Instructions to use Nhckdvrl/cvis-ablate with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Nhckdvrl/cvis-ablate with Transformers:
# pip install -U transformers accelerate # Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("Nhckdvrl/cvis-ablate", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Add files using upload-large-folder tool
Browse files- README.md +43 -0
- a4_fixed_images/round_1/chat_template.jinja +154 -0
- a4_fixed_images/round_1/config.json +113 -0
- a4_fixed_images/round_1/generation_config.json +9 -0
- a4_fixed_images/round_1/model.safetensors.index.json +768 -0
- a4_fixed_images/round_1/preprocessor_config.json +21 -0
- a4_fixed_images/round_1/processor_config.json +60 -0
- a4_fixed_images/round_1/tokenizer_config.json +33 -0
- a4_fixed_images/round_1/trainer_state.json +2454 -0
- a4_fixed_images/round_2/chat_template.jinja +154 -0
- a4_fixed_images/round_2/config.json +113 -0
- a4_fixed_images/round_2/generation_config.json +9 -0
- a4_fixed_images/round_2/model.safetensors.index.json +768 -0
- a4_fixed_images/round_2/preprocessor_config.json +21 -0
- a4_fixed_images/round_2/processor_config.json +60 -0
- a4_fixed_images/round_2/tokenizer_config.json +33 -0
- a4_fixed_images/round_2/trainer_state.json +2454 -0
- c1_no_history/round_1/chat_template.jinja +154 -0
- c1_no_history/round_1/config.json +113 -0
- c1_no_history/round_1/generation_config.json +9 -0
- c1_no_history/round_1/model.safetensors.index.json +768 -0
- c1_no_history/round_1/preprocessor_config.json +21 -0
- c1_no_history/round_1/processor_config.json +60 -0
- c1_no_history/round_1/tokenizer_config.json +33 -0
- c1_no_history/round_1/trainer_state.json +2454 -0
- c1_no_history/round_2/chat_template.jinja +154 -0
- c1_no_history/round_2/config.json +113 -0
- c1_no_history/round_2/generation_config.json +9 -0
- c1_no_history/round_2/model.safetensors.index.json +768 -0
- c1_no_history/round_2/preprocessor_config.json +21 -0
- c1_no_history/round_2/processor_config.json +60 -0
- c1_no_history/round_2/tokenizer_config.json +33 -0
- c1_no_history/round_2/trainer_state.json +2454 -0
- c1_no_history/round_3/chat_template.jinja +154 -0
- c1_no_history/round_3/config.json +113 -0
- c1_no_history/round_3/generation_config.json +9 -0
- c1_no_history/round_3/model.safetensors.index.json +768 -0
- c1_no_history/round_3/preprocessor_config.json +21 -0
- c1_no_history/round_3/processor_config.json +60 -0
- c1_no_history/round_3/tokenizer_config.json +33 -0
- c1_no_history/round_3/trainer_state.json +2454 -0
- c1_no_history/round_4/chat_template.jinja +154 -0
- c1_no_history/round_4/config.json +113 -0
- c1_no_history/round_4/generation_config.json +9 -0
- c1_no_history/round_4/model.safetensors.index.json +768 -0
- c1_no_history/round_4/preprocessor_config.json +21 -0
- c1_no_history/round_4/processor_config.json +60 -0
- c1_no_history/round_4/tokenizer_config.json +33 -0
- c1_no_history/round_4/trainer_state.json +2454 -0
- c2_uniform_difficulty/round_1/model.safetensors.index.json +768 -0
README.md
ADDED
|
@@ -0,0 +1,43 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
tags:
|
| 3 |
+
- c-vis
|
| 4 |
+
- ablation
|
| 5 |
+
- grpo
|
| 6 |
+
- qwen3.5
|
| 7 |
+
library_name: transformers
|
| 8 |
+
---
|
| 9 |
+
|
| 10 |
+
# C-VIS ablation checkpoints (cvis_ablate)
|
| 11 |
+
|
| 12 |
+
Solver weights from the C-VIS ablation study (branch `cvis_ablate`). Each folder holds one finished round: a full Qwen3.5-9B fine-tune after 160 GRPO optimizer steps, trained with ms-swift.
|
| 13 |
+
|
| 14 |
+
Layout: `<experiment>/round_<N>/` contains the safetensors shards, config, tokenizer, processor, chat template and `trainer_state.json` (training log history). Optimizer states are not included.
|
| 15 |
+
|
| 16 |
+
| Experiment | Round | MathVision | MathVista | MMMU | Folder |
|
| 17 |
+
| --- | --- | --- | --- | --- | --- |
|
| 18 |
+
| a4_fixed_images | 1 | 50.66% | 81.80% | 68.67% | `a4_fixed_images/round_1` |
|
| 19 |
+
| a4_fixed_images | 2 | 51.64% | 82.60% | 68.44% | `a4_fixed_images/round_2` |
|
| 20 |
+
| c1_no_history | 1 | 53.29% | 82.60% | 69.22% | `c1_no_history/round_1` |
|
| 21 |
+
| c1_no_history | 2 | 53.29% | 82.40% | 69.00% | `c1_no_history/round_2` |
|
| 22 |
+
| c1_no_history | 3 | 54.28% | 83.20% | 70.11% | `c1_no_history/round_3` |
|
| 23 |
+
| c1_no_history | 4 | 50.33% | 82.80% | 70.89% | `c1_no_history/round_4` |
|
| 24 |
+
| c2_uniform_difficulty | 1 | 52.63% | 81.80% | 70.33% | `c2_uniform_difficulty/round_1` |
|
| 25 |
+
| c3_uniform_category | 1 | 53.29% | 83.70% | 70.11% | `c3_uniform_category/round_1` |
|
| 26 |
+
| c3_uniform_category | 2 | 53.95% | 81.80% | 70.67% | `c3_uniform_category/round_2` |
|
| 27 |
+
| c3_uniform_category | 3 | 53.95% | 82.80% | 69.78% | `c3_uniform_category/round_3` |
|
| 28 |
+
| c4_no_difficulty_filter | 1 | 54.93% | 82.00% | 69.78% | `c4_no_difficulty_filter/round_1` |
|
| 29 |
+
| c4_no_difficulty_filter | 2 | 56.25% | 82.10% | 69.00% | `c4_no_difficulty_filter/round_2` |
|
| 30 |
+
| c4_no_difficulty_filter | 3 | 55.59% | 83.20% | 70.00% | `c4_no_difficulty_filter/round_3` |
|
| 31 |
+
| c4_no_difficulty_filter | 4 | 57.57% | 82.30% | 70.11% | `c4_no_difficulty_filter/round_4` |
|
| 32 |
+
|
| 33 |
+
Experiments:
|
| 34 |
+
|
| 35 |
+
- `c1_no_history`: curriculum without history.
|
| 36 |
+
- `c2_uniform_difficulty`: uniform difficulty curriculum.
|
| 37 |
+
- `c3_uniform_category`: uniform category curriculum.
|
| 38 |
+
- `c4_no_difficulty_filter`: no difficulty filter.
|
| 39 |
+
- `a4_fixed_images`: fixed image pool.
|
| 40 |
+
|
| 41 |
+
Scores are the full benchmark splits used by the C-VIS evaluation (MathVision 304, MathVista 1000, MMMU 900), sampled at temperature 0.7. More rounds are added as runs finish.
|
| 42 |
+
|
| 43 |
+
Load one round with `from_pretrained("<repo>", subfolder="c4_no_difficulty_filter/round_4")`.
|
a4_fixed_images/round_1/chat_template.jinja
ADDED
|
@@ -0,0 +1,154 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{%- set image_count = namespace(value=0) %}
|
| 2 |
+
{%- set video_count = namespace(value=0) %}
|
| 3 |
+
{%- macro render_content(content, do_vision_count, is_system_content=false) %}
|
| 4 |
+
{%- if content is string %}
|
| 5 |
+
{{- content }}
|
| 6 |
+
{%- elif content is iterable and content is not mapping %}
|
| 7 |
+
{%- for item in content %}
|
| 8 |
+
{%- if 'image' in item or 'image_url' in item or item.type == 'image' %}
|
| 9 |
+
{%- if is_system_content %}
|
| 10 |
+
{{- raise_exception('System message cannot contain images.') }}
|
| 11 |
+
{%- endif %}
|
| 12 |
+
{%- if do_vision_count %}
|
| 13 |
+
{%- set image_count.value = image_count.value + 1 %}
|
| 14 |
+
{%- endif %}
|
| 15 |
+
{%- if add_vision_id %}
|
| 16 |
+
{{- 'Picture ' ~ image_count.value ~ ': ' }}
|
| 17 |
+
{%- endif %}
|
| 18 |
+
{{- '<|vision_start|><|image_pad|><|vision_end|>' }}
|
| 19 |
+
{%- elif 'video' in item or item.type == 'video' %}
|
| 20 |
+
{%- if is_system_content %}
|
| 21 |
+
{{- raise_exception('System message cannot contain videos.') }}
|
| 22 |
+
{%- endif %}
|
| 23 |
+
{%- if do_vision_count %}
|
| 24 |
+
{%- set video_count.value = video_count.value + 1 %}
|
| 25 |
+
{%- endif %}
|
| 26 |
+
{%- if add_vision_id %}
|
| 27 |
+
{{- 'Video ' ~ video_count.value ~ ': ' }}
|
| 28 |
+
{%- endif %}
|
| 29 |
+
{{- '<|vision_start|><|video_pad|><|vision_end|>' }}
|
| 30 |
+
{%- elif 'text' in item %}
|
| 31 |
+
{{- item.text }}
|
| 32 |
+
{%- else %}
|
| 33 |
+
{{- raise_exception('Unexpected item type in content.') }}
|
| 34 |
+
{%- endif %}
|
| 35 |
+
{%- endfor %}
|
| 36 |
+
{%- elif content is none or content is undefined %}
|
| 37 |
+
{{- '' }}
|
| 38 |
+
{%- else %}
|
| 39 |
+
{{- raise_exception('Unexpected content type.') }}
|
| 40 |
+
{%- endif %}
|
| 41 |
+
{%- endmacro %}
|
| 42 |
+
{%- if not messages %}
|
| 43 |
+
{{- raise_exception('No messages provided.') }}
|
| 44 |
+
{%- endif %}
|
| 45 |
+
{%- if tools and tools is iterable and tools is not mapping %}
|
| 46 |
+
{{- '<|im_start|>system\n' }}
|
| 47 |
+
{{- "# Tools\n\nYou have access to the following functions:\n\n<tools>" }}
|
| 48 |
+
{%- for tool in tools %}
|
| 49 |
+
{{- "\n" }}
|
| 50 |
+
{{- tool | tojson }}
|
| 51 |
+
{%- endfor %}
|
| 52 |
+
{{- "\n</tools>" }}
|
| 53 |
+
{{- '\n\nIf you choose to call a function ONLY reply in the following format with NO suffix:\n\n<tool_call>\n<function=example_function_name>\n<parameter=example_parameter_1>\nvalue_1\n</parameter>\n<parameter=example_parameter_2>\nThis is the value for the second parameter\nthat can span\nmultiple lines\n</parameter>\n</function>\n</tool_call>\n\n<IMPORTANT>\nReminder:\n- Function calls MUST follow the specified format: an inner <function=...></function> block must be nested within <tool_call></tool_call> XML tags\n- Required parameters MUST be specified\n- You may provide optional reasoning for your function call in natural language BEFORE the function call, but NOT after\n- If there is no function call available, answer the question like normal with your current knowledge and do not tell the user about function calls\n</IMPORTANT>' }}
|
| 54 |
+
{%- if messages[0].role == 'system' %}
|
| 55 |
+
{%- set content = render_content(messages[0].content, false, true)|trim %}
|
| 56 |
+
{%- if content %}
|
| 57 |
+
{{- '\n\n' + content }}
|
| 58 |
+
{%- endif %}
|
| 59 |
+
{%- endif %}
|
| 60 |
+
{{- '<|im_end|>\n' }}
|
| 61 |
+
{%- else %}
|
| 62 |
+
{%- if messages[0].role == 'system' %}
|
| 63 |
+
{%- set content = render_content(messages[0].content, false, true)|trim %}
|
| 64 |
+
{{- '<|im_start|>system\n' + content + '<|im_end|>\n' }}
|
| 65 |
+
{%- endif %}
|
| 66 |
+
{%- endif %}
|
| 67 |
+
{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
|
| 68 |
+
{%- for message in messages[::-1] %}
|
| 69 |
+
{%- set index = (messages|length - 1) - loop.index0 %}
|
| 70 |
+
{%- if ns.multi_step_tool and message.role == "user" %}
|
| 71 |
+
{%- set content = render_content(message.content, false)|trim %}
|
| 72 |
+
{%- if not(content.startswith('<tool_response>') and content.endswith('</tool_response>')) %}
|
| 73 |
+
{%- set ns.multi_step_tool = false %}
|
| 74 |
+
{%- set ns.last_query_index = index %}
|
| 75 |
+
{%- endif %}
|
| 76 |
+
{%- endif %}
|
| 77 |
+
{%- endfor %}
|
| 78 |
+
{%- if ns.multi_step_tool %}
|
| 79 |
+
{{- raise_exception('No user query found in messages.') }}
|
| 80 |
+
{%- endif %}
|
| 81 |
+
{%- for message in messages %}
|
| 82 |
+
{%- set content = render_content(message.content, true)|trim %}
|
| 83 |
+
{%- if message.role == "system" %}
|
| 84 |
+
{%- if not loop.first %}
|
| 85 |
+
{{- raise_exception('System message must be at the beginning.') }}
|
| 86 |
+
{%- endif %}
|
| 87 |
+
{%- elif message.role == "user" %}
|
| 88 |
+
{{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
|
| 89 |
+
{%- elif message.role == "assistant" %}
|
| 90 |
+
{%- set reasoning_content = '' %}
|
| 91 |
+
{%- if message.reasoning_content is string %}
|
| 92 |
+
{%- set reasoning_content = message.reasoning_content %}
|
| 93 |
+
{%- else %}
|
| 94 |
+
{%- if '</think>' in content %}
|
| 95 |
+
{%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
|
| 96 |
+
{%- set content = content.split('</think>')[-1].lstrip('\n') %}
|
| 97 |
+
{%- endif %}
|
| 98 |
+
{%- endif %}
|
| 99 |
+
{%- set reasoning_content = reasoning_content|trim %}
|
| 100 |
+
{%- if loop.index0 > ns.last_query_index %}
|
| 101 |
+
{{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content + '\n</think>\n\n' + content }}
|
| 102 |
+
{%- else %}
|
| 103 |
+
{{- '<|im_start|>' + message.role + '\n' + content }}
|
| 104 |
+
{%- endif %}
|
| 105 |
+
{%- if message.tool_calls and message.tool_calls is iterable and message.tool_calls is not mapping %}
|
| 106 |
+
{%- for tool_call in message.tool_calls %}
|
| 107 |
+
{%- if tool_call.function is defined %}
|
| 108 |
+
{%- set tool_call = tool_call.function %}
|
| 109 |
+
{%- endif %}
|
| 110 |
+
{%- if loop.first %}
|
| 111 |
+
{%- if content|trim %}
|
| 112 |
+
{{- '\n\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
|
| 113 |
+
{%- else %}
|
| 114 |
+
{{- '<tool_call>\n<function=' + tool_call.name + '>\n' }}
|
| 115 |
+
{%- endif %}
|
| 116 |
+
{%- else %}
|
| 117 |
+
{{- '\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
|
| 118 |
+
{%- endif %}
|
| 119 |
+
{%- if tool_call.arguments is defined %}
|
| 120 |
+
{%- for args_name, args_value in tool_call.arguments|items %}
|
| 121 |
+
{{- '<parameter=' + args_name + '>\n' }}
|
| 122 |
+
{%- set args_value = args_value | tojson | safe if args_value is mapping or (args_value is sequence and args_value is not string) else args_value | string %}
|
| 123 |
+
{{- args_value }}
|
| 124 |
+
{{- '\n</parameter>\n' }}
|
| 125 |
+
{%- endfor %}
|
| 126 |
+
{%- endif %}
|
| 127 |
+
{{- '</function>\n</tool_call>' }}
|
| 128 |
+
{%- endfor %}
|
| 129 |
+
{%- endif %}
|
| 130 |
+
{{- '<|im_end|>\n' }}
|
| 131 |
+
{%- elif message.role == "tool" %}
|
| 132 |
+
{%- if loop.previtem and loop.previtem.role != "tool" %}
|
| 133 |
+
{{- '<|im_start|>user' }}
|
| 134 |
+
{%- endif %}
|
| 135 |
+
{{- '\n<tool_response>\n' }}
|
| 136 |
+
{{- content }}
|
| 137 |
+
{{- '\n</tool_response>' }}
|
| 138 |
+
{%- if not loop.last and loop.nextitem.role != "tool" %}
|
| 139 |
+
{{- '<|im_end|>\n' }}
|
| 140 |
+
{%- elif loop.last %}
|
| 141 |
+
{{- '<|im_end|>\n' }}
|
| 142 |
+
{%- endif %}
|
| 143 |
+
{%- else %}
|
| 144 |
+
{{- raise_exception('Unexpected message role.') }}
|
| 145 |
+
{%- endif %}
|
| 146 |
+
{%- endfor %}
|
| 147 |
+
{%- if add_generation_prompt %}
|
| 148 |
+
{{- '<|im_start|>assistant\n' }}
|
| 149 |
+
{%- if enable_thinking is defined and enable_thinking is false %}
|
| 150 |
+
{{- '<think>\n\n</think>\n\n' }}
|
| 151 |
+
{%- else %}
|
| 152 |
+
{{- '<think>\n' }}
|
| 153 |
+
{%- endif %}
|
| 154 |
+
{%- endif %}
|
a4_fixed_images/round_1/config.json
ADDED
|
@@ -0,0 +1,113 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"Qwen3_5ForConditionalGeneration"
|
| 4 |
+
],
|
| 5 |
+
"dtype": "bfloat16",
|
| 6 |
+
"eos_token_id": 248046,
|
| 7 |
+
"hidden_size": 4096,
|
| 8 |
+
"image_token_id": 248056,
|
| 9 |
+
"model_type": "qwen3_5",
|
| 10 |
+
"pad_token_id": 248044,
|
| 11 |
+
"text_config": {
|
| 12 |
+
"attention_bias": false,
|
| 13 |
+
"attention_dropout": 0.0,
|
| 14 |
+
"attn_output_gate": true,
|
| 15 |
+
"bos_token_id": null,
|
| 16 |
+
"dtype": "bfloat16",
|
| 17 |
+
"eos_token_id": 248044,
|
| 18 |
+
"full_attention_interval": 4,
|
| 19 |
+
"head_dim": 256,
|
| 20 |
+
"hidden_act": "silu",
|
| 21 |
+
"hidden_size": 4096,
|
| 22 |
+
"initializer_range": 0.02,
|
| 23 |
+
"intermediate_size": 12288,
|
| 24 |
+
"layer_types": [
|
| 25 |
+
"linear_attention",
|
| 26 |
+
"linear_attention",
|
| 27 |
+
"linear_attention",
|
| 28 |
+
"full_attention",
|
| 29 |
+
"linear_attention",
|
| 30 |
+
"linear_attention",
|
| 31 |
+
"linear_attention",
|
| 32 |
+
"full_attention",
|
| 33 |
+
"linear_attention",
|
| 34 |
+
"linear_attention",
|
| 35 |
+
"linear_attention",
|
| 36 |
+
"full_attention",
|
| 37 |
+
"linear_attention",
|
| 38 |
+
"linear_attention",
|
| 39 |
+
"linear_attention",
|
| 40 |
+
"full_attention",
|
| 41 |
+
"linear_attention",
|
| 42 |
+
"linear_attention",
|
| 43 |
+
"linear_attention",
|
| 44 |
+
"full_attention",
|
| 45 |
+
"linear_attention",
|
| 46 |
+
"linear_attention",
|
| 47 |
+
"linear_attention",
|
| 48 |
+
"full_attention",
|
| 49 |
+
"linear_attention",
|
| 50 |
+
"linear_attention",
|
| 51 |
+
"linear_attention",
|
| 52 |
+
"full_attention",
|
| 53 |
+
"linear_attention",
|
| 54 |
+
"linear_attention",
|
| 55 |
+
"linear_attention",
|
| 56 |
+
"full_attention"
|
| 57 |
+
],
|
| 58 |
+
"linear_conv_kernel_dim": 4,
|
| 59 |
+
"linear_key_head_dim": 128,
|
| 60 |
+
"linear_num_key_heads": 16,
|
| 61 |
+
"linear_num_value_heads": 32,
|
| 62 |
+
"linear_value_head_dim": 128,
|
| 63 |
+
"mamba_ssm_dtype": "float32",
|
| 64 |
+
"max_position_embeddings": 262144,
|
| 65 |
+
"mlp_only_layers": [],
|
| 66 |
+
"model_type": "qwen3_5_text",
|
| 67 |
+
"mtp_num_hidden_layers": 1,
|
| 68 |
+
"mtp_use_dedicated_embeddings": false,
|
| 69 |
+
"num_attention_heads": 16,
|
| 70 |
+
"num_hidden_layers": 32,
|
| 71 |
+
"num_key_value_heads": 4,
|
| 72 |
+
"pad_token_id": 248044,
|
| 73 |
+
"partial_rotary_factor": 0.25,
|
| 74 |
+
"rms_norm_eps": 1e-06,
|
| 75 |
+
"rope_parameters": {
|
| 76 |
+
"mrope_interleaved": true,
|
| 77 |
+
"mrope_section": [
|
| 78 |
+
11,
|
| 79 |
+
11,
|
| 80 |
+
10
|
| 81 |
+
],
|
| 82 |
+
"partial_rotary_factor": 0.25,
|
| 83 |
+
"rope_theta": 10000000,
|
| 84 |
+
"rope_type": "default"
|
| 85 |
+
},
|
| 86 |
+
"tie_word_embeddings": false,
|
| 87 |
+
"use_cache": false,
|
| 88 |
+
"vocab_size": 248320
|
| 89 |
+
},
|
| 90 |
+
"tie_word_embeddings": false,
|
| 91 |
+
"transformers_version": "5.10.4",
|
| 92 |
+
"use_cache": false,
|
| 93 |
+
"video_token_id": 248057,
|
| 94 |
+
"vision_config": {
|
| 95 |
+
"deepstack_visual_indexes": [],
|
| 96 |
+
"depth": 27,
|
| 97 |
+
"dtype": "bfloat16",
|
| 98 |
+
"hidden_act": "gelu_pytorch_tanh",
|
| 99 |
+
"hidden_size": 1152,
|
| 100 |
+
"in_channels": 3,
|
| 101 |
+
"initializer_range": 0.02,
|
| 102 |
+
"intermediate_size": 4304,
|
| 103 |
+
"model_type": "qwen3_5_vision",
|
| 104 |
+
"num_heads": 16,
|
| 105 |
+
"num_position_embeddings": 2304,
|
| 106 |
+
"out_hidden_size": 4096,
|
| 107 |
+
"patch_size": 16,
|
| 108 |
+
"spatial_merge_size": 2,
|
| 109 |
+
"temporal_patch_size": 2
|
| 110 |
+
},
|
| 111 |
+
"vision_end_token_id": 248054,
|
| 112 |
+
"vision_start_token_id": 248053
|
| 113 |
+
}
|
a4_fixed_images/round_1/generation_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"eos_token_id": [
|
| 4 |
+
248044,
|
| 5 |
+
248046
|
| 6 |
+
],
|
| 7 |
+
"transformers_version": "5.10.4",
|
| 8 |
+
"use_cache": true
|
| 9 |
+
}
|
a4_fixed_images/round_1/model.safetensors.index.json
ADDED
|
@@ -0,0 +1,768 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"metadata": {
|
| 3 |
+
"total_parameters": 1469680,
|
| 4 |
+
"total_size": 18819627488
|
| 5 |
+
},
|
| 6 |
+
"weight_map": {
|
| 7 |
+
"lm_head.weight": "model-00001-of-00004.safetensors",
|
| 8 |
+
"model.language_model.embed_tokens.weight": "model-00001-of-00004.safetensors",
|
| 9 |
+
"model.language_model.layers.0.input_layernorm.weight": "model-00001-of-00004.safetensors",
|
| 10 |
+
"model.language_model.layers.0.linear_attn.A_log": "model-00001-of-00004.safetensors",
|
| 11 |
+
"model.language_model.layers.0.linear_attn.conv1d.weight": "model-00001-of-00004.safetensors",
|
| 12 |
+
"model.language_model.layers.0.linear_attn.dt_bias": "model-00001-of-00004.safetensors",
|
| 13 |
+
"model.language_model.layers.0.linear_attn.in_proj_a.weight": "model-00001-of-00004.safetensors",
|
| 14 |
+
"model.language_model.layers.0.linear_attn.in_proj_b.weight": "model-00001-of-00004.safetensors",
|
| 15 |
+
"model.language_model.layers.0.linear_attn.in_proj_qkv.weight": "model-00001-of-00004.safetensors",
|
| 16 |
+
"model.language_model.layers.0.linear_attn.in_proj_z.weight": "model-00001-of-00004.safetensors",
|
| 17 |
+
"model.language_model.layers.0.linear_attn.norm.weight": "model-00001-of-00004.safetensors",
|
| 18 |
+
"model.language_model.layers.0.linear_attn.out_proj.weight": "model-00001-of-00004.safetensors",
|
| 19 |
+
"model.language_model.layers.0.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
|
| 20 |
+
"model.language_model.layers.0.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
|
| 21 |
+
"model.language_model.layers.0.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
|
| 22 |
+
"model.language_model.layers.0.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
|
| 23 |
+
"model.language_model.layers.1.input_layernorm.weight": "model-00001-of-00004.safetensors",
|
| 24 |
+
"model.language_model.layers.1.linear_attn.A_log": "model-00001-of-00004.safetensors",
|
| 25 |
+
"model.language_model.layers.1.linear_attn.conv1d.weight": "model-00001-of-00004.safetensors",
|
| 26 |
+
"model.language_model.layers.1.linear_attn.dt_bias": "model-00001-of-00004.safetensors",
|
| 27 |
+
"model.language_model.layers.1.linear_attn.in_proj_a.weight": "model-00001-of-00004.safetensors",
|
| 28 |
+
"model.language_model.layers.1.linear_attn.in_proj_b.weight": "model-00001-of-00004.safetensors",
|
| 29 |
+
"model.language_model.layers.1.linear_attn.in_proj_qkv.weight": "model-00001-of-00004.safetensors",
|
| 30 |
+
"model.language_model.layers.1.linear_attn.in_proj_z.weight": "model-00001-of-00004.safetensors",
|
| 31 |
+
"model.language_model.layers.1.linear_attn.norm.weight": "model-00001-of-00004.safetensors",
|
| 32 |
+
"model.language_model.layers.1.linear_attn.out_proj.weight": "model-00001-of-00004.safetensors",
|
| 33 |
+
"model.language_model.layers.1.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
|
| 34 |
+
"model.language_model.layers.1.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
|
| 35 |
+
"model.language_model.layers.1.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
|
| 36 |
+
"model.language_model.layers.1.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
|
| 37 |
+
"model.language_model.layers.10.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 38 |
+
"model.language_model.layers.10.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 39 |
+
"model.language_model.layers.10.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 40 |
+
"model.language_model.layers.10.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 41 |
+
"model.language_model.layers.10.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 42 |
+
"model.language_model.layers.10.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 43 |
+
"model.language_model.layers.10.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 44 |
+
"model.language_model.layers.10.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 45 |
+
"model.language_model.layers.10.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 46 |
+
"model.language_model.layers.10.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 47 |
+
"model.language_model.layers.10.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 48 |
+
"model.language_model.layers.10.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 49 |
+
"model.language_model.layers.10.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 50 |
+
"model.language_model.layers.10.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 51 |
+
"model.language_model.layers.11.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 52 |
+
"model.language_model.layers.11.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 53 |
+
"model.language_model.layers.11.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 54 |
+
"model.language_model.layers.11.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 55 |
+
"model.language_model.layers.11.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 56 |
+
"model.language_model.layers.11.self_attn.k_norm.weight": "model-00002-of-00004.safetensors",
|
| 57 |
+
"model.language_model.layers.11.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
|
| 58 |
+
"model.language_model.layers.11.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
|
| 59 |
+
"model.language_model.layers.11.self_attn.q_norm.weight": "model-00002-of-00004.safetensors",
|
| 60 |
+
"model.language_model.layers.11.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
|
| 61 |
+
"model.language_model.layers.11.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
|
| 62 |
+
"model.language_model.layers.12.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 63 |
+
"model.language_model.layers.12.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 64 |
+
"model.language_model.layers.12.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 65 |
+
"model.language_model.layers.12.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 66 |
+
