template.jinja
Normal file
89
checkpoint-2000/chat_template.jinja
Normal file
@@ -0,0 +1,89 @@
|
||||
{%- if tools %}
|
||||
{{- '<|im_start|>system\n' }}
|
||||
{%- if messages[0].role == 'system' %}
|
||||
{{- messages[0].content + '\n\n' }}
|
||||
{%- endif %}
|
||||
{{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
|
||||
{%- for tool in tools %}
|
||||
{{- "\n" }}
|
||||
{{- tool | tojson }}
|
||||
{%- endfor %}
|
||||
{{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
|
||||
{%- else %}
|
||||
{%- if messages[0].role == 'system' %}
|
||||
{{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
|
||||
{%- endif %}
|
||||
{%- endif %}
|
||||
{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
|
||||
{%- for message in messages[::-1] %}
|
||||
{%- set index = (messages|length - 1) - loop.index0 %}
|
||||
{%- if ns.multi_step_tool and message.role == "user" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
|
||||
{%- set ns.multi_step_tool = false %}
|
||||
{%- set ns.last_query_index = index %}
|
||||
{%- endif %}
|
||||
{%- endfor %}
|
||||
{%- for message in messages %}
|
||||
{%- if message.content is string %}
|
||||
{%- set content = message.content %}
|
||||
{%- else %}
|
||||
{%- set content = '' %}
|
||||
{%- endif %}
|
||||
{%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
|
||||
{{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
|
||||
{%- elif message.role == "assistant" %}
|
||||
{%- set reasoning_content = '' %}
|
||||
{%- if message.reasoning_content is string %}
|
||||
{%- set reasoning_content = message.reasoning_content %}
|
||||
{%- else %}
|
||||
{%- if '</think>' in content %}
|
||||
{%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
|
||||
{%- set content = content.split('</think>')[-1].lstrip('\n') %}
|
||||
{%- endif %}
|
||||
{%- endif %}
|
||||
{%- if loop.index0 > ns.last_query_index %}
|
||||
{%- if loop.last or (not loop.last and reasoning_content) %}
|
||||
{{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
|
||||
{%- else %}
|
||||
{{- '<|im_start|>' + message.role + '\n' + content }}
|
||||
{%- endif %}
|
||||
{%- else %}
|
||||
{{- '<|im_start|>' + message.role + '\n' + content }}
|
||||
{%- endif %}
|
||||
{%- if message.tool_calls %}
|
||||
{%- for tool_call in message.tool_calls %}
|
||||
{%- if (loop.first and content) or (not loop.first) %}
|
||||
{{- '\n' }}
|
||||
{%- endif %}
|
||||
{%- if tool_call.function %}
|
||||
{%- set tool_call = tool_call.function %}
|
||||
{%- endif %}
|
||||
{{- '<tool_call>\n{"name": "' }}
|
||||
{{- tool_call.name }}
|
||||
{{- '", "arguments": ' }}
|
||||
{%- if tool_call.arguments is string %}
|
||||
{{- tool_call.arguments }}
|
||||
{%- else %}
|
||||
{{- tool_call.arguments | tojson }}
|
||||
{%- endif %}
|
||||
{{- '}\n</tool_call>' }}
|
||||
{%- endfor %}
|
||||
{%- endif %}
|
||||
{{- '<|im_end|>\n' }}
|
||||
{%- elif message.role == "tool" %}
|
||||
{%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
|
||||
{{- '<|im_start|>user' }}
|
||||
{%- endif %}
|
||||
{{- '\n<tool_response>\n' }}
|
||||
{{- content }}
|
||||
{{- '\n</tool_response>' }}
|
||||
{%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
|
||||
{{- '<|im_end|>\n' }}
|
||||
{%- endif %}
|
||||
{%- endif %}
|
||||
{%- endfor %}
|
||||
{%- if add_generation_prompt %}
|
||||
{{- '<|im_start|>assistant\n' }}
|
||||
{%- if enable_thinking is defined and enable_thinking is false %}
|
||||
{{- '<think>\n\n</think>\n\n' }}
|
||||
{%- endif %}
|
||||
{%- endif %}
|
||||
63
checkpoint-2000/config.json
Normal file
63
checkpoint-2000/config.json
Normal file
@@ -0,0 +1,63 @@
|
||||
{
|
||||
"architectures": [
|
||||
"Qwen3ForCausalLM"
|
||||
],
|
||||
"attention_bias": false,
|
||||
"attention_dropout": 0.0,
|
||||
"bos_token_id": null,
|
||||
"dtype": "bfloat16",
|
||||
"eos_token_id": 151645,
|
||||
"head_dim": 128,
|
||||
"hidden_act": "silu",
|
||||
"hidden_size": 2048,
|
||||
"initializer_range": 0.02,
|
||||
"intermediate_size": 6144,
|
||||
"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"
|
||||
],
|
||||
"max_position_embeddings": 40960,
|
||||
"max_window_layers": 28,
|
||||
"model_type": "qwen3",
|
||||
"num_attention_heads": 16,
|
||||
"num_hidden_layers": 28,
|
||||
"num_key_value_heads": 8,
|
||||
"pad_token_id": 151643,
|
||||
"rms_norm_eps": 1e-06,
|
||||
"rope_parameters": {
|
||||
"rope_theta": 1000000,
|
||||
"rope_type": "default"
|
||||
},
|
||||
"sliding_window": null,
|
||||
"tie_word_embeddings": true,
|
||||
"transformers_version": "5.7.0",
|
||||
"use_cache": false,
|
||||
"use_sliding_window": false,
|
||||
"vocab_size": 151936
|
||||
}
|
||||
12
checkpoint-2000/generation_config.json
Normal file
12
checkpoint-2000/generation_config.json
Normal file
@@ -0,0 +1,12 @@
|
||||
{
|
||||
"do_sample": true,
|
||||
"eos_token_id": [
|
||||
151645,
|
||||
151643
|
||||
],
|
||||
"pad_token_id": 151643,
|
||||
"temperature": 0.6,
|
||||
"top_k": 20,
|
||||
"top_p": 0.95,
|
||||
"transformers_version": "5.7.0"
|
||||
}
|
||||
3
checkpoint-2000/model.safetensors
Normal file
3
checkpoint-2000/model.safetensors
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:ec88ee383178456b8f8ab7ee942be2785b0a32656809f229cae94c76dff1b810
|
||||
size 3441185608
|
||||
3
checkpoint-2000/optimizer.pt
Normal file
3
checkpoint-2000/optimizer.pt
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:ea2dc818dc38c52000eeebc5b101562caf8fb11d4790f976bb066ee636e4f14d
|
||||
size 6882572207
|
||||
3
checkpoint-2000/rng_state.pth
Normal file
3
checkpoint-2000/rng_state.pth
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:61c19bab1174704a4a4441475683bf1270277af15d2e2c95e964789128e482c4
|
||||
size 14645
|
||||
3
checkpoint-2000/scheduler.pt
Normal file
3
checkpoint-2000/scheduler.pt
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:11b128c7c709b8b18f78ca21ff87ee25957a3215823c4227b4ba298165c8d222
|
||||
size 1465
|
||||
3
checkpoint-2000/tokenizer.json
Normal file
3
checkpoint-2000/tokenizer.json
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:be75606093db2094d7cd20f3c2f385c212750648bd6ea4fb2bf507a6a4c55506
|
||||
size 11422650
|
||||
30
checkpoint-2000/tokenizer_config.json
Normal file
30
checkpoint-2000/tokenizer_config.json
Normal file
@@ -0,0 +1,30 @@
|
||||
{
|
||||
"add_prefix_space": false,
|
||||
"backend": "tokenizers",
|
||||
"bos_token": null,
|
||||
"clean_up_tokenization_spaces": false,
|
||||
"eos_token": "<|im_end|>",
|
||||
"errors": "replace",
|
||||
"extra_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|>"
|
||||
],
|
||||
"is_local": false,
|
||||
"local_files_only": false,
|
||||
"model_max_length": 131072,
|
||||
"pad_token": "<|endoftext|>",
|
||||
"split_special_tokens": false,
|
||||
"tokenizer_class": "Qwen2Tokenizer",
|
||||
"unk_token": null
|
||||
}
|
||||
434
checkpoint-2000/trainer_state.json
Normal file
434
checkpoint-2000/trainer_state.json
Normal file
@@ -0,0 +1,434 @@
|
||||
{
|
||||
"best_global_step": null,
|
||||
"best_metric": null,
|
||||
"best_model_checkpoint": null,
|
||||
"epoch": 0.32049997996875124,
|
||||
"eval_steps": 500,
|
||||
"global_step": 2000,
|
||||
"is_hyper_param_search": false,
|
||||
"is_local_process_zero": true,
|
||||
"is_world_process_zero": true,
|
||||
"log_history": [
|
||||
{
|
||||
"entropy": 2.223189059495926,
|
||||
"epoch": 0.008012499499218782,
|
||||
"grad_norm": 2.875,
|
||||
"learning_rate": 1.9842973882390643e-05,
|
||||
"loss": 2.3428109741210936,
|
||||
"mean_token_accuracy": 0.4963963569700718,
|
||||
"num_tokens": 305366.0,
|
||||
"step": 50
|
||||
},
|
||||
{
|
||||
"entropy": 2.188130117058754,
|
||||
"epoch": 0.016024998998437564,
|
||||
"grad_norm": 3.328125,
|
||||
"learning_rate": 1.96827431501362e-05,
|
||||
"loss": 2.225358734130859,
|
||||
"mean_token_accuracy": 0.5110921229422093,
|
||||
"num_tokens": 605042.0,
|
||||
"step": 100
|
||||
},
|
||||
{
|
||||
"entropy": 2.197347580194473,
|
||||
"epoch": 0.024037498497656344,
|
||||
"grad_norm": 3.453125,
|
||||
"learning_rate": 1.952251241788175e-05,
|
||||
"loss": 2.2299420166015627,
|
||||
"mean_token_accuracy": 0.5093490639328957,
|
||||
"num_tokens": 906888.0,
|
||||
"step": 150
|
||||
},
|
||||
{
|
||||
"entropy": 2.1913950395584108,
|
||||
"epoch": 0.03204999799687513,
|
||||
"grad_norm": 2.90625,
|
||||
"learning_rate": 1.9362281685627303e-05,
|
||||
"loss": 2.210995941162109,
|
||||
"mean_token_accuracy": 0.5097909501194954,
|
||||
"num_tokens": 1213125.0,
|
||||
"step": 200
|
||||
},
|
||||
{
|
||||
"entropy": 2.1611998677253723,
|
||||
"epoch": 0.040062497496093905,
|
||||
"grad_norm": 3.25,
|
||||
"learning_rate": 1.920205095337286e-05,
|
||||
"loss": 2.1867474365234374,
|
||||
"mean_token_accuracy": 0.5154410660266876,
|
||||
"num_tokens": 1509026.0,
|
||||
"step": 250
|
||||
},
|
||||
{
|
||||
"entropy": 2.168051454424858,
|
||||
"epoch": 0.04807499699531269,
|
||||
"grad_norm": 2.9375,
|
||||
"learning_rate": 1.904182022111841e-05,
|
||||
"loss": 2.190071716308594,
|
||||
"mean_token_accuracy": 0.5156884534657001,
|
||||
"num_tokens": 1806850.0,
|
||||
"step": 300
|
||||
},
|
||||
{
|
||||
"entropy": 2.1737154543399813,
|
||||
"epoch": 0.05608749649453147,
|
||||
"grad_norm": 3.421875,
|
||||
"learning_rate": 1.8881589488863967e-05,
|
||||
"loss": 2.199147186279297,
|
||||
"mean_token_accuracy": 0.514045044630766,
|
||||
"num_tokens": 2104505.0,
|
||||
"step": 350
|
||||
},
|
||||
{
|
||||
"entropy": 2.148039159178734,
|
||||
"epoch": 0.06409999599375026,
|
||||
"grad_norm": 3.109375,
|
||||
"learning_rate": 1.872135875660952e-05,
|
||||
"loss": 2.176946105957031,
|
||||
"mean_token_accuracy": 0.5177112428843975,
|
||||
"num_tokens": 2391088.0,
|
||||
"step": 400
|
||||
},
|
||||
{
|
||||
"entropy": 2.194890711903572,
|
||||
"epoch": 0.07211249549296903,
|
||||
"grad_norm": 3.28125,
|
||||
"learning_rate": 1.856112802435507e-05,
|
||||
"loss": 2.216288757324219,
|
||||
"mean_token_accuracy": 0.5110335703194141,
|
||||
"num_tokens": 2691272.0,
|
||||
"step": 450
|
||||
},
|
||||
{
|
||||
"entropy": 2.1557786524295808,
|
||||
"epoch": 0.08012499499218781,
|
||||
"grad_norm": 4.15625,
|
||||
"learning_rate": 1.8400897292100627e-05,
|
||||
"loss": 2.1812680053710936,
|
||||
"mean_token_accuracy": 0.5176697050035,
|
||||
"num_tokens": 2991855.0,
|
||||
"step": 500
|
||||
},
|
||||
{
|
||||
"entropy": 2.1708167713880537,
|
||||
"epoch": 0.08813749449140659,
|
||||
"grad_norm": 4.34375,
|
||||
"learning_rate": 1.824066655984618e-05,
|
||||
"loss": 2.1921002197265627,
|
||||
"mean_token_accuracy": 0.5156507430970669,
|
||||
"num_tokens": 3295049.0,
|
||||
"step": 550
|
||||
},
|
||||
{
|
||||
"entropy": 2.1686314886808393,
|
||||
"epoch": 0.09614999399062538,
|
||||
"grad_norm": 2.875,
|
||||
"learning_rate": 1.8080435827591735e-05,
|
||||
"loss": 2.1914617919921877,
|
||||
"mean_token_accuracy": 0.5139613692462445,
|
||||
"num_tokens": 3597206.0,
|
||||
"step": 600
|
||||
},
|
||||
{
|
||||
"entropy": 2.15742906332016,
|
||||
"epoch": 0.10416249348984416,
|
||||
"grad_norm": 3.109375,
|
||||
"learning_rate": 1.7920205095337287e-05,
|
||||
"loss": 2.174853973388672,
|
||||
"mean_token_accuracy": 0.5159518288075924,
|
||||
"num_tokens": 3905268.0,
|
||||
"step": 650
|
||||
},
|
||||
{
|
||||
"entropy": 2.162363375425339,
|
||||
"epoch": 0.11217499298906294,
|
||||
"grad_norm": 3.171875,
|
||||
"learning_rate": 1.7759974363082843e-05,
|
||||
"loss": 2.184231872558594,
|
||||
"mean_token_accuracy": 0.516631413847208,
|
||||
"num_tokens": 4207934.0,
|
||||
"step": 700
|
||||
},
|
||||
{
|
||||
"entropy": 2.1516289287805557,
|
||||
"epoch": 0.12018749248828171,
|
||||
"grad_norm": 3.609375,
|
||||
"learning_rate": 1.7599743630828395e-05,
|
||||
"loss": 2.167020263671875,
|
||||
"mean_token_accuracy": 0.516358617246151,
|
||||
"num_tokens": 4515774.0,
|
||||
"step": 750
|
||||
},
|
||||
{
|
||||
"entropy": 2.1549544781446457,
|
||||
"epoch": 0.1281999919875005,
|
||||
"grad_norm": 3.28125,
|
||||
"learning_rate": 1.7439512898573947e-05,
|
||||
"loss": 2.181917877197266,
|
||||
"mean_token_accuracy": 0.5182220858335495,
|
||||
"num_tokens": 4814709.0,
|
||||
"step": 800
|
||||
},
|
||||
{
|
||||
"entropy": 2.1532615756988527,
|
||||
"epoch": 0.1362124914867193,
|
||||
"grad_norm": 3.078125,
|
||||
"learning_rate": 1.7279282166319503e-05,
|
||||
"loss": 2.1659007263183594,
|
||||
"mean_token_accuracy": 0.5180425970256328,
|
||||
"num_tokens": 5112585.0,
|
||||
"step": 850
|
||||
},
|
||||
{
|
||||
"entropy": 2.1704017293453215,
|
||||
"epoch": 0.14422499098593805,
|
||||
"grad_norm": 2.921875,
|
||||
"learning_rate": 1.7119051434065055e-05,
|
||||
"loss": 2.1934996032714844,
|
||||
"mean_token_accuracy": 0.5154964691400528,
|
||||
"num_tokens": 5411369.0,
|
||||
"step": 900
|
||||
},
|
||||
{
|
||||
"entropy": 2.1349107217788696,
|
||||
"epoch": 0.15223749048515683,
|
||||
"grad_norm": 2.984375,
|
||||
"learning_rate": 1.6958820701810608e-05,
|
||||
"loss": 2.1598545837402345,
|
||||
"mean_token_accuracy": 0.5208920259773732,
|
||||
"num_tokens": 5707060.0,
|
||||
"step": 950
|
||||
},
|
||||
{
|
||||
"entropy": 2.1541705918312073,
|
||||
"epoch": 0.16024998998437562,
|
||||
"grad_norm": 3.203125,
|
||||
"learning_rate": 1.6798589969556163e-05,
|
||||
"loss": 2.170956573486328,
|
||||
"mean_token_accuracy": 0.5204938448965549,
|
||||
"num_tokens": 6004079.0,
|
||||
"step": 1000
|
||||
},
|
||||
{
|
||||
"entropy": 2.1523867017030716,
|
||||
"epoch": 0.1682624894835944,
|
||||
"grad_norm": 3.09375,
|
||||
"learning_rate": 1.6638359237301716e-05,
|
||||
"loss": 2.1695118713378907,
|
||||
"mean_token_accuracy": 0.5177376234531402,
|
||||
"num_tokens": 6315462.0,
|
||||
"step": 1050
|
||||
},
|
||||
{
|
||||
"entropy": 2.1356022965908052,
|
||||
"epoch": 0.17627498898281319,
|
||||
"grad_norm": 3.21875,
|
||||
"learning_rate": 1.6478128505047268e-05,
|
||||
"loss": 2.1513107299804686,
|
||||
"mean_token_accuracy": 0.5217180585861206,
|
||||
"num_tokens": 6613060.0,
|
||||
"step": 1100
|
||||
},
|
||||
{
|
||||
"entropy": 2.1483460652828215,
|
||||
"epoch": 0.18428748848203197,
|
||||
"grad_norm": 3.21875,
|
||||
"learning_rate": 1.6317897772792824e-05,
|
||||
"loss": 2.1671595764160156,
|
||||
"mean_token_accuracy": 0.5199222292006016,
|
||||
"num_tokens": 6909655.0,
|
||||
"step": 1150
|
||||
},
|
||||
{
|
||||
"entropy": 2.1074040442705155,
|
||||
"epoch": 0.19229998798125075,
|
||||
"grad_norm": 3.828125,
|
||||
"learning_rate": 1.6157667040538376e-05,
|
||||
"loss": 2.12770751953125,
|
||||
"mean_token_accuracy": 0.5252153894305229,
|
||||
"num_tokens": 7203263.0,
|
||||
"step": 1200
|
||||
},
|
||||
{
|
||||
"entropy": 2.1354027593135836,
|
||||
"epoch": 0.20031248748046954,
|
||||
"grad_norm": 2.921875,
|
||||
"learning_rate": 1.5997436308283932e-05,
|
||||
"loss": 2.156183166503906,
|
||||
"mean_token_accuracy": 0.518907299786806,
|
||||
"num_tokens": 7505411.0,
|
||||
"step": 1250
|
||||
},
|
||||
{
|
||||
"entropy": 2.142924894094467,
|
||||
"epoch": 0.20832498697968832,
|
||||
"grad_norm": 3.5625,
|
||||
"learning_rate": 1.5837205576029484e-05,
|
||||
"loss": 2.158486633300781,
|
||||
"mean_token_accuracy": 0.5205909128487111,
|
||||
"num_tokens": 7803562.0,
|
||||
"step": 1300
|
||||
},
|
||||
{
|
||||
"entropy": 2.1305291467905043,
|
||||
"epoch": 0.2163374864789071,
|
||||
"grad_norm": 3.34375,
|
||||
"learning_rate": 1.5676974843775036e-05,
|
||||
"loss": 2.151043701171875,
|
||||
"mean_token_accuracy": 0.5211617235839366,
|
||||
"num_tokens": 8108037.0,
|
||||
"step": 1350
|
||||
},
|
||||
{
|
||||
"entropy": 2.160947166085243,
|
||||
"epoch": 0.2243499859781259,
|
||||
"grad_norm": 3.234375,
|
||||
"learning_rate": 1.5516744111520592e-05,
|
||||
"loss": 2.177792663574219,
|
||||
"mean_token_accuracy": 0.5167004990577698,
|
||||
"num_tokens": 8408674.0,
|
||||
"step": 1400
|
||||
},
|
||||
{
|
||||
"entropy": 2.120403967499733,
|
||||
"epoch": 0.23236248547734467,
|
||||
"grad_norm": 3.375,
|
||||
"learning_rate": 1.5356513379266144e-05,
|
||||
"loss": 2.1462562561035154,
|
||||
"mean_token_accuracy": 0.5225985661149025,
|
||||
"num_tokens": 8707650.0,
|
||||
"step": 1450
|
||||
},
|
||||
{
|
||||
"entropy": 2.138882914185524,
|
||||
"epoch": 0.24037498497656343,
|
||||
"grad_norm": 3.171875,
|
||||
"learning_rate": 1.5196282647011698e-05,
|
||||
"loss": 2.1499905395507812,
|
||||
"mean_token_accuracy": 0.5204668046534061,
|
||||
"num_tokens": 9006296.0,
|
||||
"step": 1500
|
||||
},
|
||||
{
|
||||
"entropy": 2.131927245259285,
|
||||
"epoch": 0.2483874844757822,
|
||||
"grad_norm": 3.546875,
|
||||
"learning_rate": 1.5036051914757252e-05,
|
||||
"loss": 2.157076416015625,
|
||||
"mean_token_accuracy": 0.5220408822596073,
|
||||
"num_tokens": 9307203.0,
|
||||
"step": 1550
|
||||
},
|
||||
{
|
||||
"entropy": 2.1308567851781843,
|
||||
"epoch": 0.256399983975001,
|
||||
"grad_norm": 3.546875,
|
||||
"learning_rate": 1.4875821182502805e-05,
|
||||
"loss": 2.146601104736328,
|
||||
"mean_token_accuracy": 0.5219539681077003,
|
||||
"num_tokens": 9610070.0,
|
||||
"step": 1600
|
||||
},
|
||||
{
|
||||
"entropy": 2.1097783839702604,
|
||||
"epoch": 0.2644124834742198,
|
||||
"grad_norm": 3.84375,
|
||||
"learning_rate": 1.4715590450248359e-05,
|
||||
"loss": 2.1268209838867187,
|
||||
"mean_token_accuracy": 0.5236652019619942,
|
||||
"num_tokens": 9909117.0,
|
||||
"step": 1650
|
||||
},
|
||||
{
|
||||
"entropy": 2.123499180674553,
|
||||
"epoch": 0.2724249829734386,
|
||||
"grad_norm": 3.46875,
|
||||
"learning_rate": 1.4555359717993913e-05,
|
||||
"loss": 2.1409934997558593,
|
||||
"mean_token_accuracy": 0.5225465354323388,
|
||||
"num_tokens": 10209461.0,
|
||||
"step": 1700
|
||||
},
|
||||
{
|
||||
"entropy": 2.13052960395813,
|
||||
"epoch": 0.2804374824726573,
|
||||
"grad_norm": 2.71875,
|
||||
"learning_rate": 1.4395128985739467e-05,
|
||||
"loss": 2.151373291015625,
|
||||
"mean_token_accuracy": 0.5227928113937378,
|
||||
"num_tokens": 10509282.0,
|
||||
"step": 1750
|
||||
},
|
||||
{
|
||||
"entropy": 2.1226114094257356,
|
||||
"epoch": 0.2884499819718761,
|
||||
"grad_norm": 3.65625,
|
||||
"learning_rate": 1.4234898253485019e-05,
|
||||
"loss": 2.144451141357422,
|
||||
"mean_token_accuracy": 0.5233827362954616,
|
||||
"num_tokens": 10811056.0,
|
||||
"step": 1800
|
||||
},
|
||||
{
|
||||
"entropy": 2.126143154501915,
|
||||
"epoch": 0.2964624814710949,
|
||||
"grad_norm": 3.359375,
|
||||
"learning_rate": 1.4074667521230575e-05,
|
||||
"loss": 2.1457598876953123,
|
||||
"mean_token_accuracy": 0.5221967925131321,
|
||||
"num_tokens": 11111153.0,
|
||||
"step": 1850
|
||||
},
|
||||
{
|
||||
"entropy": 2.1048326581716537,
|
||||
"epoch": 0.30447498097031367,
|
||||
"grad_norm": 3.59375,
|
||||
"learning_rate": 1.3914436788976127e-05,
|
||||
"loss": 2.1260116577148436,
|
||||
"mean_token_accuracy": 0.5272719909250736,
|
||||
"num_tokens": 11405606.0,
|
||||
"step": 1900
|
||||
},
|
||||
{
|
||||
"entropy": 2.1236617982387545,
|
||||
"epoch": 0.31248748046953245,
|
||||
"grad_norm": 3.0,
|
||||
"learning_rate": 1.375420605672168e-05,
|
||||
"loss": 2.151782684326172,
|
||||
"mean_token_accuracy": 0.5227546173334122,
|
||||
"num_tokens": 11703943.0,
|
||||
"step": 1950
|
||||
},
|
||||
{
|
||||
"entropy": 2.121563429236412,
|
||||
"epoch": 0.32049997996875124,
|
||||
"grad_norm": 3.34375,
|
||||
"learning_rate": 1.3593975324467235e-05,
|
||||
"loss": 2.1370144653320313,
|
||||
"mean_token_accuracy": 0.5235123547911644,
|
||||
"num_tokens": 12002522.0,
|
||||
"step": 2000
|
||||
}
|
||||
],
|
||||
"logging_steps": 50,
|
||||
"max_steps": 6241,
|
||||
"num_input_tokens_seen": 0,
|
||||
"num_train_epochs": 1,
|
||||
"save_steps": 500,
|
||||
"stateful_callbacks": {
|
||||
"TrainerControl": {
|
||||
"args": {
|
||||
"should_epoch_stop": false,
|
||||
"should_evaluate": false,
|
||||
"should_log": false,
|
||||
"should_save": true,
|
||||
"should_training_stop": false
|
||||
},
|
||||
"attributes": {}
|
||||
}
|
||||
},
|
||||
"total_flos": 1.4174735647666176e+17,
|
||||
"train_batch_size": 4,
|
||||
"trial_name": null,
|
||||
"trial_params": null
|
||||
}
|
||||
3
checkpoint-2000/training_args.bin
Normal file
3
checkpoint-2000/training_args.bin
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:78622f63304ec0913ed0b4ea7a1b26c08a5621b3661a380d798c4ea426235671
|
||||
size 5713
|
||||
89
checkpoint-2500/chat_template.jinja
Normal file
89
checkpoint-2500/chat_template.jinja
Normal file
@@ -0,0 +1,89 @@
|
||||
{%- if tools %}
|
||||
{{- '<|im_start|>system\n' }}
|
||||
{%- if messages[0].role == 'system' %}
|
||||
{{- messages[0].content + '\n\n' }}
|
||||
{%- endif %}
|
||||
{{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
|
||||
{%- for tool in tools %}
|
||||
{{- "\n" }}
|
||||
{{- tool | tojson }}
|
||||
{%- endfor %}
|
||||
{{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
|
||||
{%- else %}
|
||||
{%- if messages[0].role == 'system' %}
|
||||
{{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
|
||||
{%- endif %}
|
||||
{%- endif %}
|
||||
{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
|
||||
{%- for message in messages[::-1] %}
|
||||
{%- set index = (messages|length - 1) - loop.index0 %}
|
||||
{%- if ns.multi_step_tool and message.role == "user" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
|
||||
{%- set ns.multi_step_tool = false %}
|
||||
{%- set ns.last_query_index = index %}
|
||||
{%- endif %}
|
||||
{%- endfor %}
|
||||
{%- for message in messages %}
|
||||
{%- if message.content is string %}
|
||||
{%- set content = message.content %}
|
||||
{%- else %}
|
||||
{%- set content = '' %}
|
||||
{%- endif %}
|
||||
{%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
|
||||
{{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
|
||||
{%- elif message.role == "assistant" %}
|
||||
{%- set reasoning_content = '' %}
|
||||
{%- if message.reasoning_content is string %}
|
||||
{%- set reasoning_content = message.reasoning_content %}
|
||||
{%- else %}
|
||||
{%- if '</think>' in content %}
|
||||
{%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
|
||||
{%- set content = content.split('</think>')[-1].lstrip('\n') %}
|
||||
{%- endif %}
|
||||
{%- endif %}
|
||||
{%- if loop.index0 > ns.last_query_index %}
|
||||
{%- if loop.last or (not loop.last and reasoning_content) %}
|
||||
{{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
|
||||
{%- else %}
|
||||
{{- '<|im_start|>' + message.role + '\n' + content }}
|
||||
{%- endif %}
|
||||
{%- else %}
|
||||
{{- '<|im_start|>' + message.role + '\n' + content }}
|
||||
{%- endif %}
|
||||
{%- if message.tool_calls %}
|
||||
{%- for tool_call in message.tool_calls %}
|
||||
{%- if (loop.first and content) or (not loop.first) %}
|
||||
{{- '\n' }}
|
||||
{%- endif %}
|
||||
{%- if tool_call.function %}
|
||||
{%- set tool_call = tool_call.function %}
|
||||
{%- endif %}
|
||||
{{- '<tool_call>\n{"name": "' }}
|
||||
{{- tool_call.name }}
|
||||
{{- '", "arguments": ' }}
|
||||
{%- if tool_call.arguments is string %}
|
||||
{{- tool_call.arguments }}
|
||||
{%- else %}
|
||||
{{- tool_call.arguments | tojson }}
|
||||
{%- endif %}
|
||||
{{- '}\n</tool_call>' }}
|
||||
{%- endfor %}
|
||||
{%- endif %}
|
||||
{{- '<|im_end|>\n' }}
|
||||
{%- elif message.role == "tool" %}
|
||||
{%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
|
||||
{{- '<|im_start|>user' }}
|
||||
{%- endif %}
|
||||
{{- '\n<tool_response>\n' }}
|
||||
{{- content }}
|
||||
{{- '\n</tool_response>' }}
|
||||
{%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
|
||||
{{- '<|im_end|>\n' }}
|
||||
{%- endif %}
|
||||
{%- endif %}
|
||||
{%- endfor %}
|
||||
{%- if add_generation_prompt %}
|
||||
{{- '<|im_start|>assistant\n' }}
|
||||
{%- if enable_thinking is defined and enable_thinking is false %}
|
||||
{{- '<think>\n\n</think>\n\n' }}
|
||||
{%- endif %}
|
||||
{%- endif %}
|
||||
63
checkpoint-2500/config.json
Normal file
63
checkpoint-2500/config.json
Normal file
@@ -0,0 +1,63 @@
|
||||
{
|
||||
"architectures": [
|
||||
"Qwen3ForCausalLM"
|
||||
],
|
||||
"attention_bias": false,
|
||||
"attention_dropout": 0.0,
|
||||
"bos_token_id": null,
|
||||
"dtype": "bfloat16",
|
||||
"eos_token_id": 151645,
|
||||
"head_dim": 128,
|
||||
"hidden_act": "silu",
|
||||
"hidden_size": 2048,
|
||||
"initializer_range": 0.02,
|
||||
"intermediate_size": 6144,
|
||||
"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"
|
||||
],
|
||||
"max_position_embeddings": 40960,
|
||||
"max_window_layers": 28,
|
||||
"model_type": "qwen3",
|
||||
"num_attention_heads": 16,
|
||||
"num_hidden_layers": 28,
|
||||
"num_key_value_heads": 8,
|
||||
"pad_token_id": 151643,
|
||||
"rms_norm_eps": 1e-06,
|
||||
"rope_parameters": {
|
||||
"rope_theta": 1000000,
|
||||
"rope_type": "default"
|
||||
},
|
||||
"sliding_window": null,
|
||||
"tie_word_embeddings": true,
|
||||
"transformers_version": "5.7.0",
|
||||
"use_cache": false,
|
||||
"use_sliding_window": false,
|
||||
"vocab_size": 151936
|
||||
}
|
||||
12
checkpoint-2500/generation_config.json
Normal file
12
checkpoint-2500/generation_config.json
Normal file
@@ -0,0 +1,12 @@
|
||||
{
|
||||
"do_sample": true,
|
||||
"eos_token_id": [
|
||||
151645,
|
||||
151643
|
||||
],
|
||||
"pad_token_id": 151643,
|
||||
"temperature": 0.6,
|
||||
"top_k": 20,
|
||||
"top_p": 0.95,
|
||||
"transformers_version": "5.7.0"
|
||||
}
|
||||
3
checkpoint-2500/model.safetensors
Normal file
3
checkpoint-2500/model.safetensors
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:98a04b91a11ec1c0b526f91d12bdfb28ff4c3c69e104271b9e28a2f75f44f891
|
||||
size 3441185608
|
||||
3
checkpoint-2500/optimizer.pt
Normal file
3
checkpoint-2500/optimizer.pt
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:7535b338ca59a9c376634a6fbbef04a836c5c21962e653fe7794832854643965
|
||||
size 6882572207
|
||||
3
checkpoint-2500/rng_state.pth
Normal file
3
checkpoint-2500/rng_state.pth
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:61c19bab1174704a4a4441475683bf1270277af15d2e2c95e964789128e482c4
|
||||
size 14645
|
||||
3
checkpoint-2500/scheduler.pt
Normal file
3
checkpoint-2500/scheduler.pt
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:c863cc66645dc4964c4c4f2cf30406fef13cbea85c7c7bd73c333d72670c821e
|
||||
size 1465
|
||||
3
checkpoint-2500/tokenizer.json
Normal file
3
checkpoint-2500/tokenizer.json
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:be75606093db2094d7cd20f3c2f385c212750648bd6ea4fb2bf507a6a4c55506
|
||||
size 11422650
|
||||
30
checkpoint-2500/tokenizer_config.json
Normal file
30
checkpoint-2500/tokenizer_config.json
Normal file
@@ -0,0 +1,30 @@
|
||||
{
|
||||
"add_prefix_space": false,
|
||||
"backend": "tokenizers",
|
||||
"bos_token": null,
|
||||
"clean_up_tokenization_spaces": false,
|
||||
"eos_token": "<|im_end|>",
|
||||
"errors": "replace",
|
||||
"extra_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|>"
|
||||
],
|
||||
"is_local": false,
|
||||
"local_files_only": false,
|
||||
"model_max_length": 131072,
|
||||
"pad_token": "<|endoftext|>",
|
||||
"split_special_tokens": false,
|
||||
"tokenizer_class": "Qwen2Tokenizer",
|
||||
"unk_token": null
|
||||
}
|
||||
534
checkpoint-2500/trainer_state.json
Normal file
534
checkpoint-2500/trainer_state.json
Normal file
@@ -0,0 +1,534 @@
|
||||
{
|
||||
"best_global_step": null,
|
||||
"best_metric": null,
|
||||
"best_model_checkpoint": null,
|
||||
"epoch": 0.4006249749609391,
|
||||
"eval_steps": 500,
|
||||
"global_step": 2500,
|
||||
"is_hyper_param_search": false,
|
||||
"is_local_process_zero": true,
|
||||
"is_world_process_zero": true,
|
||||
"log_history": [
|
||||
{
|
||||
"entropy": 2.223189059495926,
|
||||
"epoch": 0.008012499499218782,
|
||||
"grad_norm": 2.875,
|
||||
"learning_rate": 1.9842973882390643e-05,
|
||||
"loss": 2.3428109741210936,
|
||||
"mean_token_accuracy": 0.4963963569700718,
|
||||
"num_tokens": 305366.0,
|
||||
"step": 50
|
||||
},
|
||||
{
|
||||
"entropy": 2.188130117058754,
|
||||
"epoch": 0.016024998998437564,
|
||||
"grad_norm": 3.328125,
|
||||
"learning_rate": 1.96827431501362e-05,
|
||||
"loss": 2.225358734130859,
|
||||
"mean_token_accuracy": 0.5110921229422093,
|
||||
"num_tokens": 605042.0,
|
||||
"step": 100
|
||||
},
|
||||
{
|
||||
"entropy": 2.197347580194473,
|
||||
"epoch": 0.024037498497656344,
|
||||
"grad_norm": 3.453125,
|
||||
"learning_rate": 1.952251241788175e-05,
|
||||
"loss": 2.2299420166015627,
|
||||
"mean_token_accuracy": 0.5093490639328957,
|
||||
"num_tokens": 906888.0,
|
||||
"step": 150
|
||||
},
|
||||
{
|
||||
"entropy": 2.1913950395584108,
|
||||
"epoch": 0.03204999799687513,
|
||||
"grad_norm": 2.90625,
|
||||
"learning_rate": 1.9362281685627303e-05,
|
||||
"loss": 2.210995941162109,
|
||||
"mean_token_accuracy": 0.5097909501194954,
|
||||
"num_tokens": 1213125.0,
|
||||
"step": 200
|
||||
},
|
||||
{
|
||||
"entropy": 2.1611998677253723,
|
||||
"epoch": 0.040062497496093905,
|
||||
"grad_norm": 3.25,
|
||||
"learning_rate": 1.920205095337286e-05,
|
||||
"loss": 2.1867474365234374,
|
||||
"mean_token_accuracy": 0.5154410660266876,
|
||||
"num_tokens": 1509026.0,
|
||||
"step": 250
|
||||
},
|
||||
{
|
||||
"entropy": 2.168051454424858,
|
||||
"epoch": 0.04807499699531269,
|
||||
"grad_norm": 2.9375,
|
||||
"learning_rate": 1.904182022111841e-05,
|
||||
"loss": 2.190071716308594,
|
||||
"mean_token_accuracy": 0.5156884534657001,
|
||||
"num_tokens": 1806850.0,
|
||||
"step": 300
|
||||
},
|
||||
{
|
||||
"entropy": 2.1737154543399813,
|
||||
"epoch": 0.05608749649453147,
|
||||
"grad_norm": 3.421875,
|
||||
"learning_rate": 1.8881589488863967e-05,
|
||||
"loss": 2.199147186279297,
|
||||
"mean_token_accuracy": 0.514045044630766,
|
||||
"num_tokens": 2104505.0,
|
||||
"step": 350
|
||||
},
|
||||
{
|
||||
"entropy": 2.148039159178734,
|
||||
"epoch": 0.06409999599375026,
|
||||
"grad_norm": 3.109375,
|
||||
"learning_rate": 1.872135875660952e-05,
|
||||
"loss": 2.176946105957031,
|
||||
"mean_token_accuracy": 0.5177112428843975,
|
||||
"num_tokens": 2391088.0,
|
||||
"step": 400
|
||||
},
|
||||
{
|
||||
"entropy": 2.194890711903572,
|
||||
"epoch": 0.07211249549296903,
|
||||
"grad_norm": 3.28125,
|
||||
"learning_rate": 1.856112802435507e-05,
|
||||
"loss": 2.216288757324219,
|
||||
"mean_token_accuracy": 0.5110335703194141,
|
||||
"num_tokens": 2691272.0,
|
||||
"step": 450
|
||||
},
|
||||
{
|
||||
"entropy": 2.1557786524295808,
|
||||
"epoch": 0.08012499499218781,
|
||||
"grad_norm": 4.15625,
|
||||
"learning_rate": 1.8400897292100627e-05,
|
||||
"loss": 2.1812680053710936,
|
||||
"mean_token_accuracy": 0.5176697050035,
|
||||
"num_tokens": 2991855.0,
|
||||
"step": 500
|
||||
},
|
||||
{
|
||||
"entropy": 2.1708167713880537,
|
||||
"epoch": 0.08813749449140659,
|
||||
"grad_norm": 4.34375,
|
||||
"learning_rate": 1.824066655984618e-05,
|
||||
"loss": 2.1921002197265627,
|
||||
"mean_token_accuracy": 0.5156507430970669,
|
||||
"num_tokens": 3295049.0,
|
||||
"step": 550
|
||||
},
|
||||
{
|
||||
"entropy": 2.1686314886808393,
|
||||
"epoch": 0.09614999399062538,
|
||||
"grad_norm": 2.875,
|
||||
"learning_rate": 1.8080435827591735e-05,
|
||||
"loss": 2.1914617919921877,
|
||||
"mean_token_accuracy": 0.5139613692462445,
|
||||
"num_tokens": 3597206.0,
|
||||
"step": 600
|
||||
},
|
||||
{
|
||||
"entropy": 2.15742906332016,
|
||||
"epoch": 0.10416249348984416,
|
||||
"grad_norm": 3.109375,
|
||||
"learning_rate": 1.7920205095337287e-05,
|
||||
"loss": 2.174853973388672,
|
||||
"mean_token_accuracy": 0.5159518288075924,
|
||||
"num_tokens": 3905268.0,
|
||||
"step": 650
|
||||
},
|
||||
{
|
||||
"entropy": 2.162363375425339,
|
||||
"epoch": 0.11217499298906294,
|
||||
"grad_norm": 3.171875,
|
||||
"learning_rate": 1.7759974363082843e-05,
|
||||
"loss": 2.184231872558594,
|
||||
"mean_token_accuracy": 0.516631413847208,
|
||||
"num_tokens": 4207934.0,
|
||||
"step": 700
|
||||
},
|
||||
{
|
||||
"entropy": 2.1516289287805557,
|
||||
"epoch": 0.12018749248828171,
|
||||
"grad_norm": 3.609375,
|
||||
"learning_rate": 1.7599743630828395e-05,
|
||||
"loss": 2.167020263671875,
|
||||
"mean_token_accuracy": 0.516358617246151,
|
||||
"num_tokens": 4515774.0,
|
||||
"step": 750
|
||||
},
|
||||
{
|
||||
"entropy": 2.1549544781446457,
|
||||
"epoch": 0.1281999919875005,
|
||||
"grad_norm": 3.28125,
|
||||
"learning_rate": 1.7439512898573947e-05,
|
||||
"loss": 2.181917877197266,
|
||||
"mean_token_accuracy": 0.5182220858335495,
|
||||
"num_tokens": 4814709.0,
|
||||
"step": 800
|
||||
},
|
||||
{
|
||||
"entropy": 2.1532615756988527,
|
||||
"epoch": 0.1362124914867193,
|
||||
"grad_norm": 3.078125,
|
||||
"learning_rate": 1.7279282166319503e-05,
|
||||
"loss": 2.1659007263183594,
|
||||
"mean_token_accuracy": 0.5180425970256328,
|
||||
"num_tokens": 5112585.0,
|
||||
"step": 850
|
||||
},
|
||||
{
|
||||
"entropy": 2.1704017293453215,
|
||||
"epoch": 0.14422499098593805,
|
||||
"grad_norm": 2.921875,
|
||||
"learning_rate": 1.7119051434065055e-05,
|
||||
"loss": 2.1934996032714844,
|
||||
"mean_token_accuracy": 0.5154964691400528,
|
||||
"num_tokens": 5411369.0,
|
||||
"step": 900
|
||||
},
|
||||
{
|
||||
"entropy": 2.1349107217788696,
|
||||
"epoch": 0.15223749048515683,
|
||||
"grad_norm": 2.984375,
|
||||
"learning_rate": 1.6958820701810608e-05,
|
||||
"loss": 2.1598545837402345,
|
||||
"mean_token_accuracy": 0.5208920259773732,
|
||||
"num_tokens": 5707060.0,
|
||||
"step": 950
|
||||
},
|
||||
{
|
||||
"entropy": 2.1541705918312073,
|
||||
"epoch": 0.16024998998437562,
|
||||
"grad_norm": 3.203125,
|
||||
"learning_rate": 1.6798589969556163e-05,
|
||||
"loss": 2.170956573486328,
|
||||
"mean_token_accuracy": 0.5204938448965549,
|
||||
"num_tokens": 6004079.0,
|
||||
"step": 1000
|
||||
},
|
||||
{
|
||||
"entropy": 2.1523867017030716,
|
||||
"epoch": 0.1682624894835944,
|
||||
"grad_norm": 3.09375,
|
||||
"learning_rate": 1.6638359237301716e-05,
|
||||
"loss": 2.1695118713378907,
|
||||
"mean_token_accuracy": 0.5177376234531402,
|
||||
"num_tokens": 6315462.0,
|
||||
"step": 1050
|
||||
},
|
||||
{
|
||||
"entropy": 2.1356022965908052,
|
||||
"epoch": 0.17627498898281319,
|
||||
"grad_norm": 3.21875,
|
||||
"learning_rate": 1.6478128505047268e-05,
|
||||
"loss": 2.1513107299804686,
|
||||
"mean_token_accuracy": 0.5217180585861206,
|
||||
"num_tokens": 6613060.0,
|
||||
"step": 1100
|
||||
},
|
||||
{
|
||||
"entropy": 2.1483460652828215,
|
||||
"epoch": 0.18428748848203197,
|
||||
"grad_norm": 3.21875,
|
||||
"learning_rate": 1.6317897772792824e-05,
|
||||
"loss": 2.1671595764160156,
|
||||
"mean_token_accuracy": 0.5199222292006016,
|
||||
"num_tokens": 6909655.0,
|
||||
"step": 1150
|
||||
},
|
||||
{
|
||||
"entropy": 2.1074040442705155,
|
||||
"epoch": 0.19229998798125075,
|
||||
"grad_norm": 3.828125,
|
||||
"learning_rate": 1.6157667040538376e-05,
|
||||
"loss": 2.12770751953125,
|
||||
"mean_token_accuracy": 0.5252153894305229,
|
||||
"num_tokens": 7203263.0,
|
||||
"step": 1200
|
||||
},
|
||||
{
|
||||
"entropy": 2.1354027593135836,
|
||||
"epoch": 0.20031248748046954,
|
||||
"grad_norm": 2.921875,
|
||||
"learning_rate": 1.5997436308283932e-05,
|
||||
"loss": 2.156183166503906,
|
||||
"mean_token_accuracy": 0.518907299786806,
|
||||
"num_tokens": 7505411.0,
|
||||
"step": 1250
|
||||
},
|
||||
{
|
||||
"entropy": 2.142924894094467,
|
||||
"epoch": 0.20832498697968832,
|
||||
"grad_norm": 3.5625,
|
||||
"learning_rate": 1.5837205576029484e-05,
|
||||
"loss": 2.158486633300781,
|
||||
"mean_token_accuracy": 0.5205909128487111,
|
||||
"num_tokens": 7803562.0,
|
||||
"step": 1300
|
||||
},
|
||||
{
|
||||
"entropy": 2.1305291467905043,
|
||||
"epoch": 0.2163374864789071,
|
||||
"grad_norm": 3.34375,
|
||||
"learning_rate": 1.5676974843775036e-05,
|
||||
"loss": 2.151043701171875,
|
||||
"mean_token_accuracy": 0.5211617235839366,
|
||||
"num_tokens": 8108037.0,
|
||||
"step": 1350
|
||||
},
|
||||
{
|
||||
"entropy": 2.160947166085243,
|
||||
"epoch": 0.2243499859781259,
|
||||
"grad_norm": 3.234375,
|
||||
"learning_rate": 1.5516744111520592e-05,
|
||||
"loss": 2.177792663574219,
|
||||
"mean_token_accuracy": 0.5167004990577698,
|
||||
"num_tokens": 8408674.0,
|
||||
"step": 1400
|
||||
},
|
||||
{
|
||||
"entropy": 2.120403967499733,
|
||||
"epoch": 0.23236248547734467,
|
||||
"grad_norm": 3.375,
|
||||
"learning_rate": 1.5356513379266144e-05,
|
||||
"loss": 2.1462562561035154,
|
||||
"mean_token_accuracy": 0.5225985661149025,
|
||||
"num_tokens": 8707650.0,
|
||||
"step": 1450
|
||||
},
|
||||
{
|
||||
"entropy": 2.138882914185524,
|
||||
"epoch": 0.24037498497656343,
|
||||
"grad_norm": 3.171875,
|
||||
"learning_rate": 1.5196282647011698e-05,
|
||||
"loss": 2.1499905395507812,
|
||||
"mean_token_accuracy": 0.5204668046534061,
|
||||
"num_tokens": 9006296.0,
|
||||
"step": 1500
|
||||
},
|
||||
{
|
||||
"entropy": 2.131927245259285,
|
||||
"epoch": 0.2483874844757822,
|
||||
"grad_norm": 3.546875,
|
||||
"learning_rate": 1.5036051914757252e-05,
|
||||
"loss": 2.157076416015625,
|
||||
"mean_token_accuracy": 0.5220408822596073,
|
||||
"num_tokens": 9307203.0,
|
||||
"step": 1550
|
||||
},
|
||||
{
|
||||
"entropy": 2.1308567851781843,
|
||||
"epoch": 0.256399983975001,
|
||||
"grad_norm": 3.546875,
|
||||
"learning_rate": 1.4875821182502805e-05,
|
||||
"loss": 2.146601104736328,
|
||||
"mean_token_accuracy": 0.5219539681077003,
|
||||
"num_tokens": 9610070.0,
|
||||
"step": 1600
|
||||
},
|
||||
{
|
||||
"entropy": 2.1097783839702604,
|
||||
"epoch": 0.2644124834742198,
|
||||
"grad_norm": 3.84375,
|
||||
"learning_rate": 1.4715590450248359e-05,
|
||||
"loss": 2.1268209838867187,
|
||||
"mean_token_accuracy": 0.5236652019619942,
|
||||
"num_tokens": 9909117.0,
|
||||
"step": 1650
|
||||
},
|
||||
{
|
||||
"entropy": 2.123499180674553,
|
||||
"epoch": 0.2724249829734386,
|
||||
"grad_norm": 3.46875,
|
||||
"learning_rate": 1.4555359717993913e-05,
|
||||
"loss": 2.1409934997558593,
|
||||
"mean_token_accuracy": 0.5225465354323388,
|
||||
"num_tokens": 10209461.0,
|
||||
"step": 1700
|
||||
},
|
||||
{
|
||||
"entropy": 2.13052960395813,
|
||||
"epoch": 0.2804374824726573,
|
||||
"grad_norm": 2.71875,
|
||||
"learning_rate": 1.4395128985739467e-05,
|
||||
"loss": 2.151373291015625,
|
||||
"mean_token_accuracy": 0.5227928113937378,
|
||||
"num_tokens": 10509282.0,
|
||||
"step": 1750
|
||||
},
|
||||
{
|
||||
"entropy": 2.1226114094257356,
|
||||
"epoch": 0.2884499819718761,
|
||||
"grad_norm": 3.65625,
|
||||
"learning_rate": 1.4234898253485019e-05,
|
||||
"loss": 2.144451141357422,
|
||||
"mean_token_accuracy": 0.5233827362954616,
|
||||
"num_tokens": 10811056.0,
|
||||
"step": 1800
|
||||
},
|
||||
{
|
||||
"entropy": 2.126143154501915,
|
||||
"epoch": 0.2964624814710949,
|
||||
"grad_norm": 3.359375,
|
||||
"learning_rate": 1.4074667521230575e-05,
|
||||
"loss": 2.1457598876953123,
|
||||
"mean_token_accuracy": 0.5221967925131321,
|
||||
"num_tokens": 11111153.0,
|
||||
"step": 1850
|
||||
},
|
||||
{
|
||||
"entropy": 2.1048326581716537,
|
||||
"epoch": 0.30447498097031367,
|
||||
"grad_norm": 3.59375,
|
||||
"learning_rate": 1.3914436788976127e-05,
|
||||
"loss": 2.1260116577148436,
|
||||
"mean_token_accuracy": 0.5272719909250736,
|
||||
"num_tokens": 11405606.0,
|
||||
"step": 1900
|
||||
},
|
||||
{
|
||||
"entropy": 2.1236617982387545,
|
||||
"epoch": 0.31248748046953245,
|
||||
"grad_norm": 3.0,
|
||||
"learning_rate": 1.375420605672168e-05,
|
||||
"loss": 2.151782684326172,
|
||||
"mean_token_accuracy": 0.5227546173334122,
|
||||
"num_tokens": 11703943.0,
|
||||
"step": 1950
|
||||
},
|
||||
{
|
||||
"entropy": 2.121563429236412,
|
||||
"epoch": 0.32049997996875124,
|
||||
"grad_norm": 3.34375,
|
||||
"learning_rate": 1.3593975324467235e-05,
|
||||
"loss": 2.1370144653320313,
|
||||
"mean_token_accuracy": 0.5235123547911644,
|
||||
"num_tokens": 12002522.0,
|
||||
"step": 2000
|
||||
},
|
||||
{
|
||||
"entropy": 2.1155461168289182,
|
||||
"epoch": 0.32851247946797,
|
||||
"grad_norm": 3.46875,
|
||||
"learning_rate": 1.3433744592212787e-05,
|
||||
"loss": 2.1364064025878906,
|
||||
"mean_token_accuracy": 0.5237135806679726,
|
||||
"num_tokens": 12303098.0,
|
||||
"step": 2050
|
||||
},
|
||||
{
|
||||
"entropy": 2.1386052322387696,
|
||||
"epoch": 0.3365249789671888,
|
||||
"grad_norm": 3.484375,
|
||||
"learning_rate": 1.327351385995834e-05,
|
||||
"loss": 2.149888458251953,
|
||||
"mean_token_accuracy": 0.522669050693512,
|
||||
"num_tokens": 12605534.0,
|
||||
"step": 2100
|
||||
},
|
||||
{
|
||||
"entropy": 2.102600795030594,
|
||||
"epoch": 0.3445374784664076,
|
||||
"grad_norm": 3.796875,
|
||||
"learning_rate": 1.3113283127703895e-05,
|
||||
"loss": 2.1236566162109374,
|
||||
"mean_token_accuracy": 0.5279092155396938,
|
||||
"num_tokens": 12901218.0,
|
||||
"step": 2150
|
||||
},
|
||||
{
|
||||
"entropy": 2.132645187973976,
|
||||
"epoch": 0.35254997796562637,
|
||||
"grad_norm": 3.53125,
|
||||
"learning_rate": 1.2953052395449447e-05,
|
||||
"loss": 2.1523503112792968,
|
||||
"mean_token_accuracy": 0.5198129612207413,
|
||||
"num_tokens": 13213018.0,
|
||||
"step": 2200
|
||||
},
|
||||
{
|
||||
"entropy": 2.1204321217536926,
|
||||
"epoch": 0.36056247746484515,
|
||||
"grad_norm": 3.25,
|
||||
"learning_rate": 1.2792821663195001e-05,
|
||||
"loss": 2.1325447082519533,
|
||||
"mean_token_accuracy": 0.5243636940419674,
|
||||
"num_tokens": 13514408.0,
|
||||
"step": 2250
|
||||
},
|
||||
{
|
||||
"entropy": 2.102410731315613,
|
||||
"epoch": 0.36857497696406394,
|
||||
"grad_norm": 3.1875,
|
||||
"learning_rate": 1.2632590930940555e-05,
|
||||
"loss": 2.123565368652344,
|
||||
"mean_token_accuracy": 0.5257614709436893,
|
||||
"num_tokens": 13818272.0,
|
||||
"step": 2300
|
||||
},
|
||||
{
|
||||
"entropy": 2.0876205080747603,
|
||||
"epoch": 0.3765874764632827,
|
||||
"grad_norm": 3.328125,
|
||||
"learning_rate": 1.247236019868611e-05,
|
||||
"loss": 2.116696472167969,
|
||||
"mean_token_accuracy": 0.5275124031305313,
|
||||
"num_tokens": 14110702.0,
|
||||
"step": 2350
|
||||
},
|
||||
{
|
||||
"entropy": 2.1078518223762512,
|
||||
"epoch": 0.3845999759625015,
|
||||
"grad_norm": 3.453125,
|
||||
"learning_rate": 1.2312129466431662e-05,
|
||||
"loss": 2.1291993713378905,
|
||||
"mean_token_accuracy": 0.524967464953661,
|
||||
"num_tokens": 14407015.0,
|
||||
"step": 2400
|
||||
},
|
||||
{
|
||||
"entropy": 2.0930898267030718,
|
||||
"epoch": 0.3926124754617203,
|
||||
"grad_norm": 3.5,
|
||||
"learning_rate": 1.2151898734177216e-05,
|
||||
"loss": 2.1094123840332033,
|
||||
"mean_token_accuracy": 0.5290587748587131,
|
||||
"num_tokens": 14696565.0,
|
||||
"step": 2450
|
||||
},
|
||||
{
|
||||
"entropy": 2.1109757220745085,
|
||||
"epoch": 0.4006249749609391,
|
||||
"grad_norm": 3.5,
|
||||
"learning_rate": 1.199166800192277e-05,
|
||||
"loss": 2.1246888732910154,
|
||||
"mean_token_accuracy": 0.5258957357704639,
|
||||
"num_tokens": 14994401.0,
|
||||
"step": 2500
|
||||
}
|
||||
],
|
||||
"logging_steps": 50,
|
||||
"max_steps": 6241,
|
||||
"num_input_tokens_seen": 0,
|
||||
"num_train_epochs": 1,
|
||||
"save_steps": 500,
|
||||
"stateful_callbacks": {
|
||||
"TrainerControl": {
|
||||
"args": {
|
||||
"should_epoch_stop": false,
|
||||
"should_evaluate": false,
|
||||
"should_log": false,
|
||||
"should_save": true,
|
||||
"should_training_stop": false
|
||||
},
|
||||
"attributes": {}
|
||||
}
|
||||
},
|
||||
"total_flos": 1.773461452668027e+17,
|
||||
"train_batch_size": 4,
|
||||
"trial_name": null,
|
||||
"trial_params": null
|
||||
}
|
||||
3
checkpoint-2500/training_args.bin
Normal file
3
checkpoint-2500/training_args.bin
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:78622f63304ec0913ed0b4ea7a1b26c08a5621b3661a380d798c4ea426235671
|
||||