"model.language_model.layers.12.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 67 |
+
"model.language_model.layers.12.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 68 |
+
"model.language_model.layers.12.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 69 |
+
"model.language_model.layers.12.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 70 |
+
"model.language_model.layers.12.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 71 |
+
"model.language_model.layers.12.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 72 |
+
"model.language_model.layers.12.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 73 |
+
"model.language_model.layers.12.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 74 |
+
"model.language_model.layers.12.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 75 |
+
"model.language_model.layers.12.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 76 |
+
"model.language_model.layers.13.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 77 |
+
"model.language_model.layers.13.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 78 |
+
"model.language_model.layers.13.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 79 |
+
"model.language_model.layers.13.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 80 |
+
"model.language_model.layers.13.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 81 |
+
"model.language_model.layers.13.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 82 |
+
"model.language_model.layers.13.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 83 |
+
"model.language_model.layers.13.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 84 |
+
"model.language_model.layers.13.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 85 |
+
"model.language_model.layers.13.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 86 |
+
"model.language_model.layers.13.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 87 |
+
"model.language_model.layers.13.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 88 |
+
"model.language_model.layers.13.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 89 |
+
"model.language_model.layers.13.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 90 |
+
"model.language_model.layers.14.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 91 |
+
"model.language_model.layers.14.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 92 |
+
"model.language_model.layers.14.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 93 |
+
"model.language_model.layers.14.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 94 |
+
"model.language_model.layers.14.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 95 |
+
"model.language_model.layers.14.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 96 |
+
"model.language_model.layers.14.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 97 |
+
"model.language_model.layers.14.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 98 |
+
"model.language_model.layers.14.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 99 |
+
"model.language_model.layers.14.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 100 |
+
"model.language_model.layers.14.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 101 |
+
"model.language_model.layers.14.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 102 |
+
"model.language_model.layers.14.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 103 |
+
"model.language_model.layers.14.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 104 |
+
"model.language_model.layers.15.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 105 |
+
"model.language_model.layers.15.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 106 |
+
"model.language_model.layers.15.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 107 |
+
"model.language_model.layers.15.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 108 |
+
"model.language_model.layers.15.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 109 |
+
"model.language_model.layers.15.self_attn.k_norm.weight": "model-00003-of-00004.safetensors",
|
| 110 |
+
"model.language_model.layers.15.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
|
| 111 |
+
"model.language_model.layers.15.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
|
| 112 |
+
"model.language_model.layers.15.self_attn.q_norm.weight": "model-00003-of-00004.safetensors",
|
| 113 |
+
"model.language_model.layers.15.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
|
| 114 |
+
"model.language_model.layers.15.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
|
| 115 |
+
"model.language_model.layers.16.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 116 |
+
"model.language_model.layers.16.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 117 |
+
"model.language_model.layers.16.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 118 |
+
"model.language_model.layers.16.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 119 |
+
"model.language_model.layers.16.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 120 |
+
"model.language_model.layers.16.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 121 |
+
"model.language_model.layers.16.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 122 |
+
"model.language_model.layers.16.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 123 |
+
"model.language_model.layers.16.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 124 |
+
"model.language_model.layers.16.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 125 |
+
"model.language_model.layers.16.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 126 |
+
"model.language_model.layers.16.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 127 |
+
"model.language_model.layers.16.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 128 |
+
"model.language_model.layers.16.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 129 |
+
"model.language_model.layers.17.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 130 |
+
"model.language_model.layers.17.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 131 |
+
"model.language_model.layers.17.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 132 |
+
"model.language_model.layers.17.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 133 |
+
"model.language_model.layers.17.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 134 |
+
"model.language_model.layers.17.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 135 |
+
"model.language_model.layers.17.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 136 |
+
"model.language_model.layers.17.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 137 |
+
"model.language_model.layers.17.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 138 |
+
"model.language_model.layers.17.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 139 |
+
"model.language_model.layers.17.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 140 |
+
"model.language_model.layers.17.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 141 |
+
"model.language_model.layers.17.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 142 |
+
"model.language_model.layers.17.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 143 |
+
"model.language_model.layers.18.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 144 |
+
"model.language_model.layers.18.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 145 |
+
"model.language_model.layers.18.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 146 |
+
"model.language_model.layers.18.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 147 |
+
"model.language_model.layers.18.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 148 |
+
"model.language_model.layers.18.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 149 |
+
"model.language_model.layers.18.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 150 |
+
"model.language_model.layers.18.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 151 |
+
"model.language_model.layers.18.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 152 |
+
"model.language_model.layers.18.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 153 |
+
"model.language_model.layers.18.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 154 |
+
"model.language_model.layers.18.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 155 |
+
"model.language_model.layers.18.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 156 |
+
"model.language_model.layers.18.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 157 |
+
"model.language_model.layers.19.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 158 |
+
"model.language_model.layers.19.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 159 |
+
"model.language_model.layers.19.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 160 |
+
"model.language_model.layers.19.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 161 |
+
"model.language_model.layers.19.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 162 |
+
"model.language_model.layers.19.self_attn.k_norm.weight": "model-00003-of-00004.safetensors",
|
| 163 |
+
"model.language_model.layers.19.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
|
| 164 |
+
"model.language_model.layers.19.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
|
| 165 |
+
"model.language_model.layers.19.self_attn.q_norm.weight": "model-00003-of-00004.safetensors",
|
| 166 |
+
"model.language_model.layers.19.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
|
| 167 |
+
"model.language_model.layers.19.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
|
| 168 |
+
"model.language_model.layers.2.input_layernorm.weight": "model-00001-of-00004.safetensors",
|
| 169 |
+
"model.language_model.layers.2.linear_attn.A_log": "model-00001-of-00004.safetensors",
|
| 170 |
+
"model.language_model.layers.2.linear_attn.conv1d.weight": "model-00001-of-00004.safetensors",
|
| 171 |
+
"model.language_model.layers.2.linear_attn.dt_bias": "model-00001-of-00004.safetensors",
|
| 172 |
+
"model.language_model.layers.2.linear_attn.in_proj_a.weight": "model-00001-of-00004.safetensors",
|
| 173 |
+
"model.language_model.layers.2.linear_attn.in_proj_b.weight": "model-00001-of-00004.safetensors",
|
| 174 |
+
"model.language_model.layers.2.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 175 |
+
"model.language_model.layers.2.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 176 |
+
"model.language_model.layers.2.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 177 |
+
"model.language_model.layers.2.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 178 |
+
"model.language_model.layers.2.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 179 |
+
"model.language_model.layers.2.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 180 |
+
"model.language_model.layers.2.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 181 |
+
"model.language_model.layers.2.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 182 |
+
"model.language_model.layers.20.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 183 |
+
"model.language_model.layers.20.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 184 |
+
"model.language_model.layers.20.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 185 |
+
"model.language_model.layers.20.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 186 |
+
"model.language_model.layers.20.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 187 |
+
"model.language_model.layers.20.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 188 |
+
"model.language_model.layers.20.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 189 |
+
"model.language_model.layers.20.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 190 |
+
"model.language_model.layers.20.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 191 |
+
"model.language_model.layers.20.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 192 |
+
"model.language_model.layers.20.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 193 |
+
"model.language_model.layers.20.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 194 |
+
"model.language_model.layers.20.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 195 |
+
"model.language_model.layers.20.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 196 |
+
"model.language_model.layers.21.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 197 |
+
"model.language_model.layers.21.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 198 |
+
"model.language_model.layers.21.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 199 |
+
"model.language_model.layers.21.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 200 |
+
"model.language_model.layers.21.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 201 |
+
"model.language_model.layers.21.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 202 |
+
"model.language_model.layers.21.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 203 |
+
"model.language_model.layers.21.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 204 |
+
"model.language_model.layers.21.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 205 |
+
"model.language_model.layers.21.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 206 |
+
"model.language_model.layers.21.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 207 |
+
"model.language_model.layers.21.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 208 |
+
"model.language_model.layers.21.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 209 |
+
"model.language_model.layers.21.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 210 |
+
"model.language_model.layers.22.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 211 |
+
"model.language_model.layers.22.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 212 |
+
"model.language_model.layers.22.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 213 |
+
"model.language_model.layers.22.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 214 |
+
"model.language_model.layers.22.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 215 |
+
"model.language_model.layers.22.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 216 |
+
"model.language_model.layers.22.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 217 |
+
"model.language_model.layers.22.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 218 |
+
"model.language_model.layers.22.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 219 |
+
"model.language_model.layers.22.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 220 |
+
"model.language_model.layers.22.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 221 |
+
"model.language_model.layers.22.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 222 |
+
"model.language_model.layers.22.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 223 |
+
"model.language_model.layers.22.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 224 |
+
"model.language_model.layers.23.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 225 |
+
"model.language_model.layers.23.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 226 |
+
"model.language_model.layers.23.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 227 |
+
"model.language_model.layers.23.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 228 |
+
"model.language_model.layers.23.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 229 |
+
"model.language_model.layers.23.self_attn.k_norm.weight": "model-00003-of-00004.safetensors",
|
| 230 |
+
"model.language_model.layers.23.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
|
| 231 |
+
"model.language_model.layers.23.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
|
| 232 |
+
"model.language_model.layers.23.self_attn.q_norm.weight": "model-00003-of-00004.safetensors",
|
| 233 |
+
"model.language_model.layers.23.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
|
| 234 |
+
"model.language_model.layers.23.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
|
| 235 |
+
"model.language_model.layers.24.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 236 |
+
"model.language_model.layers.24.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 237 |
+
"model.language_model.layers.24.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 238 |
+
"model.language_model.layers.24.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 239 |
+
"model.language_model.layers.24.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 240 |
+
"model.language_model.layers.24.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 241 |
+
"model.language_model.layers.24.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 242 |
+
"model.language_model.layers.24.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 243 |
+
"model.language_model.layers.24.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 244 |
+
"model.language_model.layers.24.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 245 |
+
"model.language_model.layers.24.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 246 |
+
"model.language_model.layers.24.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 247 |
+
"model.language_model.layers.24.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 248 |
+
"model.language_model.layers.24.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 249 |
+
"model.language_model.layers.25.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 250 |
+
"model.language_model.layers.25.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 251 |
+
"model.language_model.layers.25.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 252 |
+
"model.language_model.layers.25.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 253 |
+
"model.language_model.layers.25.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 254 |
+
"model.language_model.layers.25.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 255 |
+
"model.language_model.layers.25.linear_attn.in_proj_qkv.weight": "model-00004-of-00004.safetensors",
|
| 256 |
+
"model.language_model.layers.25.linear_attn.in_proj_z.weight": "model-00004-of-00004.safetensors",
|
| 257 |
+
"model.language_model.layers.25.linear_attn.norm.weight": "model-00004-of-00004.safetensors",
|
| 258 |
+
"model.language_model.layers.25.linear_attn.out_proj.weight": "model-00004-of-00004.safetensors",
|
| 259 |
+
"model.language_model.layers.25.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 260 |
+
"model.language_model.layers.25.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 261 |
+
"model.language_model.layers.25.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 262 |
+
"model.language_model.layers.25.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 263 |
+
"model.language_model.layers.26.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 264 |
+
"model.language_model.layers.26.linear_attn.A_log": "model-00004-of-00004.safetensors",
|
| 265 |
+
"model.language_model.layers.26.linear_attn.conv1d.weight": "model-00004-of-00004.safetensors",
|
| 266 |
+
"model.language_model.layers.26.linear_attn.dt_bias": "model-00004-of-00004.safetensors",
|
| 267 |
+
"model.language_model.layers.26.linear_attn.in_proj_a.weight": "model-00004-of-00004.safetensors",
|
| 268 |
+
"model.language_model.layers.26.linear_attn.in_proj_b.weight": "model-00004-of-00004.safetensors",
|
| 269 |
+
"model.language_model.layers.26.linear_attn.in_proj_qkv.weight": "model-00004-of-00004.safetensors",
|
| 270 |
+
"model.language_model.layers.26.linear_attn.in_proj_z.weight": "model-00004-of-00004.safetensors",
|
| 271 |
+
"model.language_model.layers.26.linear_attn.norm.weight": "model-00004-of-00004.safetensors",
|
| 272 |
+
"model.language_model.layers.26.linear_attn.out_proj.weight": "model-00004-of-00004.safetensors",
|
| 273 |
+
"model.language_model.layers.26.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 274 |
+
"model.language_model.layers.26.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 275 |
+
"model.language_model.layers.26.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 276 |
+
"model.language_model.layers.26.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 277 |
+
"model.language_model.layers.27.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 278 |
+
"model.language_model.layers.27.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 279 |
+
"model.language_model.layers.27.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 280 |
+
"model.language_model.layers.27.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 281 |
+
"model.language_model.layers.27.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 282 |
+
"model.language_model.layers.27.self_attn.k_norm.weight": "model-00004-of-00004.safetensors",
|
| 283 |
+
"model.language_model.layers.27.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
|
| 284 |
+
"model.language_model.layers.27.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
|
| 285 |
+
"model.language_model.layers.27.self_attn.q_norm.weight": "model-00004-of-00004.safetensors",
|
| 286 |
+
"model.language_model.layers.27.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
|
| 287 |
+
"model.language_model.layers.27.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
|
| 288 |
+
"model.language_model.layers.28.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 289 |
+
"model.language_model.layers.28.linear_attn.A_log": "model-00004-of-00004.safetensors",
|
| 290 |
+
"model.language_model.layers.28.linear_attn.conv1d.weight": "model-00004-of-00004.safetensors",
|
| 291 |
+
"model.language_model.layers.28.linear_attn.dt_bias": "model-00004-of-00004.safetensors",
|
| 292 |
+
"model.language_model.layers.28.linear_attn.in_proj_a.weight": "model-00004-of-00004.safetensors",
|
| 293 |
+
"model.language_model.layers.28.linear_attn.in_proj_b.weight": "model-00004-of-00004.safetensors",
|
| 294 |
+
"model.language_model.layers.28.linear_attn.in_proj_qkv.weight": "model-00004-of-00004.safetensors",
|
| 295 |
+
"model.language_model.layers.28.linear_attn.in_proj_z.weight": "model-00004-of-00004.safetensors",
|
| 296 |
+
"model.language_model.layers.28.linear_attn.norm.weight": "model-00004-of-00004.safetensors",
|
| 297 |
+
"model.language_model.layers.28.linear_attn.out_proj.weight": "model-00004-of-00004.safetensors",
|
| 298 |
+
"model.language_model.layers.28.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 299 |
+
"model.language_model.layers.28.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 300 |
+
"model.language_model.layers.28.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 301 |
+
"model.language_model.layers.28.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 302 |
+
"model.language_model.layers.29.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 303 |
+
"model.language_model.layers.29.linear_attn.A_log": "model-00004-of-00004.safetensors",
|
| 304 |
+
"model.language_model.layers.29.linear_attn.conv1d.weight": "model-00004-of-00004.safetensors",
|
| 305 |
+
"model.language_model.layers.29.linear_attn.dt_bias": "model-00004-of-00004.safetensors",
|
| 306 |
+
"model.language_model.layers.29.linear_attn.in_proj_a.weight": "model-00004-of-00004.safetensors",
|
| 307 |
+
"model.language_model.layers.29.linear_attn.in_proj_b.weight": "model-00004-of-00004.safetensors",
|
| 308 |
+
"model.language_model.layers.29.linear_attn.in_proj_qkv.weight": "model-00004-of-00004.safetensors",
|
| 309 |
+
"model.language_model.layers.29.linear_attn.in_proj_z.weight": "model-00004-of-00004.safetensors",
|
| 310 |
+
"model.language_model.layers.29.linear_attn.norm.weight": "model-00004-of-00004.safetensors",
|
| 311 |
+
"model.language_model.layers.29.linear_attn.out_proj.weight": "model-00004-of-00004.safetensors",
|
| 312 |
+
"model.language_model.layers.29.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 313 |
+
"model.language_model.layers.29.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 314 |