size 5713
|
||||
89
checkpoint-3000/chat_template.jinja
Normal file
89
checkpoint-3000/chat_template.jinja
Normal file
@@ -0,0 +1,89 @@
|
||||
{%- if tools %}
|
||||
{{- '<|im_start|>system\n' }}
|
||||
{%- if messages[0].role == 'system' %}
|
||||
{{- messages[0].content + '\n\n' }}
|
||||
{%- endif %}
|
||||
{{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
|
||||
{%- for tool in tools %}
|
||||
{{- "\n" }}
|
||||
{{- tool | tojson }}
|
||||
{%- endfor %}
|
||||
{{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
|
||||
{%- else %}
|
||||
{%- if messages[0].role == 'system' %}
|
||||
{{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
|
||||
{%- endif %}
|
||||
{%- endif %}
|
||||
{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
|
||||
{%- for message in messages[::-1] %}
|
||||
{%- set index = (messages|length - 1) - loop.index0 %}
|
||||
{%- if ns.multi_step_tool and message.role == "user" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
|
||||
{%- set ns.multi_step_tool = false %}
|
||||
{%- set ns.last_query_index = index %}
|
||||
{%- endif %}
|
||||
{%- endfor %}
|
||||
{%- for message in messages %}
|
||||
{%- if message.content is string %}
|
||||
{%- set content = message.content %}
|
||||
{%- else %}
|
||||
{%- set content = '' %}
|
||||
{%- endif %}
|
||||
{%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
|
||||
{{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
|
||||
{%- elif message.role == "assistant" %}
|
||||
{%- set reasoning_content = '' %}
|
||||
{%- if message.reasoning_content is string %}
|
||||
{%- set reasoning_content = message.reasoning_content %}
|
||||
{%- else %}
|
||||
{%- if '</think>' in content %}
|
||||
{%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
|
||||
{%- set content = content.split('</think>')[-1].lstrip('\n') %}
|
||||
{%- endif %}
|
||||
{%- endif %}
|
||||
{%- if loop.index0 > ns.last_query_index %}
|
||||
{%- if loop.last or (not loop.last and reasoning_content) %}
|
||||
{{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
|
||||
{%- else %}
|
||||
{{- '<|im_start|>' + message.role + '\n' + content }}
|
||||
{%- endif %}
|
||||
{%- else %}
|
||||
{{- '<|im_start|>' + message.role + '\n' + content }}
|
||||
{%- endif %}
|
||||
{%- if message.tool_calls %}
|
||||
{%- for tool_call in message.tool_calls %}
|
||||
{%- if (loop.first and content) or (not loop.first) %}
|
||||
{{- '\n' }}
|
||||
{%- endif %}
|
||||
{%- if tool_call.function %}
|
||||
{%- set tool_call = tool_call.function %}
|
||||
{%- endif %}
|
||||
{{- '<tool_call>\n{"name": "' }}
|
||||
{{- tool_call.name }}
|
||||
{{- '", "arguments": ' }}
|
||||
{%- if tool_call.arguments is string %}
|
||||
{{- tool_call.arguments }}
|
||||
{%- else %}
|
||||
{{- tool_call.arguments | tojson }}
|
||||
{%- endif %}
|
||||
{{- '}\n</tool_call>' }}
|
||||
{%- endfor %}
|
||||
{%- endif %}
|
||||
{{- '<|im_end|>\n' }}
|
||||
{%- elif message.role == "tool" %}
|
||||
{%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
|
||||
{{- '<|im_start|>user' }}
|
||||
{%- endif %}
|
||||
{{- '\n<tool_response>\n' }}
|
||||
{{- content }}
|
||||
{{- '\n</tool_response>' }}
|
||||
{%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
|
||||
{{- '<|im_end|>\n' }}
|
||||
{%- endif %}
|
||||
{%- endif %}
|
||||
{%- endfor %}
|
||||
{%- if add_generation_prompt %}
|
||||
{{- '<|im_start|>assistant\n' }}
|
||||
{%- if enable_thinking is defined and enable_thinking is false %}
|
||||
{{- '<think>\n\n</think>\n\n' }}
|
||||
{%- endif %}
|
||||
{%- endif %}
|
||||
63
checkpoint-3000/config.json
Normal file
63
checkpoint-3000/config.json
Normal file
@@ -0,0 +1,63 @@
|
||||
{
|
||||
"architectures": [
|
||||
"Qwen3ForCausalLM"
|
||||
],
|
||||
"attention_bias": false,
|
||||
"attention_dropout": 0.0,
|
||||
"bos_token_id": null,
|
||||
"dtype": "bfloat16",
|
||||
"eos_token_id": 151645,
|
||||
"head_dim": 128,
|
||||
"hidden_act": "silu",
|
||||
"hidden_size": 2048,
|
||||
"initializer_range": 0.02,
|
||||
"intermediate_size": 6144,
|
||||
"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"
|
||||
],
|
||||
"max_position_embeddings": 40960,
|
||||
"max_window_layers": 28,
|
||||
"model_type": "qwen3",
|
||||
"num_attention_heads": 16,
|
||||
"num_hidden_layers": 28,
|
||||
"num_key_value_heads": 8,
|
||||
"pad_token_id": 151643,
|
||||
"rms_norm_eps": 1e-06,
|
||||
"rope_parameters": {
|
||||
"rope_theta": 1000000,
|
||||
"rope_type": "default"
|
||||
},
|
||||
"sliding_window": null,
|
||||
"tie_word_embeddings": true,
|
||||
"transformers_version": "5.7.0",
|
||||
"use_cache": false,
|
||||
"use_sliding_window": false,
|
||||
"vocab_size": 151936
|
||||
}
|
||||
12
checkpoint-3000/generation_config.json
Normal file
12
checkpoint-3000/generation_config.json
Normal file
@@ -0,0 +1,12 @@
|
||||
{
|
||||
"do_sample": true,
|
||||
"eos_token_id": [
|
||||
151645,
|
||||
151643
|
||||
],
|
||||
"pad_token_id": 151643,
|
||||
"temperature": 0.6,
|
||||
"top_k": 20,
|
||||
"top_p": 0.95,
|
||||
"transformers_version": "5.7.0"
|
||||
}
|
||||
3
checkpoint-3000/model.safetensors
Normal file
3
checkpoint-3000/model.safetensors
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:d5b72e5752b081abdc1994e5b2076200e5122c26cf9578d62c85a820f8ff24de
|
||||
size 3441185608
|
||||
3
checkpoint-3000/optimizer.pt
Normal file
3
checkpoint-3000/optimizer.pt
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:34d6d5e6b3baf98ae75f9122a60622751f2ca0f8eb5d5542a1fd00ed368520c2
|
||||
size 6882572207
|
||||
3
checkpoint-3000/rng_state.pth
Normal file
3
checkpoint-3000/rng_state.pth
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:61c19bab1174704a4a4441475683bf1270277af15d2e2c95e964789128e482c4
|
||||
size 14645
|
||||
3
checkpoint-3000/scheduler.pt
Normal file
3
checkpoint-3000/scheduler.pt
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:875525fc96d4b4bb5c12817e379ae11ed73b09ebd4f315c5475a43de7d437801
|
||||
size 1465
|
||||
3
checkpoint-3000/tokenizer.json
Normal file
3
checkpoint-3000/tokenizer.json
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:be75606093db2094d7cd20f3c2f385c212750648bd6ea4fb2bf507a6a4c55506
|
||||
size 11422650
|
||||
30
checkpoint-3000/tokenizer_config.json
Normal file
30
checkpoint-3000/tokenizer_config.json
Normal file
@@ -0,0 +1,30 @@
|
||||
{
|
||||
"add_prefix_space": false,
|
||||
"backend": "tokenizers",
|
||||
"bos_token": null,
|
||||
"clean_up_tokenization_spaces": false,
|
||||
"eos_token": "<|im_end|>",
|
||||
"errors": "replace",
|
||||
"extra_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|>"
|
||||
],
|
||||
"is_local": false,
|
||||
"local_files_only": false,
|
||||
"model_max_length": 131072,
|
||||
"pad_token": "<|endoftext|>",
|
||||
"split_special_tokens": false,
|
||||
"tokenizer_class": "Qwen2Tokenizer",
|
||||
"unk_token": null
|
||||
}
|
||||
634
checkpoint-3000/trainer_state.json
Normal file
634
checkpoint-3000/trainer_state.json
Normal file
@@ -0,0 +1,634 @@
|
||||
{
|
||||
"best_global_step": null,
|
||||
"best_metric": null,
|
||||
"best_model_checkpoint": null,
|
||||
"epoch": 0.48074996995312685,
|
||||
"eval_steps": 500,
|
||||
"global_step": 3000,
|
||||
"is_hyper_param_search": false,
|
||||
"is_local_process_zero": true,
|
||||
"is_world_process_zero": true,
|
||||
"log_history": [
|
||||
{
|
||||
"entropy": 2.223189059495926,
|
||||
"epoch": 0.008012499499218782,
|
||||
"grad_norm": 2.875,
|
||||
"learning_rate": 1.9842973882390643e-05,
|
||||
"loss": 2.3428109741210936,
|
||||
"mean_token_accuracy": 0.4963963569700718,
|
||||
"num_tokens": 305366.0,
|
||||
"step": 50
|
||||
},
|
||||
{
|
||||
"entropy": 2.188130117058754,
|
||||
"epoch": 0.016024998998437564,
|
||||
"grad_norm": 3.328125,
|
||||
"learning_rate": 1.96827431501362e-05,
|
||||
"loss": 2.225358734130859,
|
||||
"mean_token_accuracy": 0.5110921229422093,
|
||||
"num_tokens": 605042.0,
|
||||
"step": 100
|
||||
},
|
||||
{
|
||||
"entropy": 2.197347580194473,
|
||||
"epoch": 0.024037498497656344,
|
||||
"grad_norm": 3.453125,
|
||||
"learning_rate": 1.952251241788175e-05,
|
||||
"loss": 2.2299420166015627,
|
||||
"mean_token_accuracy": 0.5093490639328957,
|
||||
"num_tokens": 906888.0,
|
||||
"step": 150
|
||||
},
|
||||
{
|
||||
"entropy": 2.1913950395584108,
|
||||
"epoch": 0.03204999799687513,
|
||||
"grad_norm": 2.90625,
|
||||
"learning_rate": 1.9362281685627303e-05,
|
||||
"loss": 2.210995941162109,
|
||||
"mean_token_accuracy": 0.5097909501194954,
|
||||
"num_tokens": 1213125.0,
|
||||
"step": 200
|
||||
},
|
||||
{
|
||||
"entropy": 2.1611998677253723,
|
||||
"epoch": 0.040062497496093905,
|
||||
"grad_norm": 3.25,
|
||||
"learning_rate": 1.920205095337286e-05,
|
||||
"loss": 2.1867474365234374,
|
||||
"mean_token_accuracy": 0.5154410660266876,
|
||||
"num_tokens": 1509026.0,
|
||||
"step": 250
|
||||
},
|
||||
{
|
||||
"entropy": 2.168051454424858,
|
||||
"epoch": 0.04807499699531269,
|
||||
"grad_norm": 2.9375,
|
||||
"learning_rate": 1.904182022111841e-05,
|
||||
"loss": 2.190071716308594,
|
||||
"mean_token_accuracy": 0.5156884534657001,
|
||||
"num_tokens": 1806850.0,
|
||||
"step": 300
|
||||
},
|
||||
{
|
||||
"entropy": 2.1737154543399813,
|
||||
"epoch": 0.05608749649453147,
|
||||
"grad_norm": 3.421875,
|
||||
"learning_rate": 1.8881589488863967e-05,
|
||||
"loss": 2.199147186279297,
|
||||
"mean_token_accuracy": 0.514045044630766,
|
||||
"num_tokens": 2104505.0,
|
||||
"step": 350
|
||||
},
|
||||
{
|
||||
"entropy": 2.148039159178734,
|
||||
"epoch": 0.06409999599375026,
|
||||
"grad_norm": 3.109375,
|
||||
"learning_rate": 1.872135875660952e-05,
|
||||
"loss": 2.176946105957031,
|
||||
"mean_token_accuracy": 0.5177112428843975,
|
||||
"num_tokens": 2391088.0,
|
||||
"step": 400
|
||||
},
|
||||
{
|
||||
"entropy": 2.194890711903572,
|
||||
"epoch": 0.07211249549296903,
|
||||
"grad_norm": 3.28125,
|
||||
"learning_rate": 1.856112802435507e-05,
|
||||
"loss": 2.216288757324219,
|
||||
"mean_token_accuracy": 0.5110335703194141,
|
||||
"num_tokens": 2691272.0,
|
||||
"step": 450
|
||||
},
|
||||
{
|
||||
"entropy": 2.1557786524295808,
|
||||
"epoch": 0.08012499499218781,
|
||||
"grad_norm": 4.15625,
|
||||
"learning_rate": 1.8400897292100627e-05,
|
||||
"loss": 2.1812680053710936,
|
||||
"mean_token_accuracy": 0.5176697050035,
|
||||
"num_tokens": 2991855.0,
|
||||
"step": 500
|
||||
},
|
||||
{
|
||||
"entropy": 2.1708167713880537,
|
||||
"epoch": 0.08813749449140659,
|
||||
"grad_norm": 4.34375,
|
||||
"learning_rate": 1.824066655984618e-05,
|
||||
"loss": 2.1921002197265627,
|
||||
"mean_token_accuracy": 0.5156507430970669,
|
||||
"num_tokens": 3295049.0,
|
||||
"step": 550
|
||||
},
|
||||
{
|
||||
"entropy": 2.1686314886808393,
|
||||
"epoch": 0.09614999399062538,
|
||||
"grad_norm": 2.875,
|
||||
"learning_rate": 1.8080435827591735e-05,
|
||||
"loss": 2.1914617919921877,
|
||||
"mean_token_accuracy": 0.5139613692462445,
|
||||
"num_tokens": 3597206.0,
|
||||
"step": 600
|
||||
},
|
||||
{
|
||||
"entropy": 2.15742906332016,
|
||||
"epoch": 0.10416249348984416,
|
||||
"grad_norm": 3.109375,
|
||||
"learning_rate": 1.7920205095337287e-05,
|
||||
"loss": 2.174853973388672,
|
||||
"mean_token_accuracy": 0.5159518288075924,
|
||||
"num_tokens": 3905268.0,
|
||||
"step": 650
|
||||
},
|
||||
{
|
||||
"entropy": 2.162363375425339,
|
||||
"epoch": 0.11217499298906294,
|
||||
"grad_norm": 3.171875,
|
||||
"learning_rate": 1.7759974363082843e-05,
|
||||
"loss": 2.184231872558594,
|
||||
"mean_token_accuracy": 0.516631413847208,
|
||||
"num_tokens": 4207934.0,
|
||||
"step": 700
|
||||
},
|
||||
{
|
||||
"entropy": 2.1516289287805557,
|
||||
"epoch": 0.12018749248828171,
|
||||
"grad_norm": 3.609375,
|
||||
"learning_rate": 1.7599743630828395e-05,
|
||||
"loss": 2.167020263671875,
|
||||
"mean_token_accuracy": 0.516358617246151,
|
||||
"num_tokens": 4515774.0,
|
||||
"step": 750
|
||||
},
|
||||
{
|
||||
"entropy": 2.1549544781446457,
|
||||
"epoch": 0.1281999919875005,
|
||||
"grad_norm": 3.28125,
|
||||
"learning_rate": 1.7439512898573947e-05,
|
||||
"loss": 2.181917877197266,
|
||||
"mean_token_accuracy": 0.5182220858335495,
|
||||
"num_tokens": 4814709.0,
|
||||
"step": 800
|
||||
},
|
||||
{
|
||||
"entropy": 2.1532615756988527,
|
||||
"epoch": 0.1362124914867193,
|
||||
"grad_norm": 3.078125,
|
||||
"learning_rate": 1.7279282166319503e-05,
|
||||
"loss": 2.1659007263183594,
|
||||
"mean_token_accuracy": 0.5180425970256328,
|
||||
"num_tokens": 5112585.0,
|
||||
"step": 850
|
||||
},
|
||||
{
|
||||
"entropy": 2.1704017293453215,
|
||||
"epoch": 0.14422499098593805,
|
||||
"grad_norm": 2.921875,
|
||||
"learning_rate": 1.7119051434065055e-05,
|
||||
"loss": 2.1934996032714844,
|
||||
"mean_token_accuracy": 0.5154964691400528,
|
||||
"num_tokens": 5411369.0,
|
||||
"step": 900
|
||||
},
|
||||
{
|
||||
"entropy": 2.1349107217788696,
|
||||
"epoch": 0.15223749048515683,
|
||||
"grad_norm": 2.984375,
|
||||
"learning_rate": 1.6958820701810608e-05,
|
||||
"loss": 2.1598545837402345,
|
||||
"mean_token_accuracy": 0.5208920259773732,
|
||||
"num_tokens": 5707060.0,
|
||||
"step": 950
|
||||
},
|
||||
{
|
||||
"entropy": 2.1541705918312073,
|
||||
"epoch": 0.16024998998437562,
|
||||
"grad_norm": 3.203125,
|
||||
"learning_rate": 1.6798589969556163e-05,
|
||||
"loss": 2.170956573486328,
|
||||
"mean_token_accuracy": 0.5204938448965549,
|
||||
"num_tokens": 6004079.0,
|
||||
"step": 1000
|
||||
},
|
||||
{
|
||||
"entropy": 2.1523867017030716,
|
||||
"epoch": 0.1682624894835944,
|
||||
"grad_norm": 3.09375,
|
||||
"learning_rate": 1.6638359237301716e-05,
|
||||
"loss": 2.1695118713378907,
|
||||
"mean_token_accuracy": 0.5177376234531402,
|
||||
"num_tokens": 6315462.0,
|
||||
"step": 1050
|
||||
},
|
||||
{
|
||||
"entropy": 2.1356022965908052,
|
||||
"epoch": 0.17627498898281319,
|
||||
"grad_norm": 3.21875,
|
||||
"learning_rate": 1.6478128505047268e-05,
|
||||
"loss": 2.1513107299804686,
|
||||
"mean_token_accuracy": 0.5217180585861206,
|
||||
"num_tokens": 6613060.0,
|
||||
"step": 1100
|
||||
},
|
||||
{
|
||||
"entropy": 2.1483460652828215,
|
||||
"epoch": 0.18428748848203197,
|
||||
"grad_norm": 3.21875,
|
||||
"learning_rate": 1.6317897772792824e-05,
|
||||
"loss": 2.1671595764160156,
|
||||
"mean_token_accuracy": 0.5199222292006016,
|
||||
"num_tokens": 6909655.0,
|
||||
"step": 1150
|
||||
},
|
||||
{
|
||||
"entropy": 2.1074040442705155,
|
||||
"epoch": 0.19229998798125075,
|
||||
"grad_norm": 3.828125,
|
||||
"learning_rate": 1.6157667040538376e-05,
|
||||
"loss": 2.12770751953125,
|
||||
"mean_token_accuracy": 0.5252153894305229,
|
||||
"num_tokens": 7203263.0,
|
||||
"step": 1200
|
||||
},
|
||||
{
|
||||
"entropy": 2.1354027593135836,
|
||||
"epoch": 0.20031248748046954,
|
||||
"grad_norm": 2.921875,
|
||||
"learning_rate": 1.5997436308283932e-05,
|
||||
"loss": 2.156183166503906,
|
||||
"mean_token_accuracy": 0.518907299786806,
|
||||
"num_tokens": 7505411.0,
|
||||
"step": 1250
|
||||
},
|
||||
{
|
||||
"entropy": 2.142924894094467,
|
||||
"epoch": 0.20832498697968832,
|
||||
"grad_norm": 3.5625,
|
||||
"learning_rate": 1.5837205576029484e-05,
|
||||
"loss": 2.158486633300781,
|
||||
"mean_token_accuracy": 0.5205909128487111,
|
||||
"num_tokens": 7803562.0,
|
||||
"step": 1300
|
||||
},
|
||||
{
|
||||
"entropy": 2.1305291467905043,
|
||||
"epoch": 0.2163374864789071,
|
||||
"grad_norm": 3.34375,
|
||||
"learning_rate": 1.5676974843775036e-05,
|
||||
"loss": 2.151043701171875,
|
||||
"mean_token_accuracy": 0.5211617235839366,
|
||||
"num_tokens": 8108037.0,
|
||||
"step": 1350
|
||||
},
|
||||
{
|
||||
"entropy": 2.160947166085243,
|
||||
"epoch": 0.2243499859781259,
|
||||
"grad_norm": 3.234375,
|
||||
"learning_rate": 1.5516744111520592e-05,
|
||||
"loss": 2.177792663574219,
|
||||
"mean_token_accuracy": 0.5167004990577698,
|
||||
"num_tokens": 8408674.0,
|
||||
"step": 1400
|
||||
},
|
||||
{
|
||||
"entropy": 2.120403967499733,
|
||||
"epoch": 0.23236248547734467,
|
||||
"grad_norm": 3.375,
|
||||
"learning_rate": 1.5356513379266144e-05,
|
||||
"loss": 2.1462562561035154,
|
||||
"mean_token_accuracy": 0.5225985661149025,
|
||||
"num_tokens": 8707650.0,
|
||||
"step": 1450
|
||||
},
|
||||
{
|
||||
"entropy": 2.138882914185524,
|
||||
"epoch": 0.24037498497656343,
|
||||
"grad_norm": 3.171875,
|
||||
"learning_rate": 1.5196282647011698e-05,
|
||||
"loss": 2.1499905395507812,
|
||||
"mean_token_accuracy": 0.5204668046534061,
|
||||
"num_tokens": 9006296.0,
|
||||
"step": 1500
|
||||
},
|
||||
{
|
||||
"entropy": 2.131927245259285,
|
||||
"epoch": 0.2483874844757822,
|
||||
"grad_norm": 3.546875,
|
||||
"learning_rate": 1.5036051914757252e-05,
|
||||
"loss": 2.157076416015625,
|
||||
"mean_token_accuracy": 0.5220408822596073,
|
||||
"num_tokens": 9307203.0,
|
||||
"step": 1550
|
||||
},
|
||||
{
|
||||
"entropy": 2.1308567851781843,
|
||||
"epoch": 0.256399983975001,
|
||||
"grad_norm": 3.546875,
|
||||
"learning_rate": 1.4875821182502805e-05,
|
||||
"loss": 2.146601104736328,
|
||||
"mean_token_accuracy": 0.5219539681077003,
|
||||
"num_tokens": 9610070.0,
|
||||
"step": 1600
|
||||
},
|
||||
{
|
||||
"entropy": 2.1097783839702604,
|
||||
"epoch": 0.2644124834742198,
|
||||
"grad_norm": 3.84375,
|
||||
"learning_rate": 1.4715590450248359e-05,
|
||||
"loss": 2.1268209838867187,
|
||||
"mean_token_accuracy": 0.5236652019619942,
|
||||
"num_tokens": 9909117.0,
|
||||
"step": 1650
|
||||
},
|
||||
{
|
||||
"entropy": 2.123499180674553,
|
||||
"epoch": 0.2724249829734386,
|
||||
"grad_norm": 3.46875,
|
||||
"learning_rate": 1.4555359717993913e-05,
|
||||
"loss": 2.1409934997558593,
|
||||
"mean_token_accuracy": 0.5225465354323388,
|
||||
"num_tokens": 10209461.0,
|
||||
"step": 1700
|
||||
},
|
||||
{
|
||||
"entropy": 2.13052960395813,
|
||||
"epoch": 0.2804374824726573,
|
||||
"grad_norm": 2.71875,
|
||||
"learning_rate": 1.4395128985739467e-05,
|
||||
"loss": 2.151373291015625,
|
||||
"mean_token_accuracy": 0.5227928113937378,
|
||||
"num_tokens": 10509282.0,
|
||||
"step": 1750
|
||||
},
|
||||
{
|
||||
"entropy": 2.1226114094257356,
|
||||
"epoch": 0.2884499819718761,
|
||||
"grad_norm": 3.65625,
|
||||
"learning_rate": 1.4234898253485019e-05,
|
||||
"loss": 2.144451141357422,
|
||||
"mean_token_accuracy": 0.5233827362954616,
|
||||
"num_tokens": 10811056.0,
|
||||
"step": 1800
|
||||
},
|
||||
{
|
||||
"entropy": 2.126143154501915,
|
||||
"epoch": 0.2964624814710949,
|
||||
"grad_norm": 3.359375,
|
||||
"learning_rate": 1.4074667521230575e-05,
|
||||
"loss": 2.1457598876953123,
|
||||
"mean_token_accuracy": 0.5221967925131321,
|
||||
"num_tokens": 11111153.0,
|
||||
"step": 1850
|
||||
},
|
||||
{
|
||||
"entropy": 2.1048326581716537,
|
||||
"epoch": 0.30447498097031367,
|
||||
"grad_norm": 3.59375,
|
||||
"learning_rate": 1.3914436788976127e-05,
|
||||
"loss": 2.1260116577148436,
|
||||
"mean_token_accuracy": 0.5272719909250736,
|
||||
"num_tokens": 11405606.0,
|
||||
"step": 1900
|
||||
},
|
||||
{
|
||||
"entropy": 2.1236617982387545,
|
||||
"epoch": 0.31248748046953245,
|
||||
"grad_norm": 3.0,
|
||||
"learning_rate": 1.375420605672168e-05,
|
||||
"loss": 2.151782684326172,
|
||||
"mean_token_accuracy": 0.5227546173334122,
|
||||
"num_tokens": 11703943.0,
|
||||
"step": 1950
|
||||
},
|
||||
{
|
||||
"entropy": 2.121563429236412,
|
||||
"epoch": 0.32049997996875124,
|
||||
"grad_norm": 3.34375,
|
||||
"learning_rate": 1.3593975324467235e-05,
|
||||
"loss": 2.1370144653320313,
|
||||
"mean_token_accuracy": 0.5235123547911644,
|
||||
"num_tokens": 12002522.0,
|
||||
"step": 2000
|
||||
},
|
||||
{
|
||||
"entropy": 2.1155461168289182,
|
||||
"epoch": 0.32851247946797,
|
||||
"grad_norm": 3.46875,
|
||||
"learning_rate": 1.3433744592212787e-05,
|
||||
"loss": 2.1364064025878906,
|
||||
"mean_token_accuracy": 0.5237135806679726,
|
||||
"num_tokens": 12303098.0,
|
||||
"step": 2050
|
||||
},
|
||||
{
|
||||
"entropy": 2.1386052322387696,
|
||||
"epoch": 0.3365249789671888,
|
||||
"grad_norm": 3.484375,
|
||||
"learning_rate": 1.327351385995834e-05,
|
||||
"loss": 2.149888458251953,
|
||||
"mean_token_accuracy": 0.522669050693512,
|
||||
"num_tokens": 12605534.0,
|
||||
"step": 2100
|
||||
},
|
||||
{
|
||||
"entropy": 2.102600795030594,
|
||||
"epoch": 0.3445374784664076,
|
||||
"grad_norm": 3.796875,
|
||||
"learning_rate": 1.3113283127703895e-05,
|
||||
"loss": 2.1236566162109374,
|
||||
"mean_token_accuracy": 0.5279092155396938,
|
||||
"num_tokens": 12901218.0,
|
||||
"step": 2150
|
||||
},
|
||||
{
|
||||
"entropy": 2.132645187973976,
|
||||
"epoch": 0.35254997796562637,
|
||||
"grad_norm": 3.53125,
|
||||
"learning_rate": 1.2953052395449447e-05,
|
||||
"loss": 2.1523503112792968,
|
||||
"mean_token_accuracy": 0.5198129612207413,
|
||||
"num_tokens": 13213018.0,
|
||||
"step": 2200
|
||||
},
|
||||
{
|
||||
"entropy": 2.1204321217536926,
|
||||
"epoch": 0.36056247746484515,
|
||||
"grad_norm": 3.25,
|
||||
"learning_rate": 1.2792821663195001e-05,
|
||||
"loss": 2.1325447082519533,
|
||||
"mean_token_accuracy": 0.5243636940419674,
|
||||
"num_tokens": 13514408.0,
|
||||
"step": 2250
|
||||
},
|
||||
{
|
||||
"entropy": 2.102410731315613,
|
||||
"epoch": 0.36857497696406394,
|
||||
"grad_norm": 3.1875,
|
||||
"learning_rate": 1.2632590930940555e-05,
|
||||
"loss": 2.123565368652344,
|
||||
"mean_token_accuracy": 0.5257614709436893,
|
||||
"num_tokens": 13818272.0,
|
||||
"step": 2300
|
||||
},
|
||||
{
|
||||
"entropy": 2.0876205080747603,
|
||||
"epoch": 0.3765874764632827,
|
||||
"grad_norm": 3.328125,
|
||||
"learning_rate": 1.247236019868611e-05,
|
||||
"loss": 2.116696472167969,
|
||||
"mean_token_accuracy": 0.5275124031305313,
|
||||
"num_tokens": 14110702.0,
|
||||
"step": 2350
|
||||
},
|
||||
{
|
||||
"entropy": 2.1078518223762512,
|
||||
"epoch": 0.3845999759625015,
|
||||
"grad_norm": 3.453125,
|
||||
"learning_rate": 1.2312129466431662e-05,
|
||||