+
"model.language_model.layers.29.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 315 |
+
"model.language_model.layers.29.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 316 |
+
"model.language_model.layers.3.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 317 |
+
"model.language_model.layers.3.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 318 |
+
"model.language_model.layers.3.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 319 |
+
"model.language_model.layers.3.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 320 |
+
"model.language_model.layers.3.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 321 |
+
"model.language_model.layers.3.self_attn.k_norm.weight": "model-00002-of-00004.safetensors",
|
| 322 |
+
"model.language_model.layers.3.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
|
| 323 |
+
"model.language_model.layers.3.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
|
| 324 |
+
"model.language_model.layers.3.self_attn.q_norm.weight": "model-00002-of-00004.safetensors",
|
| 325 |
+
"model.language_model.layers.3.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
|
| 326 |
+
"model.language_model.layers.3.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
|
| 327 |
+
"model.language_model.layers.30.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 328 |
+
"model.language_model.layers.30.linear_attn.A_log": "model-00004-of-00004.safetensors",
|
| 329 |
+
"model.language_model.layers.30.linear_attn.conv1d.weight": "model-00004-of-00004.safetensors",
|
| 330 |
+
"model.language_model.layers.30.linear_attn.dt_bias": "model-00004-of-00004.safetensors",
|
| 331 |
+
"model.language_model.layers.30.linear_attn.in_proj_a.weight": "model-00004-of-00004.safetensors",
|
| 332 |
+
"model.language_model.layers.30.linear_attn.in_proj_b.weight": "model-00004-of-00004.safetensors",
|
| 333 |
+
"model.language_model.layers.30.linear_attn.in_proj_qkv.weight": "model-00004-of-00004.safetensors",
|
| 334 |
+
"model.language_model.layers.30.linear_attn.in_proj_z.weight": "model-00004-of-00004.safetensors",
|
| 335 |
+
"model.language_model.layers.30.linear_attn.norm.weight": "model-00004-of-00004.safetensors",
|
| 336 |
+
"model.language_model.layers.30.linear_attn.out_proj.weight": "model-00004-of-00004.safetensors",
|
| 337 |
+
"model.language_model.layers.30.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 338 |
+
"model.language_model.layers.30.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 339 |
+
"model.language_model.layers.30.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 340 |
+
"model.language_model.layers.30.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 341 |
+
"model.language_model.layers.31.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 342 |
+
"model.language_model.layers.31.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 343 |
+
"model.language_model.layers.31.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 344 |
+
"model.language_model.layers.31.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 345 |
+
"model.language_model.layers.31.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 346 |
+
"model.language_model.layers.31.self_attn.k_norm.weight": "model-00004-of-00004.safetensors",
|
| 347 |
+
"model.language_model.layers.31.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
|
| 348 |
+
"model.language_model.layers.31.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
|
| 349 |
+
"model.language_model.layers.31.self_attn.q_norm.weight": "model-00004-of-00004.safetensors",
|
| 350 |
+
"model.language_model.layers.31.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
|
| 351 |
+
"model.language_model.layers.31.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
|
| 352 |
+
"model.language_model.layers.4.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 353 |
+
"model.language_model.layers.4.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 354 |
+
"model.language_model.layers.4.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 355 |
+
"model.language_model.layers.4.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 356 |
+
"model.language_model.layers.4.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 357 |
+
"model.language_model.layers.4.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 358 |
+
"model.language_model.layers.4.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 359 |
+
"model.language_model.layers.4.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 360 |
+
"model.language_model.layers.4.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 361 |
+
"model.language_model.layers.4.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 362 |
+
"model.language_model.layers.4.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 363 |
+
"model.language_model.layers.4.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 364 |
+
"model.language_model.layers.4.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 365 |
+
"model.language_model.layers.4.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 366 |
+
"model.language_model.layers.5.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 367 |
+
"model.language_model.layers.5.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 368 |
+
"model.language_model.layers.5.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 369 |
+
"model.language_model.layers.5.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 370 |
+
"model.language_model.layers.5.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 371 |
+
"model.language_model.layers.5.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 372 |
+
"model.language_model.layers.5.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 373 |
+
"model.language_model.layers.5.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 374 |
+
"model.language_model.layers.5.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 375 |
+
"model.language_model.layers.5.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 376 |
+
"model.language_model.layers.5.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 377 |
+
"model.language_model.layers.5.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 378 |
+
"model.language_model.layers.5.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 379 |
+
"model.language_model.layers.5.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 380 |
+
"model.language_model.layers.6.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 381 |
+
"model.language_model.layers.6.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 382 |
+
"model.language_model.layers.6.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 383 |
+
"model.language_model.layers.6.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 384 |
+
"model.language_model.layers.6.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 385 |
+
"model.language_model.layers.6.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 386 |
+
"model.language_model.layers.6.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 387 |
+
"model.language_model.layers.6.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 388 |
+
"model.language_model.layers.6.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 389 |
+
"model.language_model.layers.6.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 390 |
+
"model.language_model.layers.6.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 391 |
+
"model.language_model.layers.6.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 392 |
+
"model.language_model.layers.6.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 393 |
+
"model.language_model.layers.6.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 394 |
+
"model.language_model.layers.7.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 395 |
+
"model.language_model.layers.7.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 396 |
+
"model.language_model.layers.7.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 397 |
+
"model.language_model.layers.7.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 398 |
+
"model.language_model.layers.7.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 399 |
+
"model.language_model.layers.7.self_attn.k_norm.weight": "model-00002-of-00004.safetensors",
|
| 400 |
+
"model.language_model.layers.7.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
|
| 401 |
+
"model.language_model.layers.7.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
|
| 402 |
+
"model.language_model.layers.7.self_attn.q_norm.weight": "model-00002-of-00004.safetensors",
|
| 403 |
+
"model.language_model.layers.7.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
|
| 404 |
+
"model.language_model.layers.7.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
|
| 405 |
+
"model.language_model.layers.8.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 406 |
+
"model.language_model.layers.8.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 407 |
+
"model.language_model.layers.8.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 408 |
+
"model.language_model.layers.8.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 409 |
+
"model.language_model.layers.8.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 410 |
+
"model.language_model.layers.8.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 411 |
+
"model.language_model.layers.8.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 412 |
+
"model.language_model.layers.8.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 413 |
+
"model.language_model.layers.8.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 414 |
+
"model.language_model.layers.8.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 415 |
+
"model.language_model.layers.8.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 416 |
+
"model.language_model.layers.8.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 417 |
+
"model.language_model.layers.8.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 418 |
+
"model.language_model.layers.8.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 419 |
+
"model.language_model.layers.9.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 420 |
+
"model.language_model.layers.9.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 421 |
+
"model.language_model.layers.9.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 422 |
+
"model.language_model.layers.9.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 423 |
+
"model.language_model.layers.9.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 424 |
+
"model.language_model.layers.9.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 425 |
+
"model.language_model.layers.9.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 426 |
+
"model.language_model.layers.9.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 427 |
+
"model.language_model.layers.9.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 428 |
+
"model.language_model.layers.9.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 429 |
+
"model.language_model.layers.9.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 430 |
+
"model.language_model.layers.9.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 431 |
+
"model.language_model.layers.9.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 432 |
+
"model.language_model.layers.9.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 433 |
+
"model.language_model.norm.weight": "model-00004-of-00004.safetensors",
|
| 434 |
+
"model.visual.blocks.0.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 435 |
+
"model.visual.blocks.0.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 436 |
+
"model.visual.blocks.0.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 437 |
+
"model.visual.blocks.0.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 438 |
+
"model.visual.blocks.0.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 439 |
+
"model.visual.blocks.0.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 440 |
+
"model.visual.blocks.0.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 441 |
+
"model.visual.blocks.0.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 442 |
+
"model.visual.blocks.0.norm1.bias": "model-00004-of-00004.safetensors",
|
| 443 |
+
"model.visual.blocks.0.norm1.weight": "model-00004-of-00004.safetensors",
|
| 444 |
+
"model.visual.blocks.0.norm2.bias": "model-00004-of-00004.safetensors",
|
| 445 |
+
"model.visual.blocks.0.norm2.weight": "model-00004-of-00004.safetensors",
|
| 446 |
+
"model.visual.blocks.1.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 447 |
+
"model.visual.blocks.1.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 448 |
+
"model.visual.blocks.1.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 449 |
+
"model.visual.blocks.1.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 450 |
+
"model.visual.blocks.1.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 451 |
+
"model.visual.blocks.1.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 452 |
+
"model.visual.blocks.1.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 453 |
+
"model.visual.blocks.1.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 454 |
+
"model.visual.blocks.1.norm1.bias": "model-00004-of-00004.safetensors",
|
| 455 |
+
"model.visual.blocks.1.norm1.weight": "model-00004-of-00004.safetensors",
|
| 456 |
+
"model.visual.blocks.1.norm2.bias": "model-00004-of-00004.safetensors",
|
| 457 |
+
"model.visual.blocks.1.norm2.weight": "model-00004-of-00004.safetensors",
|
| 458 |
+
"model.visual.blocks.10.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 459 |
+
"model.visual.blocks.10.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 460 |
+
"model.visual.blocks.10.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 461 |
+
"model.visual.blocks.10.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 462 |
+
"model.visual.blocks.10.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 463 |
+
"model.visual.blocks.10.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 464 |
+
"model.visual.blocks.10.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 465 |
+
"model.visual.blocks.10.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 466 |
+
"model.visual.blocks.10.norm1.bias": "model-00004-of-00004.safetensors",
|
| 467 |
+
"model.visual.blocks.10.norm1.weight": "model-00004-of-00004.safetensors",
|
| 468 |
+
"model.visual.blocks.10.norm2.bias": "model-00004-of-00004.safetensors",
|
| 469 |
+
"model.visual.blocks.10.norm2.weight": "model-00004-of-00004.safetensors",
|
| 470 |
+
"model.visual.blocks.11.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 471 |
+
"model.visual.blocks.11.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 472 |
+
"model.visual.blocks.11.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 473 |
+
"model.visual.blocks.11.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 474 |
+
"model.visual.blocks.11.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 475 |
+
"model.visual.blocks.11.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 476 |
+
"model.visual.blocks.11.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 477 |
+
"model.visual.blocks.11.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 478 |
+
"model.visual.blocks.11.norm1.bias": "model-00004-of-00004.safetensors",
|
| 479 |
+
"model.visual.blocks.11.norm1.weight": "model-00004-of-00004.safetensors",
|
| 480 |
+
"model.visual.blocks.11.norm2.bias": "model-00004-of-00004.safetensors",
|
| 481 |
+
"model.visual.blocks.11.norm2.weight": "model-00004-of-00004.safetensors",
|
| 482 |
+
"model.visual.blocks.12.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 483 |
+
"model.visual.blocks.12.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 484 |
+
"model.visual.blocks.12.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 485 |
+
"model.visual.blocks.12.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 486 |
+
"model.visual.blocks.12.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 487 |
+
"model.visual.blocks.12.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 488 |
+
"model.visual.blocks.12.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 489 |
+
"model.visual.blocks.12.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 490 |
+
"model.visual.blocks.12.norm1.bias": "model-00004-of-00004.safetensors",
|
| 491 |
+
"model.visual.blocks.12.norm1.weight": "model-00004-of-00004.safetensors",
|
| 492 |
+
"model.visual.blocks.12.norm2.bias": "model-00004-of-00004.safetensors",
|
| 493 |
+
"model.visual.blocks.12.norm2.weight": "model-00004-of-00004.safetensors",
|
| 494 |
+
"model.visual.blocks.13.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 495 |
+
"model.visual.blocks.13.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 496 |
+
"model.visual.blocks.13.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 497 |
+
"model.visual.blocks.13.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 498 |
+
"model.visual.blocks.13.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 499 |
+
"model.visual.blocks.13.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 500 |
+
"model.visual.blocks.13.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 501 |
+
"model.visual.blocks.13.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 502 |
+
"model.visual.blocks.13.norm1.bias": "model-00004-of-00004.safetensors",
|
| 503 |
+
"model.visual.blocks.13.norm1.weight": "model-00004-of-00004.safetensors",
|
| 504 |
+
"model.visual.blocks.13.norm2.bias": "model-00004-of-00004.safetensors",
|
| 505 |
+
"model.visual.blocks.13.norm2.weight": "model-00004-of-00004.safetensors",
|
| 506 |
+
"model.visual.blocks.14.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 507 |
+
"model.visual.blocks.14.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 508 |
+
"model.visual.blocks.14.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 509 |
+
"model.visual.blocks.14.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 510 |
+
"model.visual.blocks.14.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 511 |
+
"model.visual.blocks.14.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 512 |
+
"model.visual.blocks.14.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 513 |
+
"model.visual.blocks.14.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 514 |
+
"model.visual.blocks.14.norm1.bias": "model-00004-of-00004.safetensors",
|
| 515 |
+
"model.visual.blocks.14.norm1.weight": "model-00004-of-00004.safetensors",
|
| 516 |
+
"model.visual.blocks.14.norm2.bias": "model-00004-of-00004.safetensors",
|
| 517 |
+
"model.visual.blocks.14.norm2.weight": "model-00004-of-00004.safetensors",
|
| 518 |
+
"model.visual.blocks.15.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 519 |
+
"model.visual.blocks.15.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 520 |
+
"model.visual.blocks.15.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 521 |
+
"model.visual.blocks.15.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 522 |
+
"model.visual.blocks.15.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 523 |
+
"model.visual.blocks.15.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 524 |
+
"model.visual.blocks.15.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 525 |
+
"model.visual.blocks.15.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 526 |
+
"model.visual.blocks.15.norm1.bias": "model-00004-of-00004.safetensors",
|
| 527 |
+
"model.visual.blocks.15.norm1.weight": "model-00004-of-00004.safetensors",
|
| 528 |
+
"model.visual.blocks.15.norm2.bias": "model-00004-of-00004.safetensors",
|
| 529 |
+
"model.visual.blocks.15.norm2.weight": "model-00004-of-00004.safetensors",
|
| 530 |
+
"model.visual.blocks.16.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 531 |
+
"model.visual.blocks.16.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 532 |
+
"model.visual.blocks.16.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 533 |
+
"model.visual.blocks.16.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 534 |
+
"model.visual.blocks.16.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 535 |
+
"model.visual.blocks.16.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 536 |
+
"model.visual.blocks.16.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 537 |
+
"model.visual.blocks.16.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 538 |
+
"model.visual.blocks.16.norm1.bias": "model-00004-of-00004.safetensors",
|
| 539 |
+
"model.visual.blocks.16.norm1.weight": "model-00004-of-00004.safetensors",
|
| 540 |
+
"model.visual.blocks.16.norm2.bias": "model-00004-of-00004.safetensors",
|
| 541 |
+
"model.visual.blocks.16.norm2.weight": "model-00004-of-00004.safetensors",
|
| 542 |
+
"model.visual.blocks.17.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 543 |
+
"model.visual.blocks.17.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 544 |
+
"model.visual.blocks.17.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 545 |
+
"model.visual.blocks.17.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 546 |
+
"model.visual.blocks.17.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 547 |
+
"model.visual.blocks.17.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 548 |
+
"model.visual.blocks.17.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 549 |
+
"model.visual.blocks.17.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 550 |
+
"model.visual.blocks.17.norm1.bias": "model-00004-of-00004.safetensors",
|
| 551 |
+
"model.visual.blocks.17.norm1.weight": "model-00004-of-00004.safetensors",
|
| 552 |
+
"model.visual.blocks.17.norm2.bias": "model-00004-of-00004.safetensors",
|
| 553 |
+
"model.visual.blocks.17.norm2.weight": "model-00004-of-00004.safetensors",
|
| 554 |
+
"model.visual.blocks.18.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 555 |
+
"model.visual.blocks.18.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 556 |
+
"model.visual.blocks.18.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 557 |
+
"model.visual.blocks.18.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 558 |
+
"model.visual.blocks.18.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 559 |
+
"model.visual.blocks.18.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 560 |
+
"model.visual.blocks.18.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 561 |
+
"model.visual.blocks.18.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 562 |
+
"model.visual.blocks.18.norm1.bias": "model-00004-of-00004.safetensors",
|
| 563 |
+
"model.visual.blocks.18.norm1.weight": "model-00004-of-00004.safetensors",
|
| 564 |
+
"model.visual.blocks.18.norm2.bias": "model-00004-of-00004.safetensors",
|
| 565 |
+
"model.visual.blocks.18.norm2.weight": "model-00004-of-00004.safetensors",
|
| 566 |
+
"model.visual.blocks.19.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 567 |
+
"model.visual.blocks.19.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 568 |
+
"model.visual.blocks.19.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 569 |
+
"model.visual.blocks.19.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 570 |
+
"model.visual.blocks.19.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 571 |
+
"model.visual.blocks.19.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 572 |
+
"model.visual.blocks.19.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 573 |
+
"model.visual.blocks.19.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 574 |
+
"model.visual.blocks.19.norm1.bias": "model-00004-of-00004.safetensors",
|
| 575 |
+
"model.visual.blocks.19.norm1.weight": "model-00004-of-00004.safetensors",
|
| 576 |
+
"model.visual.blocks.19.norm2.bias": "model-00004-of-00004.safetensors",
|
| 577 |
+
"model.visual.blocks.19.norm2.weight": "model-00004-of-00004.safetensors",
|
| 578 |
+
"model.visual.blocks.2.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 579 |
+
"model.visual.blocks.2.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 580 |
+
"model.visual.blocks.2.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 581 |
+
"model.visual.blocks.2.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 582 |
+
"model.visual.blocks.2.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 583 |
+
"model.visual.blocks.2.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 584 |
+
"model.visual.blocks.2.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 585 |
+
"model.visual.blocks.2.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 586 |
+
"model.visual.blocks.2.norm1.bias": "model-00004-of-00004.safetensors",
|
| 587 |
+
"model.visual.blocks.2.norm1.weight": "model-00004-of-00004.safetensors",
|
| 588 |
+
"model.visual.blocks.2.norm2.bias": "model-00004-of-00004.safetensors",
|
| 589 |
+
"model.visual.blocks.2.norm2.weight": "model-00004-of-00004.safetensors",
|
| 590 |
+
"model.visual.blocks.20.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 591 |
+
"model.visual.blocks.20.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 592 |
+
"model.visual.blocks.20.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 593 |
+
"model.visual.blocks.20.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 594 |
+
"model.visual.blocks.20.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 595 |
+
"model.visual.blocks.20.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 596 |
+
"model.visual.blocks.20.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 597 |
+
"model.visual.blocks.20.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 598 |
+
"model.visual.blocks.20.norm1.bias": "model-00004-of-00004.safetensors",
|
| 599 |
+
"model.visual.blocks.20.norm1.weight": "model-00004-of-00004.safetensors",
|
| 600 |
+
"model.visual.blocks.20.norm2.bias": "model-00004-of-00004.safetensors",
|
| 601 |
+
"model.visual.blocks.20.norm2.weight": "model-00004-of-00004.safetensors",
|
| 602 |
+
"model.visual.blocks.21.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 603 |
+
"model.visual.blocks.21.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 604 |
+
"model.visual.blocks.21.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 605 |
+
"model.visual.blocks.21.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 606 |
+
"model.visual.blocks.21.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 607 |
+
"model.visual.blocks.21.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 608 |
+
"model.visual.blocks.21.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 609 |
+
"model.visual.blocks.21.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 610 |
+
"model.visual.blocks.21.norm1.bias": "model-00004-of-00004.safetensors",
|
| 611 |
+
"model.visual.blocks.21.norm1.weight": "model-00004-of-00004.safetensors",
|
| 612 |
+
"model.visual.blocks.21.norm2.bias": "model-00004-of-00004.safetensors",
|
| 613 |
+
"model.visual.blocks.21.norm2.weight": "model-00004-of-00004.safetensors",
|
| 614 |
+
"model.visual.blocks.22.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 615 |
+
"model.visual.blocks.22.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 616 |
+
"model.visual.blocks.22.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 617 |
+
"model.visual.blocks.22.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 618 |
+
"model.visual.blocks.22.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 619 |
+
"model.visual.blocks.22.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 620 |
+
"model.visual.blocks.22.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 621 |
+
"model.visual.blocks.22.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 622 |
+
"model.visual.blocks.22.norm1.bias": "model-00004-of-00004.safetensors",
|
| 623 |
+
"model.visual.blocks.22.norm1.weight": "model-00004-of-00004.safetensors",
|
| 624 |
+
"model.visual.blocks.22.norm2.bias": "model-00004-of-00004.safetensors",
|
| 625 |
+
"model.visual.blocks.22.norm2.weight": "model-00004-of-00004.safetensors",
|
| 626 |
+
"model.visual.blocks.23.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 627 |
+
"model.visual.blocks.23.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 628 |
+
"model.visual.blocks.23.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 629 |
+
"model.visual.blocks.23.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 630 |
+
"model.visual.blocks.23.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 631 |
+
"model.visual.blocks.23.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 632 |
+
"model.visual.blocks.23.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 633 |
+
"model.visual.blocks.23.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 634 |
+
"model.visual.blocks.23.norm1.bias": "model-00004-of-00004.safetensors",
|
| 635 |
+
"model.visual.blocks.23.norm1.weight": "model-00004-of-00004.safetensors",
|
| 636 |
+
"model.visual.blocks.23.norm2.bias": "model-00004-of-00004.safetensors",
|
| 637 |
+
"model.visual.blocks.23.norm2.weight": "model-00004-of-00004.safetensors",
|
| 638 |
+
"model.visual.blocks.24.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 639 |
+
"model.visual.blocks.24.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 640 |
+
"model.visual.blocks.24.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 641 |
+
"model.visual.blocks.24.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 642 |
+
"model.visual.blocks.24.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 643 |
+
"model.visual.blocks.24.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 644 |
+
"model.visual.blocks.24.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 645 |
+
"model.visual.blocks.24.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 646 |
+
"model.visual.blocks.24.norm1.bias": "model-00004-of-00004.safetensors",
|
| 647 |
+
"model.visual.blocks.24.norm1.weight": "model-00004-of-00004.safetensors",
|
| 648 |
+
"model.visual.blocks.24.norm2.bias": "model-00004-of-00004.safetensors",
|
| 649 |
+
"model.visual.blocks.24.norm2.weight": "model-00004-of-00004.safetensors",
|
| 650 |
+
"model.visual.blocks.25.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 651 |
+
"model.visual.blocks.25.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 652 |
+
"model.visual.blocks.25.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 653 |
+
"model.visual.blocks.25.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 654 |
+
"model.visual.blocks.25.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 655 |
+
"model.visual.blocks.25.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 656 |
+
"model.visual.blocks.25.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 657 |
+
"model.visual.blocks.25.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 658 |
+
"model.visual.blocks.25.norm1.bias": "model-00004-of-00004.safetensors",
|
| 659 |
+
"model.visual.blocks.25.norm1.weight": "model-00004-of-00004.safetensors",
|
| 660 |
+
"model.visual.blocks.25.norm2.bias": "model-00004-of-00004.safetensors",
|
| 661 |
+
"model.visual.blocks.25.norm2.weight": "model-00004-of-00004.safetensors",
|
| 662 |
+
"model.visual.blocks.26.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 663 |
+
"model.visual.blocks.26.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 664 |
+
"model.visual.blocks.26.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 665 |
+
"model.visual.blocks.26.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 666 |
+
"model.visual.blocks.26.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 667 |
+
"model.visual.blocks.26.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 668 |
+
"model.visual.blocks.26.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 669 |
+
"model.visual.blocks.26.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 670 |
+
"model.visual.blocks.26.norm1.bias": "model-00004-of-00004.safetensors",
|
| 671 |
+
"model.visual.blocks.26.norm1.weight": "model-00004-of-00004.safetensors",
|
| 672 |
+
"model.visual.blocks.26.norm2.bias": "model-00004-of-00004.safetensors",
|
| 673 |
+
"model.visual.blocks.26.norm2.weight": "model-00004-of-00004.safetensors",
|
| 674 |
+
"model.visual.blocks.3.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 675 |
+
"model.visual.blocks.3.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 676 |
+
"model.visual.blocks.3.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 677 |
+
"model.visual.blocks.3.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 678 |
+
"model.visual.blocks.3.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 679 |
+
"model.visual.blocks.3.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 680 |
+
"model.visual.blocks.3.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 681 |
+
"model.visual.blocks.3.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 682 |
+
"model.visual.blocks.3.norm1.bias": "model-00004-of-00004.safetensors",
|
| 683 |
+
"model.visual.blocks.3.norm1.weight": "model-00004-of-00004.safetensors",
|
| 684 |
+
"model.visual.blocks.3.norm2.bias": "model-00004-of-00004.safetensors",
|
| 685 |
+
"model.visual.blocks.3.norm2.weight": "model-00004-of-00004.safetensors",
|
| 686 |
+
"model.visual.blocks.4.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 687 |
+
"model.visual.blocks.4.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 688 |
+
"model.visual.blocks.4.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 689 |
+
"model.visual.blocks.4.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 690 |
+
"model.visual.blocks.4.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 691 |
+
"model.visual.blocks.4.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 692 |
+
"model.visual.blocks.4.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 693 |
+
"model.visual.blocks.4.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 694 |
+
"model.visual.blocks.4.norm1.bias": "model-00004-of-00004.safetensors",
|
| 695 |
+
"model.visual.blocks.4.norm1.weight": "model-00004-of-00004.safetensors",
|
| 696 |
+
"model.visual.blocks.4.norm2.bias": "model-00004-of-00004.safetensors",
|
| 697 |
+
"model.visual.blocks.4.norm2.weight": "model-00004-of-00004.safetensors",
|
| 698 |
+
"model.visual.blocks.5.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 699 |
+
"model.visual.blocks.5.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 700 |
+
"model.visual.blocks.5.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 701 |
+
"model.visual.blocks.5.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 702 |
+
"model.visual.blocks.5.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 703 |
+
"model.visual.blocks.5.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 704 |
+
"model.visual.blocks.5.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 705 |
+
"model.visual.blocks.5.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 706 |
+
"model.visual.blocks.5.norm1.bias": "model-00004-of-00004.safetensors",
|
| 707 |
+
"model.visual.blocks.5.norm1.weight": "model-00004-of-00004.safetensors",
|
| 708 |
+
"model.visual.blocks.5.norm2.bias": "model-00004-of-00004.safetensors",
|
| 709 |
+
"model.visual.blocks.5.norm2.weight": "model-00004-of-00004.safetensors",
|
| 710 |
+
"model.visual.blocks.6.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 711 |
+
"model.visual.blocks.6.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 712 |
+
"model.visual.blocks.6.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 713 |
+
"model.visual.blocks.6.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 714 |
+
"model.visual.blocks.6.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 715 |
+
"model.visual.blocks.6.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 716 |
+
"model.visual.blocks.6.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 717 |
+
"model.visual.blocks.6.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 718 |
+
"model.visual.blocks.6.norm1.bias": "model-00004-of-00004.safetensors",
|
| 719 |
+
"model.visual.blocks.6.norm1.weight": "model-00004-of-00004.safetensors",
|
| 720 |
+
"model.visual.blocks.6.norm2.bias": "model-00004-of-00004.safetensors",
|
| 721 |
+
"model.visual.blocks.6.norm2.weight": "model-00004-of-00004.safetensors",
|
| 722 |
+
"model.visual.blocks.7.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 723 |
+
"model.visual.blocks.7.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 724 |
+
"model.visual.blocks.7.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 725 |
+
"model.visual.blocks.7.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 726 |
+
"model.visual.blocks.7.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 727 |
+
"model.visual.blocks.7.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 728 |
+
"model.visual.blocks.7.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 729 |
+
"model.visual.blocks.7.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 730 |
+
"model.visual.blocks.7.norm1.bias": "model-00004-of-00004.safetensors",
|
| 731 |
+
"model.visual.blocks.7.norm1.weight": "model-00004-of-00004.safetensors",
|
| 732 |
+
"model.visual.blocks.7.norm2.bias": "model-00004-of-00004.safetensors",
|
| 733 |
+
"model.visual.blocks.7.norm2.weight": "model-00004-of-00004.safetensors",
|
| 734 |
+
"model.visual.blocks.8.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 735 |
+
"model.visual.blocks.8.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 736 |
+
"model.visual.blocks.8.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 737 |
+
"model.visual.blocks.8.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 738 |
+
"model.visual.blocks.8.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 739 |
+
"model.visual.blocks.8.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 740 |
+
"model.visual.blocks.8.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 741 |
+
"model.visual.blocks.8.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 742 |
+
"model.visual.blocks.8.norm1.bias": "model-00004-of-00004.safetensors",
|
| 743 |
+
"model.visual.blocks.8.norm1.weight": "model-00004-of-00004.safetensors",
|
| 744 |
+
"model.visual.blocks.8.norm2.bias": "model-00004-of-00004.safetensors",
|
| 745 |
+
"model.visual.blocks.8.norm2.weight": "model-00004-of-00004.safetensors",
|
| 746 |
+
"model.visual.blocks.9.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 747 |
+
"model.visual.blocks.9.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 748 |
+
"model.visual.blocks.9.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 749 |
+
"model.visual.blocks.9.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 750 |
+
"model.visual.blocks.9.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 751 |
+
"model.visual.blocks.9.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 752 |
+
"model.visual.blocks.9.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 753 |
+
"model.visual.blocks.9.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 754 |
+
"model.visual.blocks.9.norm1.bias": "model-00004-of-00004.safetensors",
|
| 755 |
+
"model.visual.blocks.9.norm1.weight": "model-00004-of-00004.safetensors",
|
| 756 |
+
"model.visual.blocks.9.norm2.bias": "model-00004-of-00004.safetensors",
|
| 757 |
+
"model.visual.blocks.9.norm2.weight": "model-00004-of-00004.safetensors",
|
| 758 |
+
"model.visual.merger.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 759 |
+
"model.visual.merger.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 760 |
+
"model.visual.merger.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 761 |
+
"model.visual.merger.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 762 |
+
"model.visual.merger.norm.bias": "model-00004-of-00004.safetensors",
|
| 763 |
+
"model.visual.merger.norm.weight": "model-00004-of-00004.safetensors",
|
| 764 |
+
"model.visual.patch_embed.proj.bias": "model-00004-of-00004.safetensors",
|
| 765 |
+
"model.visual.patch_embed.proj.weight": "model-00004-of-00004.safetensors",
|
| 766 |
+
"model.visual.pos_embed.weight": "model-00004-of-00004.safetensors"
|
| 767 |
+
}
|
| 768 |
+
}
|
a4_fixed_images/round_1/preprocessor_config.json
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"size": {
|
| 3 |
+
"longest_edge": 16777216,
|
| 4 |
+
"shortest_edge": 65536
|
| 5 |
+
},
|
| 6 |
+
"patch_size": 16,
|
| 7 |
+
"temporal_patch_size": 2,
|
| 8 |
+
"merge_size": 2,
|
| 9 |
+
"image_mean": [
|
| 10 |
+
0.5,
|
| 11 |
+
0.5,
|
| 12 |
+
0.5
|
| 13 |
+
],
|
| 14 |
+
"image_std": [
|
| 15 |
+
0.5,
|
| 16 |
+
0.5,
|
| 17 |
+
0.5
|
| 18 |
+
],
|
| 19 |
+
"processor_class": "Qwen3VLProcessor",
|
| 20 |
+
"image_processor_type": "Qwen2VLImageProcessorFast"
|
| 21 |
+
}
|
a4_fixed_images/round_1/processor_config.json
ADDED
|
@@ -0,0 +1,60 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"image_processor": {
|
| 3 |
+
"do_convert_rgb": true,
|
| 4 |
+
"do_normalize": true,
|
| 5 |
+
"do_rescale": true,
|
| 6 |
+
"do_resize": true,
|
| 7 |
+
"image_mean": [
|
| 8 |
+
0.5,
|
| 9 |
+
0.5,
|
| 10 |
+
0.5
|
| 11 |
+
],
|
| 12 |
+
"image_processor_type": "Qwen2VLImageProcessor",
|
| 13 |
+
"image_std": [
|
| 14 |
+
0.5,
|
| 15 |
+
0.5,
|
| 16 |
+
0.5
|
| 17 |
+
],
|
| 18 |
+
"merge_size": 2,
|
| 19 |
+
"patch_size": 16,
|
| 20 |
+
"resample": 3,
|
| 21 |
+
"rescale_factor": 0.00392156862745098,
|
| 22 |
+
"size": {
|
| 23 |
+
"longest_edge": 16777216,
|
| 24 |
+
"shortest_edge": 65536
|
| 25 |
+
},
|
| 26 |
+
"temporal_patch_size": 2
|
| 27 |
+
},
|
| 28 |
+
"processor_class": "Qwen3VLProcessor",
|
| 29 |
+
"video_processor": {
|
| 30 |
+
"do_convert_rgb": true,
|
| 31 |
+
"do_normalize": true,
|
| 32 |
+
"do_rescale": true,
|
| 33 |
+
"do_resize": true,
|
| 34 |
+
"do_sample_frames": true,
|
| 35 |
+
"fps": 2,
|
| 36 |
+
"image_mean": [
|
| 37 |
+
0.5,
|
| 38 |
+
0.5,
|
| 39 |
+
0.5
|
| 40 |
+
],
|
| 41 |
+
"image_std": [
|
| 42 |
+
0.5,
|
| 43 |
+
0.5,
|
| 44 |
+
0.5
|
| 45 |
+
],
|
| 46 |
+
"max_frames": 768,
|
| 47 |
+
"merge_size": 2,
|
| 48 |
+
"min_frames": 4,
|
| 49 |
+
"patch_size": 16,
|
| 50 |
+
"resample": 3,
|
| 51 |
+
"rescale_factor": 0.00392156862745098,
|
| 52 |
+
"return_metadata": false,
|
| 53 |
+
"size": {
|
| 54 |
+
"longest_edge": 25165824,
|
| 55 |
+
"shortest_edge": 4096
|
| 56 |
+
},
|
| 57 |
+
"temporal_patch_size": 2,
|
| 58 |
+
"video_processor_type": "Qwen3VLVideoProcessor"
|
| 59 |
+
}
|
| 60 |
+
}
|
a4_fixed_images/round_1/tokenizer_config.json
ADDED
|
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"audio_bos_token": "<|audio_start|>",
|
| 4 |
+
"audio_eos_token": "<|audio_end|>",
|
| 5 |
+
"audio_token": "<|audio_pad|>",
|
| 6 |
+
"backend": "tokenizers",
|
| 7 |
+
"bos_token": null,
|
| 8 |
+
"clean_up_tokenization_spaces": false,
|
| 9 |
+
"eos_token": "<|im_end|>",
|
| 10 |
+
"errors": "replace",
|
| 11 |
+
"image_token": "<|image_pad|>",
|
| 12 |
+
"is_local": true,
|
| 13 |
+
"local_files_only": true,
|
| 14 |
+
"model_max_length": 262144,
|
| 15 |
+
"model_specific_special_tokens": {
|
| 16 |
+
"audio_bos_token": "<|audio_start|>",
|
| 17 |
+
"audio_eos_token": "<|audio_end|>",
|
| 18 |
+
"audio_token": "<|audio_pad|>",
|
| 19 |
+
"image_token": "<|image_pad|>",
|
| 20 |
+
"video_token": "<|video_pad|>",
|
| 21 |
+
"vision_bos_token": "<|vision_start|>",
|
| 22 |
+
"vision_eos_token": "<|vision_end|>"
|
| 23 |
+
},
|
| 24 |
+
"pad_token": "<|endoftext|>",
|
| 25 |
+
"pretokenize_regex": "(?i:'s|'t|'re|'ve|'m|'ll|'d)|[^\\r\\n\\p{L}\\p{N}]?[\\p{L}\\p{M}]+|\\p{N}| ?[^\\s\\p{L}\\p{M}\\p{N}]+[\\r\\n]*|\\s*[\\r\\n]+|\\s+(?!\\S)|\\s+",
|
| 26 |
+
"processor_class": "Qwen3VLProcessor",
|
| 27 |
+
"split_special_tokens": false,
|
| 28 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 29 |
+
"unk_token": null,
|
| 30 |
+
"video_token": "<|video_pad|>",
|
| 31 |
+
"vision_bos_token": "<|vision_start|>",
|
| 32 |
+
"vision_eos_token": "<|vision_end|>"
|
| 33 |
+
}
|
a4_fixed_images/round_1/trainer_state.json
ADDED
|
@@ -0,0 +1,2454 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 1.0,
|
| 6 |
+
"eval_steps": 80.0,
|
| 7 |
+
"global_step": 160,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"clip_ratio/high_max": 0.0,
|
| 14 |
+
"clip_ratio/high_mean": 0.0,
|
| 15 |
+
"clip_ratio/low_mean": 0.0,
|
| 16 |
+
"clip_ratio/low_min": 0.0,
|
| 17 |
+
"clip_ratio/region_mean": 0.0,
|
| 18 |
+
"completions/clipped_ratio": 0.28125,
|
| 19 |
+
"completions/max_length": 2048.0,
|
| 20 |
+
"completions/mean_length": 990.9921875,
|
| 21 |
+
"completions/min_length": 50.0,
|
| 22 |
+
"epoch": 0.00625,
|
| 23 |
+
"frac_reward_zero_std": 0.15625,
|
| 24 |
+
"grad_norm": 2.3146628429562144,
|
| 25 |
+
"kl": 0.0,
|
| 26 |
+
"learning_rate": 2e-07,
|
| 27 |
+
"loss": 2.7939677238464355e-09,
|
| 28 |
+
"reward": 0.42578125,
|
| 29 |
+
"reward_std": 0.383084237575531,
|
| 30 |
+
"rewards/MajorityAccuracy/mean": 0.42578125,
|
| 31 |
+
"rewards/MajorityAccuracy/std": 0.49542948603630066,
|
| 32 |
+
"step": 1,
|
| 33 |
+
"step_time": 266.03369643259794
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"clip_ratio/high_max": 0.0,
|
| 37 |
+
"clip_ratio/high_mean": 0.0,
|
| 38 |
+
"clip_ratio/low_mean": 0.0,
|
| 39 |
+
"clip_ratio/low_min": 0.0,
|
| 40 |
+
"clip_ratio/region_mean": 0.0,
|
| 41 |
+
"epoch": 0.0125,
|
| 42 |
+
"grad_norm": 3.3314376852757435,
|
| 43 |
+
"kl": 0.0,
|
| 44 |
+
"learning_rate": 4e-07,
|
| 45 |
+
"loss": -4.6566128730773926e-09,
|
| 46 |
+
"step": 2,
|
| 47 |
+
"step_time": 24.759675265755504
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"clip_ratio/high_max": 0.0017515384097350761,
|
| 51 |
+
"clip_ratio/high_mean": 0.000467505455162609,
|
| 52 |
+
"clip_ratio/low_mean": 0.0002569874905020697,
|
| 53 |
+
"clip_ratio/low_min": 0.0,
|
| 54 |
+
"clip_ratio/region_mean": 0.0007244929493026575,
|
| 55 |
+
"epoch": 0.01875,
|
| 56 |
+
"grad_norm": 2.3816850900522066,
|
| 57 |
+
"kl": 0.00017331856361124665,
|
| 58 |
+
"learning_rate": 6e-07,
|
| 59 |
+
"loss": -0.00010576099157333374,
|
| 60 |
+
"step": 3,
|
| 61 |
+
"step_time": 26.1182986847125
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"clip_ratio/high_max": 0.0038417026080423966,
|
| 65 |
+
"clip_ratio/high_mean": 0.0009903872887662146,
|
| 66 |
+
"clip_ratio/low_mean": 0.0007744043687125668,
|
| 67 |
+
"clip_ratio/low_min": 0.0,
|
| 68 |
+
"clip_ratio/region_mean": 0.001764791628374951,
|
| 69 |
+
"epoch": 0.025,
|
| 70 |
+
"grad_norm": 3.5266752733988347,
|
| 71 |
+
"kl": 0.000311654122924665,
|
| 72 |
+
"learning_rate": 8e-07,
|
| 73 |
+
"loss": 0.00014978554099798203,
|
| 74 |
+
"step": 4,
|
| 75 |
+
"step_time": 24.082487512845546
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"clip_ratio/high_max": 0.00060601800942095,
|
| 79 |
+
"clip_ratio/high_mean": 0.0001515045023552375,
|
| 80 |
+
"clip_ratio/low_mean": 0.0012094452467863448,
|
| 81 |
+
"clip_ratio/low_min": 0.0,
|
| 82 |
+
"clip_ratio/region_mean": 0.0013609497054858366,
|
| 83 |
+
"epoch": 0.03125,
|
| 84 |
+
"grad_norm": 3.8938774236700344,
|
| 85 |
+
"kl": 0.00026551546034170315,
|
| 86 |
+
"learning_rate": 1e-06,
|
| 87 |
+
"loss": -0.0001112576574087143,
|
| 88 |
+
"step": 5,
|
| 89 |
+
"step_time": 24.5988663546741
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"clip_ratio/high_max": 0.0041357566879014485,
|
| 93 |
+
"clip_ratio/high_mean": 0.0010693298809201224,
|
| 94 |
+
"clip_ratio/low_mean": 0.00011160150097566657,
|
| 95 |
+
"clip_ratio/low_min": 0.0,
|
| 96 |
+
"clip_ratio/region_mean": 0.0011809314019046724,
|
| 97 |
+
"epoch": 0.0375,
|
| 98 |
+
"grad_norm": 3.3760993419665377,
|
| 99 |
+
"kl": 0.00023432613488694187,
|
| 100 |
+
"learning_rate": 9.998973021172564e-07,
|
| 101 |
+
"loss": 6.69994333293289e-05,
|
| 102 |
+
"step": 6,
|
| 103 |
+
"step_time": 24.55241093877703
|
| 104 |
+
},
|
| 105 |
+
{
|
| 106 |
+
"clip_ratio/high_max": 0.002984301303513348,
|
| 107 |
+
"clip_ratio/high_mean": 0.000746075325878337,
|
| 108 |
+
"clip_ratio/low_mean": 0.0005334175402822439,
|
| 109 |
+
"clip_ratio/low_min": 0.0,
|
| 110 |
+