"loss": 2.1291993713378905,
|
||||
"mean_token_accuracy": 0.524967464953661,
|
||||
"num_tokens": 14407015.0,
|
||||
"step": 2400
|
||||
},
|
||||
{
|
||||
"entropy": 2.0930898267030718,
|
||||
"epoch": 0.3926124754617203,
|
||||
"grad_norm": 3.5,
|
||||
"learning_rate": 1.2151898734177216e-05,
|
||||
"loss": 2.1094123840332033,
|
||||
"mean_token_accuracy": 0.5290587748587131,
|
||||
"num_tokens": 14696565.0,
|
||||
"step": 2450
|
||||
},
|
||||
{
|
||||
"entropy": 2.1109757220745085,
|
||||
"epoch": 0.4006249749609391,
|
||||
"grad_norm": 3.5,
|
||||
"learning_rate": 1.199166800192277e-05,
|
||||
"loss": 2.1246888732910154,
|
||||
"mean_token_accuracy": 0.5258957357704639,
|
||||
"num_tokens": 14994401.0,
|
||||
"step": 2500
|
||||
},
|
||||
{
|
||||
"entropy": 2.1087811160087586,
|
||||
"epoch": 0.40863747446015786,
|
||||
"grad_norm": 3.1875,
|
||||
"learning_rate": 1.1831437269668324e-05,
|
||||
"loss": 2.12890625,
|
||||
"mean_token_accuracy": 0.5269917643070221,
|
||||
"num_tokens": 15292837.0,
|
||||
"step": 2550
|
||||
},
|
||||
{
|
||||
"entropy": 2.097252429127693,
|
||||
"epoch": 0.41664997395937664,
|
||||
"grad_norm": 3.0,
|
||||
"learning_rate": 1.1671206537413878e-05,
|
||||
"loss": 2.1145870971679686,
|
||||
"mean_token_accuracy": 0.5274137698113919,
|
||||
"num_tokens": 15590647.0,
|
||||
"step": 2600
|
||||
},
|
||||
{
|
||||
"entropy": 2.111199595332146,
|
||||
"epoch": 0.4246624734585954,
|
||||
"grad_norm": 3.25,
|
||||
"learning_rate": 1.151097580515943e-05,
|
||||
"loss": 2.1274327087402343,
|
||||
"mean_token_accuracy": 0.5250249288976192,
|
||||
"num_tokens": 15892772.0,
|
||||
"step": 2650
|
||||
},
|
||||
{
|
||||
"entropy": 2.1051502913236617,
|
||||
"epoch": 0.4326749729578142,
|
||||
"grad_norm": 3.203125,
|
||||
"learning_rate": 1.1350745072904986e-05,
|
||||
"loss": 2.127935333251953,
|
||||
"mean_token_accuracy": 0.5263640700280666,
|
||||
"num_tokens": 16191611.0,
|
||||
"step": 2700
|
||||
},
|
||||
{
|
||||
"entropy": 2.1127885603904724,
|
||||
"epoch": 0.440687472457033,
|
||||
"grad_norm": 3.515625,
|
||||
"learning_rate": 1.1190514340650538e-05,
|
||||
"loss": 2.129211883544922,
|
||||
"mean_token_accuracy": 0.5250939278304577,
|
||||
"num_tokens": 16483582.0,
|
||||
"step": 2750
|
||||
},
|
||||
{
|
||||
"entropy": 2.110930626988411,
|
||||
"epoch": 0.4486999719562518,
|
||||
"grad_norm": 3.1875,
|
||||
"learning_rate": 1.103028360839609e-05,
|
||||
"loss": 2.1220079040527344,
|
||||
"mean_token_accuracy": 0.5250919266045093,
|
||||
"num_tokens": 16787253.0,
|
||||
"step": 2800
|
||||
},
|
||||
{
|
||||
"entropy": 2.1182780677080153,
|
||||
"epoch": 0.45671247145547056,
|
||||
"grad_norm": 3.8125,
|
||||
"learning_rate": 1.0870052876141646e-05,
|
||||
"loss": 2.1358837890625,
|
||||
"mean_token_accuracy": 0.5217854312062263,
|
||||
"num_tokens": 17088741.0,
|
||||
"step": 2850
|
||||
},
|
||||
{
|
||||
"entropy": 2.1117833071947096,
|
||||
"epoch": 0.46472497095468934,
|
||||
"grad_norm": 3.171875,
|
||||
"learning_rate": 1.0709822143887198e-05,
|
||||
"loss": 2.125662841796875,
|
||||
"mean_token_accuracy": 0.524649152457714,
|
||||
"num_tokens": 17389236.0,
|
||||
"step": 2900
|
||||
},
|
||||
{
|
||||
"entropy": 2.1262730568647386,
|
||||
"epoch": 0.47273747045390807,
|
||||
"grad_norm": 3.625,
|
||||
"learning_rate": 1.054959141163275e-05,
|
||||
"loss": 2.142789611816406,
|
||||
"mean_token_accuracy": 0.5236384868621826,
|
||||
"num_tokens": 17693413.0,
|
||||
"step": 2950
|
||||
},
|
||||
{
|
||||
"entropy": 2.115410739183426,
|
||||
"epoch": 0.48074996995312685,
|
||||
"grad_norm": 3.140625,
|
||||
"learning_rate": 1.0389360679378306e-05,
|
||||
"loss": 2.1337106323242185,
|
||||
"mean_token_accuracy": 0.5249222332239151,
|
||||
"num_tokens": 17990326.0,
|
||||
"step": 3000
|
||||
}
|
||||
],
|
||||
"logging_steps": 50,
|
||||
"max_steps": 6241,
|
||||
"num_input_tokens_seen": 0,
|
||||
"num_train_epochs": 1,
|
||||
"save_steps": 500,
|
||||
"stateful_callbacks": {
|
||||
"TrainerControl": {
|
||||
"args": {
|
||||
"should_epoch_stop": false,
|
||||
"should_evaluate": false,
|
||||
"should_log": false,
|
||||
"should_save": true,
|
||||
"should_training_stop": false
|
||||
},
|
||||
"attributes": {}
|
||||
}
|
||||
},
|
||||
"total_flos": 2.126267005433856e+17,
|
||||
"train_batch_size": 4,
|
||||
"trial_name": null,
|
||||
"trial_params": null
|
||||
}
|
||||
3
checkpoint-3000/training_args.bin
Normal file
3
checkpoint-3000/training_args.bin
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:78622f63304ec0913ed0b4ea7a1b26c08a5621b3661a380d798c4ea426235671
|
||||
size 5713
|
||||
89
checkpoint-3500/chat_template.jinja
Normal file
89
checkpoint-3500/chat_template.jinja
Normal file
@@ -0,0 +1,89 @@
|
||||
{%- if tools %}
|
||||
{{- '<|im_start|>system\n' }}
|
||||
{%- if messages[0].role == 'system' %}
|
||||
{{- messages[0].content + '\n\n' }}
|
||||
{%- endif %}
|
||||
{{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
|
||||
{%- for tool in tools %}
|
||||
{{- "\n" }}
|
||||
{{- tool | tojson }}
|
||||
{%- endfor %}
|
||||
{{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
|
||||
{%- else %}
|
||||
{%- if messages[0].role == 'system' %}
|
||||
{{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
|
||||
{%- endif %}
|
||||
{%- endif %}
|
||||
{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
|
||||
{%- for message in messages[::-1] %}
|
||||
{%- set index = (messages|length - 1) - loop.index0 %}
|
||||
{%- if ns.multi_step_tool and message.role == "user" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
|
||||
{%- set ns.multi_step_tool = false %}
|
||||
{%- set ns.last_query_index = index %}
|
||||
{%- endif %}
|
||||
{%- endfor %}
|
||||
{%- for message in messages %}
|
||||
{%- if message.content is string %}
|
||||
{%- set content = message.content %}
|
||||
{%- else %}
|
||||
{%- set content = '' %}
|
||||
{%- endif %}
|
||||
{%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
|
||||
{{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
|
||||
{%- elif message.role == "assistant" %}
|
||||
{%- set reasoning_content = '' %}
|
||||
{%- if message.reasoning_content is string %}
|
||||
{%- set reasoning_content = message.reasoning_content %}
|
||||
{%- else %}
|
||||
{%- if '</think>' in content %}
|
||||
{%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
|
||||
{%- set content = content.split('</think>')[-1].lstrip('\n') %}
|
||||
{%- endif %}
|
||||
{%- endif %}
|
||||
{%- if loop.index0 > ns.last_query_index %}
|
||||
{%- if loop.last or (not loop.last and reasoning_content) %}
|
||||
{{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
|
||||
{%- else %}
|
||||
{{- '<|im_start|>' + message.role + '\n' + content }}
|
||||
{%- endif %}
|
||||
{%- else %}
|
||||
{{- '<|im_start|>' + message.role + '\n' + content }}
|
||||
{%- endif %}
|
||||
{%- if message.tool_calls %}
|
||||
{%- for tool_call in message.tool_calls %}
|
||||
{%- if (loop.first and content) or (not loop.first) %}
|
||||
{{- '\n' }}
|
||||
{%- endif %}
|
||||
{%- if tool_call.function %}
|
||||
{%- set tool_call = tool_call.function %}
|
||||
{%- endif %}
|
||||
{{- '<tool_call>\n{"name": "' }}
|
||||
{{- tool_call.name }}
|
||||
{{- '", "arguments": ' }}
|
||||
{%- if tool_call.arguments is string %}
|
||||
{{- tool_call.arguments }}
|
||||
{%- else %}
|
||||
{{- tool_call.arguments | tojson }}
|
||||
{%- endif %}
|
||||
{{- '}\n</tool_call>' }}
|
||||
{%- endfor %}
|
||||
{%- endif %}
|
||||
{{- '<|im_end|>\n' }}
|
||||
{%- elif message.role == "tool" %}
|
||||
{%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
|
||||
{{- '<|im_start|>user' }}
|
||||
{%- endif %}
|
||||
{{- '\n<tool_response>\n' }}
|
||||
{{- content }}
|
||||
{{- '\n</tool_response>' }}
|
||||
{%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
|
||||
{{- '<|im_end|>\n' }}
|
||||
{%- endif %}
|
||||
{%- endif %}
|
||||
{%- endfor %}
|
||||
{%- if add_generation_prompt %}
|
||||
{{- '<|im_start|>assistant\n' }}
|
||||
{%- if enable_thinking is defined and enable_thinking is false %}
|
||||
{{- '<think>\n\n</think>\n\n' }}
|
||||
{%- endif %}
|
||||
{%- endif %}
|
||||
63
checkpoint-3500/config.json
Normal file
63
checkpoint-3500/config.json
Normal file
@@ -0,0 +1,63 @@
|
||||
{
|
||||
"architectures": [
|
||||
"Qwen3ForCausalLM"
|
||||
],
|
||||
"attention_bias": false,
|
||||
"attention_dropout": 0.0,
|
||||
"bos_token_id": null,
|
||||
"dtype": "bfloat16",
|
||||
"eos_token_id": 151645,
|
||||
"head_dim": 128,
|
||||
"hidden_act": "silu",
|
||||
"hidden_size": 2048,
|
||||
"initializer_range": 0.02,
|
||||
"intermediate_size": 6144,
|
||||
"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"
|
||||
],
|
||||
"max_position_embeddings": 40960,
|
||||
"max_window_layers": 28,
|
||||
"model_type": "qwen3",
|
||||
"num_attention_heads": 16,
|
||||
"num_hidden_layers": 28,
|
||||
"num_key_value_heads": 8,
|
||||
"pad_token_id": 151643,
|
||||
"rms_norm_eps": 1e-06,
|
||||
"rope_parameters": {
|
||||
"rope_theta": 1000000,
|
||||
"rope_type": "default"
|
||||
},
|
||||
"sliding_window": null,
|
||||
"tie_word_embeddings": true,
|
||||
"transformers_version": "5.7.0",
|
||||
"use_cache": false,
|
||||
"use_sliding_window": false,
|
||||
"vocab_size": 151936
|
||||
}
|
||||
12
checkpoint-3500/generation_config.json
Normal file
12
checkpoint-3500/generation_config.json
Normal file
@@ -0,0 +1,12 @@
|
||||
{
|
||||
"do_sample": true,
|
||||
"eos_token_id": [
|
||||
151645,
|
||||
151643
|
||||
],
|
||||
"pad_token_id": 151643,
|
||||
"temperature": 0.6,
|
||||
"top_k": 20,
|
||||
"top_p": 0.95,
|
||||
"transformers_version": "5.7.0"
|
||||
}
|
||||
3
checkpoint-3500/model.safetensors
Normal file
3
checkpoint-3500/model.safetensors
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:cd3ed08012c528778e23bfc40ce9263d1183db5fe1b344fcc375190ceff69329
|
||||
size 3441185608
|
||||
3
checkpoint-3500/optimizer.pt
Normal file
3
checkpoint-3500/optimizer.pt
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:ff2d89dd80a7f07fe151122ae98c1d25ce93198c70301c9848ea4314b9f22671
|
||||
size 6882572207
|
||||
3
checkpoint-3500/rng_state.pth
Normal file
3
checkpoint-3500/rng_state.pth
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:61c19bab1174704a4a4441475683bf1270277af15d2e2c95e964789128e482c4
|
||||
size 14645
|
||||
3
checkpoint-3500/scheduler.pt
Normal file
3
checkpoint-3500/scheduler.pt
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:5e847bb7d94a2f7987ac5f9be2becb727ffced83ae79de46ff613b03a251110e
|
||||
size 1465
|
||||
3
checkpoint-3500/tokenizer.json
Normal file
3
checkpoint-3500/tokenizer.json
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:be75606093db2094d7cd20f3c2f385c212750648bd6ea4fb2bf507a6a4c55506
|
||||
size 11422650
|
||||
30
checkpoint-3500/tokenizer_config.json
Normal file
30
checkpoint-3500/tokenizer_config.json
Normal file
@@ -0,0 +1,30 @@
|
||||
{
|
||||
"add_prefix_space": false,
|
||||
"backend": "tokenizers",
|
||||
"bos_token": null,
|
||||
"clean_up_tokenization_spaces": false,
|
||||
"eos_token": "<|im_end|>",
|
||||
"errors": "replace",
|
||||
"extra_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|>"
|
||||
],
|
||||
"is_local": false,
|
||||
"local_files_only": false,
|
||||
"model_max_length": 131072,
|
||||
"pad_token": "<|endoftext|>",
|
||||
"split_special_tokens": false,
|
||||
"tokenizer_class": "Qwen2Tokenizer",
|
||||
"unk_token": null
|
||||
}
|
||||
734
checkpoint-3500/trainer_state.json
Normal file
734
checkpoint-3500/trainer_state.json
Normal file
@@ -0,0 +1,734 @@
|
||||
{
|
||||
"best_global_step": null,
|
||||
"best_metric": null,
|
||||
"best_model_checkpoint": null,
|
||||
"epoch": 0.5608749649453146,
|
||||
"eval_steps": 500,
|
||||
"global_step": 3500,
|
||||
"is_hyper_param_search": false,
|
||||
"is_local_process_zero": true,
|
||||
"is_world_process_zero": true,
|
||||
"log_history": [
|
||||
{
|
||||
"entropy": 2.223189059495926,
|
||||
"epoch": 0.008012499499218782,
|
||||
"grad_norm": 2.875,
|
||||
"learning_rate": 1.9842973882390643e-05,
|
||||
"loss": 2.3428109741210936,
|
||||
"mean_token_accuracy": 0.4963963569700718,
|
||||
"num_tokens": 305366.0,
|
||||
"step": 50
|
||||
},
|
||||
{
|
||||
"entropy": 2.188130117058754,
|
||||
"epoch": 0.016024998998437564,
|
||||
"grad_norm": 3.328125,
|
||||
"learning_rate": 1.96827431501362e-05,
|
||||
"loss": 2.225358734130859,
|
||||
"mean_token_accuracy": 0.5110921229422093,
|
||||
"num_tokens": 605042.0,
|
||||
"step": 100
|
||||
},
|
||||
{
|
||||
"entropy": 2.197347580194473,
|
||||
"epoch": 0.024037498497656344,
|
||||
"grad_norm": 3.453125,
|
||||
"learning_rate": 1.952251241788175e-05,
|
||||
"loss": 2.2299420166015627,
|
||||
"mean_token_accuracy": 0.5093490639328957,
|
||||
"num_tokens": 906888.0,
|
||||
"step": 150
|
||||
},
|
||||
{
|
||||
"entropy": 2.1913950395584108,
|
||||
"epoch": 0.03204999799687513,
|
||||
"grad_norm": 2.90625,
|
||||
"learning_rate": 1.9362281685627303e-05,
|
||||
"loss": 2.210995941162109,
|
||||
"mean_token_accuracy": 0.5097909501194954,
|
||||
"num_tokens": 1213125.0,
|
||||
"step": 200
|
||||
},
|
||||
{
|
||||
"entropy": 2.1611998677253723,
|
||||
"epoch": 0.040062497496093905,
|
||||
"grad_norm": 3.25,
|
||||
"learning_rate": 1.920205095337286e-05,
|
||||
"loss": 2.1867474365234374,
|
||||
"mean_token_accuracy": 0.5154410660266876,
|
||||
"num_tokens": 1509026.0,
|
||||
"step": 250
|
||||
},
|
||||
{
|
||||
"entropy": 2.168051454424858,
|
||||
"epoch": 0.04807499699531269,
|
||||
"grad_norm": 2.9375,
|
||||
"learning_rate": 1.904182022111841e-05,
|
||||
"loss": 2.190071716308594,
|
||||
"mean_token_accuracy": 0.5156884534657001,
|
||||
"num_tokens": 1806850.0,
|
||||
"step": 300
|
||||
},
|
||||
{
|
||||
"entropy": 2.1737154543399813,
|
||||
"epoch": 0.05608749649453147,
|
||||
"grad_norm": 3.421875,
|
||||
"learning_rate": 1.8881589488863967e-05,
|
||||
"loss": 2.199147186279297,
|
||||
"mean_token_accuracy": 0.514045044630766,
|
||||
"num_tokens": 2104505.0,
|
||||
"step": 350
|
||||
},
|
||||
{
|
||||
"entropy": 2.148039159178734,
|
||||
"epoch": 0.06409999599375026,
|
||||
"grad_norm": 3.109375,
|
||||
"learning_rate": 1.872135875660952e-05,
|
||||
"loss": 2.176946105957031,
|
||||
"mean_token_accuracy": 0.5177112428843975,
|
||||
"num_tokens": 2391088.0,
|
||||
"step": 400
|
||||
},
|
||||
{
|
||||
"entropy": 2.194890711903572,
|
||||
"epoch": 0.07211249549296903,
|
||||
"grad_norm": 3.28125,
|
||||
"learning_rate": 1.856112802435507e-05,
|
||||
"loss": 2.216288757324219,
|
||||
"mean_token_accuracy": 0.5110335703194141,
|
||||
"num_tokens": 2691272.0,
|
||||
"step": 450
|
||||
},
|
||||
{
|
||||
"entropy": 2.1557786524295808,
|
||||
"epoch": 0.08012499499218781,
|
||||
"grad_norm": 4.15625,
|
||||
"learning_rate": 1.8400897292100627e-05,
|
||||
"loss": 2.1812680053710936,
|
||||
"mean_token_accuracy": 0.5176697050035,
|
||||
"num_tokens": 2991855.0,
|
||||
"step": 500
|
||||
},
|
||||
{
|
||||
"entropy": 2.1708167713880537,
|
||||
"epoch": 0.08813749449140659,
|
||||
"grad_norm": 4.34375,
|
||||
"learning_rate": 1.824066655984618e-05,
|
||||
"loss": 2.1921002197265627,
|
||||
"mean_token_accuracy": 0.5156507430970669,
|
||||
"num_tokens": 3295049.0,
|
||||
"step": 550
|
||||
},
|
||||
{
|
||||
"entropy": 2.1686314886808393,
|
||||
"epoch": 0.09614999399062538,
|
||||
"grad_norm": 2.875,
|
||||
"learning_rate": 1.8080435827591735e-05,
|
||||
"loss": 2.1914617919921877,
|
||||
"mean_token_accuracy": 0.5139613692462445,
|
||||
"num_tokens": 3597206.0,
|
||||
"step": 600
|
||||
},
|
||||
{
|
||||
"entropy": 2.15742906332016,
|
||||
"epoch": 0.10416249348984416,
|
||||
"grad_norm": 3.109375,
|
||||
"learning_rate": 1.7920205095337287e-05,
|
||||
"loss": 2.174853973388672,
|
||||
"mean_token_accuracy": 0.5159518288075924,
|
||||
"num_tokens": 3905268.0,
|
||||
"step": 650
|
||||
},
|
||||
{
|
||||
"entropy": 2.162363375425339,
|
||||
"epoch": 0.11217499298906294,
|
||||
"grad_norm": 3.171875,
|
||||
"learning_rate": 1.7759974363082843e-05,
|
||||
"loss": 2.184231872558594,
|
||||
"mean_token_accuracy": 0.516631413847208,
|
||||
"num_tokens": 4207934.0,
|
||||
"step": 700
|
||||
},
|
||||
{
|
||||
"entropy": 2.1516289287805557,
|
||||
"epoch": 0.12018749248828171,
|
||||
"grad_norm": 3.609375,
|
||||
"learning_rate": 1.7599743630828395e-05,
|
||||
"loss": 2.167020263671875,
|
||||
"mean_token_accuracy": 0.516358617246151,
|
||||
"num_tokens": 4515774.0,
|
||||
"step": 750
|
||||
},
|
||||
{
|
||||
"entropy": 2.1549544781446457,
|
||||
"epoch": 0.1281999919875005,
|
||||
"grad_norm": 3.28125,
|
||||
"learning_rate": 1.7439512898573947e-05,
|
||||
"loss": 2.181917877197266,
|
||||
"mean_token_accuracy": 0.5182220858335495,
|
||||
"num_tokens": 4814709.0,
|
||||
"step": 800
|
||||
},
|
||||
{
|
||||
"entropy": 2.1532615756988527,
|
||||
"epoch": 0.1362124914867193,
|
||||
"grad_norm": 3.078125,
|
||||
"learning_rate": 1.7279282166319503e-05,
|
||||
"loss": 2.1659007263183594,
|
||||
"mean_token_accuracy": 0.5180425970256328,
|
||||
"num_tokens": 5112585.0,
|
||||
"step": 850
|
||||
},
|
||||
{
|
||||
"entropy": 2.1704017293453215,
|
||||
"epoch": 0.14422499098593805,
|
||||
"grad_norm": 2.921875,
|
||||
"learning_rate": 1.7119051434065055e-05,
|
||||
"loss": 2.1934996032714844,
|
||||
"mean_token_accuracy": 0.5154964691400528,
|
||||
"num_tokens": 5411369.0,
|
||||
"step": 900
|
||||
},
|
||||
{
|
||||
"entropy": 2.1349107217788696,
|
||||
"epoch": 0.15223749048515683,
|
||||
"grad_norm": 2.984375,
|
||||
"learning_rate": 1.6958820701810608e-05,
|
||||
"loss": 2.1598545837402345,
|
||||
"mean_token_accuracy": 0.5208920259773732,
|
||||
"num_tokens": 5707060.0,
|
||||
"step": 950
|
||||
},
|
||||
{
|
||||
"entropy": 2.1541705918312073,
|
||||
"epoch": 0.16024998998437562,
|
||||
"grad_norm": 3.203125,
|
||||
"learning_rate": 1.6798589969556163e-05,
|
||||
"loss": 2.170956573486328,
|
||||
"mean_token_accuracy": 0.5204938448965549,
|
||||
"num_tokens": 6004079.0,
|
||||
"step": 1000
|
||||
},
|
||||
{
|
||||
"entropy": 2.1523867017030716,
|
||||
"epoch": 0.1682624894835944,
|
||||
"grad_norm": 3.09375,
|
||||
"learning_rate": 1.6638359237301716e-05,
|
||||
"loss": 2.1695118713378907,
|
||||
"mean_token_accuracy": 0.5177376234531402,
|
||||
"num_tokens": 6315462.0,
|
||||
"step": 1050
|
||||
},
|
||||
{
|
||||
"entropy": 2.1356022965908052,
|
||||
"epoch": 0.17627498898281319,
|
||||
"grad_norm": 3.21875,
|
||||
"learning_rate": 1.6478128505047268e-05,
|
||||
"loss": 2.1513107299804686,
|
||||
"mean_token_accuracy": 0.5217180585861206,
|
||||
"num_tokens": 6613060.0,
|
||||
"step": 1100
|
||||
},
|
||||
{
|
||||
"entropy": 2.1483460652828215,
|
||||
"epoch": 0.18428748848203197,
|
||||
"grad_norm": 3.21875,
|
||||
"learning_rate": 1.6317897772792824e-05,
|
||||
"loss": 2.1671595764160156,
|
||||
"mean_token_accuracy": 0.5199222292006016,
|
||||
"num_tokens": 6909655.0,
|
||||
"step": 1150
|
||||
},
|
||||
{
|
||||
"entropy": 2.1074040442705155,
|
||||
"epoch": 0.19229998798125075,
|
||||
"grad_norm": 3.828125,
|
||||
"learning_rate": 1.6157667040538376e-05,
|
||||
"loss": 2.12770751953125,
|
||||
"mean_token_accuracy": 0.5252153894305229,
|
||||
"num_tokens": 7203263.0,
|
||||
"step": 1200
|
||||
},
|
||||
{
|
||||
"entropy": 2.1354027593135836,
|
||||
"epoch": 0.20031248748046954,
|
||||
"grad_norm": 2.921875,
|
||||
"learning_rate": 1.5997436308283932e-05,
|
||||
"loss": 2.156183166503906,
|
||||
"mean_token_accuracy": 0.518907299786806,
|
||||
"num_tokens": 7505411.0,
|
||||
"step": 1250
|
||||
},
|
||||
{
|
||||
"entropy": 2.142924894094467,
|
||||
"epoch": 0.20832498697968832,
|
||||
"grad_norm": 3.5625,
|
||||
"learning_rate": 1.5837205576029484e-05,
|
||||
"loss": 2.158486633300781,
|
||||
"mean_token_accuracy": 0.5205909128487111,
|
||||
"num_tokens": 7803562.0,
|
||||
"step": 1300
|
||||
},
|
||||
{
|
||||
"entropy": 2.1305291467905043,
|
||||
"epoch": 0.2163374864789071,
|
||||
"grad_norm": 3.34375,
|
||||
"learning_rate": 1.5676974843775036e-05,
|
||||
"loss": 2.151043701171875,
|
||||
"mean_token_accuracy": 0.5211617235839366,
|
||||
"num_tokens": 8108037.0,
|
||||
"step": 1350
|
||||
},
|
||||
{
|
||||
"entropy": 2.160947166085243,
|
||||
"epoch": 0.2243499859781259,
|
||||
"grad_norm": 3.234375,
|
||||
"learning_rate": 1.5516744111520592e-05,
|
||||
"loss": 2.177792663574219,
|
||||
"mean_token_accuracy": 0.5167004990577698,
|
||||
"num_tokens": 8408674.0,
|
||||
"step": 1400
|
||||
},
|
||||
{
|
||||
"entropy": 2.120403967499733,
|
||||
"epoch": 0.23236248547734467,
|
||||
"grad_norm": 3.375,
|
||||
"learning_rate": 1.5356513379266144e-05,
|
||||
"loss": 2.1462562561035154,
|
||||
"mean_token_accuracy": 0.5225985661149025,
|
||||
"num_tokens": 8707650.0,
|
||||
"step": 1450
|
||||
},
|
||||
{
|
||||
"entropy": 2.138882914185524,
|
||||
"epoch": 0.24037498497656343,
|
||||
"grad_norm": 3.171875,
|
||||
"learning_rate": 1.5196282647011698e-05,
|
||||
"loss": 2.1499905395507812,
|
||||
"mean_token_accuracy": 0.5204668046534061,
|
||||
"num_tokens": 9006296.0,
|
||||
"step": 1500
|
||||
},
|
||||
{
|
||||
"entropy": 2.131927245259285,
|
||||
"epoch": 0.2483874844757822,
|
||||
"grad_norm": 3.546875,
|
||||
"learning_rate": 1.5036051914757252e-05,
|
||||
"loss": 2.157076416015625,
|
||||
"mean_token_accuracy": 0.5220408822596073,
|
||||
"num_tokens": 9307203.0,
|
||||
"step": 1550
|
||||
},
|
||||
{
|
||||
"entropy": 2.1308567851781843,
|
||||
"epoch": 0.256399983975001,
|
||||
"grad_norm": 3.546875,
|
||||
"learning_rate": 1.4875821182502805e-05,
|
||||
"loss": 2.146601104736328,
|
||||
"mean_token_accuracy": 0.5219539681077003,
|
||||
"num_tokens": 9610070.0,
|
||||
"step": 1600
|
||||
},
|
||||
{
|
||||
"entropy": 2.1097783839702604,
|
||||
"epoch": 0.2644124834742198,
|
||||
"grad_norm": 3.84375,
|
||||
"learning_rate": 1.4715590450248359e-05,
|
||||