"clip_ratio/region_mean": 0.001279492862522602,
|
| 111 |
+
"epoch": 0.04375,
|
| 112 |
+
"grad_norm": 3.9320516580069276,
|
| 113 |
+
"kl": 0.00047141046525212005,
|
| 114 |
+
"learning_rate": 9.99589250656446e-07,
|
| 115 |
+
"loss": 0.00011583759624045342,
|
| 116 |
+
"step": 7,
|
| 117 |
+
"step_time": 24.706196520943195
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"clip_ratio/high_max": 0.0004567746291286312,
|
| 121 |
+
"clip_ratio/high_mean": 0.0001141936572821578,
|
| 122 |
+
"clip_ratio/low_mean": 0.00017007338465191424,
|
| 123 |
+
"clip_ratio/low_min": 0.0,
|
| 124 |
+
"clip_ratio/region_mean": 0.00028426704375306144,
|
| 125 |
+
"epoch": 0.05,
|
| 126 |
+
"grad_norm": 2.3650649873180645,
|
| 127 |
+
"kl": 0.00027120417507831007,
|
| 128 |
+
"learning_rate": 9.990759721625005e-07,
|
| 129 |
+
"loss": 0.0002014606143347919,
|
| 130 |
+
"step": 8,
|
| 131 |
+
"step_time": 25.505722932051867
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"clip_ratio/high_max": 0.0,
|
| 135 |
+
"clip_ratio/high_mean": 0.0,
|
| 136 |
+
"clip_ratio/low_mean": 0.0,
|
| 137 |
+
"clip_ratio/low_min": 0.0,
|
| 138 |
+
"clip_ratio/region_mean": 0.0,
|
| 139 |
+
"completions/clipped_ratio": 0.26953125,
|
| 140 |
+
"completions/max_length": 2048.0,
|
| 141 |
+
"completions/mean_length": 961.80078125,
|
| 142 |
+
"completions/min_length": 106.0,
|
| 143 |
+
"epoch": 0.05625,
|
| 144 |
+
"frac_reward_zero_std": 0.15625,
|
| 145 |
+
"grad_norm": 1.654219816693279,
|
| 146 |
+
"kl": 0.00034527726347732823,
|
| 147 |
+
"learning_rate": 9.983576774858775e-07,
|
| 148 |
+
"loss": 3.453743602221948e-06,
|
| 149 |
+
"reward": 0.359375,
|
| 150 |
+
"reward_std": 0.3738928437232971,
|
| 151 |
+
"rewards/MajorityAccuracy/mean": 0.359375,
|
| 152 |
+
"rewards/MajorityAccuracy/std": 0.4807571768760681,
|
| 153 |
+
"step": 9,
|
| 154 |
+
"step_time": 187.7896640417166
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"clip_ratio/high_max": 0.0004339167207945138,
|
| 158 |
+
"clip_ratio/high_mean": 0.00015386770974146202,
|
| 159 |
+
"clip_ratio/low_mean": 0.0005834486219100654,
|
| 160 |
+
"clip_ratio/low_min": 6.103515625e-05,
|
| 161 |
+
"clip_ratio/region_mean": 0.0007373163316515274,
|
| 162 |
+
"epoch": 0.0625,
|
| 163 |
+
"grad_norm": 2.6226070942515345,
|
| 164 |
+
"kl": 0.0005461564724100754,
|
| 165 |
+
"learning_rate": 9.974346616959475e-07,
|
| 166 |
+
"loss": -0.0002195201814174652,
|
| 167 |
+
"step": 10,
|
| 168 |
+
"step_time": 24.52841843292117
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"clip_ratio/high_max": 0.005432300516986288,
|
| 172 |
+
"clip_ratio/high_mean": 0.0017704719830362592,
|
| 173 |
+
"clip_ratio/low_mean": 0.000974005539319478,
|
| 174 |
+
"clip_ratio/low_min": 6.103515625e-05,
|
| 175 |
+
"clip_ratio/region_mean": 0.0027444775550975464,
|
| 176 |
+
"epoch": 0.06875,
|
| 177 |
+
"grad_norm": 4.897518768275102,
|
| 178 |
+
"kl": 0.0008094267250271514,
|
| 179 |
+
"learning_rate": 9.963073039597796e-07,
|
| 180 |
+
"loss": 7.1672722697258e-05,
|
| 181 |
+
"step": 11,
|
| 182 |
+
"step_time": 24.61875235196203
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"clip_ratio/high_max": 0.011929096246603876,
|
| 186 |
+
"clip_ratio/high_mean": 0.004107613043743186,
|
| 187 |
+
"clip_ratio/low_mean": 0.0019726502941921353,
|
| 188 |
+
"clip_ratio/low_min": 0.0,
|
| 189 |
+
"clip_ratio/region_mean": 0.0060802632360719144,
|
| 190 |
+
"epoch": 0.075,
|
| 191 |
+
"grad_norm": 4.365340041024201,
|
| 192 |
+
"kl": 0.004728821892058477,
|
| 193 |
+
"learning_rate": 9.949760673863846e-07,
|
| 194 |
+
"loss": 0.0002971747890114784,
|
| 195 |
+
"step": 12,
|
| 196 |
+
"step_time": 22.031336394604295
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"clip_ratio/high_max": 0.001864024205133319,
|
| 200 |
+
"clip_ratio/high_mean": 0.0004660060512833297,
|
| 201 |
+
"clip_ratio/low_mean": 0.0003192784461134579,
|
| 202 |
+
"clip_ratio/low_min": 0.0,
|
| 203 |
+
"clip_ratio/region_mean": 0.0007852844955777982,
|
| 204 |
+
"epoch": 0.08125,
|
| 205 |
+
"grad_norm": 1.9379422104235469,
|
| 206 |
+
"kl": 0.0005666740653396118,
|
| 207 |
+
"learning_rate": 9.934414988364722e-07,
|
| 208 |
+
"loss": 4.03852500312496e-05,
|
| 209 |
+
"step": 13,
|
| 210 |
+
"step_time": 24.699771769344807
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"clip_ratio/high_max": 0.004800671944394708,
|
| 214 |
+
"clip_ratio/high_mean": 0.0014583842566935346,
|
| 215 |
+
"clip_ratio/low_mean": 0.00453876843675971,
|
| 216 |
+
"clip_ratio/low_min": 0.00054931640625,
|
| 217 |
+
"clip_ratio/region_mean": 0.0059971527371089906,
|
| 218 |
+
"epoch": 0.0875,
|
| 219 |
+
"grad_norm": 2.6905511453472157,
|
| 220 |
+
"kl": 0.0034446166828274727,
|
| 221 |
+
"learning_rate": 9.917042286978063e-07,
|
| 222 |
+
"loss": 0.0006697401404380798,
|
| 223 |
+
"step": 14,
|
| 224 |
+
"step_time": 25.569942705798894
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"clip_ratio/high_max": 0.009010291862068698,
|
| 228 |
+
"clip_ratio/high_mean": 0.002318086590094026,
|
| 229 |
+
"clip_ratio/low_mean": 0.002421274475636892,
|
| 230 |
+
"clip_ratio/low_min": 0.0,
|
| 231 |
+
"clip_ratio/region_mean": 0.004739361058454961,
|
| 232 |
+
"epoch": 0.09375,
|
| 233 |
+
"grad_norm": 2.2287388440660787,
|
| 234 |
+
"kl": 0.0012414219745551236,
|
| 235 |
+
"learning_rate": 9.897649706262473e-07,
|
| 236 |
+
"loss": 0.0004444615333341062,
|
| 237 |
+
"step": 15,
|
| 238 |
+
"step_time": 24.764580907300115
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"clip_ratio/high_max": 0.0028284954896662384,
|
| 242 |
+
"clip_ratio/high_mean": 0.0012942545872647315,
|
| 243 |
+
"clip_ratio/low_mean": 0.002123031023074873,
|
| 244 |
+
"clip_ratio/low_min": 0.0,
|
| 245 |
+
"clip_ratio/region_mean": 0.00341728565399535,
|
| 246 |
+
"epoch": 0.1,
|
| 247 |
+
"grad_norm": 1.7380753656284031,
|
| 248 |
+
"kl": 0.001310213265242055,
|
| 249 |
+
"learning_rate": 9.87624521252587e-07,
|
| 250 |
+
"loss": 0.0004873822326771915,
|
| 251 |
+
"step": 16,
|
| 252 |
+
"step_time": 25.09028876433149
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"clip_ratio/high_max": 0.0,
|
| 256 |
+
"clip_ratio/high_mean": 0.0,
|
| 257 |
+
"clip_ratio/low_mean": 0.0,
|
| 258 |
+
"clip_ratio/low_min": 0.0,
|
| 259 |
+
"clip_ratio/region_mean": 0.0,
|
| 260 |
+
"completions/clipped_ratio": 0.3203125,
|
| 261 |
+
"completions/max_length": 2048.0,
|
| 262 |
+
"completions/mean_length": 1044.890625,
|
| 263 |
+
"completions/min_length": 56.0,
|
| 264 |
+
"epoch": 0.10625,
|
| 265 |
+
"frac_reward_zero_std": 0.25,
|
| 266 |
+
"grad_norm": 6.805789030752616,
|
| 267 |
+
"kl": 0.0031931393168633804,
|
| 268 |
+
"learning_rate": 9.852837598553008e-07,
|
| 269 |
+
"loss": 3.19555401802063e-05,
|
| 270 |
+
"reward": 0.328125,
|
| 271 |
+
"reward_std": 0.31353163719177246,
|
| 272 |
+
"rewards/MajorityAccuracy/mean": 0.328125,
|
| 273 |
+
"rewards/MajorityAccuracy/std": 0.47045034170150757,
|
| 274 |
+
"step": 17,
|
| 275 |
+
"step_time": 189.1639983127825
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"clip_ratio/high_max": 0.0025206318823620677,
|
| 279 |
+
"clip_ratio/high_mean": 0.0007696669199503958,
|
| 280 |
+
"clip_ratio/low_mean": 0.0001068115234375,
|
| 281 |
+
"clip_ratio/low_min": 0.0,
|
| 282 |
+
"clip_ratio/region_mean": 0.0008764784433878958,
|
| 283 |
+
"epoch": 0.1125,
|
| 284 |
+
"grad_norm": 2.1558921319875557,
|
| 285 |
+
"kl": 0.003207290865248069,
|
| 286 |
+
"learning_rate": 9.827436479993468e-07,
|
| 287 |
+
"loss": 0.0004057286714669317,
|
| 288 |
+
"step": 18,
|
| 289 |
+
"step_time": 24.79515982652083
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"clip_ratio/high_max": 0.0023629064962733537,
|
| 293 |
+
"clip_ratio/high_mean": 0.0006597685860469937,
|
| 294 |
+
"clip_ratio/low_mean": 0.00025854663363134023,
|
| 295 |
+
"clip_ratio/low_min": 0.0,
|
| 296 |
+
"clip_ratio/region_mean": 0.0009183152214973234,
|
| 297 |
+
"epoch": 0.11875,
|
| 298 |
+
"grad_norm": 2.7133155884706346,
|
| 299 |
+
"kl": 0.0027510237559909,
|
| 300 |
+
"learning_rate": 9.800052291411629e-07,
|
| 301 |
+
"loss": 0.00012091256212443113,
|
| 302 |
+
"step": 19,
|
| 303 |
+
"step_time": 25.075154999271035
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"clip_ratio/high_max": 0.0027903425216209143,
|
| 307 |
+
"clip_ratio/high_mean": 0.0006975856304052286,
|
| 308 |
+
"clip_ratio/low_mean": 0.0013582724204752594,
|
| 309 |
+
"clip_ratio/low_min": 0.0,
|
| 310 |
+
"clip_ratio/region_mean": 0.0020558580363285728,
|
| 311 |
+
"epoch": 0.125,
|
| 312 |
+
"grad_norm": 3.1789194424254177,
|
| 313 |
+
"kl": 0.0046101817570161074,
|
| 314 |
+
"learning_rate": 9.770696282000244e-07,
|
| 315 |
+
"loss": 0.0003015280526597053,
|
| 316 |
+
"step": 20,
|
| 317 |
+
"step_time": 25.26610900228843
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"clip_ratio/high_max": 0.001730706644593738,
|
| 321 |
+
"clip_ratio/high_mean": 0.0005691395454050507,
|
| 322 |
+
"clip_ratio/low_mean": 0.0007820180398994125,
|
| 323 |
+
"clip_ratio/low_min": 0.0,
|
| 324 |
+
"clip_ratio/region_mean": 0.001351157588942442,
|
| 325 |
+
"epoch": 0.13125,
|
| 326 |
+
"grad_norm": 3.3185535770089905,
|
| 327 |
+
"kl": 0.0020417476480361074,
|
| 328 |
+
"learning_rate": 9.739380510959364e-07,
|
| 329 |
+
"loss": -6.424075399991125e-05,
|
| 330 |
+
"step": 21,
|
| 331 |
+
"step_time": 24.835173186380416
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"clip_ratio/high_max": 0.0060917426308151335,
|
| 335 |
+
"clip_ratio/high_mean": 0.002575812402938027,
|
| 336 |
+
"clip_ratio/low_mean": 0.005773131979367463,
|
| 337 |
+
"clip_ratio/low_min": 0.0,
|
| 338 |
+
"clip_ratio/region_mean": 0.008348944335011765,
|
| 339 |
+
"epoch": 0.1375,
|
| 340 |
+
"grad_norm": 7.237481476471507,
|
| 341 |
+
"kl": 0.010612010140903294,
|
| 342 |
+
"learning_rate": 9.706117842542516e-07,
|
| 343 |
+
"loss": 0.00047139544039964676,
|
| 344 |
+
"step": 22,
|
| 345 |
+
"step_time": 23.643094944301993
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"clip_ratio/high_max": 0.0015243901871144772,
|
| 349 |
+
"clip_ratio/high_mean": 0.0003810975467786193,
|
| 350 |
+
"clip_ratio/low_mean": 0.001761873543728143,
|
| 351 |
+
"clip_ratio/low_min": 0.0,
|
| 352 |
+
"clip_ratio/region_mean": 0.0021429710905067623,
|
| 353 |
+
"epoch": 0.14375,
|
| 354 |
+
"grad_norm": 1.7607782035622226,
|
| 355 |
+
"kl": 0.0021426881285151467,
|
| 356 |
+
"learning_rate": 9.670921940772186e-07,
|
| 357 |
+
"loss": -4.4550040911417454e-05,
|
| 358 |
+
"step": 23,
|
| 359 |
+
"step_time": 25.049217922147363
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"clip_ratio/high_max": 0.017371104273479432,
|
| 363 |
+
"clip_ratio/high_mean": 0.005230348950135522,
|
| 364 |
+
"clip_ratio/low_mean": 0.006319820415228605,
|
| 365 |
+
"clip_ratio/low_min": 0.00042724609375,
|
| 366 |
+
"clip_ratio/region_mean": 0.011550169321708381,
|
| 367 |
+
"epoch": 0.15,
|
| 368 |
+
"grad_norm": 4.067056457080058,
|
| 369 |
+
"kl": 0.009810720104724169,
|
| 370 |
+
"learning_rate": 9.633807263826744e-07,
|
| 371 |
+
"loss": 0.0026018209755420685,
|
| 372 |
+
"step": 24,
|
| 373 |
+
"step_time": 24.847860149573535
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"clip_ratio/high_max": 0.0,
|
| 377 |
+
"clip_ratio/high_mean": 0.0,
|
| 378 |
+
"clip_ratio/low_mean": 0.0,
|
| 379 |
+
"clip_ratio/low_min": 0.0,
|
| 380 |
+
"clip_ratio/region_mean": 0.0,
|
| 381 |
+
"completions/clipped_ratio": 0.25390625,
|
| 382 |
+
"completions/max_length": 2048.0,
|
| 383 |
+
"completions/mean_length": 868.828125,
|
| 384 |
+
"completions/min_length": 59.0,
|
| 385 |
+
"epoch": 0.15625,
|
| 386 |
+
"frac_reward_zero_std": 0.3125,
|
| 387 |
+
"grad_norm": 1.9682809072812284,
|
| 388 |
+
"kl": 0.008483687415719032,
|
| 389 |
+
"learning_rate": 9.594789058101153e-07,
|
| 390 |
+
"loss": 8.484808495268226e-05,
|
| 391 |
+
"reward": 0.3359375,
|
| 392 |
+
"reward_std": 0.28289473056793213,
|
| 393 |
+
"rewards/MajorityAccuracy/mean": 0.3359375,
|
| 394 |
+
"rewards/MajorityAccuracy/std": 0.4732423424720764,
|
| 395 |
+
"step": 25,
|
| 396 |
+
"step_time": 187.20624068798497
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"clip_ratio/high_max": 0.0013789869553875178,
|
| 400 |
+
"clip_ratio/high_mean": 0.00034474673884687945,
|
| 401 |
+
"clip_ratio/low_mean": 0.0003730371972778812,
|
| 402 |
+
"clip_ratio/low_min": 0.0,
|
| 403 |
+
"clip_ratio/region_mean": 0.0007177839215728454,
|
| 404 |
+
"epoch": 0.1625,
|
| 405 |
+
"grad_norm": 1.738499353945647,
|
| 406 |
+
"kl": 0.009463958733249456,
|
| 407 |
+
"learning_rate": 9.55388335194388e-07,
|
| 408 |
+
"loss": -6.897425191709772e-05,
|
| 409 |
+
"step": 26,
|
| 410 |
+
"step_time": 23.515622918959707
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"clip_ratio/high_max": 0.0022338806884363294,
|
| 414 |
+
"clip_ratio/high_mean": 0.0005584701721090823,
|
| 415 |
+
"clip_ratio/low_mean": 0.0,
|
| 416 |
+
"clip_ratio/low_min": 0.0,
|
| 417 |
+
"clip_ratio/region_mean": 0.0005584701721090823,
|
| 418 |
+
"epoch": 0.16875,
|
| 419 |
+
"grad_norm": 1.2534798394261215,
|
| 420 |
+
"kl": 0.0056053881999105215,
|
| 421 |
+
"learning_rate": 9.511106949072587e-07,
|
| 422 |
+
"loss": -0.00011148105841130018,
|
| 423 |
+
"step": 27,
|
| 424 |
+
"step_time": 25.057520766276866
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"clip_ratio/high_max": 0.004345959328929894,
|
| 428 |
+
"clip_ratio/high_mean": 0.0011435153828642797,
|
| 429 |
+
"clip_ratio/low_mean": 0.001302871391089866,
|
| 430 |
+
"clip_ratio/low_min": 0.0,
|
| 431 |
+
"clip_ratio/region_mean": 0.0024463868176098913,
|
| 432 |
+
"epoch": 0.175,
|
| 433 |
+
"grad_norm": 3.3748451006420703,
|
| 434 |
+
"kl": 0.006861920468509197,
|
| 435 |
+
"learning_rate": 9.466477421671295e-07,
|
| 436 |
+
"loss": 0.0005592242814600468,
|
| 437 |
+
"step": 28,
|
| 438 |
+
"step_time": 22.2740297941491
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"clip_ratio/high_max": 0.0044106631248723716,
|
| 442 |
+
"clip_ratio/high_mean": 0.00122898932750104,
|
| 443 |
+
"clip_ratio/low_mean": 0.004488740567467175,
|
| 444 |
+
"clip_ratio/low_min": 0.00016567263810429722,
|
| 445 |
+
"clip_ratio/region_mean": 0.005717729902244173,
|
| 446 |
+
"epoch": 0.18125,
|
| 447 |
+
"grad_norm": 2.036538276757882,
|
| 448 |
+
"kl": 0.008326348819537088,
|
| 449 |
+
"learning_rate": 9.420013103171891e-07,
|
| 450 |
+
"loss": 0.000503048300743103,
|
| 451 |
+
"step": 29,
|
| 452 |
+
"step_time": 24.970204835291952
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"clip_ratio/high_max": 0.009880052995868027,
|
| 456 |
+
"clip_ratio/high_mean": 0.0030476697720587254,
|
| 457 |
+
"clip_ratio/low_mean": 0.0027299574576318264,
|
| 458 |
+
"clip_ratio/low_min": 0.0,
|
| 459 |
+
"clip_ratio/region_mean": 0.00577762714237906,
|
| 460 |
+
"epoch": 0.1875,
|
| 461 |
+
"grad_norm": 3.514981883029243,
|
| 462 |
+
"kl": 0.011836993391625583,
|
| 463 |
+
"learning_rate": 9.37173308072291e-07,
|
| 464 |
+
"loss": 0.0015928818611428142,
|
| 465 |
+
"step": 30,
|
| 466 |
+
"step_time": 23.238717501517385
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"clip_ratio/high_max": 0.0012295081978663802,
|
| 470 |
+
"clip_ratio/high_mean": 0.0004754781839437783,
|
| 471 |
+
"clip_ratio/low_mean": 0.0018195335433119908,
|
| 472 |
+
"clip_ratio/low_min": 0.0,
|
| 473 |
+
"clip_ratio/region_mean": 0.002295011727255769,
|
| 474 |
+
"epoch": 0.19375,
|
| 475 |
+
"grad_norm": 1.4800569225714675,
|
| 476 |
+
"kl": 0.004693487600889057,
|
| 477 |
+
"learning_rate": 9.321657187348688e-07,
|
| 478 |
+
"loss": 0.00036853610072284937,
|
| 479 |
+
"step": 31,
|
| 480 |
+
"step_time": 25.424959293566644
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"clip_ratio/high_max": 0.009604780178051442,
|
| 484 |
+
"clip_ratio/high_mean": 0.003546650448697619,
|
| 485 |
+
"clip_ratio/low_mean": 0.0058690582372946665,
|
| 486 |
+
"clip_ratio/low_min": 0.00030517578125,
|
| 487 |
+
"clip_ratio/region_mean": 0.009415708598680794,
|
| 488 |
+
"epoch": 0.2,
|
| 489 |
+
"grad_norm": 4.714669825331098,
|
| 490 |
+
"kl": 0.008068033959716558,
|
| 491 |
+
"learning_rate": 9.269805993802128e-07,
|
| 492 |
+
"loss": 0.0011119255796074867,
|
| 493 |
+
"step": 32,
|
| 494 |
+
"step_time": 24.574170005507767
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"clip_ratio/high_max": 0.0,
|
| 498 |
+
"clip_ratio/high_mean": 0.0,
|
| 499 |
+
"clip_ratio/low_mean": 0.0,
|
| 500 |
+
"clip_ratio/low_min": 0.0,
|
| 501 |
+
"clip_ratio/region_mean": 0.0,
|
| 502 |
+
"completions/clipped_ratio": 0.23046875,
|
| 503 |
+
"completions/max_length": 2048.0,
|
| 504 |
+
"completions/mean_length": 834.5234375,
|
| 505 |
+
"completions/min_length": 61.0,
|
| 506 |
+
"epoch": 0.20625,
|
| 507 |
+
"frac_reward_zero_std": 0.3125,
|
| 508 |
+
"grad_norm": 3.3104570337163537,
|
| 509 |
+
"kl": 0.009443770570214838,
|
| 510 |
+
"learning_rate": 9.216200800114411e-07,
|
| 511 |
+
"loss": 9.43961349548772e-05,
|
| 512 |
+
"reward": 0.34765625,
|
| 513 |
+
"reward_std": 0.32495206594467163,
|
| 514 |
+
"rewards/MajorityAccuracy/mean": 0.34765625,
|
| 515 |
+
"rewards/MajorityAccuracy/std": 0.4771590530872345,
|
| 516 |
+
"step": 33,
|
| 517 |
+
"step_time": 184.52237364277244
|
| 518 |
+
},
|
| 519 |
+
{
|
| 520 |
+
"clip_ratio/high_max": 0.0032373879657825455,
|
| 521 |
+
"clip_ratio/high_mean": 0.0008093469914456364,
|
| 522 |
+
"clip_ratio/low_mean": 0.0011052026238758117,
|
| 523 |
+
"clip_ratio/low_min": 0.0,
|
| 524 |
+
"clip_ratio/region_mean": 0.0019145496225974057,
|
| 525 |
+
"epoch": 0.2125,
|
| 526 |
+
"grad_norm": 3.592766649586248,
|
| 527 |
+
"kl": 0.0066102222190238535,
|
| 528 |
+
"learning_rate": 9.160863626845119e-07,
|
| 529 |
+
"loss": 0.00014210771769285202,
|
| 530 |
+
"step": 34,
|
| 531 |
+
"step_time": 23.26577059738338
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"clip_ratio/high_max": 0.0014600879221688956,
|
| 535 |
+
"clip_ratio/high_mean": 0.0003650219805422239,
|
| 536 |
+
"clip_ratio/low_mean": 0.00035214892704971135,
|
| 537 |
+
"clip_ratio/low_min": 0.0,
|
| 538 |
+
"clip_ratio/region_mean": 0.0007171709003159776,
|
| 539 |
+
"epoch": 0.21875,
|
| 540 |
+
"grad_norm": 1.5816741032239476,
|
| 541 |
+
"kl": 0.009619420394301414,
|
| 542 |
+
"learning_rate": 9.103817206036382e-07,
|
| 543 |
+
"loss": 0.0003968477831222117,
|
| 544 |
+
"step": 35,
|
| 545 |
+
"step_time": 25.560568588785827
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"clip_ratio/high_max": 0.0035340442846063524,
|
| 549 |
+
"clip_ratio/high_mean": 0.0009705584161565639,
|
| 550 |
+
"clip_ratio/low_mean": 0.0013776864361716434,
|
| 551 |
+
"clip_ratio/low_min": 0.0,
|
| 552 |
+
"clip_ratio/region_mean": 0.00234824487415608,
|
| 553 |
+
"epoch": 0.225,
|
| 554 |
+
"grad_norm": 3.2005491954421776,
|
| 555 |
+
"kl": 0.007910015410743654,
|
| 556 |
+
"learning_rate": 9.045084971874737e-07,
|
| 557 |
+
"loss": 0.002017698949202895,
|
| 558 |
+
"step": 36,
|
| 559 |
+
"step_time": 25.161365089938045
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"clip_ratio/high_max": 0.0017148912520497106,
|
| 563 |
+
"clip_ratio/high_mean": 0.00042872281301242765,
|
| 564 |
+
"clip_ratio/low_mean": 0.0014712320880789775,
|
| 565 |
+
"clip_ratio/low_min": 0.0,
|
| 566 |
+
"clip_ratio/region_mean": 0.0018999548519786913,
|
| 567 |
+
"epoch": 0.23125,
|
| 568 |
+
"grad_norm": 2.3730163270416056,
|
| 569 |
+
"kl": 0.005281591234961525,
|
| 570 |
+
"learning_rate": 8.984691051064574e-07,
|
| 571 |
+
"loss": 0.0003937923174817115,
|
| 572 |
+
"step": 37,
|
| 573 |
+
"step_time": 23.593793312553316
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"clip_ratio/high_max": 0.008605309878475964,
|
| 577 |
+
"clip_ratio/high_mean": 0.002151327469618991,
|
| 578 |
+
"clip_ratio/low_mean": 0.003125456307316199,
|
| 579 |
+
"clip_ratio/low_min": 0.0,
|
| 580 |
+
"clip_ratio/region_mean": 0.00527678380603902,
|
| 581 |
+
"epoch": 0.2375,
|
| 582 |
+
"grad_norm": 5.775254733292857,
|
| 583 |
+
"kl": 0.015758318244479597,
|
| 584 |
+
"learning_rate": 8.922660252917087e-07,
|
| 585 |
+
"loss": 0.0004318291903473437,
|
| 586 |
+
"step": 38,
|
| 587 |
+
"step_time": 25.020181828644127
|
| 588 |
+
},
|
| 589 |
+
{
|
| 590 |
+
"clip_ratio/high_max": 0.001855421403888613,
|
| 591 |
+
"clip_ratio/high_mean": 0.00046385535097215325,
|
| 592 |
+
"clip_ratio/low_mean": 0.0006370346745825373,
|
| 593 |
+
"clip_ratio/low_min": 0.0,
|
| 594 |
+
"clip_ratio/region_mean": 0.0011008900255546905,
|
| 595 |
+
"epoch": 0.24375,
|
| 596 |
+
"grad_norm": 0.9028194747565357,
|
| 597 |
+
"kl": 0.005992896214593202,
|
| 598 |
+
"learning_rate": 8.859018059158809e-07,
|
| 599 |
+
"loss": 0.00045393695472739637,
|
| 600 |
+
"step": 39,
|
| 601 |
+
"step_time": 25.349115116521716
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"clip_ratio/high_max": 0.016045166761614382,
|
| 605 |
+
"clip_ratio/high_mean": 0.005169198557268828,
|
| 606 |
+
"clip_ratio/low_mean": 0.003129285527393222,
|
| 607 |
+
"clip_ratio/low_min": 0.0,
|
| 608 |
+
"clip_ratio/region_mean": 0.008298484201077372,
|
| 609 |
+
"epoch": 0.25,
|
| 610 |
+
"grad_norm": 2.6278499582638393,
|
| 611 |
+
"kl": 0.015806431707460433,
|
| 612 |
+
"learning_rate": 8.793790613463954e-07,
|
| 613 |
+
"loss": 0.0004564480041153729,
|
| 614 |
+
"step": 40,
|
| 615 |
+
"step_time": 24.322583317756653
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"clip_ratio/high_max": 0.0,
|
| 619 |
+
"clip_ratio/high_mean": 0.0,
|
| 620 |
+
"clip_ratio/low_mean": 0.0,
|
| 621 |
+
"clip_ratio/low_min": 0.0,
|
| 622 |
+
"clip_ratio/region_mean": 0.0,
|
| 623 |
+
"completions/clipped_ratio": 0.20703125,
|
| 624 |
+
"completions/max_length": 2048.0,
|
| 625 |
+
"completions/mean_length": 830.5859375,
|
| 626 |
+
"completions/min_length": 52.0,
|
| 627 |
+
"epoch": 0.25625,
|
| 628 |
+
"frac_reward_zero_std": 0.3125,
|
| 629 |
+
"grad_norm": 4.435675504264506,
|
| 630 |
+
"kl": 0.01965052611194551,
|
| 631 |
+
"learning_rate": 8.727004710714798e-07,
|
| 632 |
+
"loss": 0.00019637649529613554,
|
| 633 |
+
"reward": 0.375,
|
| 634 |
+
"reward_std": 0.30969613790512085,
|
| 635 |
+
"rewards/MajorityAccuracy/mean": 0.375,
|
| 636 |
+
"rewards/MajorityAccuracy/std": 0.48507124185562134,
|
| 637 |
+
"step": 41,
|
| 638 |
+
"step_time": 186.1208955203183
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"clip_ratio/high_max": 0.0020988013202440925,
|
| 642 |
+
"clip_ratio/high_mean": 0.0005776684938609833,
|
| 643 |
+
"clip_ratio/low_mean": 0.0001068115234375,
|
| 644 |
+
"clip_ratio/low_min": 0.0,
|
| 645 |
+
"clip_ratio/region_mean": 0.0006844800009275787,
|
| 646 |
+
"epoch": 0.2625,
|
| 647 |
+
"grad_norm": 2.3767152343469546,
|
| 648 |
+
"kl": 0.011339177552144974,
|
| 649 |
+
"learning_rate": 8.658687785994578e-07,
|
| 650 |
+
"loss": -9.881815640255809e-05,
|
| 651 |
+
"step": 42,
|
| 652 |
+
"step_time": 25.273622174281627
|
| 653 |
+
},
|
| 654 |
+
{
|
| 655 |
+
"clip_ratio/high_max": 0.0001229709741892293,
|
| 656 |
+
"clip_ratio/high_mean": 3.074274354730733e-05,
|
| 657 |
+
"clip_ratio/low_mean": 0.00019929899826820474,
|
| 658 |
+
"clip_ratio/low_min": 0.0,
|
| 659 |
+
"clip_ratio/region_mean": 0.00023004174181551207,
|
| 660 |
+
"epoch": 0.26875,
|
| 661 |
+
"grad_norm": 0.601131966837198,
|
| 662 |
+
"kl": 0.004688937624450773,
|
| 663 |
+
"learning_rate": 8.588867903317394e-07,
|
| 664 |
+
"loss": 1.4267490769270808e-05,
|
| 665 |
+
"step": 43,
|
| 666 |
+
"step_time": 25.10782348830253
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"clip_ratio/high_max": 0.00360600691055879,
|
| 670 |
+
"clip_ratio/high_mean": 0.0009015017276396975,
|
| 671 |
+
"clip_ratio/low_mean": 0.0007220024872367503,
|
| 672 |
+
"clip_ratio/low_min": 0.0,
|
| 673 |
+
"clip_ratio/region_mean": 0.0016235042148764478,
|
| 674 |
+
"epoch": 0.275,
|
| 675 |
+
"grad_norm": 1.873107512648794,
|
| 676 |
+
"kl": 0.008548963931389153,
|
| 677 |
+
"learning_rate": 8.51757374409974e-07,