"loss": 2.1268209838867187,
|
||||
"mean_token_accuracy": 0.5236652019619942,
|
||||
"num_tokens": 9909117.0,
|
||||
"step": 1650
|
||||
},
|
||||
{
|
||||
"entropy": 2.123499180674553,
|
||||
"epoch": 0.2724249829734386,
|
||||
"grad_norm": 3.46875,
|
||||
"learning_rate": 1.4555359717993913e-05,
|
||||
"loss": 2.1409934997558593,
|
||||
"mean_token_accuracy": 0.5225465354323388,
|
||||
"num_tokens": 10209461.0,
|
||||
"step": 1700
|
||||
},
|
||||
{
|
||||
"entropy": 2.13052960395813,
|
||||
"epoch": 0.2804374824726573,
|
||||
"grad_norm": 2.71875,
|
||||
"learning_rate": 1.4395128985739467e-05,
|
||||
"loss": 2.151373291015625,
|
||||
"mean_token_accuracy": 0.5227928113937378,
|
||||
"num_tokens": 10509282.0,
|
||||
"step": 1750
|
||||
},
|
||||
{
|
||||
"entropy": 2.1226114094257356,
|
||||
"epoch": 0.2884499819718761,
|
||||
"grad_norm": 3.65625,
|
||||
"learning_rate": 1.4234898253485019e-05,
|
||||
"loss": 2.144451141357422,
|
||||
"mean_token_accuracy": 0.5233827362954616,
|
||||
"num_tokens": 10811056.0,
|
||||
"step": 1800
|
||||
},
|
||||
{
|
||||
"entropy": 2.126143154501915,
|
||||
"epoch": 0.2964624814710949,
|
||||
"grad_norm": 3.359375,
|
||||
"learning_rate": 1.4074667521230575e-05,
|
||||
"loss": 2.1457598876953123,
|
||||
"mean_token_accuracy": 0.5221967925131321,
|
||||
"num_tokens": 11111153.0,
|
||||
"step": 1850
|
||||
},
|
||||
{
|
||||
"entropy": 2.1048326581716537,
|
||||
"epoch": 0.30447498097031367,
|
||||
"grad_norm": 3.59375,
|
||||
"learning_rate": 1.3914436788976127e-05,
|
||||
"loss": 2.1260116577148436,
|
||||
"mean_token_accuracy": 0.5272719909250736,
|
||||
"num_tokens": 11405606.0,
|
||||
"step": 1900
|
||||
},
|
||||
{
|
||||
"entropy": 2.1236617982387545,
|
||||
"epoch": 0.31248748046953245,
|
||||
"grad_norm": 3.0,
|
||||
"learning_rate": 1.375420605672168e-05,
|
||||
"loss": 2.151782684326172,
|
||||
"mean_token_accuracy": 0.5227546173334122,
|
||||
"num_tokens": 11703943.0,
|
||||
"step": 1950
|
||||
},
|
||||
{
|
||||
"entropy": 2.121563429236412,
|
||||
"epoch": 0.32049997996875124,
|
||||
"grad_norm": 3.34375,
|
||||
"learning_rate": 1.3593975324467235e-05,
|
||||
"loss": 2.1370144653320313,
|
||||
"mean_token_accuracy": 0.5235123547911644,
|
||||
"num_tokens": 12002522.0,
|
||||
"step": 2000
|
||||
},
|
||||
{
|
||||
"entropy": 2.1155461168289182,
|
||||
"epoch": 0.32851247946797,
|
||||
"grad_norm": 3.46875,
|
||||
"learning_rate": 1.3433744592212787e-05,
|
||||
"loss": 2.1364064025878906,
|
||||
"mean_token_accuracy": 0.5237135806679726,
|
||||
"num_tokens": 12303098.0,
|
||||
"step": 2050
|
||||
},
|
||||
{
|
||||
"entropy": 2.1386052322387696,
|
||||
"epoch": 0.3365249789671888,
|
||||
"grad_norm": 3.484375,
|
||||
"learning_rate": 1.327351385995834e-05,
|
||||
"loss": 2.149888458251953,
|
||||
"mean_token_accuracy": 0.522669050693512,
|
||||
"num_tokens": 12605534.0,
|
||||
"step": 2100
|
||||
},
|
||||
{
|
||||
"entropy": 2.102600795030594,
|
||||
"epoch": 0.3445374784664076,
|
||||
"grad_norm": 3.796875,
|
||||
"learning_rate": 1.3113283127703895e-05,
|
||||
"loss": 2.1236566162109374,
|
||||
"mean_token_accuracy": 0.5279092155396938,
|
||||
"num_tokens": 12901218.0,
|
||||
"step": 2150
|
||||
},
|
||||
{
|
||||
"entropy": 2.132645187973976,
|
||||
"epoch": 0.35254997796562637,
|
||||
"grad_norm": 3.53125,
|
||||
"learning_rate": 1.2953052395449447e-05,
|
||||
"loss": 2.1523503112792968,
|
||||
"mean_token_accuracy": 0.5198129612207413,
|
||||
"num_tokens": 13213018.0,
|
||||
"step": 2200
|
||||
},
|
||||
{
|
||||
"entropy": 2.1204321217536926,
|
||||
"epoch": 0.36056247746484515,
|
||||
"grad_norm": 3.25,
|
||||
"learning_rate": 1.2792821663195001e-05,
|
||||
"loss": 2.1325447082519533,
|
||||
"mean_token_accuracy": 0.5243636940419674,
|
||||
"num_tokens": 13514408.0,
|
||||
"step": 2250
|
||||
},
|
||||
{
|
||||
"entropy": 2.102410731315613,
|
||||
"epoch": 0.36857497696406394,
|
||||
"grad_norm": 3.1875,
|
||||
"learning_rate": 1.2632590930940555e-05,
|
||||
"loss": 2.123565368652344,
|
||||
"mean_token_accuracy": 0.5257614709436893,
|
||||
"num_tokens": 13818272.0,
|
||||
"step": 2300
|
||||
},
|
||||
{
|
||||
"entropy": 2.0876205080747603,
|
||||
"epoch": 0.3765874764632827,
|
||||
"grad_norm": 3.328125,
|
||||
"learning_rate": 1.247236019868611e-05,
|
||||
"loss": 2.116696472167969,
|
||||
"mean_token_accuracy": 0.5275124031305313,
|
||||
"num_tokens": 14110702.0,
|
||||
"step": 2350
|
||||
},
|
||||
{
|
||||
"entropy": 2.1078518223762512,
|
||||
"epoch": 0.3845999759625015,
|
||||
"grad_norm": 3.453125,
|
||||
"learning_rate": 1.2312129466431662e-05,
|
||||
"loss": 2.1291993713378905,
|
||||
"mean_token_accuracy": 0.524967464953661,
|
||||
"num_tokens": 14407015.0,
|
||||
"step": 2400
|
||||
},
|
||||
{
|
||||
"entropy": 2.0930898267030718,
|
||||
"epoch": 0.3926124754617203,
|
||||
"grad_norm": 3.5,
|
||||
"learning_rate": 1.2151898734177216e-05,
|
||||
"loss": 2.1094123840332033,
|
||||
"mean_token_accuracy": 0.5290587748587131,
|
||||
"num_tokens": 14696565.0,
|
||||
"step": 2450
|
||||
},
|
||||
{
|
||||
"entropy": 2.1109757220745085,
|
||||
"epoch": 0.4006249749609391,
|
||||
"grad_norm": 3.5,
|
||||
"learning_rate": 1.199166800192277e-05,
|
||||
"loss": 2.1246888732910154,
|
||||
"mean_token_accuracy": 0.5258957357704639,
|
||||
"num_tokens": 14994401.0,
|
||||
"step": 2500
|
||||
},
|
||||
{
|
||||
"entropy": 2.1087811160087586,
|
||||
"epoch": 0.40863747446015786,
|
||||
"grad_norm": 3.1875,
|
||||
"learning_rate": 1.1831437269668324e-05,
|
||||
"loss": 2.12890625,
|
||||
"mean_token_accuracy": 0.5269917643070221,
|
||||
"num_tokens": 15292837.0,
|
||||
"step": 2550
|
||||
},
|
||||
{
|
||||
"entropy": 2.097252429127693,
|
||||
"epoch": 0.41664997395937664,
|
||||
"grad_norm": 3.0,
|
||||
"learning_rate": 1.1671206537413878e-05,
|
||||
"loss": 2.1145870971679686,
|
||||
"mean_token_accuracy": 0.5274137698113919,
|
||||
"num_tokens": 15590647.0,
|
||||
"step": 2600
|
||||
},
|
||||
{
|
||||
"entropy": 2.111199595332146,
|
||||
"epoch": 0.4246624734585954,
|
||||
"grad_norm": 3.25,
|
||||
"learning_rate": 1.151097580515943e-05,
|
||||
"loss": 2.1274327087402343,
|
||||
"mean_token_accuracy": 0.5250249288976192,
|
||||
"num_tokens": 15892772.0,
|
||||
"step": 2650
|
||||
},
|
||||
{
|
||||
"entropy": 2.1051502913236617,
|
||||
"epoch": 0.4326749729578142,
|
||||
"grad_norm": 3.203125,
|
||||
"learning_rate": 1.1350745072904986e-05,
|
||||
"loss": 2.127935333251953,
|
||||
"mean_token_accuracy": 0.5263640700280666,
|
||||
"num_tokens": 16191611.0,
|
||||
"step": 2700
|
||||
},
|
||||
{
|
||||
"entropy": 2.1127885603904724,
|
||||
"epoch": 0.440687472457033,
|
||||
"grad_norm": 3.515625,
|
||||
"learning_rate": 1.1190514340650538e-05,
|
||||
"loss": 2.129211883544922,
|
||||
"mean_token_accuracy": 0.5250939278304577,
|
||||
"num_tokens": 16483582.0,
|
||||
"step": 2750
|
||||
},
|
||||
{
|
||||
"entropy": 2.110930626988411,
|
||||
"epoch": 0.4486999719562518,
|
||||
"grad_norm": 3.1875,
|
||||
"learning_rate": 1.103028360839609e-05,
|
||||
"loss": 2.1220079040527344,
|
||||
"mean_token_accuracy": 0.5250919266045093,
|
||||
"num_tokens": 16787253.0,
|
||||
"step": 2800
|
||||
},
|
||||
{
|
||||
"entropy": 2.1182780677080153,
|
||||
"epoch": 0.45671247145547056,
|
||||
"grad_norm": 3.8125,
|
||||
"learning_rate": 1.0870052876141646e-05,
|
||||
"loss": 2.1358837890625,
|
||||
"mean_token_accuracy": 0.5217854312062263,
|
||||
"num_tokens": 17088741.0,
|
||||
"step": 2850
|
||||
},
|
||||
{
|
||||
"entropy": 2.1117833071947096,
|
||||
"epoch": 0.46472497095468934,
|
||||
"grad_norm": 3.171875,
|
||||
"learning_rate": 1.0709822143887198e-05,
|
||||
"loss": 2.125662841796875,
|
||||
"mean_token_accuracy": 0.524649152457714,
|
||||
"num_tokens": 17389236.0,
|
||||
"step": 2900
|
||||
},
|
||||
{
|
||||
"entropy": 2.1262730568647386,
|
||||
"epoch": 0.47273747045390807,
|
||||
"grad_norm": 3.625,
|
||||
"learning_rate": 1.054959141163275e-05,
|
||||
"loss": 2.142789611816406,
|
||||
"mean_token_accuracy": 0.5236384868621826,
|
||||
"num_tokens": 17693413.0,
|
||||
"step": 2950
|
||||
},
|
||||
{
|
||||
"entropy": 2.115410739183426,
|
||||
"epoch": 0.48074996995312685,
|
||||
"grad_norm": 3.140625,
|
||||
"learning_rate": 1.0389360679378306e-05,
|
||||
"loss": 2.1337106323242185,
|
||||
"mean_token_accuracy": 0.5249222332239151,
|
||||
"num_tokens": 17990326.0,
|
||||
"step": 3000
|
||||
},
|
||||
{
|
||||
"entropy": 2.0951381689310073,
|
||||
"epoch": 0.48876246945234564,
|
||||
"grad_norm": 3.234375,
|
||||
"learning_rate": 1.0229129947123859e-05,
|
||||
"loss": 2.108144073486328,
|
||||
"mean_token_accuracy": 0.5269971822202205,
|
||||
"num_tokens": 18293136.0,
|
||||
"step": 3050
|
||||
},
|
||||
{
|
||||
"entropy": 2.116977259516716,
|
||||
"epoch": 0.4967749689515644,
|
||||
"grad_norm": 3.421875,
|
||||
"learning_rate": 1.0068899214869413e-05,
|
||||
"loss": 2.1327830505371095,
|
||||
"mean_token_accuracy": 0.5238270902633667,
|
||||
"num_tokens": 18591481.0,
|
||||
"step": 3100
|
||||
},
|
||||
{
|
||||
"entropy": 2.08598925113678,
|
||||
"epoch": 0.5047874684507833,
|
||||
"grad_norm": 2.96875,
|
||||
"learning_rate": 9.908668482614967e-06,
|
||||
"loss": 2.1032344055175782,
|
||||
"mean_token_accuracy": 0.5289211909472943,
|
||||
"num_tokens": 18885221.0,
|
||||
"step": 3150
|
||||
},
|
||||
{
|
||||
"entropy": 2.097712662220001,
|
||||
"epoch": 0.512799967950002,
|
||||
"grad_norm": 3.40625,
|
||||
"learning_rate": 9.74843775036052e-06,
|
||||
"loss": 2.1150633239746095,
|
||||
"mean_token_accuracy": 0.5288272814452648,
|
||||
"num_tokens": 19184784.0,
|
||||
"step": 3200
|
||||
},
|
||||
{
|
||||
"entropy": 2.1063760107755662,
|
||||
"epoch": 0.5208124674492208,
|
||||
"grad_norm": 3.28125,
|
||||
"learning_rate": 9.588207018106075e-06,
|
||||
"loss": 2.1217109680175783,
|
||||
"mean_token_accuracy": 0.526290989369154,
|
||||
"num_tokens": 19482538.0,
|
||||
"step": 3250
|
||||
},
|
||||
{
|
||||
"entropy": 2.090182801485062,
|
||||
"epoch": 0.5288249669484396,
|
||||
"grad_norm": 2.875,
|
||||
"learning_rate": 9.427976285851627e-06,
|
||||
"loss": 2.1130311584472654,
|
||||
"mean_token_accuracy": 0.5285439921915531,
|
||||
"num_tokens": 19782544.0,
|
||||
"step": 3300
|
||||
},
|
||||
{
|
||||
"entropy": 2.103260806798935,
|
||||
"epoch": 0.5368374664476584,
|
||||
"grad_norm": 3.59375,
|
||||
"learning_rate": 9.267745553597181e-06,
|
||||
"loss": 2.116189727783203,
|
||||
"mean_token_accuracy": 0.5273195558786392,
|
||||
"num_tokens": 20086408.0,
|
||||
"step": 3350
|
||||
},
|
||||
{
|
||||
"entropy": 2.0673747235536575,
|
||||
"epoch": 0.5448499659468772,
|
||||
"grad_norm": 3.125,
|
||||
"learning_rate": 9.107514821342735e-06,
|
||||
"loss": 2.087482147216797,
|
||||
"mean_token_accuracy": 0.532562478929758,
|
||||
"num_tokens": 20380429.0,
|
||||
"step": 3400
|
||||
},
|
||||
{
|
||||
"entropy": 2.1030789083242416,
|
||||
"epoch": 0.552862465446096,
|
||||
"grad_norm": 3.296875,
|
||||
"learning_rate": 8.947284089088289e-06,
|
||||
"loss": 2.122591247558594,
|
||||
"mean_token_accuracy": 0.5256703786551953,
|
||||
"num_tokens": 20682849.0,
|
||||
"step": 3450
|
||||
},
|
||||
{
|
||||
"entropy": 2.0988716661930082,
|
||||
"epoch": 0.5608749649453146,
|
||||
"grad_norm": 3.546875,
|
||||
"learning_rate": 8.787053356833841e-06,
|
||||
"loss": 2.115867156982422,
|
||||
"mean_token_accuracy": 0.5266730943322182,
|
||||
"num_tokens": 20979225.0,
|
||||
"step": 3500
|
||||
}
|
||||
],
|
||||
"logging_steps": 50,
|
||||
"max_steps": 6241,
|
||||
"num_input_tokens_seen": 0,
|
||||
"num_train_epochs": 1,
|
||||
"save_steps": 500,
|
||||
"stateful_callbacks": {
|
||||
"TrainerControl": {
|
||||
"args": {
|
||||
"should_epoch_stop": false,
|
||||
"should_evaluate": false,
|
||||
"should_log": false,
|
||||
"should_save": true,
|
||||
"should_training_stop": false
|
||||
},
|
||||
"attributes": {}
|
||||
}
|
||||
},
|
||||
"total_flos": 2.4793245607162675e+17,
|
||||
"train_batch_size": 4,
|
||||
"trial_name": null,
|
||||
"trial_params": null
|
||||
}
|
||||
3
checkpoint-3500/training_args.bin
Normal file
3
checkpoint-3500/training_args.bin
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:78622f63304ec0913ed0b4ea7a1b26c08a5621b3661a380d798c4ea426235671
|
||||
size 5713
|
||||
89
checkpoint-4000/chat_template.jinja
Normal file
89
checkpoint-4000/chat_template.jinja
Normal file
@@ -0,0 +1,89 @@
|
||||
{%- if tools %}
|
||||
{{- '<|im_start|>system\n' }}
|
||||
{%- if messages[0].role == 'system' %}
|
||||
{{- messages[0].content + '\n\n' }}
|
||||
{%- endif %}
|
||||
{{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
|
||||
{%- for tool in tools %}
|
||||
{{- "\n" }}
|
||||
{{- tool | tojson }}
|
||||
{%- endfor %}
|
||||
{{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
|
||||
{%- else %}
|
||||
{%- if messages[0].role == 'system' %}
|
||||
{{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
|
||||
{%- endif %}
|
||||
{%- endif %}
|
||||
{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
|
||||
{%- for message in messages[::-1] %}
|
||||
{%- set index = (messages|length - 1) - loop.index0 %}
|
||||
{%- if ns.multi_step_tool and message.role == "user" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
|
||||
{%- set ns.multi_step_tool = false %}
|
||||
{%- set ns.last_query_index = index %}
|
||||
{%- endif %}
|
||||
{%- endfor %}
|
||||
{%- for message in messages %}
|
||||
{%- if message.content is string %}
|
||||
{%- set content = message.content %}
|
||||
{%- else %}
|
||||
{%- set content = '' %}
|
||||
{%- endif %}
|
||||
{%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
|
||||
{{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
|
||||
{%- elif message.role == "assistant" %}
|
||||
{%- set reasoning_content = '' %}
|
||||
{%- if message.reasoning_content is string %}
|
||||
{%- set reasoning_content = message.reasoning_content %}
|
||||
{%- else %}
|
||||
{%- if '</think>' in content %}
|
||||
{%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
|
||||
{%- set content = content.split('</think>')[-1].lstrip('\n') %}
|
||||
{%- endif %}
|
||||
{%- endif %}
|
||||
{%- if loop.index0 > ns.last_query_index %}
|
||||
{%- if loop.last or (not loop.last and reasoning_content) %}
|
||||
{{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
|
||||
{%- else %}
|
||||
{{- '<|im_start|>' + message.role + '\n' + content }}
|
||||
{%- endif %}
|
||||
{%- else %}
|
||||
{{- '<|im_start|>' + message.role + '\n' + content }}
|
||||
{%- endif %}
|
||||
{%- if message.tool_calls %}
|
||||
{%- for tool_call in message.tool_calls %}
|
||||
{%- if (loop.first and content) or (not loop.first) %}
|
||||
{{- '\n' }}
|
||||
{%- endif %}
|
||||
{%- if tool_call.function %}
|
||||
{%- set tool_call = tool_call.function %}
|
||||
{%- endif %}
|
||||
{{- '<tool_call>\n{"name": "' }}
|
||||
{{- tool_call.name }}
|
||||
{{- '", "arguments": ' }}
|
||||
{%- if tool_call.arguments is string %}
|
||||
{{- tool_call.arguments }}
|
||||
{%- else %}
|
||||
{{- tool_call.arguments | tojson }}
|
||||
{%- endif %}
|
||||
{{- '}\n</tool_call>' }}
|
||||
{%- endfor %}
|
||||
{%- endif %}
|
||||
{{- '<|im_end|>\n' }}
|
||||
{%- elif message.role == "tool" %}
|
||||
{%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
|
||||
{{- '<|im_start|>user' }}
|
||||
{%- endif %}
|
||||
{{- '\n<tool_response>\n' }}
|
||||
{{- content }}
|
||||
{{- '\n</tool_response>' }}
|
||||
{%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
|
||||
{{- '<|im_end|>\n' }}
|
||||
{%- endif %}
|
||||
{%- endif %}
|
||||
{%- endfor %}
|
||||
{%- if add_generation_prompt %}
|
||||
{{- '<|im_start|>assistant\n' }}
|
||||
{%- if enable_thinking is defined and enable_thinking is false %}
|
||||
{{- '<think>\n\n</think>\n\n' }}
|
||||
{%- endif %}
|
||||
{%- endif %}
|
||||
63
checkpoint-4000/config.json
Normal file
63
checkpoint-4000/config.json
Normal file
@@ -0,0 +1,63 @@
|
||||
{
|
||||
"architectures": [
|
||||
"Qwen3ForCausalLM"
|
||||
],
|
||||
"attention_bias": false,
|
||||
"attention_dropout": 0.0,
|
||||
"bos_token_id": null,
|
||||
"dtype": "bfloat16",
|
||||
"eos_token_id": 151645,
|
||||
"head_dim": 128,
|
||||
"hidden_act": "silu",
|
||||
"hidden_size": 2048,
|
||||
"initializer_range": 0.02,
|
||||
"intermediate_size": 6144,
|
||||
"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"
|
||||
],
|
||||
"max_position_embeddings": 40960,
|
||||
"max_window_layers": 28,
|
||||
"model_type": "qwen3",
|
||||
"num_attention_heads": 16,
|
||||
"num_hidden_layers": 28,
|
||||
"num_key_value_heads": 8,
|
||||
"pad_token_id": 151643,
|
||||
"rms_norm_eps": 1e-06,
|
||||
"rope_parameters": {
|
||||
"rope_theta": 1000000,
|
||||
"rope_type": "default"
|
||||
},
|
||||
"sliding_window": null,
|
||||
"tie_word_embeddings": true,
|
||||
"transformers_version": "5.7.0",
|
||||
"use_cache": false,
|
||||
"use_sliding_window": false,
|
||||
"vocab_size": 151936
|
||||
}
|
||||
12
checkpoint-4000/generation_config.json
Normal file
12
checkpoint-4000/generation_config.json
Normal file
@@ -0,0 +1,12 @@
|
||||
{
|
||||
"do_sample": true,
|
||||
"eos_token_id": [
|
||||
151645,
|
||||
151643
|
||||
],
|
||||
"pad_token_id": 151643,
|
||||
"temperature": 0.6,
|
||||
"top_k": 20,
|
||||
"top_p": 0.95,
|
||||
"transformers_version": "5.7.0"
|
||||
}
|
||||
3
checkpoint-4000/model.safetensors
Normal file
3
checkpoint-4000/model.safetensors
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:59f52f23dfe48c7547727ea25697ac3a9d18e4d016c8fa4c5c9e1b4dff8c31cb
|
||||
size 3441185608
|
||||
3
checkpoint-4000/optimizer.pt
Normal file
3
checkpoint-4000/optimizer.pt
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:5c313769ae550920040bf3021b0ea64cd98c9f52d74cedfbdb319bf5c5369138
|
||||
size 6882572207
|
||||
3
checkpoint-4000/rng_state.pth
Normal file
3
checkpoint-4000/rng_state.pth
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:61c19bab1174704a4a4441475683bf1270277af15d2e2c95e964789128e482c4
|
||||
size 14645
|
||||
3
checkpoint-4000/scheduler.pt
Normal file
3
checkpoint-4000/scheduler.pt
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:78d0e196db341b3518b3d3e05e16be9cc71c297b4603d00fde73052aafae3e28
|
||||
size 1465
|
||||
3
checkpoint-4000/tokenizer.json
Normal file
3
checkpoint-4000/tokenizer.json
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:be75606093db2094d7cd20f3c2f385c212750648bd6ea4fb2bf507a6a4c55506
|
||||
size 11422650
|
||||
30
checkpoint-4000/tokenizer_config.json
Normal file
30
checkpoint-4000/tokenizer_config.json
Normal file
@@ -0,0 +1,30 @@
|
||||
{
|
||||
"add_prefix_space": false,
|
||||
"backend": "tokenizers",
|
||||
"bos_token": null,
|
||||
"clean_up_tokenization_spaces": false,
|
||||
"eos_token": "<|im_end|>",
|
||||
"errors": "replace",
|
||||
"extra_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|>"
|
||||
],
|
||||
"is_local": false,
|
||||
"local_files_only": false,
|
||||
"model_max_length": 131072,
|
||||
"pad_token": "<|endoftext|>",
|
||||
"split_special_tokens": false,
|
||||
"tokenizer_class": "Qwen2Tokenizer",
|
||||
"unk_token": null
|
||||
}
|
||||
834
checkpoint-4000/trainer_state.json
Normal file
834
checkpoint-4000/trainer_state.json
Normal file
@@ -0,0 +1,834 @@
|
||||
{
|
||||
"best_global_step": null,
|
||||
"best_metric": null,
|
||||
"best_model_checkpoint": null,
|
||||
"epoch": 0.6409999599375025,
|
||||
"eval_steps": 500,
|
||||
"global_step": 4000,
|
||||
"is_hyper_param_search": false,
|
||||
"is_local_process_zero": true,
|
||||
"is_world_process_zero": true,
|
||||
"log_history": [
|
||||
{
|
||||
"entropy": 2.223189059495926,
|
||||
"epoch": 0.008012499499218782,
|
||||
"grad_norm": 2.875,
|
||||
"learning_rate": 1.9842973882390643e-05,
|
||||
"loss": 2.3428109741210936,
|
||||
"mean_token_accuracy": 0.4963963569700718,
|
||||
"num_tokens": 305366.0,
|
||||
"step": 50
|
||||
},
|
||||
{
|
||||
"entropy": 2.188130117058754,
|
||||
"epoch": 0.016024998998437564,
|
||||
"grad_norm": 3.328125,
|
||||
"learning_rate": 1.96827431501362e-05,
|
||||
"loss": 2.225358734130859,
|
||||
"mean_token_accuracy": 0.5110921229422093,
|
||||
"num_tokens": 605042.0,
|
||||
"step": 100
|
||||
},
|
||||
{
|
||||
"entropy": 2.197347580194473,
|
||||
"epoch": 0.024037498497656344,
|
||||
"grad_norm": 3.453125,
|
||||
"learning_rate": 1.952251241788175e-05,
|
||||
"loss": 2.2299420166015627,
|
||||
"mean_token_accuracy": 0.5093490639328957,
|
||||
"num_tokens": 906888.0,
|
||||
"step": 150
|
||||
},
|
||||
{
|
||||
"entropy": 2.1913950395584108,
|
||||
"epoch": 0.03204999799687513,
|
||||
"grad_norm": 2.90625,
|
||||
"learning_rate": 1.9362281685627303e-05,
|
||||
"loss": 2.210995941162109,
|
||||
"mean_token_accuracy": 0.5097909501194954,
|
||||
"num_tokens": 1213125.0,
|
||||
"step": 200
|
||||
},
|
||||
{
|
||||
"entropy": 2.1611998677253723,
|
||||
"epoch": 0.040062497496093905,
|
||||
"grad_norm": 3.25,
|
||||
"learning_rate": 1.920205095337286e-05,
|
||||
"loss": 2.1867474365234374,
|
||||
"mean_token_accuracy": 0.5154410660266876,
|
||||
"num_tokens": 1509026.0,
|
||||
"step": 250
|
||||
},
|
||||
{
|
||||
"entropy": 2.168051454424858,
|
||||
"epoch": 0.04807499699531269,
|
||||
"grad_norm": 2.9375,
|
||||
"learning_rate": 1.904182022111841e-05,
|
||||
"loss": 2.190071716308594,
|
||||
"mean_token_accuracy": 0.5156884534657001,
|
||||
"num_tokens": 1806850.0,
|
||||
"step": 300
|
||||
},
|
||||
{
|
||||
"entropy": 2.1737154543399813,
|
||||
"epoch": 0.05608749649453147,
|
||||
"grad_norm": 3.421875,
|
||||
"learning_rate": 1.8881589488863967e-05,
|
||||
"loss": 2.199147186279297,
|
||||
"mean_token_accuracy": 0.514045044630766,
|
||||
"num_tokens": 2104505.0,
|
||||
"step": 350
|
||||
},
|
||||
{
|
||||
"entropy": 2.148039159178734,
|
||||
"epoch": 0.06409999599375026,
|
||||
"grad_norm": 3.109375,
|
||||
"learning_rate": 1.872135875660952e-05,