|
| 678 |
+
"loss": 0.0001463306980440393,
|
| 679 |
+
"step": 44,
|
| 680 |
+
"step_time": 25.522821152582765
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"clip_ratio/high_max": 0.004330091382144019,
|
| 684 |
+
"clip_ratio/high_mean": 0.0014082388006499968,
|
| 685 |
+
"clip_ratio/low_mean": 0.0006772555061616004,
|
| 686 |
+
"clip_ratio/low_min": 0.0,
|
| 687 |
+
"clip_ratio/region_mean": 0.002085494306811597,
|
| 688 |
+
"epoch": 0.28125,
|
| 689 |
+
"grad_norm": 4.599381036371639,
|
| 690 |
+
"kl": 0.022559158969670534,
|
| 691 |
+
"learning_rate": 8.444834595378433e-07,
|
| 692 |
+
"loss": 0.00041219330159947276,
|
| 693 |
+
"step": 45,
|
| 694 |
+
"step_time": 23.75522142741829
|
| 695 |
+
},
|
| 696 |
+
{
|
| 697 |
+
"clip_ratio/high_max": 0.004391021211631596,
|
| 698 |
+
"clip_ratio/high_mean": 0.0011394219764042646,
|
| 699 |
+
"clip_ratio/low_mean": 0.00252378593722824,
|
| 700 |
+
"clip_ratio/low_min": 0.0,
|
| 701 |
+
"clip_ratio/region_mean": 0.003663207906356547,
|
| 702 |
+
"epoch": 0.2875,
|
| 703 |
+
"grad_norm": 2.7716800375688253,
|
| 704 |
+
"kl": 0.015145306009799242,
|
| 705 |
+
"learning_rate": 8.370680337779736e-07,
|
| 706 |
+
"loss": 0.0004911655560135841,
|
| 707 |
+
"step": 46,
|
| 708 |
+
"step_time": 23.005441825836897
|
| 709 |
+
},
|
| 710 |
+
{
|
| 711 |
+
"clip_ratio/high_max": 0.00517284162924625,
|
| 712 |
+
"clip_ratio/high_mean": 0.0014013953696121462,
|
| 713 |
+
"clip_ratio/low_mean": 0.0018433599980198778,
|
| 714 |
+
"clip_ratio/low_min": 0.0,
|
| 715 |
+
"clip_ratio/region_mean": 0.0032447553821839392,
|
| 716 |
+
"epoch": 0.29375,
|
| 717 |
+
"grad_norm": 2.4036114929278,
|
| 718 |
+
"kl": 0.015055993222631514,
|
| 719 |
+
"learning_rate": 8.295141433244659e-07,
|
| 720 |
+
"loss": -5.273032002151012e-06,
|
| 721 |
+
"step": 47,
|
| 722 |
+
"step_time": 23.920041020493954
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"clip_ratio/high_max": 0.006011248624417931,
|
| 726 |
+
"clip_ratio/high_mean": 0.002285138485603966,
|
| 727 |
+
"clip_ratio/low_mean": 0.002038680664554704,
|
| 728 |
+
"clip_ratio/low_min": 0.0001220703125,
|
| 729 |
+
"clip_ratio/region_mean": 0.004323819128330797,
|
| 730 |
+
"epoch": 0.3,
|
| 731 |
+
"grad_norm": 3.108941874020213,
|
| 732 |
+
"kl": 0.012410315102897584,
|
| 733 |
+
"learning_rate": 8.218248912515442e-07,
|
| 734 |
+
"loss": 0.0005759168416261673,
|
| 735 |
+
"step": 48,
|
| 736 |
+
"step_time": 24.512430033180863
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"clip_ratio/high_max": 0.0,
|
| 740 |
+
"clip_ratio/high_mean": 0.0,
|
| 741 |
+
"clip_ratio/low_mean": 0.0,
|
| 742 |
+
"clip_ratio/low_min": 0.0,
|
| 743 |
+
"clip_ratio/region_mean": 0.0,
|
| 744 |
+
"completions/clipped_ratio": 0.33203125,
|
| 745 |
+
"completions/max_length": 2048.0,
|
| 746 |
+
"completions/mean_length": 1103.98046875,
|
| 747 |
+
"completions/min_length": 81.0,
|
| 748 |
+
"epoch": 0.30625,
|
| 749 |
+
"frac_reward_zero_std": 0.25,
|
| 750 |
+
"grad_norm": 2.730843043457364,
|
| 751 |
+
"kl": 0.006533110630698502,
|
| 752 |
+
"learning_rate": 8.140034362388397e-07,
|
| 753 |
+
"loss": 6.536219007102773e-05,
|
| 754 |
+
"reward": 0.33984375,
|
| 755 |
+
"reward_std": 0.35494479537010193,
|
| 756 |
+
"rewards/MajorityAccuracy/mean": 0.33984375,
|
| 757 |
+
"rewards/MajorityAccuracy/std": 0.47458380460739136,
|
| 758 |
+
"step": 49,
|
| 759 |
+
"step_time": 189.34944529784843
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"clip_ratio/high_max": 0.003355264023412019,
|
| 763 |
+
"clip_ratio/high_mean": 0.0010740415309555829,
|
| 764 |
+
"clip_ratio/low_mean": 0.001083829702110961,
|
| 765 |
+
"clip_ratio/low_min": 0.0,
|
| 766 |
+
"clip_ratio/region_mean": 0.0021578712548944168,
|
| 767 |
+
"epoch": 0.3125,
|
| 768 |
+
"grad_norm": 2.709807168302933,
|
| 769 |
+
"kl": 0.009958796901628375,
|
| 770 |
+
"learning_rate": 8.060529912738314e-07,
|
| 771 |
+
"loss": 0.00023437291383743286,
|
| 772 |
+
"step": 50,
|
| 773 |
+
"step_time": 24.296555117238313
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"clip_ratio/high_max": 0.0005043650744482875,
|
| 777 |
+
"clip_ratio/high_mean": 0.00012609126861207187,
|
| 778 |
+
"clip_ratio/low_mean": 0.000831386205391027,
|
| 779 |
+
"clip_ratio/low_min": 0.0,
|
| 780 |
+
"clip_ratio/region_mean": 0.0009574774667271413,
|
| 781 |
+
"epoch": 0.31875,
|
| 782 |
+
"grad_norm": 4.708793363396933,
|
| 783 |
+
"kl": 0.009627448802348226,
|
| 784 |
+
"learning_rate": 7.979768223319785e-07,
|
| 785 |
+
"loss": 0.0008214515983127058,
|
| 786 |
+
"step": 51,
|
| 787 |
+
"step_time": 25.24472848745063
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"clip_ratio/high_max": 0.005039593728724867,
|
| 791 |
+
"clip_ratio/high_mean": 0.001322398442425765,
|
| 792 |
+
"clip_ratio/low_mean": 0.001815359697502572,
|
| 793 |
+
"clip_ratio/low_min": 0.0,
|
| 794 |
+
"clip_ratio/region_mean": 0.003137758103548549,
|
| 795 |
+
"epoch": 0.325,
|
| 796 |
+
"grad_norm": 2.1085325724309687,
|
| 797 |
+
"kl": 0.01827312121167779,
|
| 798 |
+
"learning_rate": 7.897782470350849e-07,
|
| 799 |
+
"loss": 0.0006650196737609804,
|
| 800 |
+
"step": 52,
|
| 801 |
+
"step_time": 25.206268513575196
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"clip_ratio/high_max": 0.00302134090452455,
|
| 805 |
+
"clip_ratio/high_mean": 0.0007553352261311375,
|
| 806 |
+
"clip_ratio/low_mean": 0.0005035400390625,
|
| 807 |
+
"clip_ratio/low_min": 0.0,
|
| 808 |
+
"clip_ratio/region_mean": 0.0012588752579176798,
|
| 809 |
+
"epoch": 0.33125,
|
| 810 |
+
"grad_norm": 1.979577048243759,
|
| 811 |
+
"kl": 0.015566267189569771,
|
| 812 |
+
"learning_rate": 7.814606332884488e-07,
|
| 813 |
+
"loss": 0.0006761232507415116,
|
| 814 |
+
"step": 53,
|
| 815 |
+
"step_time": 25.19498684629798
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"clip_ratio/high_max": 0.0034592138254083693,
|
| 819 |
+
"clip_ratio/high_mean": 0.0010057364197564311,
|
| 820 |
+
"clip_ratio/low_mean": 0.00030979466100689024,
|
| 821 |
+
"clip_ratio/low_min": 0.0,
|
| 822 |
+
"clip_ratio/region_mean": 0.0013155310589354485,
|
| 823 |
+
"epoch": 0.3375,
|
| 824 |
+
"grad_norm": 1.8885285401530618,
|
| 825 |
+
"kl": 0.011069664673414081,
|
| 826 |
+
"learning_rate": 7.730273978973552e-07,
|
| 827 |
+
"loss": 0.0002471886691637337,
|
| 828 |
+
"step": 54,
|
| 829 |
+
"step_time": 24.97195475222543
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"clip_ratio/high_max": 0.001962186273885891,
|
| 833 |
+
"clip_ratio/high_mean": 0.0005515817247214727,
|
| 834 |
+
"clip_ratio/low_mean": 0.001602885065949522,
|
| 835 |
+
"clip_ratio/low_min": 0.0,
|
| 836 |
+
"clip_ratio/region_mean": 0.002154466783395037,
|
| 837 |
+
"epoch": 0.34375,
|
| 838 |
+
"grad_norm": 1.7644331666417563,
|
| 839 |
+
"kl": 0.004976838390575722,
|
| 840 |
+
"learning_rate": 7.644820051634812e-07,
|
| 841 |
+
"loss": 0.0003386082826182246,
|
| 842 |
+
"step": 55,
|
| 843 |
+
"step_time": 24.92111270967871
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"clip_ratio/high_max": 0.007073798013152555,
|
| 847 |
+
"clip_ratio/high_mean": 0.0024187457893276587,
|
| 848 |
+
"clip_ratio/low_mean": 0.002348580601392314,
|
| 849 |
+
"clip_ratio/low_min": 0.00018310546875,
|
| 850 |
+
"clip_ratio/region_mean": 0.004767326463479549,
|
| 851 |
+
"epoch": 0.35,
|
| 852 |
+
"grad_norm": 2.5480165644609083,
|
| 853 |
+
"kl": 0.0065385435009375215,
|
| 854 |
+
"learning_rate": 7.558279654617912e-07,
|
| 855 |
+
"loss": 0.000597221776843071,
|
| 856 |
+
"step": 56,
|
| 857 |
+
"step_time": 24.623682228848338
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"clip_ratio/high_max": 0.0,
|
| 861 |
+
"clip_ratio/high_mean": 0.0,
|
| 862 |
+
"clip_ratio/low_mean": 0.0,
|
| 863 |
+
"clip_ratio/low_min": 0.0,
|
| 864 |
+
"clip_ratio/region_mean": 0.0,
|
| 865 |
+
"completions/clipped_ratio": 0.34765625,
|
| 866 |
+
"completions/max_length": 2048.0,
|
| 867 |
+
"completions/mean_length": 1143.0859375,
|
| 868 |
+
"completions/min_length": 79.0,
|
| 869 |
+
"epoch": 0.35625,
|
| 870 |
+
"frac_reward_zero_std": 0.3125,
|
| 871 |
+
"grad_norm": 2.9666692446453657,
|
| 872 |
+
"kl": 0.01279032719321549,
|
| 873 |
+
"learning_rate": 7.470688337985029e-07,
|
| 874 |
+
"loss": 0.0001279786229133606,
|
| 875 |
+
"reward": 0.27734375,
|
| 876 |
+
"reward_std": 0.3135252594947815,
|
| 877 |
+
"rewards/MajorityAccuracy/mean": 0.27734375,
|
| 878 |
+
"rewards/MajorityAccuracy/std": 0.4485645592212677,
|
| 879 |
+
"step": 57,
|
| 880 |
+
"step_time": 190.7737743984908
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"clip_ratio/high_max": 0.0004636314115487039,
|
| 884 |
+
"clip_ratio/high_mean": 0.00011590785288717598,
|
| 885 |
+
"clip_ratio/low_mean": 0.00016501231584697962,
|
| 886 |
+
"clip_ratio/low_min": 0.0,
|
| 887 |
+
"clip_ratio/region_mean": 0.0002809201760101132,
|
| 888 |
+
"epoch": 0.3625,
|
| 889 |
+
"grad_norm": 1.9326018373776388,
|
| 890 |
+
"kl": 0.011659832438454032,
|
| 891 |
+
"learning_rate": 7.382082083507225e-07,
|
| 892 |
+
"loss": 0.00020526288426481187,
|
| 893 |
+
"step": 58,
|
| 894 |
+
"step_time": 25.164008419029415
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"clip_ratio/high_max": 0.002798642875859514,
|
| 898 |
+
"clip_ratio/high_mean": 0.0009598941032891162,
|
| 899 |
+
"clip_ratio/low_mean": 0.00010335335900890641,
|
| 900 |
+
"clip_ratio/low_min": 0.0,
|
| 901 |
+
"clip_ratio/region_mean": 0.0010632474622980226,
|
| 902 |
+
"epoch": 0.36875,
|
| 903 |
+
"grad_norm": 2.519912179336201,
|
| 904 |
+
"kl": 0.020991226192563772,
|
| 905 |
+
"learning_rate": 7.292497289883432e-07,
|
| 906 |
+
"loss": 0.00015528060612268746,
|
| 907 |
+
"step": 59,
|
| 908 |
+
"step_time": 24.06486978009343
|
| 909 |
+
},
|
| 910 |
+
{
|
| 911 |
+
"clip_ratio/high_max": 0.0011771252829930745,
|
| 912 |
+
"clip_ratio/high_mean": 0.0003370309841557173,
|
| 913 |
+
"clip_ratio/low_mean": 0.00015273837198037654,
|
| 914 |
+
"clip_ratio/low_min": 0.0,
|
| 915 |
+
"clip_ratio/region_mean": 0.0004897693579550833,
|
| 916 |
+
"epoch": 0.375,
|
| 917 |
+
"grad_norm": 1.241801476654187,
|
| 918 |
+
"kl": 0.0053275886457413435,
|
| 919 |
+
"learning_rate": 7.201970757788171e-07,
|
| 920 |
+
"loss": 8.38044798001647e-05,
|
| 921 |
+
"step": 60,
|
| 922 |
+
"step_time": 25.250475091394037
|
| 923 |
+
},
|
| 924 |
+
{
|
| 925 |
+
"clip_ratio/high_max": 0.008908138319384307,
|
| 926 |
+
"clip_ratio/high_mean": 0.0025468366948189214,
|
| 927 |
+
"clip_ratio/low_mean": 0.0010164195809920784,
|
| 928 |
+
"clip_ratio/low_min": 0.0,
|
| 929 |
+
"clip_ratio/region_mean": 0.0035632563194667455,
|
| 930 |
+
"epoch": 0.38125,
|
| 931 |
+
"grad_norm": 2.7323930424254206,
|
| 932 |
+
"kl": 0.011100393487140536,
|
| 933 |
+
"learning_rate": 7.110539674754159e-07,
|
| 934 |
+
"loss": -9.282343671657145e-05,
|
| 935 |
+
"step": 61,
|
| 936 |
+
"step_time": 25.087209795601666
|
| 937 |
+
},
|
| 938 |
+
{
|
| 939 |
+
"clip_ratio/high_max": 0.004926497786073014,
|
| 940 |
+
"clip_ratio/high_mean": 0.0014753090217709541,
|
| 941 |
+
"clip_ratio/low_mean": 0.0008577648477512412,
|
| 942 |
+
"clip_ratio/low_min": 0.0,
|
| 943 |
+
"clip_ratio/region_mean": 0.0023330738476943225,
|
| 944 |
+
"epoch": 0.3875,
|
| 945 |
+
"grad_norm": 8.332830261684553,
|
| 946 |
+
"kl": 0.016185397282242775,
|
| 947 |
+
"learning_rate": 7.018241599895973e-07,
|
| 948 |
+
"loss": 0.00034105314989574254,
|
| 949 |
+
"step": 62,
|
| 950 |
+
"step_time": 25.33527609333396
|
| 951 |
+
},
|
| 952 |
+
{
|
| 953 |
+
"clip_ratio/high_max": 0.004454142908798531,
|
| 954 |
+
"clip_ratio/high_mean": 0.0013302986990311183,
|
| 955 |
+
"clip_ratio/low_mean": 0.004791011211636942,
|
| 956 |
+
"clip_ratio/low_min": 0.0,
|
| 957 |
+
"clip_ratio/region_mean": 0.006121309939771891,
|
| 958 |
+
"epoch": 0.39375,
|
| 959 |
+
"grad_norm": 4.232858212614428,
|
| 960 |
+
"kl": 0.02442008175421506,
|
| 961 |
+
"learning_rate": 6.925114448481088e-07,
|
| 962 |
+
"loss": 0.002828963566571474,
|
| 963 |
+
"step": 63,
|
| 964 |
+
"step_time": 25.49894282594323
|
| 965 |
+
},
|
| 966 |
+
{
|
| 967 |
+
"clip_ratio/high_max": 0.0022660501417703927,
|
| 968 |
+
"clip_ratio/high_mean": 0.0006519417220260948,
|
| 969 |
+
"clip_ratio/low_mean": 0.0008030155004234985,
|
| 970 |
+
"clip_ratio/low_min": 0.0,
|
| 971 |
+
"clip_ratio/region_mean": 0.0014549572151736356,
|
| 972 |
+
"epoch": 0.4,
|
| 973 |
+
"grad_norm": 2.93560472805829,
|
| 974 |
+
"kl": 0.00991360453190282,
|
| 975 |
+
"learning_rate": 6.831196476354614e-07,
|
| 976 |
+
"loss": 0.000980147859081626,
|
| 977 |
+
"step": 64,
|
| 978 |
+
"step_time": 25.155366223771125
|
| 979 |
+
},
|
| 980 |
+
{
|
| 981 |
+
"clip_ratio/high_max": 0.0,
|
| 982 |
+
"clip_ratio/high_mean": 0.0,
|
| 983 |
+
"clip_ratio/low_mean": 0.0,
|
| 984 |
+
"clip_ratio/low_min": 0.0,
|
| 985 |
+
"clip_ratio/region_mean": 0.0,
|
| 986 |
+
"completions/clipped_ratio": 0.28515625,
|
| 987 |
+
"completions/max_length": 2048.0,
|
| 988 |
+
"completions/mean_length": 1043.77734375,
|
| 989 |
+
"completions/min_length": 62.0,
|
| 990 |
+
"epoch": 0.40625,
|
| 991 |
+
"frac_reward_zero_std": 0.28125,
|
| 992 |
+
"grad_norm": 1.8720285423703602,
|
| 993 |
+
"kl": 0.013166663469746709,
|
| 994 |
+
"learning_rate": 6.7365262642241e-07,
|
| 995 |
+
"loss": 0.00013170995225664228,
|
| 996 |
+
"reward": 0.3125,
|
| 997 |
+
"reward_std": 0.324573814868927,
|
| 998 |
+
"rewards/MajorityAccuracy/mean": 0.3125,
|
| 999 |
+
"rewards/MajorityAccuracy/std": 0.4644203782081604,
|
| 1000 |
+
"step": 65,
|
| 1001 |
+
"step_time": 188.755262802355
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"clip_ratio/high_max": 0.0013784582260996103,
|
| 1005 |
+
"clip_ratio/high_mean": 0.0003446145565249026,
|
| 1006 |
+
"clip_ratio/low_mean": 0.0011673199987853877,
|
| 1007 |
+
"clip_ratio/low_min": 0.0,
|
| 1008 |
+
"clip_ratio/region_mean": 0.001511934540758375,
|
| 1009 |
+
"epoch": 0.4125,
|
| 1010 |
+
"grad_norm": 3.947332496313382,
|
| 1011 |
+
"kl": 0.023555473308078945,
|
| 1012 |
+
"learning_rate": 6.641142701810931e-07,
|
| 1013 |
+
"loss": 9.53255221247673e-05,
|
| 1014 |
+
"step": 66,
|
| 1015 |
+
"step_time": 23.481388201005757
|
| 1016 |
+
},
|
| 1017 |
+
{
|
| 1018 |
+
"clip_ratio/high_max": 0.0,
|
| 1019 |
+
"clip_ratio/high_mean": 0.0,
|
| 1020 |
+
"clip_ratio/low_mean": 0.0004651642157114111,
|
| 1021 |
+
"clip_ratio/low_min": 0.0,
|
| 1022 |
+
"clip_ratio/region_mean": 0.0004651642157114111,
|
| 1023 |
+
"epoch": 0.41875,
|
| 1024 |
+
"grad_norm": 1.835942889874058,
|
| 1025 |
+
"kl": 0.006090442155255005,
|
| 1026 |
+
"learning_rate": 6.545084971874736e-07,
|
| 1027 |
+
"loss": 0.00024211208801716566,
|
| 1028 |
+
"step": 67,
|
| 1029 |
+
"step_time": 24.860420756507665
|
| 1030 |
+
},
|
| 1031 |
+
{
|
| 1032 |
+
"clip_ratio/high_max": 0.0008088483591564,
|
| 1033 |
+
"clip_ratio/high_mean": 0.00021960218509775586,
|
| 1034 |
+
"clip_ratio/low_mean": 0.0008722166548977839,
|
| 1035 |
+
"clip_ratio/low_min": 0.0,
|
| 1036 |
+
"clip_ratio/region_mean": 0.0010918188490904868,
|
| 1037 |
+
"epoch": 0.425,
|
| 1038 |
+
"grad_norm": 1.5897330334448532,
|
| 1039 |
+
"kl": 0.011180203640833497,
|
| 1040 |
+
"learning_rate": 6.448392534117461e-07,
|
| 1041 |
+
"loss": -0.00017426944395992905,
|
| 1042 |
+
"step": 68,
|
| 1043 |
+
"step_time": 24.696536101400852
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"clip_ratio/high_max": 0.00032044506224337965,
|
| 1047 |
+
"clip_ratio/high_mean": 8.011126556084491e-05,
|
| 1048 |
+
"clip_ratio/low_mean": 0.0008242859767051414,
|
| 1049 |
+
"clip_ratio/low_min": 0.0,
|
| 1050 |
+
"clip_ratio/region_mean": 0.0009043972459039651,
|
| 1051 |
+
"epoch": 0.43125,
|
| 1052 |
+
"grad_norm": 1.539846896719152,
|
| 1053 |
+
"kl": 0.0075706454226747155,
|
| 1054 |
+
"learning_rate": 6.351105108973644e-07,
|
| 1055 |
+
"loss": 8.448302105534822e-05,
|
| 1056 |
+
"step": 69,
|
| 1057 |
+
"step_time": 25.091761415358633
|
| 1058 |
+
},
|
| 1059 |
+
{
|
| 1060 |
+
"clip_ratio/high_max": 0.010265647782944143,
|
| 1061 |
+
"clip_ratio/high_mean": 0.0031366994080599397,
|
| 1062 |
+
"clip_ratio/low_mean": 0.0010798340081237257,
|
| 1063 |
+
"clip_ratio/low_min": 0.0,
|
| 1064 |
+
"clip_ratio/region_mean": 0.004216533387079835,
|
| 1065 |
+
"epoch": 0.4375,
|
| 1066 |
+
"grad_norm": 8.56758722719742,
|
| 1067 |
+
"kl": 0.01484630221966654,
|
| 1068 |
+
"learning_rate": 6.253262661293602e-07,
|
| 1069 |
+
"loss": 0.0009232594165951014,
|
| 1070 |
+
"step": 70,
|
| 1071 |
+
"step_time": 24.143507443368435
|
| 1072 |
+
},
|
| 1073 |
+
{
|
| 1074 |
+
"clip_ratio/high_max": 0.00421601822017692,
|
| 1075 |
+
"clip_ratio/high_mean": 0.0011167555567226373,
|
| 1076 |
+
"clip_ratio/low_mean": 0.0026076857320731506,
|
| 1077 |
+
"clip_ratio/low_min": 0.0,
|
| 1078 |
+
"clip_ratio/region_mean": 0.003724441325175576,
|
| 1079 |
+
"epoch": 0.44375,
|
| 1080 |
+
"grad_norm": 4.120724314480742,
|
| 1081 |
+
"kl": 0.01342374226078391,
|
| 1082 |
+
"learning_rate": 6.154905383926216e-07,
|
| 1083 |
+
"loss": 7.47879603295587e-05,
|
| 1084 |
+
"step": 71,
|
| 1085 |
+
"step_time": 25.081223375629634
|
| 1086 |
+
},
|
| 1087 |
+
{
|
| 1088 |
+
"clip_ratio/high_max": 0.012967722141183913,
|
| 1089 |
+
"clip_ratio/high_mean": 0.003681420726934448,
|
| 1090 |
+
"clip_ratio/low_mean": 0.002078761159282294,
|
| 1091 |
+
"clip_ratio/low_min": 0.0,
|
| 1092 |
+
"clip_ratio/region_mean": 0.005760181898949668,
|
| 1093 |
+
"epoch": 0.45,
|
| 1094 |
+
"grad_norm": 2.629674246528432,
|
| 1095 |
+
"kl": 0.020645775948651135,
|
| 1096 |
+
"learning_rate": 6.056073681208037e-07,
|
| 1097 |
+
"loss": 0.0004600215470418334,
|
| 1098 |
+
"step": 72,
|
| 1099 |
+
"step_time": 25.150920601096004
|
| 1100 |
+
},
|
| 1101 |
+
{
|
| 1102 |
+
"clip_ratio/high_max": 0.0,
|
| 1103 |
+
"clip_ratio/high_mean": 0.0,
|
| 1104 |
+
"clip_ratio/low_mean": 0.0,
|
| 1105 |
+
"clip_ratio/low_min": 0.0,
|
| 1106 |
+
"clip_ratio/region_mean": 0.0,
|
| 1107 |
+
"completions/clipped_ratio": 0.2578125,
|
| 1108 |
+
"completions/max_length": 2048.0,
|
| 1109 |
+
"completions/mean_length": 959.28515625,
|
| 1110 |
+
"completions/min_length": 77.0,
|
| 1111 |
+
"epoch": 0.45625,
|
| 1112 |
+
"frac_reward_zero_std": 0.40625,
|
| 1113 |
+
"grad_norm": 1.6524036798452102,
|
| 1114 |
+
"kl": 0.018394945189356804,
|
| 1115 |
+
"learning_rate": 5.956808152365532e-07,
|
| 1116 |
+
"loss": 0.0001839824835769832,
|
| 1117 |
+
"reward": 0.3046875,
|
| 1118 |
+
"reward_std": 0.26329511404037476,
|
| 1119 |
+
"rewards/MajorityAccuracy/mean": 0.3046875,
|
| 1120 |
+
"rewards/MajorityAccuracy/std": 0.4611765742301941,
|
| 1121 |
+
"step": 73,
|
| 1122 |
+
"step_time": 187.0246638902463
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"clip_ratio/high_max": 0.0012619348999578506,
|
| 1126 |
+
"clip_ratio/high_mean": 0.00037067890298203565,
|
| 1127 |
+
"clip_ratio/low_mean": 0.00013171192222216632,
|
| 1128 |
+
"clip_ratio/low_min": 0.0,
|
| 1129 |
+
"clip_ratio/region_mean": 0.000502390834299149,
|
| 1130 |
+
"epoch": 0.4625,
|
| 1131 |
+
"grad_norm": 1.8834989402918971,
|
| 1132 |
+
"kl": 0.01169894658960402,
|
| 1133 |
+
"learning_rate": 5.857149574837268e-07,
|
| 1134 |
+
"loss": 0.00023947088629938662,
|
| 1135 |
+
"step": 74,
|
| 1136 |
+
"step_time": 25.26510666217655
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"clip_ratio/high_max": 0.0012876391774625517,
|
| 1140 |
+
"clip_ratio/high_mean": 0.000340769178365008,
|
| 1141 |
+
"clip_ratio/low_mean": 0.00272836798831122,
|
| 1142 |
+
"clip_ratio/low_min": 0.0,
|
| 1143 |
+
"clip_ratio/region_mean": 0.0030691372012370266,
|
| 1144 |
+
"epoch": 0.46875,
|
| 1145 |
+
"grad_norm": 5.186623443467292,
|
| 1146 |
+
"kl": 0.02987821865826845,
|
| 1147 |
+
"learning_rate": 5.757138887522883e-07,
|
| 1148 |
+
"loss": 0.00042566563934087753,
|
| 1149 |
+
"step": 75,
|
| 1150 |
+
"step_time": 24.878897356800735
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"clip_ratio/high_max": 0.0011904762359336019,
|
| 1154 |
+
"clip_ratio/high_mean": 0.00029761905898340046,
|
| 1155 |
+
"clip_ratio/low_mean": 3.0517578125e-05,
|
| 1156 |
+
"clip_ratio/low_min": 0.0,
|
| 1157 |
+
"clip_ratio/region_mean": 0.00032813663710840046,
|
| 1158 |
+
"epoch": 0.475,
|
| 1159 |
+
"grad_norm": 0.7290932019702344,
|
| 1160 |
+
"kl": 0.013780927285552025,
|
| 1161 |
+
"learning_rate": 5.656817173965732e-07,
|
| 1162 |
+
"loss": 7.455376908183098e-05,
|
| 1163 |
+
"step": 76,
|
| 1164 |
+
"step_time": 24.861102600581944
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"clip_ratio/high_max": 0.0,
|
| 1168 |
+
"clip_ratio/high_mean": 0.0,
|
| 1169 |
+
"clip_ratio/low_mean": 0.0,
|
| 1170 |
+
"clip_ratio/low_min": 0.0,
|
| 1171 |
+
"clip_ratio/region_mean": 0.0,
|
| 1172 |
+
"epoch": 0.48125,
|
| 1173 |
+
"grad_norm": 0.054841499540253906,
|
| 1174 |
+
"kl": 0.016748676542192698,
|
| 1175 |
+
"learning_rate": 5.556225645476118e-07,
|
| 1176 |
+
"loss": 0.00016755085380282253,
|
| 1177 |
+
"step": 77,
|
| 1178 |
+
"step_time": 24.9227648046799
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"clip_ratio/high_max": 0.00029895758780185133,
|
| 1182 |
+
"clip_ratio/high_mean": 7.473939695046283e-05,
|
| 1183 |
+
"clip_ratio/low_mean": 0.0011211319506401196,
|
| 1184 |
+
"clip_ratio/low_min": 0.0,
|
| 1185 |
+
"clip_ratio/region_mean": 0.0011958713475905824,
|
| 1186 |
+
"epoch": 0.4875,
|
| 1187 |
+
"grad_norm": 2.5976583105048854,
|
| 1188 |
+
"kl": 0.027776528848335147,
|
| 1189 |
+
"learning_rate": 5.455405624202032e-07,
|
| 1190 |
+
"loss": 0.0005805623950436711,
|
| 1191 |
+
"step": 78,
|
| 1192 |
+
"step_time": 24.20325362170115
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"clip_ratio/high_max": 0.006348784867441282,
|
| 1196 |
+
"clip_ratio/high_mean": 0.0022835059571661986,
|
| 1197 |
+
"clip_ratio/low_mean": 0.000630342197837308,
|
| 1198 |
+
"clip_ratio/low_min": 0.0,
|
| 1199 |
+
"clip_ratio/region_mean": 0.002913848147727549,
|
| 1200 |
+
"epoch": 0.49375,
|
| 1201 |
+
"grad_norm": 2.6125841408693127,
|
| 1202 |
+
"kl": 0.019668483408167958,
|
| 1203 |
+
"learning_rate": 5.354398526154365e-07,
|
| 1204 |
+
"loss": 0.0006650089635513723,
|
| 1205 |
+
"step": 79,
|
| 1206 |
+
"step_time": 24.482460925821215
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"clip_ratio/high_max": 0.0009377771784784272,
|
| 1210 |
+
"clip_ratio/high_mean": 0.0002344442946196068,
|
| 1211 |
+
"clip_ratio/low_mean": 0.0008538617839803919,
|
| 1212 |
+
"clip_ratio/low_min": 0.0,
|
| 1213 |
+
"clip_ratio/region_mean": 0.00108830607496202,
|
| 1214 |
+
"epoch": 0.5,
|
| 1215 |
+
"grad_norm": 1.2329915808573657,
|
| 1216 |
+
"kl": 0.009494279162026942,
|
| 1217 |
+
"learning_rate": 5.253245844193564e-07,
|
| 1218 |
+
"loss": 0.00024856001255102456,
|
| 1219 |
+
"step": 80,
|
| 1220 |
+
"step_time": 25.045087489765137
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"clip_ratio/high_max": 0.0,
|
| 1224 |
+
"clip_ratio/high_mean": 0.0,
|
| 1225 |
+
"clip_ratio/low_mean": 0.0,
|
| 1226 |
+
"clip_ratio/low_min": 0.0,
|
| 1227 |
+
"clip_ratio/region_mean": 0.0,
|
| 1228 |
+
"completions/clipped_ratio": 0.24609375,
|
| 1229 |
+
"completions/max_length": 2048.0,
|
| 1230 |
+
"completions/mean_length": 896.82421875,
|
| 1231 |
+
"completions/min_length": 68.0,
|
| 1232 |
+
"epoch": 0.50625,
|
| 1233 |
+
"frac_reward_zero_std": 0.28125,
|
| 1234 |
+
"grad_norm": 3.3974921992697706,
|
| 1235 |
+
"kl": 0.023021922912448645,
|
| 1236 |
+
"learning_rate": 5.151989130984714e-07,
|
| 1237 |
+
"loss": 0.00023039562802296132,
|
| 1238 |
+
"reward": 0.328125,
|
| 1239 |
+
"reward_std": 0.33653348684310913,
|
| 1240 |
+
"rewards/MajorityAccuracy/mean": 0.328125,
|
| 1241 |
+
"rewards/MajorityAccuracy/std": 0.47045034170150757,
|
| 1242 |
+
"step": 81,
|
| 1243 |
+
"step_time": 186.45555065572262
|
| 1244 |
+
},
|
| 1245 |
+
{
|
| 1246 |
+
"clip_ratio/high_max": 0.001850122818723321,
|
| 1247 |
+
"clip_ratio/high_mean": 0.0005660075403284281,
|
| 1248 |
+
"clip_ratio/low_mean": 0.0007283160302904435,
|
| 1249 |
+
"clip_ratio/low_min": 0.0,
|
| 1250 |
+
"clip_ratio/region_mean": 0.0012943235706188716,
|
| 1251 |
+
"epoch": 0.5125,
|
| 1252 |
+
"grad_norm": 6.1506998835379125,
|
| 1253 |
+
"kl": 0.023190242471173406,
|
| 1254 |
+
"learning_rate": 5.050669981928055e-07,
|
| 1255 |
+
"loss": 0.0005143824964761734,
|
| 1256 |
+
"step": 82,
|
| 1257 |
+
"step_time": 24.80176503676921
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"clip_ratio/high_max": 0.008243723947089165,
|
| 1261 |
+
"clip_ratio/high_mean": 0.0029613986262120306,
|
| 1262 |
+
"clip_ratio/low_mean": 0.0013813759542244952,
|
| 1263 |
+
"clip_ratio/low_min": 0.0,
|
| 1264 |
+
"clip_ratio/region_mean": 0.00434277462773025,
|
| 1265 |
+
"epoch": 0.51875,
|
| 1266 |
+
"grad_norm": 6.092074924720109,
|
| 1267 |
+
"kl": 0.04293838003650308,
|
| 1268 |
+
"learning_rate": 4.949330018071946e-07,
|
| 1269 |
+
"loss": 0.0019269995391368866,
|
| 1270 |
+
"step": 83,
|
| 1271 |
+
"step_time": 23.338271051179618
|
| 1272 |
+
},
|
| 1273 |
+
{
|
| 1274 |
+
"clip_ratio/high_max": 0.0015348413871834055,
|
| 1275 |
+
"clip_ratio/high_mean": 0.00038371034679585136,
|
| 1276 |
+
"clip_ratio/low_mean": 0.0012249260580574628,
|
| 1277 |
+
"clip_ratio/low_min": 0.0,
|
| 1278 |
+
"clip_ratio/region_mean": 0.0016086363830254413,
|
| 1279 |
+
"epoch": 0.525,
|
| 1280 |
+
"grad_norm": 5.846199849884313,
|
| 1281 |
+
"kl": 0.0193571075797081,
|
| 1282 |
+
"learning_rate": 4.848010869015287e-07,
|
| 1283 |
+
"loss": 0.00062585185514763,
|
| 1284 |
+
"step": 84,
|
| 1285 |
+
"step_time": 24.295556480064988
|
| 1286 |
+
},
|
| 1287 |
+
{
|
| 1288 |
+
"clip_ratio/high_max": 0.0037318404065445065,
|
| 1289 |
+
"clip_ratio/high_mean": 0.001191170020319987,
|
| 1290 |
+
"clip_ratio/low_mean": 0.00024115185260598082,
|
| 1291 |
+
"clip_ratio/low_min": 0.0,
|
| 1292 |
+
"clip_ratio/region_mean": 0.001432321863831021,
|
| 1293 |
+
"epoch": 0.53125,
|
| 1294 |
+
"grad_norm": 2.847888132693557,
|
| 1295 |
+
"kl": 0.016478966339491308,
|
| 1296 |
+
"learning_rate": 4.746754155806437e-07,
|
| 1297 |
+
"loss": 0.0005441484972834587,
|
| 1298 |
+
"step": 85,
|
| 1299 |
+
"step_time": 25.023430220317096
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"clip_ratio/high_max": 0.001742735446896404,
|
| 1303 |
+
"clip_ratio/high_mean": 0.0004585776332532987,
|
| 1304 |
+
"clip_ratio/low_mean": 0.0003279163865954615,
|
| 1305 |
+
"clip_ratio/low_min": 0.0,
|
| 1306 |
+
"clip_ratio/region_mean": 0.0007864940271247178,
|
| 1307 |
+
"epoch": 0.5375,
|
| 1308 |
+
"grad_norm": 1.3459034547155788,
|
| 1309 |
+
"kl": 0.021491907769814134,
|
| 1310 |
+
"learning_rate": 4.645601473845635e-07,
|
| 1311 |
+
"loss": 0.00029047258431091905,
|
| 1312 |
+
"step": 86,
|
| 1313 |
+
"step_time": 25.05843911226839
|
| 1314 |
+
},
|
| 1315 |
+
{
|
| 1316 |
+
"clip_ratio/high_max": 0.003957026056014001,
|
| 1317 |
+
"clip_ratio/high_mean": 0.0009892565140035003,
|
| 1318 |
+
"clip_ratio/low_mean": 0.002246821954031475,
|
| 1319 |
+
"clip_ratio/low_min": 0.0,
|
| 1320 |
+
"clip_ratio/region_mean": 0.00323607845348306,
|
| 1321 |
+
"epoch": 0.54375,
|
| 1322 |
+
"grad_norm": 3.1928224199586235,
|
| 1323 |
+
"kl": 0.04086883505806327,
|
| 1324 |
+
"learning_rate": 4.544594375797968e-07,
|
| 1325 |
+
"loss": 6.373366340994835e-05,
|
| 1326 |
+
"step": 87,
|
| 1327 |
+
"step_time": 25.11815517488867
|
| 1328 |
+
},
|
| 1329 |
+
{
|
| 1330 |
+
"clip_ratio/high_max": 0.0007317073177546263,
|
| 1331 |
+
"clip_ratio/high_mean": 0.00018292682943865657,
|
| 1332 |
+
"clip_ratio/low_mean": 0.0003973472266807221,
|
| 1333 |
+
"clip_ratio/low_min": 0.0,
|
| 1334 |
+
"clip_ratio/region_mean": 0.0005802740561193787,
|
| 1335 |
+
"epoch": 0.55,
|
| 1336 |
+
"grad_norm": 1.0467356293291883,
|
| 1337 |
+
"kl": 0.021052672062069178,
|
| 1338 |
+
"learning_rate": 4.443774354523882e-07,
|
| 1339 |
+
"loss": 0.00045237556332722306,
|
| 1340 |
+
"step": 88,
|
| 1341 |
+
"step_time": 25.01378091145307
|
| 1342 |
+
},
|
| 1343 |
+
{
|
| 1344 |
+
"clip_ratio/high_max": 0.0,
|
| 1345 |
+
"clip_ratio/high_mean": 0.0,
|
| 1346 |
+
"clip_ratio/low_mean": 0.0,
|
| 1347 |
+
"clip_ratio/low_min": 0.0,
|
| 1348 |
+
"clip_ratio/region_mean": 0.0,
|
| 1349 |
+
"completions/clipped_ratio": 0.3203125,
|
| 1350 |
+
"completions/max_length": 2048.0,
|
| 1351 |
+
"completions/mean_length": 1111.6484375,
|
| 1352 |
+
"completions/min_length": 67.0,
|
| 1353 |
+
"epoch": 0.55625,
|
| 1354 |
+
"frac_reward_zero_std": 0.3125,
|
| 1355 |
+
"grad_norm": 2.1027611797554,
|
| 1356 |
+
"kl": 0.016956934006884694,
|
| 1357 |
+
"learning_rate": 4.3431828260342675e-07,
|
| 1358 |
+
"loss": 0.00016954075545072556,
|
| 1359 |
+
"reward": 0.31640625,
|
| 1360 |
+
"reward_std": 0.30946123600006104,
|
| 1361 |
+
"rewards/MajorityAccuracy/mean": 0.31640625,
|
| 1362 |
+
"rewards/MajorityAccuracy/std": 0.4659844934940338,
|
| 1363 |
+
"step": 89,
|
| 1364 |
+
"step_time": 189.546223285608
|
| 1365 |
+
},
|
| 1366 |
+
{
|
| 1367 |
+
"clip_ratio/high_max": 0.0009415200038347393,
|
| 1368 |
+
"clip_ratio/high_mean": 0.0002600640436867252,
|
| 1369 |
+
"clip_ratio/low_mean": 7.831944822100922e-05,
|
| 1370 |
+
"clip_ratio/low_min": 0.0,
|
| 1371 |
+
"clip_ratio/region_mean": 0.0003383834882697556,
|
| 1372 |
+
"epoch": 0.5625,
|
| 1373 |
+
"grad_norm": 2.0243598583907025,
|
| 1374 |
+
"kl": 0.012310028774663806,
|
| 1375 |
+
"learning_rate": 4.242861112477118e-07,
|
| 1376 |
+
"loss": 0.0001713670790195465,
|
| 1377 |
+
"step": 90,
|
| 1378 |
+
"step_time": 24.38176016509533
|
| 1379 |
+
},
|
| 1380 |
+
{
|
| 1381 |
+
"clip_ratio/high_max": 0.004336754092946649,
|
| 1382 |
+
"clip_ratio/high_mean": 0.0013549783689086325,
|
| 1383 |
+
"clip_ratio/low_mean": 0.00024161423789337277,
|
| 1384 |
+
"clip_ratio/low_min": 0.0,
|
| 1385 |
+
"clip_ratio/region_mean": 0.0015965925995260477,
|
| 1386 |
+
"epoch": 0.56875,
|
| 1387 |
+
"grad_norm": 3.811972694550947,
|
| 1388 |
+
"kl": 0.022398795350454748,
|
| 1389 |
+
"learning_rate": 4.1428504251627325e-07,
|
| 1390 |
+
"loss": 0.00045355409383773804,
|
| 1391 |
+
"step": 91,
|
| 1392 |
+
"step_time": 24.92947938479483
|
| 1393 |
+
},
|
| 1394 |
+
{
|
| 1395 |
+
"clip_ratio/high_max": 0.002847232979547698,
|
| 1396 |
+
"clip_ratio/high_mean": 0.0007913247991382377,
|
| 1397 |
+
"clip_ratio/low_mean": 0.0006336197020573309,
|
| 1398 |
+
"clip_ratio/low_min": 0.0,
|
| 1399 |
+
"clip_ratio/region_mean": 0.0014249444830056746,
|
| 1400 |
+
"epoch": 0.575,
|
| 1401 |
+
"grad_norm": 3.1538847356735937,
|
| 1402 |
+
"kl": 0.013934758258983493,
|
| 1403 |
+
"learning_rate": 4.0431918476344685e-07,
|
| 1404 |
+
"loss": -0.0001462756044929847,
|
| 1405 |
+
"step": 92,
|
| 1406 |
+
"step_time": 25.495253500062972
|
| 1407 |
+
},
|
| 1408 |
+
{
|
| 1409 |
+
"clip_ratio/high_max": 0.003849103974062018,
|
| 1410 |
+
"clip_ratio/high_mean": 0.0009828622969507705,
|
| 1411 |
+
"clip_ratio/low_mean": 0.0010413822892587632,
|
| 1412 |
+
"clip_ratio/low_min": 0.0,
|
| 1413 |
+
"clip_ratio/region_mean": 0.0020242445898475125,
|
| 1414 |
+
"epoch": 0.58125,
|
| 1415 |
+
"grad_norm": 2.7587271329499923,
|
| 1416 |
+
"kl": 0.017253325204364955,
|
| 1417 |
+
"learning_rate": 3.943926318791963e-07,
|
| 1418 |
+
"loss": -0.0001343645271845162,
|
| 1419 |
+
"step": 93,
|
| 1420 |
+
"step_time": 24.8436949881725
|
| 1421 |
+
},
|
| 1422 |
+
{
|
| 1423 |
+
"clip_ratio/high_max": 0.0056405535724479705,
|
| 1424 |
+
"clip_ratio/high_mean": 0.001793539966456592,
|
| 1425 |
+
"clip_ratio/low_mean": 0.0005771984760940541,
|
| 1426 |
+
"clip_ratio/low_min": 0.0,
|
| 1427 |
+
"clip_ratio/region_mean": 0.0023707384134468157,
|
| 1428 |
+
"epoch": 0.5875,
|
| 1429 |
+
"grad_norm": 2.083172522930478,
|
| 1430 |
+
"kl": 0.011975540430285037,
|
| 1431 |
+
"learning_rate": 3.845094616073783e-07,
|
| 1432 |
+
"loss": 0.00028084847144782543,
|
| 1433 |
+
"step": 94,
|
| 1434 |
+
"step_time": 25.06255929172039
|
| 1435 |
+
},
|
| 1436 |
+
{
|
| 1437 |
+
"clip_ratio/high_max": 0.002470113628078252,
|
| 1438 |
+
"clip_ratio/high_mean": 0.000617528407019563,
|
| 1439 |
+
"clip_ratio/low_mean": 0.00036924509731761646,
|
| 1440 |
+
"clip_ratio/low_min": 0.0,
|
| 1441 |
+
"clip_ratio/region_mean": 0.0009867735043371795,
|
| 1442 |
+
"epoch": 0.59375,
|
| 1443 |
+
"grad_norm": 1.7426336400174607,
|
| 1444 |
+
"kl": 0.02157336857635528,
|
| 1445 |
+
"learning_rate": 3.7467373387063964e-07,
|
| 1446 |
+
"loss": 6.145448423922062e-06,
|
| 1447 |
+
"step": 95,
|
| 1448 |
+
"step_time": 25.26688278792426
|
| 1449 |
+
},
|
| 1450 |
+
{
|
| 1451 |
+
"clip_ratio/high_max": 0.0008942883214331232,
|
| 1452 |
+
"clip_ratio/high_mean": 0.0002235720803582808,
|
| 1453 |
+
"clip_ratio/low_mean": 0.00017997144459513947,
|
| 1454 |
+
"clip_ratio/low_min": 0.0,
|
| 1455 |
+
"clip_ratio/region_mean": 0.0004035435249534203,
|
| 1456 |
+
"epoch": 0.6,
|
| 1457 |
+
"grad_norm": 0.8577462305943301,
|
| 1458 |
+
"kl": 0.01962153031490743,
|
| 1459 |
+
"learning_rate": 3.648894891026358e-07,
|
| 1460 |
+
"loss": 0.0003303435514681041,
|
| 1461 |
+
"step": 96,
|
| 1462 |
+
"step_time": 24.010909103322774
|
| 1463 |
+
},
|
| 1464 |
+
{
|
| 1465 |
+
"clip_ratio/high_max": 0.0,
|
| 1466 |
+
"clip_ratio/high_mean": 0.0,
|
| 1467 |
+
"clip_ratio/low_mean": 0.0,
|
| 1468 |
+
"clip_ratio/low_min": 0.0,
|
| 1469 |
+
"clip_ratio/region_mean": 0.0,
|
| 1470 |
+
"completions/clipped_ratio": 0.28125,
|
| 1471 |
+
"completions/max_length": 2048.0,
|
| 1472 |
+
"completions/mean_length": 971.265625,
|
| 1473 |
+
"completions/min_length": 34.0,
|
| 1474 |
+
"epoch": 0.60625,
|
| 1475 |
+
"frac_reward_zero_std": 0.40625,
|
| 1476 |
+
"grad_norm": 4.033850316769576,
|
| 1477 |
+
"kl": 0.019453940098173916,
|
| 1478 |
+
"learning_rate": 3.5516074658825397e-07,
|
| 1479 |
+
"loss": 0.00019451367552392185,
|
| 1480 |
+
"reward": 0.36328125,
|
| 1481 |
+
"reward_std": 0.2752572298049927,
|
| 1482 |
+
"rewards/MajorityAccuracy/mean": 0.36328125,
|
| 1483 |
+
"rewards/MajorityAccuracy/std": 0.48188698291778564,
|
| 1484 |
+
"step": 97,
|
| 1485 |
+
"step_time": 186.63202212378383
|
| 1486 |
+
},
|
| 1487 |
+
{
|
| 1488 |
+
"clip_ratio/high_max": 0.0001915441098390147,
|
| 1489 |
+
"clip_ratio/high_mean": 4.788602745975368e-05,
|
| 1490 |
+
"clip_ratio/low_mean": 0.00014835370166110806,
|
| 1491 |
+
"clip_ratio/low_min": 0.0,
|
| 1492 |
+
"clip_ratio/region_mean": 0.00019623972912086174,
|
| 1493 |
+
"epoch": 0.6125,
|
| 1494 |
+
"grad_norm": 2.4460375390871016,
|
| 1495 |
+
"kl": 0.017091197543777525,
|
| 1496 |
+
"learning_rate": 3.454915028125263e-07,
|
| 1497 |
+
"loss": 0.000285599468043074,
|
| 1498 |
+
"step": 98,
|
| 1499 |
+
"step_time": 24.969417478423566
|
| 1500 |
+
},
|
| 1501 |
+
{
|
| 1502 |
+
"clip_ratio/high_max": 0.0016793739196145907,
|
| 1503 |
+
"clip_ratio/high_mean": 0.0004489132516027894,
|
| 1504 |
+
"clip_ratio/low_mean": 0.0002453713314025663,
|
| 1505 |
+
"clip_ratio/low_min": 0.0,
|
| 1506 |
+
"clip_ratio/region_mean": 0.0006942845720914192,
|
| 1507 |
+
"epoch": 0.61875,
|
| 1508 |
+
"grad_norm": 2.4262962347716126,
|
| 1509 |
+
"kl": 0.027859210735186934,
|
| 1510 |
+
"learning_rate": 3.3588572981890684e-07,
|
| 1511 |
+
"loss": 0.00011016274220310152,
|
| 1512 |
+
"step": 99,
|
| 1513 |
+
"step_time": 25.10024201963097
|
| 1514 |
+
},
|
| 1515 |
+
{
|
| 1516 |
+
"clip_ratio/high_max": 0.0032153439533431083,
|
| 1517 |
+
"clip_ratio/high_mean": 0.0008038359883357771,
|
| 1518 |
+
"clip_ratio/low_mean": 0.0005821525555802509,
|
| 1519 |
+
"clip_ratio/low_min": 0.0,
|
| 1520 |
+
"clip_ratio/region_mean": 0.0013859885220881552,
|
| 1521 |
+
"epoch": 0.625,
|
| 1522 |
+
"grad_norm": 2.871092365741541,
|
| 1523 |
+
"kl": 0.01657767640426755,
|
| 1524 |
+
"learning_rate": 3.263473735775899e-07,
|
| 1525 |
+
"loss": -9.723560651764274e-05,
|
| 1526 |
+
"step": 100,
|
| 1527 |
+
"step_time": 24.98642905522138
|
| 1528 |
+
},
|
| 1529 |
+
{
|
| 1530 |
+
"clip_ratio/high_max": 0.0016026148150558583,
|
| 1531 |
+
"clip_ratio/high_mean": 0.0004330036008468596,
|
| 1532 |
+
"clip_ratio/low_mean": 0.00020753904573211912,
|
| 1533 |
+
"clip_ratio/low_min": 0.0,
|
| 1534 |
+
"clip_ratio/region_mean": 0.0006405426465789787,
|
| 1535 |
+
"epoch": 0.63125,
|
| 1536 |
+
"grad_norm": 1.4251135135242259,
|
| 1537 |
+
"kl": 0.02586570894345641,
|
| 1538 |
+
"learning_rate": 3.1688035236453865e-07,
|
| 1539 |
+
"loss": 0.000234480612562038,
|
| 1540 |
+
"step": 101,
|
| 1541 |
+
"step_time": 25.3615639237687
|
| 1542 |
+
},
|
| 1543 |
+
{
|
| 1544 |
+
"clip_ratio/high_max": 0.0017492205079179257,
|
| 1545 |
+
"clip_ratio/high_mean": 0.00043730512697948143,
|
| 1546 |
+
"clip_ratio/low_mean": 0.00014153804113448132,
|
| 1547 |
+
"clip_ratio/low_min": 0.0,
|
| 1548 |
+
"clip_ratio/region_mean": 0.0005788431681139627,
|
| 1549 |
+
"epoch": 0.6375,
|
| 1550 |
+
"grad_norm": 1.4482799165238307,
|
| 1551 |
+
"kl": 0.013480304391123354,
|
| 1552 |
+
"learning_rate": 3.0748855515189096e-07,
|
| 1553 |
+
"loss": 0.00023842553491704166,
|
| 1554 |
+
"step": 102,
|
| 1555 |
+
"step_time": 24.615313773043454
|
| 1556 |
+
},
|
| 1557 |
+
{
|
| 1558 |
+
"clip_ratio/high_max": 0.002139671050827019,
|
| 1559 |
+
"clip_ratio/high_mean": 0.0005349177627067547,
|
| 1560 |
+
"clip_ratio/low_mean": 0.0007519550272263587,
|
| 1561 |
+
"clip_ratio/low_min": 0.0,
|
| 1562 |
+
"clip_ratio/region_mean": 0.0012868728081230074,
|
| 1563 |
+
"epoch": 0.64375,
|
| 1564 |
+
"grad_norm": 1.9377679145055535,
|
| 1565 |
+
"kl": 0.016755527933128178,
|
| 1566 |
+
"learning_rate": 2.981758400104028e-07,
|
| 1567 |
+
"loss": 0.0003511752001941204,
|
| 1568 |
+
"step": 103,
|
| 1569 |
+
"step_time": 24.814407263416797
|
| 1570 |
+
},
|
| 1571 |
+
{
|
| 1572 |
+
"clip_ratio/high_max": 0.009579537610989064,
|
| 1573 |
+
"clip_ratio/high_mean": 0.0025748268089955673,
|
| 1574 |
+
"clip_ratio/low_mean": 0.0008987587207229808,
|
| 1575 |
+
"clip_ratio/low_min": 0.0,
|
| 1576 |
+
"clip_ratio/region_mean": 0.0034735855006147176,
|
| 1577 |
+
"epoch": 0.65,
|
| 1578 |
+
"grad_norm": 2.7705509955512326,
|
| 1579 |
+
"kl": 0.033528828993439674,
|
| 1580 |
+
"learning_rate": 2.8894603252458403e-07,
|
| 1581 |
+
"loss": 0.0006886814953759313,
|
| 1582 |
+
"step": 104,
|
| 1583 |
+
"step_time": 24.53277084371075
|
| 1584 |
+
},
|
| 1585 |
+
{
|
| 1586 |
+
"clip_ratio/high_max": 0.0,
|
| 1587 |
+
"clip_ratio/high_mean": 0.0,
|
| 1588 |
+
"clip_ratio/low_mean": 0.0,
|
| 1589 |
+
"clip_ratio/low_min": 0.0,
|
| 1590 |
+
"clip_ratio/region_mean": 0.0,
|
| 1591 |
+
"completions/clipped_ratio": 0.27734375,
|
| 1592 |
+
"completions/max_length": 2048.0,
|
| 1593 |
+
"completions/mean_length": 957.72265625,
|
| 1594 |
+
"completions/min_length": 68.0,
|
| 1595 |
+
"epoch": 0.65625,
|
| 1596 |
+
"frac_reward_zero_std": 0.4375,
|
| 1597 |
+
"grad_norm": 2.28875755490129,
|
| 1598 |
+
"kl": 0.019718763884156942,
|
| 1599 |
+
"learning_rate": 2.7980292422118277e-07,
|
| 1600 |
+
"loss": 0.00019719981355592608,
|
| 1601 |
+
"reward": 0.3984375,
|
| 1602 |
+
"reward_std": 0.24830123782157898,
|
| 1603 |
+
"rewards/MajorityAccuracy/mean": 0.3984375,
|
| 1604 |
+
"rewards/MajorityAccuracy/std": 0.4905354082584381,
|
| 1605 |
+
"step": 105,
|
| 1606 |
+
"step_time": 192.05276277009398
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"clip_ratio/high_max": 0.0013536567421397194,
|
| 1610 |
+
"clip_ratio/high_mean": 0.0004177289047220256,
|
| 1611 |
+
"clip_ratio/low_mean": 8.144665116560645e-05,
|
| 1612 |
+
"clip_ratio/low_min": 0.0,
|
| 1613 |
+
"clip_ratio/region_mean": 0.0004991755558876321,
|
| 1614 |
+
"epoch": 0.6625,
|
| 1615 |
+
"grad_norm": 1.5128694147397466,
|
| 1616 |
+
"kl": 0.009830390801653266,
|
| 1617 |
+
"learning_rate": 2.70750271011657e-07,
|
| 1618 |
+
"loss": 0.00011343969526933506,
|
| 1619 |
+
"step": 106,
|
| 1620 |
+
"step_time": 25.149025632068515
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"clip_ratio/high_max": 0.00331838964484632,
|
| 1624 |
+
"clip_ratio/high_mean": 0.00082959741121158,
|
| 1625 |
+
"clip_ratio/low_mean": 0.0,
|
| 1626 |
+
"clip_ratio/low_min": 0.0,
|
| 1627 |
+
"clip_ratio/region_mean": 0.00082959741121158,
|
| 1628 |
+
"epoch": 0.66875,
|
| 1629 |
+
"grad_norm": 3.099293767401613,
|
| 1630 |
+
"kl": 0.01998239499516785,
|
| 1631 |
+
"learning_rate": 2.6179179164927754e-07,
|
| 1632 |
+
"loss": 0.0008740659686736763,
|
| 1633 |
+
"step": 107,
|
| 1634 |
+
"step_time": 24.994523694738746
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"clip_ratio/high_max": 0.0006780688054277562,
|
| 1638 |
+
"clip_ratio/high_mean": 0.00019130939290334936,
|
| 1639 |
+
"clip_ratio/low_mean": 0.00028616029885597527,
|
| 1640 |
+
"clip_ratio/low_min": 0.0,
|
| 1641 |
+
"clip_ratio/region_mean": 0.00047746969175932463,
|
| 1642 |
+
"epoch": 0.675,
|
| 1643 |
+
"grad_norm": 1.3628905983566921,
|
| 1644 |
+
"kl": 0.011274957447312772,
|
| 1645 |
+
"learning_rate": 2.529311662014972e-07,
|
| 1646 |
+
"loss": 6.51412847219035e-05,
|
| 1647 |
+
"step": 108,
|
| 1648 |
+
"step_time": 25.129499250091612
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"clip_ratio/high_max": 0.0017718233284540474,
|
| 1652 |
+
"clip_ratio/high_mean": 0.0005352473235689104,
|
| 1653 |
+
"clip_ratio/low_mean": 0.00017861263950180728,
|
| 1654 |
+
"clip_ratio/low_min": 0.0,
|
| 1655 |
+
"clip_ratio/region_mean": 0.0007138599612517282,
|
| 1656 |
+
"epoch": 0.68125,
|
| 1657 |
+
"grad_norm": 1.3052075076564047,
|
| 1658 |
+
"kl": 0.016214093775488436,
|
| 1659 |
+
"learning_rate": 2.441720345382089e-07,
|
| 1660 |
+
"loss": 0.00022750305652152747,
|
| 1661 |
+
"step": 109,
|
| 1662 |
+
"step_time": 25.220700857695192
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"clip_ratio/high_max": 0.0034421810414642096,
|
| 1666 |
+
"clip_ratio/high_mean": 0.0008605452603660524,
|
| 1667 |
+
"clip_ratio/low_mean": 0.00041510329174343497,
|
| 1668 |
+
"clip_ratio/low_min": 0.0,
|
| 1669 |
+
"clip_ratio/region_mean": 0.0012756485521094874,
|
| 1670 |
+
"epoch": 0.6875,
|
| 1671 |
+
"grad_norm": 3.244673504087267,
|
| 1672 |
+
"kl": 0.02611761330626905,
|
| 1673 |
+
"learning_rate": 2.355179948365189e-07,
|
| 1674 |
+
"loss": 0.0005276810843497515,
|
| 1675 |
+
"step": 110,
|
| 1676 |
+
"step_time": 24.69555734610185
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"clip_ratio/high_max": 0.0015879357160883956,
|
| 1680 |
+
"clip_ratio/high_mean": 0.00046121826562739443,
|
| 1681 |
+
"clip_ratio/low_mean": 0.0006464043981395662,
|
| 1682 |
+
"clip_ratio/low_min": 0.0,
|
| 1683 |
+
"clip_ratio/region_mean": 0.0011076226637669606,
|
| 1684 |
+
"epoch": 0.69375,
|
| 1685 |
+
"grad_norm": 1.916367298862318,
|
| 1686 |
+
"kl": 0.025136363692581654,
|
| 1687 |
+
"learning_rate": 2.26972602102645e-07,
|
| 1688 |
+
"loss": 0.00031128450063988566,
|
| 1689 |
+
"step": 111,
|
| 1690 |
+
"step_time": 24.158066716976464
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"clip_ratio/high_max": 0.0003559427859727293,
|
| 1694 |
+
"clip_ratio/high_mean": 0.00015505334158660844,
|
| 1695 |
+
"clip_ratio/low_mean": 0.000769265134294983,
|
| 1696 |
+
"clip_ratio/low_min": 0.0,
|
| 1697 |
+
"clip_ratio/region_mean": 0.0009243184758815914,
|
| 1698 |
+
"epoch": 0.7,
|
| 1699 |
+
"grad_norm": 2.4164971183922743,
|
| 1700 |
+
"kl": 0.03256253874860704,
|
| 1701 |
+
"learning_rate": 2.1853936671155127e-07,
|
| 1702 |
+
"loss": 0.0008035096107050776,
|
| 1703 |
+
"step": 112,
|
| 1704 |
+
"step_time": 23.435303118545562
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"clip_ratio/high_max": 0.0,
|
| 1708 |
+
"clip_ratio/high_mean": 0.0,
|
| 1709 |
+
"clip_ratio/low_mean": 0.0,
|
| 1710 |
+
"clip_ratio/low_min": 0.0,
|
| 1711 |
+
"clip_ratio/region_mean": 0.0,
|
| 1712 |
+
"completions/clipped_ratio": 0.2734375,
|
| 1713 |
+
"completions/max_length": 2048.0,
|
| 1714 |
+
"completions/mean_length": 975.16796875,
|
| 1715 |
+
"completions/min_length": 66.0,
|
| 1716 |
+
"epoch": 0.70625,
|
| 1717 |
+
"frac_reward_zero_std": 0.3125,
|
| 1718 |
+
"grad_norm": 1.9859750666638591,
|
| 1719 |
+
"kl": 0.02636914746835828,
|
| 1720 |
+
"learning_rate": 2.1022175296491512e-07,
|
| 1721 |
+
"loss": 0.0002639134181663394,
|
| 1722 |
+
"reward": 0.30859375,
|
| 1723 |
+
"reward_std": 0.3141743838787079,
|
| 1724 |
+
"rewards/MajorityAccuracy/mean": 0.30859375,
|
| 1725 |
+
"rewards/MajorityAccuracy/std": 0.46281784772872925,
|
| 1726 |
+
"step": 113,
|
| 1727 |
+
"step_time": 190.86060116020963
|
| 1728 |
+
},
|
| 1729 |
+
{
|
| 1730 |
+
"clip_ratio/high_max": 0.0018672945152502507,
|
| 1731 |
+
"clip_ratio/high_mean": 0.0005297008901834488,
|
| 1732 |
+
"clip_ratio/low_mean": 0.00029539769457187504,
|
| 1733 |
+
"clip_ratio/low_min": 0.0,
|
| 1734 |
+
"clip_ratio/region_mean": 0.0008250985847553238,
|
| 1735 |
+
"epoch": 0.7125,
|
| 1736 |
+
"grad_norm": 3.63644805248669,
|
| 1737 |
+
"kl": 0.01319556636735797,
|
| 1738 |
+
"learning_rate": 2.0202317766802152e-07,
|
| 1739 |
+
"loss": 0.00039296108298003674,
|
| 1740 |
+
"step": 114,
|
| 1741 |
+
"step_time": 25.281277362722903
|
| 1742 |
+
},
|
| 1743 |
+
{
|
| 1744 |
+
"clip_ratio/high_max": 0.0009505764755886048,
|
| 1745 |
+
"clip_ratio/high_mean": 0.0002376441188971512,
|
| 1746 |
+
"clip_ratio/low_mean": 0.00015623746912751812,
|
| 1747 |
+
"clip_ratio/low_min": 0.0,
|
| 1748 |
+
"clip_ratio/region_mean": 0.0003938815789297223,
|
| 1749 |
+
"epoch": 0.71875,
|
| 1750 |
+
"grad_norm": 1.362298040082264,
|
| 1751 |
+
"kl": 0.015483090071938932,
|
| 1752 |
+
"learning_rate": 1.9394700872616853e-07,
|
| 1753 |
+
"loss": 0.00013875345757696778,
|
| 1754 |
+
"step": 115,
|
| 1755 |
+
"step_time": 25.173529450781643
|
| 1756 |
+
},
|
| 1757 |
+
{
|
| 1758 |
+
"clip_ratio/high_max": 0.0020688906661234796,
|
| 1759 |
+
"clip_ratio/high_mean": 0.0005958390102023259,
|
| 1760 |
+
"clip_ratio/low_mean": 0.0007943460586830042,
|
| 1761 |
+
"clip_ratio/low_min": 0.0,
|
| 1762 |
+
"clip_ratio/region_mean": 0.0013901850688853301,
|
| 1763 |
+
"epoch": 0.725,
|
| 1764 |
+
"grad_norm": 2.8909850252389484,
|
| 1765 |
+
"kl": 0.031830734107643366,
|
| 1766 |
+
"learning_rate": 1.8599656376116024e-07,
|
| 1767 |
+
"loss": 7.592457404825836e-05,
|
| 1768 |
+
"step": 116,
|
| 1769 |
+
"step_time": 24.382398823741823
|
| 1770 |
+
},
|
| 1771 |
+
{
|
| 1772 |
+
"clip_ratio/high_max": 0.002094647556077689,
|
| 1773 |
+
"clip_ratio/high_mean": 0.0006006190669722855,
|
| 1774 |
+
"clip_ratio/low_mean": 0.0007526215340476483,
|
| 1775 |
+
"clip_ratio/low_min": 0.0,
|
| 1776 |
+
"clip_ratio/region_mean": 0.0013532406082958914,
|
| 1777 |
+
"epoch": 0.73125,
|
| 1778 |
+
"grad_norm": 3.2674748181751934,
|
| 1779 |
+
"kl": 0.04125896911136806,
|
| 1780 |
+
"learning_rate": 1.7817510874845582e-07,
|
| 1781 |
+
"loss": 0.0008493869681842625,
|
| 1782 |
+
"step": 117,
|
| 1783 |
+
"step_time": 23.59787268610671
|
| 1784 |
+
},
|
| 1785 |
+
{
|
| 1786 |
+
"clip_ratio/high_max": 0.00026301946491003036,
|
| 1787 |
+
"clip_ratio/high_mean": 6.575486622750759e-05,
|
| 1788 |
+
"clip_ratio/low_mean": 6.103515625e-05,
|
| 1789 |
+
"clip_ratio/low_min": 0.0,
|
| 1790 |
+
"clip_ratio/region_mean": 0.0001267900224775076,
|
| 1791 |
+
"epoch": 0.7375,
|
| 1792 |
+
"grad_norm": 0.6903988940946875,
|
| 1793 |
+
"kl": 0.030178327346220613,
|
| 1794 |
+
"learning_rate": 1.7048585667553412e-07,
|
| 1795 |
+
"loss": 0.0003295271017123014,
|
| 1796 |
+
"step": 118,
|
| 1797 |
+
"step_time": 24.630872033536434
|
| 1798 |
+
},
|
| 1799 |
+
{
|
| 1800 |
+
"clip_ratio/high_max": 0.003531721231411211,
|
| 1801 |
+
"clip_ratio/high_mean": 0.0010159049670619424,
|
| 1802 |
+
"clip_ratio/low_mean": 0.0013609976522275247,
|
| 1803 |
+
"clip_ratio/low_min": 0.0,
|
| 1804 |
+
"clip_ratio/region_mean": 0.002376902659307234,
|
| 1805 |
+
"epoch": 0.74375,
|
| 1806 |
+
"grad_norm": 3.717369604902626,
|
| 1807 |
+
"kl": 0.017110652406699955,
|
| 1808 |
+
"learning_rate": 1.6293196622202632e-07,
|
| 1809 |