|
||||
"loss": 2.176946105957031,
|
||||
"mean_token_accuracy": 0.5177112428843975,
|
||||
"num_tokens": 2391088.0,
|
||||
"step": 400
|
||||
},
|
||||
{
|
||||
"entropy": 2.194890711903572,
|
||||
"epoch": 0.07211249549296903,
|
||||
"grad_norm": 3.28125,
|
||||
"learning_rate": 1.856112802435507e-05,
|
||||
"loss": 2.216288757324219,
|
||||
"mean_token_accuracy": 0.5110335703194141,
|
||||
"num_tokens": 2691272.0,
|
||||
"step": 450
|
||||
},
|
||||
{
|
||||
"entropy": 2.1557786524295808,
|
||||
"epoch": 0.08012499499218781,
|
||||
"grad_norm": 4.15625,
|
||||
"learning_rate": 1.8400897292100627e-05,
|
||||
"loss": 2.1812680053710936,
|
||||
"mean_token_accuracy": 0.5176697050035,
|
||||
"num_tokens": 2991855.0,
|
||||
"step": 500
|
||||
},
|
||||
{
|
||||
"entropy": 2.1708167713880537,
|
||||
"epoch": 0.08813749449140659,
|
||||
"grad_norm": 4.34375,
|
||||
"learning_rate": 1.824066655984618e-05,
|
||||
"loss": 2.1921002197265627,
|
||||
"mean_token_accuracy": 0.5156507430970669,
|
||||
"num_tokens": 3295049.0,
|
||||
"step": 550
|
||||
},
|
||||
{
|
||||
"entropy": 2.1686314886808393,
|
||||
"epoch": 0.09614999399062538,
|
||||
"grad_norm": 2.875,
|
||||
"learning_rate": 1.8080435827591735e-05,
|
||||
"loss": 2.1914617919921877,
|
||||
"mean_token_accuracy": 0.5139613692462445,
|
||||
"num_tokens": 3597206.0,
|
||||
"step": 600
|
||||
},
|
||||
{
|
||||
"entropy": 2.15742906332016,
|
||||
"epoch": 0.10416249348984416,
|
||||
"grad_norm": 3.109375,
|
||||
"learning_rate": 1.7920205095337287e-05,
|
||||
"loss": 2.174853973388672,
|
||||
"mean_token_accuracy": 0.5159518288075924,
|
||||
"num_tokens": 3905268.0,
|
||||
"step": 650
|
||||
},
|
||||
{
|
||||
"entropy": 2.162363375425339,
|
||||
"epoch": 0.11217499298906294,
|
||||
"grad_norm": 3.171875,
|
||||
"learning_rate": 1.7759974363082843e-05,
|
||||
"loss": 2.184231872558594,
|
||||
"mean_token_accuracy": 0.516631413847208,
|
||||
"num_tokens": 4207934.0,
|
||||
"step": 700
|
||||
},
|
||||
{
|
||||
"entropy": 2.1516289287805557,
|
||||
"epoch": 0.12018749248828171,
|
||||
"grad_norm": 3.609375,
|
||||
"learning_rate": 1.7599743630828395e-05,
|
||||
"loss": 2.167020263671875,
|
||||
"mean_token_accuracy": 0.516358617246151,
|
||||
"num_tokens": 4515774.0,
|
||||
"step": 750
|
||||
},
|
||||
{
|
||||
"entropy": 2.1549544781446457,
|
||||
"epoch": 0.1281999919875005,
|
||||
"grad_norm": 3.28125,
|
||||
"learning_rate": 1.7439512898573947e-05,
|
||||
"loss": 2.181917877197266,
|
||||
"mean_token_accuracy": 0.5182220858335495,
|
||||
"num_tokens": 4814709.0,
|
||||
"step": 800
|
||||
},
|
||||
{
|
||||
"entropy": 2.1532615756988527,
|
||||
"epoch": 0.1362124914867193,
|
||||
"grad_norm": 3.078125,
|
||||
"learning_rate": 1.7279282166319503e-05,
|
||||
"loss": 2.1659007263183594,
|
||||
"mean_token_accuracy": 0.5180425970256328,
|
||||
"num_tokens": 5112585.0,
|
||||
"step": 850
|
||||
},
|
||||
{
|
||||
"entropy": 2.1704017293453215,
|
||||
"epoch": 0.14422499098593805,
|
||||
"grad_norm": 2.921875,
|
||||
"learning_rate": 1.7119051434065055e-05,
|
||||
"loss": 2.1934996032714844,
|
||||
"mean_token_accuracy": 0.5154964691400528,
|
||||
"num_tokens": 5411369.0,
|
||||
"step": 900
|
||||
},
|
||||
{
|
||||
"entropy": 2.1349107217788696,
|
||||
"epoch": 0.15223749048515683,
|
||||
"grad_norm": 2.984375,
|
||||
"learning_rate": 1.6958820701810608e-05,
|
||||
"loss": 2.1598545837402345,
|
||||
"mean_token_accuracy": 0.5208920259773732,
|
||||
"num_tokens": 5707060.0,
|
||||
"step": 950
|
||||
},
|
||||
{
|
||||
"entropy": 2.1541705918312073,
|
||||
"epoch": 0.16024998998437562,
|
||||
"grad_norm": 3.203125,
|
||||
"learning_rate": 1.6798589969556163e-05,
|
||||
"loss": 2.170956573486328,
|
||||
"mean_token_accuracy": 0.5204938448965549,
|
||||
"num_tokens": 6004079.0,
|
||||
"step": 1000
|
||||
},
|
||||
{
|
||||
"entropy": 2.1523867017030716,
|
||||
"epoch": 0.1682624894835944,
|
||||
"grad_norm": 3.09375,
|
||||
"learning_rate": 1.6638359237301716e-05,
|
||||
"loss": 2.1695118713378907,
|
||||
"mean_token_accuracy": 0.5177376234531402,
|
||||
"num_tokens": 6315462.0,
|
||||
"step": 1050
|
||||
},
|
||||
{
|
||||
"entropy": 2.1356022965908052,
|
||||
"epoch": 0.17627498898281319,
|
||||
"grad_norm": 3.21875,
|
||||
"learning_rate": 1.6478128505047268e-05,
|
||||
"loss": 2.1513107299804686,
|
||||
"mean_token_accuracy": 0.5217180585861206,
|
||||
"num_tokens": 6613060.0,
|
||||
"step": 1100
|
||||
},
|
||||
{
|
||||
"entropy": 2.1483460652828215,
|
||||
"epoch": 0.18428748848203197,
|
||||
"grad_norm": 3.21875,
|
||||
"learning_rate": 1.6317897772792824e-05,
|
||||
"loss": 2.1671595764160156,
|
||||
"mean_token_accuracy": 0.5199222292006016,
|
||||
"num_tokens": 6909655.0,
|
||||
"step": 1150
|
||||
},
|
||||
{
|
||||
"entropy": 2.1074040442705155,
|
||||
"epoch": 0.19229998798125075,
|
||||
"grad_norm": 3.828125,
|
||||
"learning_rate": 1.6157667040538376e-05,
|
||||
"loss": 2.12770751953125,
|
||||
"mean_token_accuracy": 0.5252153894305229,
|
||||
"num_tokens": 7203263.0,
|
||||
"step": 1200
|
||||
},
|
||||
{
|
||||
"entropy": 2.1354027593135836,
|
||||
"epoch": 0.20031248748046954,
|
||||
"grad_norm": 2.921875,
|
||||
"learning_rate": 1.5997436308283932e-05,
|
||||
"loss": 2.156183166503906,
|
||||
"mean_token_accuracy": 0.518907299786806,
|
||||
"num_tokens": 7505411.0,
|
||||
"step": 1250
|
||||
},
|
||||
{
|
||||
"entropy": 2.142924894094467,
|
||||
"epoch": 0.20832498697968832,
|
||||
"grad_norm": 3.5625,
|
||||
"learning_rate": 1.5837205576029484e-05,
|
||||
"loss": 2.158486633300781,
|
||||
"mean_token_accuracy": 0.5205909128487111,
|
||||
"num_tokens": 7803562.0,
|
||||
"step": 1300
|
||||
},
|
||||
{
|
||||
"entropy": 2.1305291467905043,
|
||||
"epoch": 0.2163374864789071,
|
||||
"grad_norm": 3.34375,
|
||||
"learning_rate": 1.5676974843775036e-05,
|
||||
"loss": 2.151043701171875,
|
||||
"mean_token_accuracy": 0.5211617235839366,
|
||||
"num_tokens": 8108037.0,
|
||||
"step": 1350
|
||||
},
|
||||
{
|
||||
"entropy": 2.160947166085243,
|
||||
"epoch": 0.2243499859781259,
|
||||
"grad_norm": 3.234375,
|
||||
"learning_rate": 1.5516744111520592e-05,
|
||||
"loss": 2.177792663574219,
|
||||
"mean_token_accuracy": 0.5167004990577698,
|
||||
"num_tokens": 8408674.0,
|
||||
"step": 1400
|
||||
},
|
||||
{
|
||||
"entropy": 2.120403967499733,
|
||||
"epoch": 0.23236248547734467,
|
||||
"grad_norm": 3.375,
|
||||
"learning_rate": 1.5356513379266144e-05,
|
||||
"loss": 2.1462562561035154,
|
||||
"mean_token_accuracy": 0.5225985661149025,
|
||||
"num_tokens": 8707650.0,
|
||||
"step": 1450
|
||||
},
|
||||
{
|
||||
"entropy": 2.138882914185524,
|
||||
"epoch": 0.24037498497656343,
|
||||
"grad_norm": 3.171875,
|
||||
"learning_rate": 1.5196282647011698e-05,
|
||||
"loss": 2.1499905395507812,
|
||||
"mean_token_accuracy": 0.5204668046534061,
|
||||
"num_tokens": 9006296.0,
|
||||
"step": 1500
|
||||
},
|
||||
{
|
||||
"entropy": 2.131927245259285,
|
||||
"epoch": 0.2483874844757822,
|
||||
"grad_norm": 3.546875,
|
||||
"learning_rate": 1.5036051914757252e-05,
|
||||
"loss": 2.157076416015625,
|
||||
"mean_token_accuracy": 0.5220408822596073,
|
||||
"num_tokens": 9307203.0,
|
||||
"step": 1550
|
||||
},
|
||||
{
|
||||
"entropy": 2.1308567851781843,
|
||||
"epoch": 0.256399983975001,
|
||||
"grad_norm": 3.546875,
|
||||
"learning_rate": 1.4875821182502805e-05,
|
||||
"loss": 2.146601104736328,
|
||||
"mean_token_accuracy": 0.5219539681077003,
|
||||
"num_tokens": 9610070.0,
|
||||
"step": 1600
|
||||
},
|
||||
{
|
||||
"entropy": 2.1097783839702604,
|
||||
"epoch": 0.2644124834742198,
|
||||
"grad_norm": 3.84375,
|
||||
"learning_rate": 1.4715590450248359e-05,
|
||||
"loss": 2.1268209838867187,
|
||||
"mean_token_accuracy": 0.5236652019619942,
|
||||
"num_tokens": 9909117.0,
|
||||
"step": 1650
|
||||
},
|
||||
{
|
||||
"entropy": 2.123499180674553,
|
||||
"epoch": 0.2724249829734386,
|
||||
"grad_norm": 3.46875,
|
||||
"learning_rate": 1.4555359717993913e-05,
|
||||
"loss": 2.1409934997558593,
|
||||
"mean_token_accuracy": 0.5225465354323388,
|
||||
"num_tokens": 10209461.0,
|
||||
"step": 1700
|
||||
},
|
||||
{
|
||||
"entropy": 2.13052960395813,
|
||||
"epoch": 0.2804374824726573,
|
||||
"grad_norm": 2.71875,
|
||||
"learning_rate": 1.4395128985739467e-05,
|
||||
"loss": 2.151373291015625,
|
||||
"mean_token_accuracy": 0.5227928113937378,
|
||||
"num_tokens": 10509282.0,
|
||||
"step": 1750
|
||||
},
|
||||
{
|
||||
"entropy": 2.1226114094257356,
|
||||
"epoch": 0.2884499819718761,
|
||||
"grad_norm": 3.65625,
|
||||
"learning_rate": 1.4234898253485019e-05,
|
||||
"loss": 2.144451141357422,
|
||||
"mean_token_accuracy": 0.5233827362954616,
|
||||
"num_tokens": 10811056.0,
|
||||
"step": 1800
|
||||
},
|
||||
{
|
||||
"entropy": 2.126143154501915,
|
||||
"epoch": 0.2964624814710949,
|
||||
"grad_norm": 3.359375,
|
||||
"learning_rate": 1.4074667521230575e-05,
|
||||
"loss": 2.1457598876953123,
|
||||
"mean_token_accuracy": 0.5221967925131321,
|
||||
"num_tokens": 11111153.0,
|
||||
"step": 1850
|
||||
},
|
||||
{
|
||||
"entropy": 2.1048326581716537,
|
||||
"epoch": 0.30447498097031367,
|
||||
"grad_norm": 3.59375,
|
||||
"learning_rate": 1.3914436788976127e-05,
|
||||
"loss": 2.1260116577148436,
|
||||
"mean_token_accuracy": 0.5272719909250736,
|
||||
"num_tokens": 11405606.0,
|
||||
"step": 1900
|
||||
},
|
||||
{
|
||||
"entropy": 2.1236617982387545,
|
||||
"epoch": 0.31248748046953245,
|
||||
"grad_norm": 3.0,
|
||||
"learning_rate": 1.375420605672168e-05,
|
||||
"loss": 2.151782684326172,
|
||||
"mean_token_accuracy": 0.5227546173334122,
|
||||
"num_tokens": 11703943.0,
|
||||
"step": 1950
|
||||
},
|
||||
{
|
||||
"entropy": 2.121563429236412,
|
||||
"epoch": 0.32049997996875124,
|
||||
"grad_norm": 3.34375,
|
||||
"learning_rate": 1.3593975324467235e-05,
|
||||
"loss": 2.1370144653320313,
|
||||
"mean_token_accuracy": 0.5235123547911644,
|
||||
"num_tokens": 12002522.0,
|
||||
"step": 2000
|
||||
},
|
||||
{
|
||||
"entropy": 2.1155461168289182,
|
||||
"epoch": 0.32851247946797,
|
||||
"grad_norm": 3.46875,
|
||||
"learning_rate": 1.3433744592212787e-05,
|
||||
"loss": 2.1364064025878906,
|
||||
"mean_token_accuracy": 0.5237135806679726,
|
||||
"num_tokens": 12303098.0,
|
||||
"step": 2050
|
||||
},
|
||||
{
|
||||
"entropy": 2.1386052322387696,
|
||||
"epoch": 0.3365249789671888,
|
||||
"grad_norm": 3.484375,
|
||||
"learning_rate": 1.327351385995834e-05,
|
||||
"loss": 2.149888458251953,
|
||||
"mean_token_accuracy": 0.522669050693512,
|
||||
"num_tokens": 12605534.0,
|
||||
"step": 2100
|
||||
},
|
||||
{
|
||||
"entropy": 2.102600795030594,
|
||||
"epoch": 0.3445374784664076,
|
||||
"grad_norm": 3.796875,
|
||||
"learning_rate": 1.3113283127703895e-05,
|
||||
"loss": 2.1236566162109374,
|
||||
"mean_token_accuracy": 0.5279092155396938,
|
||||
"num_tokens": 12901218.0,
|
||||
"step": 2150
|
||||
},
|
||||
{
|
||||
"entropy": 2.132645187973976,
|
||||
"epoch": 0.35254997796562637,
|
||||
"grad_norm": 3.53125,
|
||||
"learning_rate": 1.2953052395449447e-05,
|
||||
"loss": 2.1523503112792968,
|
||||
"mean_token_accuracy": 0.5198129612207413,
|
||||
"num_tokens": 13213018.0,
|
||||
"step": 2200
|
||||
},
|
||||
{
|
||||
"entropy": 2.1204321217536926,
|
||||
"epoch": 0.36056247746484515,
|
||||
"grad_norm": 3.25,
|
||||
"learning_rate": 1.2792821663195001e-05,
|
||||
"loss": 2.1325447082519533,
|
||||
"mean_token_accuracy": 0.5243636940419674,
|
||||
"num_tokens": 13514408.0,
|
||||
"step": 2250
|
||||
},
|
||||
{
|
||||
"entropy": 2.102410731315613,
|
||||
"epoch": 0.36857497696406394,
|
||||
"grad_norm": 3.1875,
|
||||
"learning_rate": 1.2632590930940555e-05,
|
||||
"loss": 2.123565368652344,
|
||||
"mean_token_accuracy": 0.5257614709436893,
|
||||
"num_tokens": 13818272.0,
|
||||
"step": 2300
|
||||
},
|
||||
{
|
||||
"entropy": 2.0876205080747603,
|
||||
"epoch": 0.3765874764632827,
|
||||
"grad_norm": 3.328125,
|
||||
"learning_rate": 1.247236019868611e-05,
|
||||
"loss": 2.116696472167969,
|
||||
"mean_token_accuracy": 0.5275124031305313,
|
||||
"num_tokens": 14110702.0,
|
||||
"step": 2350
|
||||
},
|
||||
{
|
||||
"entropy": 2.1078518223762512,
|
||||
"epoch": 0.3845999759625015,
|
||||
"grad_norm": 3.453125,
|
||||
"learning_rate": 1.2312129466431662e-05,
|
||||
"loss": 2.1291993713378905,
|
||||
"mean_token_accuracy": 0.524967464953661,
|
||||
"num_tokens": 14407015.0,
|
||||
"step": 2400
|
||||
},
|
||||
{
|
||||
"entropy": 2.0930898267030718,
|
||||
"epoch": 0.3926124754617203,
|
||||
"grad_norm": 3.5,
|
||||
"learning_rate": 1.2151898734177216e-05,
|
||||
"loss": 2.1094123840332033,
|
||||
"mean_token_accuracy": 0.5290587748587131,
|
||||
"num_tokens": 14696565.0,
|
||||
"step": 2450
|
||||
},
|
||||
{
|
||||
"entropy": 2.1109757220745085,
|
||||
"epoch": 0.4006249749609391,
|
||||
"grad_norm": 3.5,
|
||||
"learning_rate": 1.199166800192277e-05,
|
||||
"loss": 2.1246888732910154,
|
||||
"mean_token_accuracy": 0.5258957357704639,
|
||||
"num_tokens": 14994401.0,
|
||||
"step": 2500
|
||||
},
|
||||
{
|
||||
"entropy": 2.1087811160087586,
|
||||
"epoch": 0.40863747446015786,
|
||||
"grad_norm": 3.1875,
|
||||
"learning_rate": 1.1831437269668324e-05,
|
||||
"loss": 2.12890625,
|
||||
"mean_token_accuracy": 0.5269917643070221,
|
||||
"num_tokens": 15292837.0,
|
||||
"step": 2550
|
||||
},
|
||||
{
|
||||
"entropy": 2.097252429127693,
|
||||
"epoch": 0.41664997395937664,
|
||||
"grad_norm": 3.0,
|
||||
"learning_rate": 1.1671206537413878e-05,
|
||||
"loss": 2.1145870971679686,
|
||||
"mean_token_accuracy": 0.5274137698113919,
|
||||
"num_tokens": 15590647.0,
|
||||
"step": 2600
|
||||
},
|
||||
{
|
||||
"entropy": 2.111199595332146,
|
||||
"epoch": 0.4246624734585954,
|
||||
"grad_norm": 3.25,
|
||||
"learning_rate": 1.151097580515943e-05,
|
||||
"loss": 2.1274327087402343,
|
||||
"mean_token_accuracy": 0.5250249288976192,
|
||||
"num_tokens": 15892772.0,
|
||||
"step": 2650
|
||||
},
|
||||
{
|
||||
"entropy": 2.1051502913236617,
|
||||
"epoch": 0.4326749729578142,
|
||||
"grad_norm": 3.203125,
|
||||
"learning_rate": 1.1350745072904986e-05,
|
||||
"loss": 2.127935333251953,
|
||||
"mean_token_accuracy": 0.5263640700280666,
|
||||
"num_tokens": 16191611.0,
|
||||
"step": 2700
|
||||
},
|
||||
{
|
||||
"entropy": 2.1127885603904724,
|
||||
"epoch": 0.440687472457033,
|
||||
"grad_norm": 3.515625,
|
||||
"learning_rate": 1.1190514340650538e-05,
|
||||
"loss": 2.129211883544922,
|
||||
"mean_token_accuracy": 0.5250939278304577,
|
||||
"num_tokens": 16483582.0,
|
||||
"step": 2750
|
||||
},
|
||||
{
|
||||
"entropy": 2.110930626988411,
|
||||
"epoch": 0.4486999719562518,
|
||||
"grad_norm": 3.1875,
|
||||
"learning_rate": 1.103028360839609e-05,
|
||||
"loss": 2.1220079040527344,
|
||||
"mean_token_accuracy": 0.5250919266045093,
|
||||
"num_tokens": 16787253.0,
|
||||
"step": 2800
|
||||
},
|
||||
{
|
||||
"entropy": 2.1182780677080153,
|
||||
"epoch": 0.45671247145547056,
|
||||
"grad_norm": 3.8125,
|
||||
"learning_rate": 1.0870052876141646e-05,
|
||||
"loss": 2.1358837890625,
|
||||
"mean_token_accuracy": 0.5217854312062263,
|
||||
"num_tokens": 17088741.0,
|
||||
"step": 2850
|
||||
},
|
||||
{
|
||||
"entropy": 2.1117833071947096,
|
||||
"epoch": 0.46472497095468934,
|
||||
"grad_norm": 3.171875,
|
||||
"learning_rate": 1.0709822143887198e-05,
|
||||
"loss": 2.125662841796875,
|
||||
"mean_token_accuracy": 0.524649152457714,
|
||||
"num_tokens": 17389236.0,
|
||||
"step": 2900
|
||||
},
|
||||
{
|
||||
"entropy": 2.1262730568647386,
|
||||
"epoch": 0.47273747045390807,
|
||||
"grad_norm": 3.625,
|
||||
"learning_rate": 1.054959141163275e-05,
|
||||
"loss": 2.142789611816406,
|
||||
"mean_token_accuracy": 0.5236384868621826,
|
||||
"num_tokens": 17693413.0,
|
||||
"step": 2950
|
||||
},
|
||||
{
|
||||
"entropy": 2.115410739183426,
|
||||
"epoch": 0.48074996995312685,
|
||||
"grad_norm": 3.140625,
|
||||
"learning_rate": 1.0389360679378306e-05,
|
||||
"loss": 2.1337106323242185,
|
||||
"mean_token_accuracy": 0.5249222332239151,
|
||||
"num_tokens": 17990326.0,
|
||||
"step": 3000
|
||||
},
|
||||
{
|
||||
"entropy": 2.0951381689310073,
|
||||
"epoch": 0.48876246945234564,
|
||||
"grad_norm": 3.234375,
|
||||
"learning_rate": 1.0229129947123859e-05,
|
||||
"loss": 2.108144073486328,
|
||||
"mean_token_accuracy": 0.5269971822202205,
|
||||
"num_tokens": 18293136.0,
|
||||
"step": 3050
|
||||
},
|
||||
{
|
||||
"entropy": 2.116977259516716,
|
||||
"epoch": 0.4967749689515644,
|
||||
"grad_norm": 3.421875,
|
||||
"learning_rate": 1.0068899214869413e-05,
|
||||
"loss": 2.1327830505371095,
|
||||
"mean_token_accuracy": 0.5238270902633667,
|
||||
"num_tokens": 18591481.0,
|
||||
"step": 3100
|
||||
},
|
||||
{
|
||||
"entropy": 2.08598925113678,
|
||||
"epoch": 0.5047874684507833,
|
||||
"grad_norm": 2.96875,
|
||||
"learning_rate": 9.908668482614967e-06,
|
||||
"loss": 2.1032344055175782,
|
||||
"mean_token_accuracy": 0.5289211909472943,
|
||||
"num_tokens": 18885221.0,
|
||||
"step": 3150
|
||||
},
|
||||
{
|
||||
"entropy": 2.097712662220001,
|
||||
"epoch": 0.512799967950002,
|
||||
"grad_norm": 3.40625,
|
||||
"learning_rate": 9.74843775036052e-06,
|
||||
"loss": 2.1150633239746095,
|
||||
"mean_token_accuracy": 0.5288272814452648,
|
||||
"num_tokens": 19184784.0,
|
||||
"step": 3200
|
||||
},
|
||||
{
|
||||
"entropy": 2.1063760107755662,
|
||||
"epoch": 0.5208124674492208,
|
||||
"grad_norm": 3.28125,
|
||||
"learning_rate": 9.588207018106075e-06,
|
||||
"loss": 2.1217109680175783,
|
||||
"mean_token_accuracy": 0.526290989369154,
|
||||
"num_tokens": 19482538.0,
|
||||
"step": 3250
|
||||
},
|
||||
{
|
||||
"entropy": 2.090182801485062,
|
||||
"epoch": 0.5288249669484396,
|
||||
"grad_norm": 2.875,
|
||||
"learning_rate": 9.427976285851627e-06,
|
||||
"loss": 2.1130311584472654,
|
||||
"mean_token_accuracy": 0.5285439921915531,
|
||||
"num_tokens": 19782544.0,
|
||||
"step": 3300
|
||||
},
|
||||
{
|
||||
"entropy": 2.103260806798935,
|
||||
"epoch": 0.5368374664476584,
|
||||
"grad_norm": 3.59375,
|
||||
"learning_rate": 9.267745553597181e-06,
|
||||
"loss": 2.116189727783203,
|
||||
"mean_token_accuracy": 0.5273195558786392,
|
||||
"num_tokens": 20086408.0,
|
||||
"step": 3350
|
||||
},
|
||||
{
|
||||
"entropy": 2.0673747235536575,
|
||||
"epoch": 0.5448499659468772,
|
||||
"grad_norm": 3.125,
|
||||
"learning_rate": 9.107514821342735e-06,
|
||||
"loss": 2.087482147216797,
|
||||
"mean_token_accuracy": 0.532562478929758,
|
||||
"num_tokens": 20380429.0,
|
||||
"step": 3400
|
||||
},
|
||||
{
|
||||
"entropy": 2.1030789083242416,
|
||||
"epoch": 0.552862465446096,
|
||||
"grad_norm": 3.296875,
|
||||
"learning_rate": 8.947284089088289e-06,
|
||||
"loss": 2.122591247558594,
|
||||
"mean_token_accuracy": 0.5256703786551953,
|
||||
"num_tokens": 20682849.0,
|
||||
"step": 3450
|
||||
},
|
||||
{
|
||||
"entropy": 2.0988716661930082,
|
||||
"epoch": 0.5608749649453146,
|
||||
"grad_norm": 3.546875,
|
||||
"learning_rate": 8.787053356833841e-06,
|
||||
"loss": 2.115867156982422,
|
||||
"mean_token_accuracy": 0.5266730943322182,
|
||||
"num_tokens": 20979225.0,
|
||||
"step": 3500
|
||||
},
|
||||
{
|
||||
"entropy": 2.0954072701931,
|
||||
"epoch": 0.5688874644445334,
|
||||
"grad_norm": 3.859375,
|
||||
"learning_rate": 8.626822624579395e-06,
|
||||
"loss": 2.1116868591308595,
|
||||
"mean_token_accuracy": 0.5267495161294937,
|
||||
"num_tokens": 21284166.0,
|
||||
"step": 3550
|
||||
},
|
||||
{
|
||||
"entropy": 2.11238471031189,
|
||||
"epoch": 0.5768999639437522,
|
||||
"grad_norm": 3.484375,
|
||||
"learning_rate": 8.466591892324949e-06,
|
||||
"loss": 2.1292352294921875,
|
||||
"mean_token_accuracy": 0.5249243360757828,
|
||||
"num_tokens": 21584612.0,
|
||||
"step": 3600
|
||||
},
|
||||
{
|
||||
"entropy": 2.091464368700981,
|
||||
"epoch": 0.584912463442971,
|
||||
"grad_norm": 3.359375,
|
||||
"learning_rate": 8.306361160070501e-06,
|
||||
"loss": 2.1099319458007812,
|
||||
"mean_token_accuracy": 0.5292035435140133,
|
||||
"num_tokens": 21879345.0,
|
||||
"step": 3650
|
||||
},
|
||||
{
|
||||
"entropy": 2.095204224586487,
|
||||
"epoch": 0.5929249629421898,
|
||||
"grad_norm": 2.96875,
|
||||
"learning_rate": 8.146130427816055e-06,
|
||||
"loss": 2.1198716735839844,
|
||||
"mean_token_accuracy": 0.5255991271138192,
|
||||
"num_tokens": 22183486.0,
|
||||
"step": 3700
|
||||
},
|
||||
{
|
||||
"entropy": 2.0666573679447175,
|
||||
"epoch": 0.6009374624414086,
|
||||
"grad_norm": 2.90625,
|
||||
"learning_rate": 7.98589969556161e-06,
|
||||
"loss": 2.08689697265625,
|
||||
"mean_token_accuracy": 0.5318664258718491,
|
||||
"num_tokens": 22476438.0,
|
||||
"step": 3750
|
||||
},
|
||||
{
|
||||
"entropy": 2.122153177857399,
|
||||
"epoch": 0.6089499619406273,
|
||||
"grad_norm": 3.40625,
|
||||
"learning_rate": 7.825668963307162e-06,
|
||||
"loss": 2.138288879394531,
|
||||
"mean_token_accuracy": 0.5232584895193577,
|
||||
"num_tokens": 22780385.0,
|
||||
"step": 3800
|
||||
},
|
||||
{
|
||||
"entropy": 2.1190748488903046,
|
||||
"epoch": 0.6169624614398461,
|
||||
"grad_norm": 3.46875,
|
||||
"learning_rate": 7.665438231052716e-06,
|
||||
"loss": 2.139790344238281,
|
||||
"mean_token_accuracy": 0.5243287613987923,
|
||||
"num_tokens": 23080847.0,
|
||||
"step": 3850
|
||||
},
|
||||
{
|
||||
"entropy": 2.1312570214271545,
|
||||
"epoch": 0.6249749609390649,
|
||||
"grad_norm": 3.765625,
|
||||
"learning_rate": 7.5052074987982706e-06,
|
||||
"loss": 2.1471080017089843,
|
||||
"mean_token_accuracy": 0.5232046774029732,
|
||||
"num_tokens": 23384738.0,
|
||||
"step": 3900
|
||||
},
|
||||
{
|
||||
"entropy": 2.0779008239507677,
|
||||
"epoch": 0.6329874604382837,
|
||||
"grad_norm": 3.171875,
|
||||
"learning_rate": 7.344976766543823e-06,
|
||||
"loss": 2.0974729919433592,
|
||||
"mean_token_accuracy": 0.5305448272824287,
|
||||
"num_tokens": 23682750.0,
|
||||
"step": 3950
|
||||
},
|
||||
{
|
||||
"entropy": 2.1121455317735673,
|
||||
"epoch": 0.6409999599375025,
|
||||
"grad_norm": 3.921875,
|
||||
"learning_rate": 7.184746034289377e-06,
|
||||
"loss": 2.127069549560547,
|
||||
"mean_token_accuracy": 0.5252936625480652,
|
||||
"num_tokens": 23989029.0,
|
||||
"step": 4000
|
||||
}
|
||||
],
|
||||
"logging_steps": 50,
|
||||
"max_steps": 6241,
|
||||
"num_input_tokens_seen": 0,
|
||||
"num_train_epochs": 1,
|
||||
"save_steps": 500,
|
||||
"stateful_callbacks": {
|
||||
"TrainerControl": {
|
||||
"args": {
|
||||
"should_epoch_stop": false,
|
||||
"should_evaluate": false,
|
||||
"should_log": false,
|
||||
"should_save": true,
|
||||
"should_training_stop": false
|
||||
},
|
||||
"attributes": {}
|
||||
}
|
||||
},
|
||||
"total_flos": 2.832566128574546e+17,
|
||||
"train_batch_size": 4,
|
||||
"trial_name": null,
|
||||
"trial_params": null
|
||||
}
|
||||
3
checkpoint-4000/training_args.bin
Normal file
3
checkpoint-4000/training_args.bin
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:78622f63304ec0913ed0b4ea7a1b26c08a5621b3661a380d798c4ea426235671
|
||||
size 5713
|
||||
89
checkpoint-4500/chat_template.jinja
Normal file
89
checkpoint-4500/chat_template.jinja
Normal file
@@ -0,0 +1,89 @@
|
||||
{%- if tools %}
|
||||
{{- '<|im_start|>system\n' }}
|
||||
{%- if messages[0].role == 'system' %}
|
||||
{{- messages[0].content + '\n\n' }}
|
||||
{%- endif %}
|
||||
{{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
|
||||
{%- for tool in tools %}
|
||||
{{- "\n" }}
|
||||
{{- tool | tojson }}
|
||||
{%- endfor %}
|
||||
{{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
|
||||
{%- else %}
|
||||
{%- if messages[0].role == 'system' %}
|
||||
{{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
|
||||
{%- endif %}
|
||||
{%- endif %}
|
||||
{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
|
||||
{%- for message in messages[::-1] %}
|
||||
{%- set index = (messages|length - 1) - loop.index0 %}
|
||||
{%- if ns.multi_step_tool and message.role == "user" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
|
||||
{%- set ns.multi_step_tool = false %}
|
||||
{%- set ns.last_query_index = index %}
|
||||
{%- endif %}
|
||||
{%- endfor %}
|
||||
{%- for message in messages %}
|
||||
{%- if message.content is string %}
|
||||
{%- set content = message.content %}
|
||||
{%- else %}
|
||||
{%- set content = '' %}
|
||||
{%- endif %}
|
||||
{%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
|
||||
{{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
|
||||
{%- elif message.role == "assistant" %}
|
||||
{%- set reasoning_content = '' %}
|
||||
{%- if message.reasoning_content is string %}
|
||||
{%- set reasoning_content = message.reasoning_content %}
|
||||
{%- else %}
|
||||
{%- if '</think>' in content %}
|
||||
{%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
|
||||
{%- set content = content.split('</think>')[-1].lstrip('\n') %}
|
||||
{%- endif %}
|
||||
{%- endif %}
|
||||
{%- if loop.index0 > ns.last_query_index %}
|
||||
{%- if loop.last or (not loop.last and reasoning_content) %}
|
||||
{{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
|
||||
{%- else %}
|
||||
{{- '<|im_start|>' + message.role + '\n' + content }}
|
||||
{%- endif %}
|
||||
{%- else %}
|
||||
{{- '<|im_start|>' + message.role + '\n' + content }}
|
||||
{%- endif %}
|
||||
{%- if message.tool_calls %}
|
||||
{%- for tool_call in message.tool_calls %}
|
||||
{%- if (loop.first and content) or (not loop.first) %}
|
||||
{{- '\n' }}
|
||||
{%- endif %}
|
||||
{%- if tool_call.function %}
|
||||
{%- set tool_call = tool_call.function %}
|
||||
{%- endif %}
|
||||
{{- '<tool_call>\n{"name": "' }}
|
||||
{{- tool_call.name }}
|
||||
{{- '", "arguments": ' }}
|
||||
{%- if tool_call.arguments is string %}
|
||||
{{- tool_call.arguments }}
|
||||
{%- else %}
|
||||
{{- tool_call.arguments | tojson }}
|
||||
{%- endif %}
|
||||
{{- '}\n</tool_call>' }}
|
||||
{%- endfor %}
|
||||
{%- endif %}
|
||||
{{- '<|im_end|>\n' }}
|
||||
{%- elif message.role == "tool" %}
|
||||
{%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
|
||||
{{- '<|im_start|>user' }}
|
||||
{%- endif %}
|
||||
{{- '\n<tool_response>\n' }}
|
||||
{{- content }}
|
||||
{{- '\n</tool_response>' }}
|
||||
{%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
|
||||
{{- '<|im_end|>\n' }}
|
||||
{%- endif %}
|
||||
{%- endif %}
|
||||
{%- endfor %}
|
||||
{%- if add_generation_prompt %}
|
||||
{{- '<|im_start|>assistant\n' }}
|
||||
{%- if enable_thinking is defined and enable_thinking is false %}
|
||||
{{- '<think>\n\n</think>\n\n' }}
|
||||
{%- endif %}
|
||||
{%- endif %}
|
||||
63
checkpoint-4500/config.json
Normal file
63
checkpoint-4500/config.json
Normal file
@@ -0,0 +1,63 @@
|
||||
{
|
||||
"architectures": [
|
||||
"Qwen3ForCausalLM"
|
||||
],
|
||||
"attention_bias": false,
|
||||
"attention_dropout": 0.0,
|
||||
"bos_token_id": null,
|
||||
"dtype": "bfloat16",
|
||||
"eos_token_id": 151645,
|
||||
"head_dim": 128,
|
||||
"hidden_act": "silu",
|
||||
"hidden_size": 2048,
|
||||
"initializer_range": 0.02,
|
||||
"intermediate_size": 6144,
|
||||
"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"
|
||||
],
|
||||
"max_position_embeddings": 40960,
|
||||
"max_window_layers": 28,
|
||||
"model_type": "qwen3",
|
||||
"num_attention_heads": 16,
|
||||
"num_hidden_layers": 28,
|
||||
"num_key_value_heads": 8,
|
||||
"pad_token_id": 151643,
|
||||
"rms_norm_eps": 1e-06,
|
||||
"rope_parameters": {
|
||||
"rope_theta": 1000000,
|
||||
"rope_type": "default"
|
||||
},
|
||||
"sliding_window": null,
|
||||
"tie_word_embeddings": true,
|
||||
"transformers_version": "5.7.0",
|
||||
"use_cache": false,
|
||||
"use_sliding_window": false,
|
||||
"vocab_size": 151936
|
||||
}
|
||||
12
checkpoint-4500/generation_config.json
Normal file
12
checkpoint-4500/generation_config.json
Normal file
@@ -0,0 +1,12 @@
|
||||
{
|
||||
"do_sample": true,
|
||||
"eos_token_id": [
|
||||
151645,
|
||||
151643
|
||||
],
|
||||
"pad_token_id": 151643,
|
||||
"temperature": 0.6,
|
||||
"top_k": 20,
|
||||
"top_p": 0.95,
|
||||
"transformers_version": "5.7.0"
|
||||
}
|
||||
3
checkpoint-4500/model.safetensors
Normal file
3
checkpoint-4500/model.safetensors
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:748cbfd96fce4131cccae2cfbbf7e5943bb67a7960feaffb683370fbb897f84b
|
||||
size 3441185608
|
||||
3
checkpoint-4500/optimizer.pt
Normal file
3
checkpoint-4500/optimizer.pt
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:504fdf18c65054e63b9b27fd2ef1b64e410a2d62984669334f0bbfee95f68807
|
||||
size 6882572207
|
||||
3
checkpoint-4500/rng_state.pth
Normal file
3
checkpoint-4500/rng_state.pth
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:61c19bab1174704a4a4441475683bf1270277af15d2e2c95e964789128e482c4
|
||||
size 14645
|
||||
3
checkpoint-4500/scheduler.pt
Normal file
3
checkpoint-4500/scheduler.pt
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:f0c65d41abe4fed1f917d1493ac9e89afd04deb14307279675e66aa4431576fd
|
||||
size 1465
|
||||
3
checkpoint-4500/tokenizer.json
Normal file
3
checkpoint-4500/tokenizer.json
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:be75606093db2094d7cd20f3c2f385c212750648bd6ea4fb2bf507a6a4c55506
|
||||
size 11422650
|
||||
30
checkpoint-4500/tokenizer_config.json
Normal file
30
checkpoint-4500/tokenizer_config.json
Normal file
@@ -0,0 +1,30 @@
|
||||
{
|
||||
"add_prefix_space": false,
|
||||
"backend": "tokenizers",
|
||||
"bos_token": null,
|
||||
"clean_up_tokenization_spaces": false,
|
||||
"eos_token": "<|im_end|>",
|
||||
"errors": "replace",
|
||||
"extra_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|>"
|
||||
],
|
||||
"is_local": false,
|
||||
"local_files_only": false,
|
||||
"model_max_length": 131072,
|
||||
"pad_token": "<|endoftext|>",
|
||||
"split_special_tokens": false,
|
||||
"tokenizer_class": "Qwen2Tokenizer",
|
||||
"unk_token": null
|
||||
}
|
||||
934
checkpoint-4500/trainer_state.json
Normal file
934
checkpoint-4500/trainer_state.json
Normal file
@@ -0,0 +1,934 @@
|
||||
{
|
||||
"best_global_step": null,
|
||||
"best_metric": null,
|
||||
"best_model_checkpoint": null,
|
||||
"epoch": 0.7211249549296903,
|
||||
"eval_steps": 500,
|
||||
"global_step": 4500,
|
||||
"is_hyper_param_search": false,
|
||||
"is_local_process_zero": true,
|
||||
"is_world_process_zero": true,
|
||||
"log_history": [
|
||||
{
|
||||
"entropy": 2.223189059495926,
|
||||
"epoch": 0.008012499499218782,
|
||||
"grad_norm": 2.875,
|
||||
"learning_rate": 1.9842973882390643e-05,
|
||||
"loss": 2.3428109741210936,
|
||||
"mean_token_accuracy": 0.4963963569700718,
|
||||
"num_tokens": 305366.0,
|
||||
"step": 50
|
||||
},
|
||||
{
|
||||
"entropy": 2.188130117058754,
|
||||
"epoch": 0.016024998998437564,
|
||||
"grad_norm": 3.328125,
|
||||
"learning_rate": 1.96827431501362e-05,
|
||||
"loss": 2.225358734130859,
|
||||
"mean_token_accuracy": 0.5110921229422093,
|
||||
"num_tokens": 605042.0,
|
||||
"step": 100
|
||||
},
|
||||
{
|
||||
"entropy": 2.197347580194473,
|
||||
"epoch": 0.024037498497656344,
|
||||
"grad_norm": 3.453125,
|
||||
"learning_rate": 1.952251241788175e-05,
|
||||
"loss": 2.2299420166015627,
|
||||
"mean_token_accuracy": 0.5093490639328957,
|
||||
"num_tokens": 906888.0,
|
||||
"step": 150
|
||||
},
|
||||
{
|
||||
"entropy": 2.1913950395584108,
|
||||
"epoch": 0.03204999799687513,
|
||||
"grad_norm": 2.90625,
|
||||
"learning_rate": 1.9362281685627303e-05,
|
||||
"loss": 2.210995941162109,
|
||||
"mean_token_accuracy": 0.5097909501194954,
|
||||
"num_tokens": 1213125.0,
|
||||
"step": 200
|
||||
},
|
||||
{
|
||||
"entropy": 2.1611998677253723,
|
||||
"epoch": 0.040062497496093905,
|
||||
"grad_norm": 3.25,
|
||||
"learning_rate": 1.920205095337286e-05,
|
||||
"loss": 2.1867474365234374,
|
||||
"mean_token_accuracy": 0.5154410660266876,
|
||||
"num_tokens": 1509026.0,
|
||||
"step": 250
|
||||
},
|
||||
{
|
||||
"entropy": 2.168051454424858,
|
||||
"epoch": 0.04807499699531269,
|
||||
"grad_norm": 2.9375,
|
||||
"learning_rate": 1.904182022111841e-05,
|
||||
"loss": 2.190071716308594,
|
||||
"mean_token_accuracy": 0.5156884534657001,
|
||||
"num_tokens": 1806850.0,
|
||||
"step": 300
|
||||
},
|
||||
{
|
||||
"entropy": 2.1737154543399813,
|
||||
"epoch": 0.05608749649453147,
|
||||
"grad_norm": 3.421875,
|
||||
"learning_rate": 1.8881589488863967e-05,
|
||||
"loss": 2.199147186279297,
|
||||
"mean_token_accuracy": 0.514045044630766,
|
||||
"num_tokens": 2104505.0,
|
||||
"step": 350
|
||||
},
|
||||
{
|
||||
"entropy": 2.148039159178734,
|
||||
"epoch": 0.06409999599375026,
|
||||
"grad_norm": 3.109375,
|
||||
"learning_rate": 1.872135875660952e-05,
|
||||
"loss": 2.176946105957031,
|
||||
"mean_token_accuracy": 0.5177112428843975,
|
||||
"num_tokens": 2391088.0,
|
||||
"step": 400
|
||||
},
|
||||
{
|
||||
"entropy": 2.194890711903572,
|
||||
"epoch": 0.07211249549296903,
|
||||
"grad_norm": 3.28125,
|
||||
"learning_rate": 1.856112802435507e-05,
|
||||
"loss": 2.216288757324219,
|
||||
"mean_token_accuracy": 0.5110335703194141,
|
||||
"num_tokens": 2691272.0,
|
||||
"step": 450
|
||||
},
|
||||
{
|
||||
"entropy": 2.1557786524295808,
|
||||
"epoch": 0.08012499499218781,
|
||||
"grad_norm": 4.15625,
|
||||
"learning_rate": 1.8400897292100627e-05,
|
||||
"loss": 2.1812680053710936,
|
||||
"mean_token_accuracy": 0.5176697050035,
|
||||
"num_tokens": 2991855.0,
|
||||
"step": 500
|
||||
},
|
||||
{
|
||||
"entropy": 2.1708167713880537,
|
||||
"epoch": 0.08813749449140659,
|
||||
"grad_norm": 4.34375,
|
||||
"learning_rate": 1.824066655984618e-05,
|
||||
"loss": 2.1921002197265627,
|
||||
"mean_token_accuracy": 0.5156507430970669,
|
||||
"num_tokens": 3295049.0,
|
||||
"step": 550
|
||||
},
|
||||
{
|
||||
"entropy": 2.1686314886808393,
|
||||
"epoch": 0.09614999399062538,
|
||||
"grad_norm": 2.875,
|
||||
"learning_rate": 1.8080435827591735e-05,
|
||||
"loss": 2.1914617919921877,
|
||||
"mean_token_accuracy": 0.5139613692462445,
|
||||
"num_tokens": 3597206.0,
|
||||
"step": 600
|
||||
},
|
||||
{
|
||||
"entropy": 2.15742906332016,
|
||||
"epoch": 0.10416249348984416,
|
||||
"grad_norm": 3.109375,
|
||||
"learning_rate": 1.7920205095337287e-05,
|
||||
"loss": 2.174853973388672,
|
||||
"mean_token_accuracy": 0.5159518288075924,
|
||||
"num_tokens": 3905268.0,
|
||||
"step": 650
|
||||
},
|
||||
{
|
||||
"entropy": 2.162363375425339,
|
||||
"epoch": 0.11217499298906294,
|
||||
"grad_norm": 3.171875,
|
||||
"learning_rate": 1.7759974363082843e-05,
|
||||
"loss": 2.184231872558594,
|
||||
"mean_token_accuracy": 0.516631413847208,
|
||||
"num_tokens": 4207934.0,
|
||||
"step": 700
|
||||
},
|
||||
{
|
||||
"entropy": 2.1516289287805557,
|
||||
"epoch": 0.12018749248828171,
|
||||
"grad_norm": 3.609375,
|
||||
"learning_rate": 1.7599743630828395e-05,
|
||||
"loss": 2.167020263671875,
|
||||
"mean_token_accuracy": 0.516358617246151,
|
||||
"num_tokens": 4515774.0,
|
||||
"step": 750
|
||||
},
|
||||
{
|
||||
"entropy": 2.1549544781446457,
|
||||
"epoch": 0.1281999919875005,
|
||||
"grad_norm": 3.28125,
|
||||
"learning_rate": 1.7439512898573947e-05,
|
||||
"loss": 2.181917877197266,
|
||||
"mean_token_accuracy": 0.5182220858335495,
|
||||
"num_tokens": 4814709.0,
|
||||
"step": 800
|
||||
},
|
||||
{
|
||||
"entropy": 2.1532615756988527,
|
||||
"epoch": 0.1362124914867193,
|
||||
"grad_norm": 3.078125,
|
||||
"learning_rate": 1.7279282166319503e-05,
|
||||
"loss": 2.1659007263183594,
|
||||
"mean_token_accuracy": 0.5180425970256328,
|
||||
"num_tokens": 5112585.0,
|
||||
"step": 850
|
||||
},
|
||||
{
|
||||
"entropy": 2.1704017293453215,
|
||||
"epoch": 0.14422499098593805,
|
||||
"grad_norm": 2.921875,
|
||||
"learning_rate": 1.7119051434065055e-05,
|
||||
"loss": 2.1934996032714844,
|
||||
"mean_token_accuracy": 0.5154964691400528,
|
||||
"num_tokens": 5411369.0,
|
||||
"step": 900
|
||||
},
|
||||
{
|
||||
"entropy": 2.1349107217788696,
|
||||
"epoch": 0.15223749048515683,
|
||||
"grad_norm": 2.984375,
|
||||
"learning_rate": 1.6958820701810608e-05,
|
||||
"loss": 2.1598545837402345,
|
||||
"mean_token_accuracy": 0.5208920259773732,
|
||||
"num_tokens": 5707060.0,
|
||||
"step": 950
|
||||
},
|
||||
{
|
||||
"entropy": 2.1541705918312073,
|
||||
"epoch": 0.16024998998437562,
|
||||
"grad_norm": 3.203125,
|
||||
"learning_rate": 1.6798589969556163e-05,
|
||||
"loss": 2.170956573486328,
|
||||
"mean_token_accuracy": 0.5204938448965549,
|
||||
"num_tokens": 6004079.0,
|
||||
"step": 1000
|
||||
},
|
||||
{
|
||||
"entropy": 2.1523867017030716,
|
||||
"epoch": 0.1682624894835944,
|
||||
"grad_norm": 3.09375,
|
||||
"learning_rate": 1.6638359237301716e-05,
|
||||
"loss": 2.1695118713378907,
|
||||
"mean_token_accuracy": 0.5177376234531402,
|
||||
"num_tokens": 6315462.0,
|
||||
"step": 1050
|
||||
},
|
||||
{
|
||||
"entropy": 2.1356022965908052,
|
||||
"epoch": 0.17627498898281319,
|
||||
"grad_norm": 3.21875,
|
||||
"learning_rate": 1.6478128505047268e-05,
|
||||
"loss": 2.1513107299804686,
|
||||
"mean_token_accuracy": 0.5217180585861206,
|
||||
"num_tokens": 6613060.0,
|
||||
"step": 1100
|
||||
},
|
||||
{
|
||||
"entropy": 2.1483460652828215,
|
||||
"epoch": 0.18428748848203197,
|
||||
"grad_norm": 3.21875,
|
||||
"learning_rate": 1.6317897772792824e-05,
|
||||
"loss": 2.1671595764160156,
|
||||
"mean_token_accuracy": 0.5199222292006016,
|
||||
"num_tokens": 6909655.0,
|
||||
"step": 1150
|
||||
},
|
||||
{
|
||||
"entropy": 2.1074040442705155,
|
||||
"epoch": 0.19229998798125075,
|
||||
"grad_norm": 3.828125,
|
||||
"learning_rate": 1.6157667040538376e-05,
|
||||
"loss": 2.12770751953125,
|
||||
"mean_token_accuracy": 0.5252153894305229,
|
||||
"num_tokens": 7203263.0,
|
||||
"step": 1200
|
||||
},
|
||||
{
|
||||
"entropy": 2.1354027593135836,
|
||||
"epoch": 0.20031248748046954,
|
||||
"grad_norm": 2.921875,
|
||||
"learning_rate": 1.5997436308283932e-05,
|
||||
"loss": 2.156183166503906,
|
||||
"mean_token_accuracy": 0.518907299786806,
|
||||
"num_tokens": 7505411.0,
|
||||
"step": 1250
|
||||
},
|
||||
{
|
||||
"entropy": 2.142924894094467,
|
||||
"epoch": 0.20832498697968832,
|
||||
"grad_norm": 3.5625,
|
||||
"learning_rate": 1.5837205576029484e-05,
|
||||
"loss": 2.158486633300781,
|
||||
"mean_token_accuracy": 0.5205909128487111,
|
||||
"num_tokens": 7803562.0,
|
||||
"step": 1300
|
||||
},
|
||||
{
|
||||
"entropy": 2.1305291467905043,
|
||||
"epoch": 0.2163374864789071,
|
||||
"grad_norm": 3.34375,
|
||||
"learning_rate": 1.5676974843775036e-05,
|
||||
"loss": 2.151043701171875,
|
||||
"mean_token_accuracy": 0.5211617235839366,
|
||||
"num_tokens": 8108037.0,
|
||||
"step": 1350
|
||||
},
|
||||
{
|
||||
"entropy": 2.160947166085243,
|
||||
"epoch": 0.2243499859781259,
|
||||
"grad_norm": 3.234375,
|
||||
"learning_rate": 1.5516744111520592e-05,
|
||||
"loss": 2.177792663574219,
|
||||
"mean_token_accuracy": 0.5167004990577698,
|
||||
"num_tokens": 8408674.0,
|
||||
"step": 1400
|
||||
},
|
||||
{
|
||||
"entropy": 2.120403967499733,
|
||||
"epoch": 0.23236248547734467,
|
||||
"grad_norm": 3.375,
|
||||
"learning_rate": 1.5356513379266144e-05,
|
||||
"loss": 2.1462562561035154,
|
||||
"mean_token_accuracy": 0.5225985661149025,
|
||||
"num_tokens": 8707650.0,
|
||||
"step": 1450
|
||||
},
|
||||
{
|
||||
"entropy": 2.138882914185524,
|
||||
"epoch": 0.24037498497656343,
|
||||
"grad_norm": 3.171875,
|
||||
"learning_rate": 1.5196282647011698e-05,
|
||||
"loss": 2.1499905395507812,
|
||||
"mean_token_accuracy": 0.5204668046534061,
|
||||
"num_tokens": 9006296.0,
|
||||
"step": 1500
|
||||
},
|
||||
{
|
||||
"entropy": 2.131927245259285,
|
||||
"epoch": 0.2483874844757822,
|
||||
"grad_norm": 3.546875,
|
||||
"learning_rate": 1.5036051914757252e-05,
|
||||
"loss": 2.157076416015625,
|
||||
"mean_token_accuracy": 0.5220408822596073,
|
||||
"num_tokens": 9307203.0,
|
||||
"step": 1550
|
||||
},
|
||||
{
|
||||
"entropy": 2.1308567851781843,
|
||||
"epoch": 0.256399983975001,
|
||||
"grad_norm": 3.546875,
|
||||
"learning_rate": 1.4875821182502805e-05,
|
||||
"loss": 2.146601104736328,
|
||||
"mean_token_accuracy": 0.5219539681077003,
|
||||
"num_tokens": 9610070.0,
|
||||
"step": 1600
|
||||
},
|
||||
{
|
||||
"entropy": 2.1097783839702604,
|
||||
"epoch": 0.2644124834742198,
|
||||
"grad_norm": 3.84375,
|
||||
"learning_rate": 1.4715590450248359e-05,
|
||||
"loss": 2.1268209838867187,
|
||||
"mean_token_accuracy": 0.5236652019619942,
|
||||
"num_tokens": 9909117.0,
|
||||
"step": 1650
|
||||
},
|
||||
{
|
||||
"entropy": 2.123499180674553,
|
||||
"epoch": 0.2724249829734386,
|
||||
"grad_norm": 3.46875,
|
||||
"learning_rate": 1.4555359717993913e-05,
|
||||
"loss": 2.1409934997558593,
|
||||
"mean_token_accuracy": 0.5225465354323388,
|
||||
"num_tokens": 10209461.0,
|
||||
"step": 1700
|
||||
},
|
||||
{
|
||||
"entropy": 2.13052960395813,
|
||||
"epoch": 0.2804374824726573,
|
||||
"grad_norm": 2.71875,
|
||||
"learning_rate": 1.4395128985739467e-05,
|
||||
"loss": 2.151373291015625,
|
||||
"mean_token_accuracy": 0.5227928113937378,
|
||||
"num_tokens": 10509282.0,
|
||||
"step": 1750
|
||||
},
|
||||
{
|
||||
"entropy": 2.1226114094257356,
|
||||
"epoch": 0.2884499819718761,
|
||||
"grad_norm": 3.65625,
|
||||
"learning_rate": 1.4234898253485019e-05,
|
||||
"loss": 2.144451141357422,
|
||||
"mean_token_accuracy": 0.5233827362954616,
|
||||
"num_tokens": 10811056.0,
|
||||
"step": 1800
|
||||
},
|
||||
{
|
||||
"entropy": 2.126143154501915,
|
||||
"epoch": 0.2964624814710949,
|
||||
"grad_norm": 3.359375,
|
||||
"learning_rate": 1.4074667521230575e-05,
|
||||
"loss": 2.1457598876953123,
|
||||
"mean_token_accuracy": 0.5221967925131321,
|
||||
"num_tokens": 11111153.0,
|
||||
"step": 1850
|
||||
},
|
||||
{
|
||||
"entropy": 2.1048326581716537,
|
||||
"epoch": 0.30447498097031367,
|
||||
"grad_norm": 3.59375,
|
||||
"learning_rate": 1.3914436788976127e-05,
|
||||
"loss": 2.1260116577148436,
|
||||
"mean_token_accuracy": 0.5272719909250736,
|
||||
"num_tokens": 11405606.0,
|
||||
"step": 1900
|
||||
},
|
||||
{
|
||||
"entropy": 2.1236617982387545,
|
||||
"epoch": 0.31248748046953245,
|
||||
"grad_norm": 3.0,
|
||||
"learning_rate": 1.375420605672168e-05,
|
||||
"loss": 2.151782684326172,
|
||||
"mean_token_accuracy": 0.5227546173334122,
|
||||
"num_tokens": 11703943.0,
|
||||
"step": 1950
|
||||
},
|
||||
{
|
||||
"entropy": 2.121563429236412,
|
||||
"epoch": 0.32049997996875124,
|
||||
"grad_norm": 3.34375,
|
||||
"learning_rate": 1.3593975324467235e-05,
|
||||
"loss": 2.1370144653320313,
|
||||
"mean_token_accuracy": 0.5235123547911644,
|
||||
"num_tokens": 12002522.0,
|
||||
"step": 2000
|
||||
},
|
||||
{
|
||||
"entropy": 2.1155461168289182,
|
||||
"epoch": 0.32851247946797,
|
||||
"grad_norm": 3.46875,
|
||||
"learning_rate": 1.3433744592212787e-05,
|
||||
"loss": 2.1364064025878906,
|
||||
"mean_token_accuracy": 0.5237135806679726,
|
||||
"num_tokens": 12303098.0,
|
||||
"step": 2050
|
||||
},
|
||||
{
|
||||
"entropy": 2.1386052322387696,
|
||||
"epoch": 0.3365249789671888,
|
||||
"grad_norm": 3.484375,
|
||||
"learning_rate": 1.327351385995834e-05,
|
||||
"loss": 2.149888458251953,
|
||||
"mean_token_accuracy": 0.522669050693512,
|
||||
"num_tokens": 12605534.0,
|
||||
"step": 2100
|
||||
},
|
||||
{
|
||||
"entropy": 2.102600795030594,
|
||||
"epoch": 0.3445374784664076,
|
||||
"grad_norm": 3.796875,
|
||||
"learning_rate": 1.3113283127703895e-05,
|
||||
"loss": 2.1236566162109374,
|
||||
"mean_token_accuracy": 0.5279092155396938,
|
||||
"num_tokens": 12901218.0,
|
||||
"step": 2150
|
||||
},
|
||||
{
|
||||
"entropy": 2.132645187973976,
|
||||
"epoch": 0.35254997796562637,
|
||||
"grad_norm": 3.53125,
|
||||
"learning_rate": 1.2953052395449447e-05,
|
||||
"loss": 2.1523503112792968,
|
||||
"mean_token_accuracy": 0.5198129612207413,
|
||||
"num_tokens": 13213018.0,
|
||||
"step": 2200
|
||||
},
|
||||
{
|
||||
"entropy": 2.1204321217536926,
|
||||
"epoch": 0.36056247746484515,
|
||||
"grad_norm": 3.25,
|
||||
"learning_rate": 1.2792821663195001e-05,
|
||||
"loss": 2.1325447082519533,
|
||||
"mean_token_accuracy": 0.5243636940419674,
|
||||
"num_tokens": 13514408.0,
|
||||
"step": 2250
|
||||
},
|
||||
{
|
||||
"entropy": 2.102410731315613,
|
||||
"epoch": 0.36857497696406394,
|
||||
"grad_norm": 3.1875,
|
||||
"learning_rate": 1.2632590930940555e-05,
|
||||
"loss": 2.123565368652344,
|
||||
"mean_token_accuracy": 0.5257614709436893,
|
||||
"num_tokens": 13818272.0,
|
||||
"step": 2300
|
||||
},
|
||||
{
|
||||
"entropy": 2.0876205080747603,
|
||||
"epoch": 0.3765874764632827,
|
||||
"grad_norm": 3.328125,
|
||||
"learning_rate": 1.247236019868611e-05,
|
||||
"loss": 2.116696472167969,
|
||||
"mean_token_accuracy": 0.5275124031305313,
|
||||
"num_tokens": 14110702.0,
|
||||
"step": 2350
|
||||
},
|
||||
{
|
||||
"entropy": 2.1078518223762512,
|
||||
"epoch": 0.3845999759625015,
|
||||
"grad_norm": 3.453125,
|
||||
"learning_rate": 1.2312129466431662e-05,
|
||||
"loss": 2.1291993713378905,
|
||||
"mean_token_accuracy": 0.524967464953661,
|
||||
"num_tokens": 14407015.0,
|
||||
"step": 2400
|
||||
},
|
||||
{
|
||||
"entropy": 2.0930898267030718,
|
||||
"epoch": 0.3926124754617203,
|
||||
"grad_norm": 3.5,
|
||||
"learning_rate": 1.2151898734177216e-05,
|
||||
"loss": 2.1094123840332033,
|
||||
"mean_token_accuracy": 0.5290587748587131,
|
||||
"num_tokens": 14696565.0,
|
||||
"step": 2450
|
||||
},
|
||||
{
|
||||
"entropy": 2.1109757220745085,
|
||||
"epoch": 0.4006249749609391,
|
||||
"grad_norm": 3.5,
|
||||
"learning_rate": 1.199166800192277e-05,
|
||||
"loss": 2.1246888732910154,
|
||||
"mean_token_accuracy": 0.5258957357704639,
|
||||
"num_tokens": 14994401.0,
|
||||
"step": 2500
|
||||
},
|
||||
{
|
||||
"entropy": 2.1087811160087586,
|
||||
"epoch": 0.40863747446015786,
|
||||
"grad_norm": 3.1875,
|
||||
"learning_rate": 1.1831437269668324e-05,
|
||||
"loss": 2.12890625,
|
||||
"mean_token_accuracy": 0.5269917643070221,
|
||||
"num_tokens": 15292837.0,
|
||||
"step": 2550
|
||||
},
|
||||
{
|
||||
"entropy": 2.097252429127693,
|
||||
"epoch": 0.41664997395937664,
|
||||
"grad_norm": 3.0,
|
||||
"learning_rate": 1.1671206537413878e-05,
|
||||
"loss": 2.1145870971679686,
|
||||
"mean_token_accuracy": 0.5274137698113919,
|
||||
"num_tokens": 15590647.0,
|
||||
"step": 2600
|
||||
},
|
||||
{
|
||||
"entropy": 2.111199595332146,
|
||||
"epoch": 0.4246624734585954,
|
||||
"grad_norm": 3.25,
|
||||
"learning_rate": 1.151097580515943e-05,
|
||||
"loss": 2.1274327087402343,
|
||||
"mean_token_accuracy": 0.5250249288976192,
|
||||
"num_tokens": 15892772.0,
|
||||
"step": 2650
|
||||
},
|
||||
{
|
||||
"entropy": 2.1051502913236617,
|
||||
"epoch": 0.4326749729578142,
|
||||
"grad_norm": 3.203125,
|
||||
"learning_rate": 1.1350745072904986e-05,
|
||||
"loss": 2.127935333251953,
|
||||
"mean_token_accuracy": 0.5263640700280666,
|
||||
"num_tokens": 16191611.0,
|
||||
"step": 2700
|
||||
},
|
||||
{
|
||||
"entropy": 2.1127885603904724,
|
||||
"epoch": 0.440687472457033,
|
||||
"grad_norm": 3.515625,
|
||||
"learning_rate": 1.1190514340650538e-05,
|
||||
"loss": 2.129211883544922,
|
||||
"mean_token_accuracy": 0.5250939278304577,
|
||||
"num_tokens": 16483582.0,
|
||||
"step": 2750
|
||||
},
|
||||
{
|
||||
"entropy": 2.110930626988411,
|
||||
"epoch": 0.4486999719562518,
|
||||
"grad_norm": 3.1875,
|
||||
"learning_rate": 1.103028360839609e-05,
|
||||
"loss": 2.1220079040527344,
|
||||
"mean_token_accuracy": 0.5250919266045093,
|
||||
"num_tokens": 16787253.0,
|
||||
"step": 2800
|
||||
},
|
||||
{
|
||||
"entropy": 2.1182780677080153,
|
||||
"epoch": 0.45671247145547056,
|
||||
"grad_norm": 3.8125,
|
||||
"learning_rate": 1.0870052876141646e-05,
|
||||
"loss": 2.1358837890625,
|
||||
"mean_token_accuracy": 0.5217854312062263,
|
||||
"num_tokens": 17088741.0,
|
||||
"step": 2850
|
||||
},
|
||||
{
|
||||
"entropy": 2.1117833071947096,
|
||||
"epoch": 0.46472497095468934,
|
||||
"grad_norm": 3.171875,
|
||||
"learning_rate": 1.0709822143887198e-05,
|
||||
"loss": 2.125662841796875,
|
||||
"mean_token_accuracy": 0.524649152457714,
|
||||
"num_tokens": 17389236.0,
|
||||
"step": 2900
|
||||
},
|
||||
{
|
||||
"entropy": 2.1262730568647386,
|
||||
"epoch": 0.47273747045390807,
|
||||
"grad_norm": 3.625,
|
||||
"learning_rate": 1.054959141163275e-05,
|
||||
"loss": 2.142789611816406,
|
||||
"mean_token_accuracy": 0.5236384868621826,
|
||||
"num_tokens": 17693413.0,
|
||||
"step": 2950
|
||||
},
|
||||
{
|
||||
"entropy": 2.115410739183426,
|
||||
"epoch": 0.48074996995312685,
|
||||
"grad_norm": 3.140625,
|
||||
"learning_rate": 1.0389360679378306e-05,
|
||||
"loss": 2.1337106323242185,
|
||||
"mean_token_accuracy": 0.5249222332239151,
|
||||
"num_tokens": 17990326.0,
|
||||
"step": 3000
|
||||
},
|
||||
{
|
||||
"entropy": 2.0951381689310073,
|
||||
"epoch": 0.48876246945234564,
|
||||
"grad_norm": 3.234375,
|
||||
"learning_rate": 1.0229129947123859e-05,
|
||||
"loss": 2.108144073486328,
|
||||
"mean_token_accuracy": 0.5269971822202205,
|
||||
"num_tokens": 18293136.0,
|
||||
"step": 3050
|
||||
},
|
||||
{
|
||||
"entropy": 2.116977259516716,
|
||||
"epoch": 0.4967749689515644,
|
||||
"grad_norm": 3.421875,
|
||||
"learning_rate": 1.0068899214869413e-05,
|
||||
"loss": 2.1327830505371095,
|
||||
"mean_token_accuracy": 0.5238270902633667,
|
||||
"num_tokens": 18591481.0,
|
||||
"step": 3100
|
||||
},
|
||||
{
|
||||
"entropy": 2.08598925113678,
|
||||
"epoch": 0.5047874684507833,
|
||||
"grad_norm": 2.96875,
|
||||
"learning_rate": 9.908668482614967e-06,
|
||||
"loss": 2.1032344055175782,
|
||||
"mean_token_accuracy": 0.5289211909472943,
|
||||
"num_tokens": 18885221.0,
|
||||
"step": 3150
|
||||
},
|
||||
{
|
||||
"entropy": 2.097712662220001,
|
||||
"epoch": 0.512799967950002,
|
||||
"grad_norm": 3.40625,
|
||||
"learning_rate": 9.74843775036052e-06,
|
||||
"loss": 2.1150633239746095,
|
||||
"mean_token_accuracy": 0.5288272814452648,
|
||||
"num_tokens": 19184784.0,
|
||||
"step": 3200
|
||||
},
|
||||
{
|
||||
"entropy": 2.1063760107755662,
|
||||
"epoch": 0.5208124674492208,
|
||||
"grad_norm": 3.28125,
|
||||
"learning_rate": 9.588207018106075e-06,
|
||||
"loss": 2.1217109680175783,
|
||||
"mean_token_accuracy": 0.526290989369154,
|
||||
"num_tokens": 19482538.0,
|
||||
"step": 3250
|
||||
},
|
||||
{
|
||||
"entropy": 2.090182801485062,
|
||||
"epoch": 0.5288249669484396,
|
||||
"grad_norm": 2.875,
|
||||
"learning_rate": 9.427976285851627e-06,
|
||||
"loss": 2.1130311584472654,
|
||||
"mean_token_accuracy": 0.5285439921915531,
|
||||
"num_tokens": 19782544.0,
|
||||
"step": 3300
|
||||
},
|
||||
{
|
||||
"entropy": 2.103260806798935,
|
||||
"epoch": 0.5368374664476584,
|
||||
"grad_norm": 3.59375,
|
||||
"learning_rate": 9.267745553597181e-06,
|
||||
"loss": 2.116189727783203,
|
||||
"mean_token_accuracy": 0.5273195558786392,
|
||||
"num_tokens": 20086408.0,
|
||||
"step": 3350
|
||||
},
|
||||
{
|
||||
"entropy": 2.0673747235536575,
|
||||
"epoch": 0.5448499659468772,
|
||||
"grad_norm": 3.125,
|
||||
"learning_rate": 9.107514821342735e-06,
|
||||
"loss": 2.087482147216797,
|
||||
"mean_token_accuracy": 0.532562478929758,
|
||||
"num_tokens": 20380429.0,
|
||||
"step": 3400
|
||||
},
|
||||
{
|
||||
"entropy": 2.1030789083242416,
|
||||
"epoch": 0.552862465446096,
|
||||
"grad_norm": 3.296875,
|
||||
"learning_rate": 8.947284089088289e-06,
|
||||
"loss": 2.122591247558594,
|
||||
"mean_token_accuracy": 0.5256703786551953,
|
||||
"num_tokens": 20682849.0,
|
||||
"step": 3450
|
||||
},
|
||||
{
|
||||
"entropy": 2.0988716661930082,
|
||||
"epoch": 0.5608749649453146,
|
||||
"grad_norm": 3.546875,
|
||||
"learning_rate": 8.787053356833841e-06,
|
||||
"loss": 2.115867156982422,
|
||||
"mean_token_accuracy": 0.5266730943322182,
|
||||
"num_tokens": 20979225.0,
|
||||
"step": 3500
|
||||
},
|
||||
{
|
||||
"entropy": 2.0954072701931,
|
||||
"epoch": 0.5688874644445334,
|
||||
"grad_norm": 3.859375,
|
||||
"learning_rate": 8.626822624579395e-06,
|
||||
"loss": 2.1116868591308595,
|
||||
"mean_token_accuracy": 0.5267495161294937,
|
||||
"num_tokens": 21284166.0,
|
||||
"step": 3550
|
||||
},
|
||||
{
|
||||
"entropy": 2.11238471031189,
|
||||
"epoch": 0.5768999639437522,
|
||||
"grad_norm": 3.484375,
|
||||
"learning_rate": 8.466591892324949e-06,
|
||||
"loss": 2.1292352294921875,
|
||||
"mean_token_accuracy": 0.5249243360757828,
|
||||
"num_tokens": 21584612.0,
|
||||
"step": 3600
|
||||
},
|
||||
{
|
||||
"entropy": 2.091464368700981,
|
||||
"epoch": 0.584912463442971,
|
||||
"grad_norm": 3.359375,
|
||||
"learning_rate": 8.306361160070501e-06,
|
||||
"loss": 2.1099319458007812,
|
||||
"mean_token_accuracy": 0.5292035435140133,
|
||||
"num_tokens": 21879345.0,
|
||||
"step": 3650
|
||||
},
|
||||
{
|
||||
"entropy": 2.095204224586487,
|
||||
"epoch": 0.5929249629421898,
|
||||
"grad_norm": 2.96875,
|
||||
"learning_rate": 8.146130427816055e-06,
|
||||
"loss": 2.1198716735839844,
|
||||
"mean_token_accuracy": 0.5255991271138192,
|
||||
"num_tokens": 22183486.0,
|
||||
"step": 3700
|
||||
},
|
||||
{
|
||||
"entropy": 2.0666573679447175,
|
||||
"epoch": 0.6009374624414086,
|
||||
"grad_norm": 2.90625,
|
||||
"learning_rate": 7.98589969556161e-06,
|
||||
"loss": 2.08689697265625,
|
||||
"mean_token_accuracy": 0.5318664258718491,
|
||||
"num_tokens": 22476438.0,
|
||||
"step": 3750
|
||||
},
|
||||
{
|
||||
"entropy": 2.122153177857399,
|
||||
"epoch": 0.6089499619406273,
|
||||
"grad_norm": 3.40625,
|
||||
"learning_rate": 7.825668963307162e-06,
|
||||
"loss": 2.138288879394531,
|
||||
"mean_token_accuracy": 0.5232584895193577,
|
||||
"num_tokens": 22780385.0,
|
||||
"step": 3800
|
||||
},
|
||||
{
|
||||
"entropy": 2.1190748488903046,
|
||||
"epoch": 0.6169624614398461,
|
||||
"grad_norm": 3.46875,
|
||||
"learning_rate": 7.665438231052716e-06,
|
||||
"loss": 2.139790344238281,
|
||||
"mean_token_accuracy": 0.5243287613987923,
|
||||
"num_tokens": 23080847.0,
|
||||
"step": 3850
|
||||
},
|
||||
{
|
||||
"entropy": 2.1312570214271545,
|
||||
"epoch": 0.6249749609390649,
|
||||
"grad_norm": 3.765625,
|
||||
"learning_rate": 7.5052074987982706e-06,
|
||||
"loss": 2.1471080017089843,
|
||||
"mean_token_accuracy": 0.5232046774029732,
|
||||
"num_tokens": 23384738.0,
|
||||
"step": 3900
|
||||
},
|
||||
{
|
||||
"entropy": 2.0779008239507677,
|
||||
"epoch": 0.6329874604382837,
|
||||
"grad_norm": 3.171875,
|
||||
"learning_rate": 7.344976766543823e-06,
|
||||
"loss": 2.0974729919433592,
|
||||
"mean_token_accuracy": 0.5305448272824287,
|
||||
"num_tokens": 23682750.0,
|
||||
"step": 3950
|
||||
},
|
||||
{
|
||||
"entropy": 2.1121455317735673,
|
||||
"epoch": 0.6409999599375025,
|
||||
"grad_norm": 3.921875,
|
||||
"learning_rate": 7.184746034289377e-06,
|
||||
"loss": 2.127069549560547,
|
||||
"mean_token_accuracy": 0.5252936625480652,
|
||||
"num_tokens": 23989029.0,
|
||||
"step": 4000
|
||||
},
|
||||
{
|
||||
"entropy": 2.0982888478040693,
|
||||
"epoch": 0.6490124594367213,
|
||||
"grad_norm": 4.0625,
|
||||
"learning_rate": 7.024515302034931e-06,
|
||||
"loss": 2.112042236328125,
|
||||
"mean_token_accuracy": 0.5281323495507241,
|
||||
"num_tokens": 24292063.0,
|
||||
"step": 4050
|
||||
},
|
||||
{
|
||||
"entropy": 2.098363963365555,
|
||||
"epoch": 0.65702495893594,
|
||||
"grad_norm": 3.046875,
|
||||
"learning_rate": 6.864284569780485e-06,
|
||||
"loss": 2.1156489562988283,
|
||||
"mean_token_accuracy": 0.5275120075047016,
|
||||
"num_tokens": 24590201.0,
|
||||
"step": 4100
|
||||
},
|
||||
{
|
||||
"entropy": 2.1065632277727127,
|
||||
"epoch": 0.6650374584351588,
|
||||
"grad_norm": 3.375,
|
||||
"learning_rate": 6.704053837526038e-06,
|
||||
"loss": 2.12911376953125,
|
||||
"mean_token_accuracy": 0.5262684133648873,
|
||||
"num_tokens": 24892122.0,
|
||||
"step": 4150
|
||||
},
|
||||
{
|
||||
"entropy": 2.1014072370529173,
|
||||
"epoch": 0.6730499579343776,
|
||||
"grad_norm": 2.84375,
|
||||
"learning_rate": 6.543823105271592e-06,
|
||||
"loss": 2.1130335998535155,
|
||||
"mean_token_accuracy": 0.5262750512361527,
|
||||
"num_tokens": 25194969.0,
|
||||
"step": 4200
|
||||
},
|
||||
{
|
||||
"entropy": 2.0905236744880678,
|
||||
"epoch": 0.6810624574335964,
|
||||
"grad_norm": 3.03125,
|
||||
"learning_rate": 6.383592373017146e-06,
|
||||
"loss": 2.1153533935546873,
|
||||
"mean_token_accuracy": 0.527137616276741,
|
||||
"num_tokens": 25499521.0,
|
||||
"step": 4250
|
||||
},
|
||||
{
|
||||
"entropy": 2.1032627063989637,
|
||||
"epoch": 0.6890749569328152,
|
||||
"grad_norm": 3.125,
|
||||
"learning_rate": 6.223361640762698e-06,
|
||||
"loss": 2.122831115722656,
|
||||
"mean_token_accuracy": 0.5254572249948979,
|
||||
"num_tokens": 25798933.0,
|
||||
"step": 4300
|
||||
},
|
||||
{
|
||||
"entropy": 2.118139310479164,
|
||||
"epoch": 0.697087456432034,
|
||||
"grad_norm": 3.28125,
|
||||
"learning_rate": 6.063130908508252e-06,
|
||||
"loss": 2.1376751708984374,
|
||||
"mean_token_accuracy": 0.5243131402134895,
|
||||
"num_tokens": 26096619.0,
|
||||
"step": 4350
|
||||
},
|
||||
{
|
||||
"entropy": 2.082554380297661,
|
||||
"epoch": 0.7050999559312527,
|
||||
"grad_norm": 3.390625,
|
||||
"learning_rate": 5.902900176253806e-06,
|
||||
"loss": 2.0949493408203126,
|
||||
"mean_token_accuracy": 0.5316410692036152,
|
||||
"num_tokens": 26394395.0,
|
||||
"step": 4400
|
||||
},
|
||||
{
|
||||
"entropy": 2.09870474755764,
|
||||
"epoch": 0.7131124554304715,
|
||||
"grad_norm": 3.421875,
|
||||
"learning_rate": 5.7426694439993594e-06,
|
||||
"loss": 2.1184346008300783,
|
||||
"mean_token_accuracy": 0.5267573061585427,
|
||||
"num_tokens": 26692333.0,
|
||||
"step": 4450
|
||||
},
|
||||
{
|
||||
"entropy": 2.0866885662078856,
|
||||
"epoch": 0.7211249549296903,
|
||||
"grad_norm": 3.71875,
|
||||
"learning_rate": 5.5824387117449134e-06,
|
||||
"loss": 2.102904968261719,
|
||||
"mean_token_accuracy": 0.529726449996233,
|
||||
"num_tokens": 26989384.0,
|
||||
"step": 4500
|
||||
}
|
||||
],
|
||||
"logging_steps": 50,
|
||||
"max_steps": 6241,
|
||||
"num_input_tokens_seen": 0,
|
||||
"num_train_epochs": 1,
|
||||
"save_steps": 500,
|
||||
"stateful_callbacks": {
|
||||
"TrainerControl": {
|
||||
"args": {
|
||||
"should_epoch_stop": false,
|
||||
"should_evaluate": false,
|
||||
"should_log": false,
|
||||
"should_save": true,
|
||||
"should_training_stop": false
|
||||
},
|
||||
"attributes": {}
|
||||
}
|
||||
},
|
||||
"total_flos": 3.1866411651588096e+17,
|
||||
"train_batch_size": 4,
|
||||
"trial_name": null,
|
||||
"trial_params": null
|
||||
}
|
||||
3
checkpoint-4500/training_args.bin
Normal file
3
checkpoint-4500/training_args.bin
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:78622f63304ec0913ed0b4ea7a1b26c08a5621b3661a380d798c4ea426235671
|
||||
size 5713
|
||||
89
checkpoint-500/chat_template.jinja
Normal file
89
checkpoint-500/chat_template.jinja
Normal file
@@ -0,0 +1,89 @@
|
||||
{%- if tools %}
|
||||
{{- '<|im_start|>system\n' }}
|
||||
{%- if messages[0].role == 'system' %}
|
||||
{{- messages[0].content + '\n\n' }}
|
||||
{%- endif %}
|
||||
{{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
|
||||
{%- for tool in tools %}
|
||||
{{- "\n" }}
|
||||
{{- tool | tojson }}
|
||||
{%- endfor %}
|
||||
{{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
|
||||
{%- else %}
|
||||
{%- if messages[0].role == 'system' %}
|
||||
{{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
|
||||
{%- endif %}
|
||||
{%- endif %}
|
||||
{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
|
||||
{%- for message in messages[::-1] %}
|
||||
{%- set index = (messages|length - 1) - loop.index0 %}
|
||||
{%- if ns.multi_step_tool and message.role == "user" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
|
||||
{%- set ns.multi_step_tool = false %}
|
||||
{%- set ns.last_query_index = index %}
|
||||
{%- endif %}
|
||||
{%- endfor %}
|
||||
{%- for message in messages %}
|
||||
{%- if message.content is string %}
|
||||
{%- set content = message.content %}
|
||||
{%- else %}
|
||||
{%- set content = '' %}
|
||||
{%- endif %}
|
||||
{%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
|
||||
{{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
|
||||
{%- elif message.role == "assistant" %}
|
||||
{%- set reasoning_content = '' %}
|
||||
{%- if message.reasoning_content is string %}
|
||||
{%- set reasoning_content = message.reasoning_content %}
|
||||
{%- else %}
|
||||
{%- if '</think>' in content %}
|
||||
{%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
|
||||
{%- set content = content.split('</think>')[-1].lstrip('\n') %}
|
||||
{%- endif %}
|
||||
{%- endif %}
|
||||
{%- if loop.index0 > ns.last_query_index %}
|
||||
{%- if loop.last or (not loop.last and reasoning_content) %}
|
||||
{{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
|
||||
{%- else %}
|
||||
{{- '<|im_start|>' + message.role + '\n' + content }}
|
||||
{%- endif %}
|
||||
{%- else %}
|
||||
{{- '<|im_start|>' + message.role + '\n' + content }}
|
||||
{%- endif %}
|
||||
{%- if message.tool_calls %}
|
||||
{%- for tool_call in message.tool_calls %}
|
||||
{%- if (loop.first and content) or (not loop.first) %}
|
||||
{{- '\n' }}
|
||||
{%- endif %}
|
||||
{%- if tool_call.function %}
|
||||
{%- set tool_call = tool_call.function %}
|
||||
{%- endif %}
|
||||
{{- '<tool_call>\n{"name": "' }}
|
||||
{{- tool_call.name }}
|
||||
{{- '", "arguments": ' }}
|
||||
{%- if tool_call.arguments is string %}
|
||||
{{- tool_call.arguments }}
|
||||
{%- else %}
|
||||
{{- tool_call.arguments | tojson }}
|
||||
{%- endif %}
|
||||
{{- '}\n</tool_call>' }}
|
||||
{%- endfor %}
|
||||
{%- endif %}
|
||||
{{- '<|im_end|>\n' }}
|
||||
{%- elif message.role == "tool" %}
|
||||
{%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
|
||||
{{- '<|im_start|>user' }}
|
||||
{%- endif %}
|
||||
{{- '\n<tool_response>\n' }}
|
||||
{{- content }}
|
||||
{{- '\n</tool_response>' }}
|
||||
{%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
|
||||
{{- '<|im_end|>\n' }}
|
||||
{%- endif %}
|
||||
{%- endif %}
|
||||
{%- endfor %}
|
||||
{%- if add_generation_prompt %}
|
||||
{{- '<|im_start|>assistant\n' }}
|
||||
{%- if enable_thinking is defined and enable_thinking is false %}
|
||||
{{- '<think>\n\n</think>\n\n' }}
|
||||
{%- endif %}
|
||||
{%- endif %}
|
||||
63
checkpoint-500/config.json
Normal file
63
checkpoint-500/config.json
Normal file
@@ -0,0 +1,63 @@
|
||||
{
|
||||
"architectures": [
|
||||
"Qwen3ForCausalLM"
|
||||
],
|
||||
"attention_bias": false,
|
||||
"attention_dropout": 0.0,
|
||||
"bos_token_id": null,
|
||||
"dtype": "bfloat16",
|
||||
"eos_token_id": 151645,
|
||||
"head_dim": 128,
|
||||
"hidden_act": "silu",
|
||||
"hidden_size": 2048,
|
||||
"initializer_range": 0.02,
|
||||
"intermediate_size": 6144,
|
||||
"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"
|
||||
],
|
||||
"max_position_embeddings": 40960,
|
||||
"max_window_layers": 28,
|
||||
"model_type": "qwen3",
|
||||
"num_attention_heads": 16,
|
||||
"num_hidden_layers": 28,
|
||||
"num_key_value_heads": 8,
|
||||
"pad_token_id": 151643,
|
||||
"rms_norm_eps": 1e-06,
|
||||
"rope_parameters": {
|
||||
"rope_theta": 1000000,
|
||||
"rope_type": "default"
|
||||
},
|
||||
"sliding_window": null,
|
||||
"tie_word_embeddings": true,
|
||||
"transformers_version": "5.7.0",
|
||||
"use_cache": false,
|
||||
"use_sliding_window": false,
|
||||
"vocab_size": 151936
|
||||
}
|
||||
12
checkpoint-500/generation_config.json
Normal file
12
checkpoint-500/generation_config.json
Normal file
@@ -0,0 +1,12 @@
|
||||
{
|
||||
"do_sample": true,
|
||||
"eos_token_id": [
|
||||
151645,
|
||||
151643
|
||||
],
|
||||
"pad_token_id": 151643,
|
||||
"temperature": 0.6,
|
||||
"top_k": 20,
|
||||
"top_p": 0.95,
|
||||
"transformers_version": "5.7.0"
|
||||
}
|
||||
3
checkpoint-500/model.safetensors
Normal file
3
checkpoint-500/model.safetensors
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:510bdc2405d027342acd0e2047289831efc0b3f25febcd5c74681c16385dd5c2
|
||||
size 3441185608
|
||||
3
checkpoint-500/optimizer.pt
Normal file
3
checkpoint-500/optimizer.pt
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:3466f40535b1f6a6dd5a3dfad0962141622b90ac955b344d5278efbb20cd2ba1
|
||||
size 6882572207
|
||||
3
checkpoint-500/rng_state.pth
Normal file
3
checkpoint-500/rng_state.pth
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:61c19bab1174704a4a4441475683bf1270277af15d2e2c95e964789128e482c4
|
||||
size 14645
|
||||
3
checkpoint-500/scheduler.pt
Normal file
3
checkpoint-500/scheduler.pt
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:e36e50838b6f150abbf126d12ba88d9ca9bcc1b4d6927bad9696043424255e41
|
||||
size 1465
|
||||
3
checkpoint-500/tokenizer.json
Normal file
3
checkpoint-500/tokenizer.json
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:be75606093db2094d7cd20f3c2f385c212750648bd6ea4fb2bf507a6a4c55506
|
||||
size 11422650
|
||||
30
checkpoint-500/tokenizer_config.json
Normal file
30
checkpoint-500/tokenizer_config.json
Normal file
@@ -0,0 +1,30 @@
|
||||
{
|
||||
"add_prefix_space": false,
|
||||
"backend": "tokenizers",
|
||||
"bos_token": null,
|
||||
"clean_up_tokenization_spaces": false,
|
||||
"eos_token": "<|im_end|>",
|
||||
"errors": "replace",
|
||||
"extra_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|>"
|
||||
],
|
||||
"is_local": false,
|
||||
"local_files_only": false,
|
||||
"model_max_length": 131072,
|
||||
"pad_token": "<|endoftext|>",
|
||||
"split_special_tokens": false,
|
||||
"tokenizer_class": "Qwen2Tokenizer",
|
||||
"unk_token": null
|
||||
}
|
||||
Some files were not shown because too many files have changed in this diff Show More
Reference in New Issue
Block a user