+
"loss": 0.0001672329381108284,
|
| 1810 |
+
"step": 119,
|
| 1811 |
+
"step_time": 24.86356428777799
|
| 1812 |
+
},
|
| 1813 |
+
{
|
| 1814 |
+
"clip_ratio/high_max": 0.0026056855858769268,
|
| 1815 |
+
"clip_ratio/high_mean": 0.0006958106023375876,
|
| 1816 |
+
"clip_ratio/low_mean": 0.0006228400452528149,
|
| 1817 |
+
"clip_ratio/low_min": 0.0,
|
| 1818 |
+
"clip_ratio/region_mean": 0.0013186506548663601,
|
| 1819 |
+
"epoch": 0.75,
|
| 1820 |
+
"grad_norm": 2.550545344860883,
|
| 1821 |
+
"kl": 0.030713965417817235,
|
| 1822 |
+
"learning_rate": 1.555165404621567e-07,
|
| 1823 |
+
"loss": 0.0006629219278693199,
|
| 1824 |
+
"step": 120,
|
| 1825 |
+
"step_time": 24.70534085528925
|
| 1826 |
+
},
|
| 1827 |
+
{
|
| 1828 |
+
"clip_ratio/high_max": 0.0,
|
| 1829 |
+
"clip_ratio/high_mean": 0.0,
|
| 1830 |
+
"clip_ratio/low_mean": 0.0,
|
| 1831 |
+
"clip_ratio/low_min": 0.0,
|
| 1832 |
+
"clip_ratio/region_mean": 0.0,
|
| 1833 |
+
"completions/clipped_ratio": 0.26953125,
|
| 1834 |
+
"completions/max_length": 2048.0,
|
| 1835 |
+
"completions/mean_length": 954.19921875,
|
| 1836 |
+
"completions/min_length": 74.0,
|
| 1837 |
+
"epoch": 0.75625,
|
| 1838 |
+
"frac_reward_zero_std": 0.34375,
|
| 1839 |
+
"grad_norm": 2.295066473981048,
|
| 1840 |
+
"kl": 0.04109575576148927,
|
| 1841 |
+
"learning_rate": 1.4824262559002592e-07,
|
| 1842 |
+
"loss": 0.0004110461741220206,
|
| 1843 |
+
"reward": 0.29296875,
|
| 1844 |
+
"reward_std": 0.27526116371154785,
|
| 1845 |
+
"rewards/MajorityAccuracy/mean": 0.29296875,
|
| 1846 |
+
"rewards/MajorityAccuracy/std": 0.45601576566696167,
|
| 1847 |
+
"step": 121,
|
| 1848 |
+
"step_time": 189.11332981148735
|
| 1849 |
+
},
|
| 1850 |
+
{
|
| 1851 |
+
"clip_ratio/high_max": 0.002691173256607726,
|
| 1852 |
+
"clip_ratio/high_mean": 0.0006727933141519316,
|
| 1853 |
+
"clip_ratio/low_mean": 0.00010486577230039984,
|
| 1854 |
+
"clip_ratio/low_min": 0.0,
|
| 1855 |
+
"clip_ratio/region_mean": 0.0007776590719004162,
|
| 1856 |
+
"epoch": 0.7625,
|
| 1857 |
+
"grad_norm": 3.3589411886045992,
|
| 1858 |
+
"kl": 0.024765971349552274,
|
| 1859 |
+
"learning_rate": 1.4111320966826057e-07,
|
| 1860 |
+
"loss": 0.000181948475074023,
|
| 1861 |
+
"step": 122,
|
| 1862 |
+
"step_time": 24.114926864393055
|
| 1863 |
+
},
|
| 1864 |
+
{
|
| 1865 |
+
"clip_ratio/high_max": 0.0007919766358099878,
|
| 1866 |
+
"clip_ratio/high_mean": 0.00019799415895249695,
|
| 1867 |
+
"clip_ratio/low_mean": 0.00046339991240529343,
|
| 1868 |
+
"clip_ratio/low_min": 0.0,
|
| 1869 |
+
"clip_ratio/region_mean": 0.0006613940749957692,
|
| 1870 |
+
"epoch": 0.76875,
|
| 1871 |
+
"grad_norm": 2.8148155409034836,
|
| 1872 |
+
"kl": 0.019490653416141868,
|
| 1873 |
+
"learning_rate": 1.3413122140054217e-07,
|
| 1874 |
+
"loss": 0.0004310149815864861,
|
| 1875 |
+
"step": 123,
|
| 1876 |
+
"step_time": 25.226889363955706
|
| 1877 |
+
},
|
| 1878 |
+
{
|
| 1879 |
+
"clip_ratio/high_max": 0.001019900882965885,
|
| 1880 |
+
"clip_ratio/high_mean": 0.00027850684637087397,
|
| 1881 |
+
"clip_ratio/low_mean": 0.0005645751953125,
|
| 1882 |
+
"clip_ratio/low_min": 0.0,
|
| 1883 |
+
"clip_ratio/region_mean": 0.000843082059873268,
|
| 1884 |
+
"epoch": 0.775,
|
| 1885 |
+
"grad_norm": 1.3352343529344102,
|
| 1886 |
+
"kl": 0.01114171277731657,
|
| 1887 |
+
"learning_rate": 1.272995289285202e-07,
|
| 1888 |
+
"loss": 0.00017396478506270796,
|
| 1889 |
+
"step": 124,
|
| 1890 |
+
"step_time": 25.77579073328525
|
| 1891 |
+
},
|
| 1892 |
+
{
|
| 1893 |
+
"clip_ratio/high_max": 0.003423288813792169,
|
| 1894 |
+
"clip_ratio/high_mean": 0.0008558222034480423,
|
| 1895 |
+
"clip_ratio/low_mean": 0.0002621775001898641,
|
| 1896 |
+
"clip_ratio/low_min": 0.0,
|
| 1897 |
+
"clip_ratio/region_mean": 0.0011179996890859911,
|
| 1898 |
+
"epoch": 0.78125,
|
| 1899 |
+
"grad_norm": 2.560915871196629,
|
| 1900 |
+
"kl": 0.03662781580351293,
|
| 1901 |
+
"learning_rate": 1.2062093865360457e-07,
|
| 1902 |
+
"loss": 0.0003918518195860088,
|
| 1903 |
+
"step": 125,
|
| 1904 |
+
"step_time": 24.434862991329283
|
| 1905 |
+
},
|
| 1906 |
+
{
|
| 1907 |
+
"clip_ratio/high_max": 0.0006894102989463136,
|
| 1908 |
+
"clip_ratio/high_mean": 0.0001723525747365784,
|
| 1909 |
+
"clip_ratio/low_mean": 3.672170169011224e-05,
|
| 1910 |
+
"clip_ratio/low_min": 0.0,
|
| 1911 |
+
"clip_ratio/region_mean": 0.00020907427824568003,
|
| 1912 |
+
"epoch": 0.7875,
|
| 1913 |
+
"grad_norm": 1.2872196914841973,
|
| 1914 |
+
"kl": 0.014173673465847969,
|
| 1915 |
+
"learning_rate": 1.1409819408411897e-07,
|
| 1916 |
+
"loss": 0.0001859133190009743,
|
| 1917 |
+
"step": 126,
|
| 1918 |
+
"step_time": 25.5073729949072
|
| 1919 |
+
},
|
| 1920 |
+
{
|
| 1921 |
+
"clip_ratio/high_max": 0.010582421557046473,
|
| 1922 |
+
"clip_ratio/high_mean": 0.003131385281449184,
|
| 1923 |
+
"clip_ratio/low_mean": 0.0001568958250572905,
|
| 1924 |
+
"clip_ratio/low_min": 0.0,
|
| 1925 |
+
"clip_ratio/region_mean": 0.0032882810846786015,
|
| 1926 |
+
"epoch": 0.79375,
|
| 1927 |
+
"grad_norm": 6.078395553934866,
|
| 1928 |
+
"kl": 0.034252620767802,
|
| 1929 |
+
"learning_rate": 1.0773397470829143e-07,
|
| 1930 |
+
"loss": 0.0006722395773977041,
|
| 1931 |
+
"step": 127,
|
| 1932 |
+
"step_time": 25.89566485211253
|
| 1933 |
+
},
|
| 1934 |
+
{
|
| 1935 |
+
"clip_ratio/high_max": 0.0003007804189110175,
|
| 1936 |
+
"clip_ratio/high_mean": 7.519510472775437e-05,
|
| 1937 |
+
"clip_ratio/low_mean": 0.0016314827371388674,
|
| 1938 |
+
"clip_ratio/low_min": 0.0,
|
| 1939 |
+
"clip_ratio/region_mean": 0.0017066778455046006,
|
| 1940 |
+
"epoch": 0.8,
|
| 1941 |
+
"grad_norm": 4.146853791672376,
|
| 1942 |
+
"kl": 0.022625454468652606,
|
| 1943 |
+
"learning_rate": 1.0153089489354256e-07,
|
| 1944 |
+
"loss": 0.0004904419183731079,
|
| 1945 |
+
"step": 128,
|
| 1946 |
+
"step_time": 23.750699387397617
|
| 1947 |
+
},
|
| 1948 |
+
{
|
| 1949 |
+
"clip_ratio/high_max": 0.0,
|
| 1950 |
+
"clip_ratio/high_mean": 0.0,
|
| 1951 |
+
"clip_ratio/low_mean": 0.0,
|
| 1952 |
+
"clip_ratio/low_min": 0.0,
|
| 1953 |
+
"clip_ratio/region_mean": 0.0,
|
| 1954 |
+
"completions/clipped_ratio": 0.2109375,
|
| 1955 |
+
"completions/max_length": 2048.0,
|
| 1956 |
+
"completions/mean_length": 863.640625,
|
| 1957 |
+
"completions/min_length": 71.0,
|
| 1958 |
+
"epoch": 0.80625,
|
| 1959 |
+
"frac_reward_zero_std": 0.25,
|
| 1960 |
+
"grad_norm": 1.9383471989800944,
|
| 1961 |
+
"kl": 0.017135215923190117,
|
| 1962 |
+
"learning_rate": 9.549150281252632e-08,
|
| 1963 |
+
"loss": 0.00017131016647908837,
|
| 1964 |
+
"reward": 0.41015625,
|
| 1965 |
+
"reward_std": 0.33850935101509094,
|
| 1966 |
+
"rewards/MajorityAccuracy/mean": 0.41015625,
|
| 1967 |
+
"rewards/MajorityAccuracy/std": 0.49282535910606384,
|
| 1968 |
+
"step": 129,
|
| 1969 |
+
"step_time": 188.2599949403666
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"clip_ratio/high_max": 0.002011611344642006,
|
| 1973 |
+
"clip_ratio/high_mean": 0.0007117059794836678,
|
| 1974 |
+
"clip_ratio/low_mean": 0.00011473522681626491,
|
| 1975 |
+
"clip_ratio/low_min": 0.0,
|
| 1976 |
+
"clip_ratio/region_mean": 0.0008264412026619539,
|
| 1977 |
+
"epoch": 0.8125,
|
| 1978 |
+
"grad_norm": 2.3009548516409577,
|
| 1979 |
+
"kl": 0.014620095142163336,
|
| 1980 |
+
"learning_rate": 8.961827939636196e-08,
|
| 1981 |
+
"loss": 0.00010867882519960403,
|
| 1982 |
+
"step": 130,
|
| 1983 |
+
"step_time": 24.58518546447158
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"clip_ratio/high_max": 0.0012733404873870313,
|
| 1987 |
+
"clip_ratio/high_mean": 0.00047625156003050506,
|
| 1988 |
+
"clip_ratio/low_mean": 0.0010886941672652029,
|
| 1989 |
+
"clip_ratio/low_min": 0.0,
|
| 1990 |
+
"clip_ratio/region_mean": 0.0015649457200197503,
|
| 1991 |
+
"epoch": 0.81875,
|
| 1992 |
+
"grad_norm": 4.3042506500131115,
|
| 1993 |
+
"kl": 0.017423331388272345,
|
| 1994 |
+
"learning_rate": 8.391363731548811e-08,
|
| 1995 |
+
"loss": 0.0006831632927060127,
|
| 1996 |
+
"step": 131,
|
| 1997 |
+
"step_time": 23.56351294508204
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"clip_ratio/high_max": 0.001178193066152744,
|
| 2001 |
+
"clip_ratio/high_mean": 0.000294548266538186,
|
| 2002 |
+
"clip_ratio/low_mean": 0.0,
|
| 2003 |
+
"clip_ratio/low_min": 0.0,
|
| 2004 |
+
"clip_ratio/region_mean": 0.000294548266538186,
|
| 2005 |
+
"epoch": 0.825,
|
| 2006 |
+
"grad_norm": 2.1399281983677323,
|
| 2007 |
+
"kl": 0.015660963719710708,
|
| 2008 |
+
"learning_rate": 7.837991998855897e-08,
|
| 2009 |
+
"loss": 0.0005850719171576202,
|
| 2010 |
+
"step": 132,
|
| 2011 |
+
"step_time": 25.023899918422103
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"clip_ratio/high_max": 0.005895672831684351,
|
| 2015 |
+
"clip_ratio/high_mean": 0.0016358352731913328,
|
| 2016 |
+
"clip_ratio/low_mean": 0.0003604320954764262,
|
| 2017 |
+
"clip_ratio/low_min": 0.0,
|
| 2018 |
+
"clip_ratio/region_mean": 0.001996267368667759,
|
| 2019 |
+
"epoch": 0.83125,
|
| 2020 |
+
"grad_norm": 4.0838796896349026,
|
| 2021 |
+
"kl": 0.024278505239635706,
|
| 2022 |
+
"learning_rate": 7.301940061978722e-08,
|
| 2023 |
+
"loss": -1.4914650819264352e-05,
|
| 2024 |
+
"step": 133,
|
| 2025 |
+
"step_time": 24.47555505251512
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"clip_ratio/high_max": 0.0012338610977167264,
|
| 2029 |
+
"clip_ratio/high_mean": 0.0003635475331975613,
|
| 2030 |
+
"clip_ratio/low_mean": 0.0004865375340159517,
|
| 2031 |
+
"clip_ratio/low_min": 0.0,
|
| 2032 |
+
"clip_ratio/region_mean": 0.0008500850599375553,
|
| 2033 |
+
"epoch": 0.8375,
|
| 2034 |
+
"grad_norm": 2.113449490986628,
|
| 2035 |
+
"kl": 0.029349149204790592,
|
| 2036 |
+
"learning_rate": 6.783428126513125e-08,
|
| 2037 |
+
"loss": 0.0005446458235383034,
|
| 2038 |
+
"step": 134,
|
| 2039 |
+
"step_time": 24.94025549478829
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"clip_ratio/high_max": 0.0002969121269416064,
|
| 2043 |
+
"clip_ratio/high_mean": 0.00012119696475565434,
|
| 2044 |
+
"clip_ratio/low_mean": 7.780917621857952e-05,
|
| 2045 |
+
"clip_ratio/low_min": 0.0,
|
| 2046 |
+
"clip_ratio/region_mean": 0.00019900614097423386,
|
| 2047 |
+
"epoch": 0.84375,
|
| 2048 |
+
"grad_norm": 2.174909072256445,
|
| 2049 |
+
"kl": 0.013756011612713337,
|
| 2050 |
+
"learning_rate": 6.282669192770895e-08,
|
| 2051 |
+
"loss": 0.00029106857255101204,
|
| 2052 |
+
"step": 135,
|
| 2053 |
+
"step_time": 25.155068232677877
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"clip_ratio/high_max": 0.0009322202677140012,
|
| 2057 |
+
"clip_ratio/high_mean": 0.0002564633068686817,
|
| 2058 |
+
"clip_ratio/low_mean": 0.0007215233945316868,
|
| 2059 |
+
"clip_ratio/low_min": 0.0,
|
| 2060 |
+
"clip_ratio/region_mean": 0.0009779866977623897,
|
| 2061 |
+
"epoch": 0.85,
|
| 2062 |
+
"grad_norm": 2.6232752079948165,
|
| 2063 |
+
"kl": 0.018263249890878797,
|
| 2064 |
+
"learning_rate": 5.799868968281074e-08,
|
| 2065 |
+
"loss": 0.0003974251449108124,
|
| 2066 |
+
"step": 136,
|
| 2067 |
+
"step_time": 24.37575152190402
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"clip_ratio/high_max": 0.0,
|
| 2071 |
+
"clip_ratio/high_mean": 0.0,
|
| 2072 |
+
"clip_ratio/low_mean": 0.0,
|
| 2073 |
+
"clip_ratio/low_min": 0.0,
|
| 2074 |
+
"clip_ratio/region_mean": 0.0,
|
| 2075 |
+
"completions/clipped_ratio": 0.36328125,
|
| 2076 |
+
"completions/max_length": 2048.0,
|
| 2077 |
+
"completions/mean_length": 1145.296875,
|
| 2078 |
+
"completions/min_length": 75.0,
|
| 2079 |
+
"epoch": 0.85625,
|
| 2080 |
+
"frac_reward_zero_std": 0.3125,
|
| 2081 |
+
"grad_norm": 1.9561079503209793,
|
| 2082 |
+
"kl": 0.02927554491907358,
|
| 2083 |
+
"learning_rate": 5.335225783287051e-08,
|
| 2084 |
+
"loss": 0.00029286928474903107,
|
| 2085 |
+
"reward": 0.26953125,
|
| 2086 |
+
"reward_std": 0.3051016926765442,
|
| 2087 |
+
"rewards/MajorityAccuracy/mean": 0.26953125,
|
| 2088 |
+
"rewards/MajorityAccuracy/std": 0.4445854723453522,
|
| 2089 |
+
"step": 137,
|
| 2090 |
+
"step_time": 192.8925414616242
|
| 2091 |
+
},
|
| 2092 |
+
{
|
| 2093 |
+
"clip_ratio/high_max": 0.0008992919465526938,
|
| 2094 |
+
"clip_ratio/high_mean": 0.00022482298663817346,
|
| 2095 |
+
"clip_ratio/low_mean": 0.00019592476019170135,
|
| 2096 |
+
"clip_ratio/low_min": 0.0,
|
| 2097 |
+
"clip_ratio/region_mean": 0.0004207477468298748,
|
| 2098 |
+
"epoch": 0.8625,
|
| 2099 |
+
"grad_norm": 2.6141996833797996,
|
| 2100 |
+
"kl": 0.016634974628686905,
|
| 2101 |
+
"learning_rate": 4.888930509274125e-08,
|
| 2102 |
+
"loss": -0.00017158743867184967,
|
| 2103 |
+
"step": 138,
|
| 2104 |
+
"step_time": 25.26775275170803
|
| 2105 |
+
},
|
| 2106 |
+
{
|
| 2107 |
+
"clip_ratio/high_max": 0.0006848368502687663,
|
| 2108 |
+
"clip_ratio/high_mean": 0.00017120921256719157,
|
| 2109 |
+
"clip_ratio/low_mean": 0.00019458217138890177,
|
| 2110 |
+
"clip_ratio/low_min": 0.0,
|
| 2111 |
+
"clip_ratio/region_mean": 0.00036579138395609334,
|
| 2112 |
+
"epoch": 0.86875,
|
| 2113 |
+
"grad_norm": 1.776746501400259,
|
| 2114 |
+
"kl": 0.018070745281875134,
|
| 2115 |
+
"learning_rate": 4.4611664805611794e-08,
|
| 2116 |
+
"loss": 0.00029547285521402955,
|
| 2117 |
+
"step": 139,
|
| 2118 |
+
"step_time": 25.11653576651588
|
| 2119 |
+
},
|
| 2120 |
+
{
|
| 2121 |
+
"clip_ratio/high_max": 0.0055776299923309125,
|
| 2122 |
+
"clip_ratio/high_mean": 0.0013944074980827281,
|
| 2123 |
+
"clip_ratio/low_mean": 0.00017622481573198456,
|
| 2124 |
+
"clip_ratio/low_min": 0.0,
|
| 2125 |
+
"clip_ratio/region_mean": 0.0015706323029007763,
|
| 2126 |
+
"epoch": 0.875,
|
| 2127 |
+
"grad_norm": 2.8978415243855835,
|
| 2128 |
+
"kl": 0.022847897023893893,
|
| 2129 |
+
"learning_rate": 4.0521094189884696e-08,
|
| 2130 |
+
"loss": 0.0005900425603613257,
|
| 2131 |
+
"step": 140,
|
| 2132 |
+
"step_time": 25.054628142621368
|
| 2133 |
+
},
|
| 2134 |
+
{
|
| 2135 |
+
"clip_ratio/high_max": 0.0009842519648373127,
|
| 2136 |
+
"clip_ratio/high_mean": 0.0002460629912093282,
|
| 2137 |
+
"clip_ratio/low_mean": 0.0004762319440487772,
|
| 2138 |
+
"clip_ratio/low_min": 0.0,
|
| 2139 |
+
"clip_ratio/region_mean": 0.0007222949352581054,
|
| 2140 |
+
"epoch": 0.88125,
|
| 2141 |
+
"grad_norm": 3.413591205952619,
|
| 2142 |
+
"kl": 0.022365630487911403,
|
| 2143 |
+
"learning_rate": 3.6619273617325695e-08,
|
| 2144 |
+
"loss": 0.0006395417149178684,
|
| 2145 |
+
"step": 141,
|
| 2146 |
+
"step_time": 24.42295361496508
|
| 2147 |
+
},
|
| 2148 |
+
{
|
| 2149 |
+
"clip_ratio/high_max": 0.0007843145285733044,
|
| 2150 |
+
"clip_ratio/high_mean": 0.0002351411312702112,
|
| 2151 |
+
"clip_ratio/low_mean": 0.0007846314329071902,
|
| 2152 |
+
"clip_ratio/low_min": 0.0,
|
| 2153 |
+
"clip_ratio/region_mean": 0.0010197725678153802,
|
| 2154 |
+
"epoch": 0.8875,
|
| 2155 |
+
"grad_norm": 1.569088547557081,
|
| 2156 |
+
"kl": 0.028394629480317235,
|
| 2157 |
+
"learning_rate": 3.2907805922781476e-08,
|
| 2158 |
+
"loss": 0.00022586526756640524,
|
| 2159 |
+
"step": 142,
|
| 2160 |
+
"step_time": 25.045972133986652
|
| 2161 |
+
},
|
| 2162 |
+
{
|
| 2163 |
+
"clip_ratio/high_max": 0.0,
|
| 2164 |
+
"clip_ratio/high_mean": 0.0,
|
| 2165 |
+
"clip_ratio/low_mean": 7.975521657499485e-05,
|
| 2166 |
+
"clip_ratio/low_min": 0.0,
|
| 2167 |
+
"clip_ratio/region_mean": 7.975521657499485e-05,
|
| 2168 |
+
"epoch": 0.89375,
|
| 2169 |
+
"grad_norm": 1.6509010485567466,
|
| 2170 |
+
"kl": 0.01939802709966898,
|
| 2171 |
+
"learning_rate": 2.9388215745748345e-08,
|
| 2172 |
+
"loss": 0.0003978288732469082,
|
| 2173 |
+
"step": 143,
|
| 2174 |
+
"step_time": 24.988902838900685
|
| 2175 |
+
},
|
| 2176 |
+
{
|
| 2177 |
+
"clip_ratio/high_max": 0.0022813693212810904,
|
| 2178 |
+
"clip_ratio/high_mean": 0.0007333235771511681,
|
| 2179 |
+
"clip_ratio/low_mean": 0.0003949884994653985,
|
| 2180 |
+
"clip_ratio/low_min": 0.0,
|
| 2181 |
+
"clip_ratio/region_mean": 0.0011283120475127362,
|
| 2182 |
+
"epoch": 0.9,
|
| 2183 |
+
"grad_norm": 2.318566558018227,
|
| 2184 |
+
"kl": 0.010722269886173308,
|
| 2185 |
+
"learning_rate": 2.6061948904063658e-08,
|
| 2186 |
+
"loss": -0.0001780148595571518,
|
| 2187 |
+
"step": 144,
|
| 2188 |
+
"step_time": 24.384330032393336
|
| 2189 |
+
},
|
| 2190 |
+
{
|
| 2191 |
+
"clip_ratio/high_max": 0.0,
|
| 2192 |
+
"clip_ratio/high_mean": 0.0,
|
| 2193 |
+
"clip_ratio/low_mean": 0.0,
|
| 2194 |
+
"clip_ratio/low_min": 0.0,
|
| 2195 |
+
"clip_ratio/region_mean": 0.0,
|
| 2196 |
+
"completions/clipped_ratio": 0.18359375,
|
| 2197 |
+
"completions/max_length": 2048.0,
|
| 2198 |
+
"completions/mean_length": 835.53125,
|
| 2199 |
+
"completions/min_length": 68.0,
|
| 2200 |
+
"epoch": 0.90625,
|
| 2201 |
+
"frac_reward_zero_std": 0.34375,
|
| 2202 |
+
"grad_norm": 2.5832926085179686,
|
| 2203 |
+
"kl": 0.02262058388441801,
|
| 2204 |
+
"learning_rate": 2.293037179997559e-08,
|
| 2205 |
+
"loss": 0.00022622865799348801,
|
| 2206 |
+
"reward": 0.40625,
|
| 2207 |
+
"reward_std": 0.2886909246444702,
|
| 2208 |
+
"rewards/MajorityAccuracy/mean": 0.40625,
|
| 2209 |
+
"rewards/MajorityAccuracy/std": 0.49209436774253845,
|
| 2210 |
+
"step": 145,
|
| 2211 |
+
"step_time": 187.30761687643826
|
| 2212 |
+
},
|
| 2213 |
+
{
|
| 2214 |
+
"clip_ratio/high_max": 0.0007988402430783026,
|
| 2215 |
+
"clip_ratio/high_mean": 0.00019971006076957565,
|
| 2216 |
+
"clip_ratio/low_mean": 3.371089405845851e-05,
|
| 2217 |
+
"clip_ratio/low_min": 0.0,
|
| 2218 |
+
"clip_ratio/region_mean": 0.00023342095482803416,
|
| 2219 |
+
"epoch": 0.9125,
|
| 2220 |
+
"grad_norm": 2.452989163155721,
|
| 2221 |
+
"kl": 0.022664929856546223,
|
| 2222 |
+
"learning_rate": 1.9994770858837107e-08,
|
| 2223 |
+
"loss": 0.00020722296903841197,
|
| 2224 |
+
"step": 146,
|
| 2225 |
+
"step_time": 25.073296445421875
|
| 2226 |
+
},
|
| 2227 |
+
{
|
| 2228 |
+
"clip_ratio/high_max": 0.0015718622016720474,
|
| 2229 |
+
"clip_ratio/high_mean": 0.0005881927208974957,
|
| 2230 |
+
"clip_ratio/low_mean": 8.017171421670355e-05,
|
| 2231 |
+
"clip_ratio/low_min": 0.0,
|
| 2232 |
+
"clip_ratio/region_mean": 0.0006683644351141993,
|
| 2233 |
+
"epoch": 0.91875,
|
| 2234 |
+
"grad_norm": 3.4914952193381774,
|
| 2235 |
+
"kl": 0.043645494151860476,
|
| 2236 |
+
"learning_rate": 1.725635200065323e-08,
|
| 2237 |
+
"loss": 0.0003801044076681137,
|
| 2238 |
+
"step": 147,
|
| 2239 |
+
"step_time": 25.019193139858544
|
| 2240 |
+
},
|
| 2241 |
+
{
|
| 2242 |
+
"clip_ratio/high_max": 0.0,
|
| 2243 |
+
"clip_ratio/high_mean": 0.0,
|
| 2244 |
+
"clip_ratio/low_mean": 0.0001221914808411384,
|
| 2245 |
+
"clip_ratio/low_min": 0.0,
|
| 2246 |
+
"clip_ratio/region_mean": 0.0001221914808411384,
|
| 2247 |
+
"epoch": 0.925,
|
| 2248 |
+
"grad_norm": 1.2652140491422241,
|
| 2249 |
+
"kl": 0.01965733477845788,
|
| 2250 |
+
"learning_rate": 1.4716240144699187e-08,
|
| 2251 |
+
"loss": 0.0002354808384552598,
|
| 2252 |
+
"step": 148,
|
| 2253 |
+
"step_time": 25.726078127045184
|
| 2254 |
+
},
|
| 2255 |
+
{
|
| 2256 |
+
"clip_ratio/high_max": 0.0019392838585190475,
|
| 2257 |
+
"clip_ratio/high_mean": 0.0005334649904398248,
|
| 2258 |
+
"clip_ratio/low_mean": 9.1552734375e-05,
|
| 2259 |
+
"clip_ratio/low_min": 0.0,
|
| 2260 |
+
"clip_ratio/region_mean": 0.0006250177248148248,
|
| 2261 |
+
"epoch": 0.93125,
|
| 2262 |
+
"grad_norm": 2.568356597678313,
|
| 2263 |
+
"kl": 0.016452013631351292,
|
| 2264 |
+
"learning_rate": 1.2375478747413015e-08,
|
| 2265 |
+
"loss": -1.7919963283929974e-05,
|
| 2266 |
+
"step": 149,
|
| 2267 |
+
"step_time": 25.044571021571755
|
| 2268 |
+
},
|
| 2269 |
+
{
|
| 2270 |
+
"clip_ratio/high_max": 0.0010685963425203227,
|
| 2271 |
+
"clip_ratio/high_mean": 0.00026714908563008066,
|
| 2272 |
+
"clip_ratio/low_mean": 0.0002847484502126463,
|
| 2273 |
+
"clip_ratio/low_min": 0.0,
|
| 2274 |
+
"clip_ratio/region_mean": 0.0005518975412996951,
|
| 2275 |
+
"epoch": 0.9375,
|
| 2276 |
+
"grad_norm": 2.2352143312818034,
|
| 2277 |
+
"kl": 0.0124446633271873,
|
| 2278 |
+
"learning_rate": 1.0235029373752757e-08,
|
| 2279 |
+
"loss": 0.00018786174769047648,
|
| 2280 |
+
"step": 150,
|
| 2281 |
+
"step_time": 25.779688292648643
|
| 2282 |
+
},
|
| 2283 |
+
{
|
| 2284 |
+
"clip_ratio/high_max": 0.006216227542608976,
|
| 2285 |
+
"clip_ratio/high_mean": 0.0018017938709817827,
|
| 2286 |
+
"clip_ratio/low_mean": 0.0004595588252414018,
|
| 2287 |
+
"clip_ratio/low_min": 0.0,
|
| 2288 |
+
"clip_ratio/region_mean": 0.0022613526962231845,
|
| 2289 |
+
"epoch": 0.94375,
|
| 2290 |
+
"grad_norm": 4.7098163841077785,
|
| 2291 |
+
"kl": 0.03684628428891301,
|
| 2292 |
+
"learning_rate": 8.295771302193721e-09,
|
| 2293 |
+
"loss": 0.00014680475578643382,
|
| 2294 |
+
"step": 151,
|
| 2295 |
+
"step_time": 23.19183904491365
|
| 2296 |
+
},
|
| 2297 |
+
{
|
| 2298 |
+
"clip_ratio/high_max": 0.0005615729314740747,
|
| 2299 |
+
"clip_ratio/high_mean": 0.00014039323286851868,
|
| 2300 |
+
"clip_ratio/low_mean": 6.103515625e-05,
|
| 2301 |
+
"clip_ratio/low_min": 0.0,
|
| 2302 |
+
"clip_ratio/region_mean": 0.00020142838911851868,
|
| 2303 |
+
"epoch": 0.95,
|
| 2304 |
+
"grad_norm": 1.3259749274459292,
|
| 2305 |
+
"kl": 0.03128535416908562,
|
| 2306 |
+
"learning_rate": 6.558501163527963e-09,
|
| 2307 |
+
"loss": 0.00012140729813836515,
|
| 2308 |
+
"step": 152,
|
| 2309 |
+
"step_time": 24.266863436903805
|
| 2310 |
+
},
|
| 2311 |
+
{
|
| 2312 |
+
"clip_ratio/high_max": 0.0,
|
| 2313 |
+
"clip_ratio/high_mean": 0.0,
|
| 2314 |
+
"clip_ratio/low_mean": 0.0,
|
| 2315 |
+
"clip_ratio/low_min": 0.0,
|
| 2316 |
+
"clip_ratio/region_mean": 0.0,
|
| 2317 |
+
"completions/clipped_ratio": 0.1953125,
|
| 2318 |
+
"completions/max_length": 2048.0,
|
| 2319 |
+
"completions/mean_length": 792.77734375,
|
| 2320 |
+
"completions/min_length": 77.0,
|
| 2321 |
+
"epoch": 0.95625,
|
| 2322 |
+
"frac_reward_zero_std": 0.15625,
|
| 2323 |
+
"grad_norm": 3.0298100000957255,
|
| 2324 |
+
"kl": 0.0191346884239465,
|
| 2325 |
+
"learning_rate": 5.023932613615445e-09,
|
| 2326 |
+
"loss": 0.00019115488976240158,
|
| 2327 |
+
"reward": 0.515625,
|
| 2328 |
+
"reward_std": 0.379666805267334,
|
| 2329 |
+
"rewards/MajorityAccuracy/mean": 0.515625,
|
| 2330 |
+
"rewards/MajorityAccuracy/std": 0.5007347464561462,
|
| 2331 |
+
"step": 153,
|
| 2332 |
+
"step_time": 183.35863576876
|
| 2333 |
+
},
|
| 2334 |
+
{
|
| 2335 |
+
"clip_ratio/high_max": 0.0024290561996167526,
|
| 2336 |
+
"clip_ratio/high_mean": 0.0007921676224214025,
|
| 2337 |
+
"clip_ratio/low_mean": 0.0001479005441069603,
|
| 2338 |
+
"clip_ratio/low_min": 0.0,
|
| 2339 |
+
"clip_ratio/region_mean": 0.0009400681738043204,
|
| 2340 |
+
"epoch": 0.9625,
|
| 2341 |
+
"grad_norm": 2.8726326972136333,
|
| 2342 |
+
"kl": 0.02365651843138039,
|
| 2343 |
+
"learning_rate": 3.6926960402202666e-09,
|
| 2344 |
+
"loss": 0.00026881787925958633,
|
| 2345 |
+
"step": 154,
|
| 2346 |
+
"step_time": 24.344831645023078
|
| 2347 |
+
},
|
| 2348 |
+
{
|
| 2349 |
+
"clip_ratio/high_max": 0.0,
|
| 2350 |
+
"clip_ratio/high_mean": 0.0,
|
| 2351 |
+
"clip_ratio/low_mean": 0.00014156303950585425,
|
| 2352 |
+
"clip_ratio/low_min": 0.0,
|
| 2353 |
+
"clip_ratio/region_mean": 0.00014156303950585425,
|
| 2354 |
+
"epoch": 0.96875,
|
| 2355 |
+
"grad_norm": 1.5081686581055254,
|
| 2356 |
+
"kl": 0.015255122911185026,
|
| 2357 |
+
"learning_rate": 2.5653383040524224e-09,
|
| 2358 |
+
"loss": 0.000211803984711878,
|
| 2359 |
+
"step": 155,
|
| 2360 |
+
"step_time": 23.872454029042274
|
| 2361 |
+
},
|
| 2362 |
+
{
|
| 2363 |
+
"clip_ratio/high_max": 0.0013207430893089622,
|
| 2364 |
+
"clip_ratio/high_mean": 0.0005413344115368091,
|
| 2365 |
+
"clip_ratio/low_mean": 0.00044826819066656753,
|
| 2366 |
+
"clip_ratio/low_min": 0.0,
|
| 2367 |
+
"clip_ratio/region_mean": 0.0009896026022033766,
|
| 2368 |
+
"epoch": 0.975,
|
| 2369 |
+
"grad_norm": 5.4098475523322405,
|
| 2370 |
+
"kl": 0.012522216304205358,
|
| 2371 |
+
"learning_rate": 1.642322514122385e-09,
|
| 2372 |
+
"loss": 0.0004043988883495331,
|
| 2373 |
+
"step": 156,
|
| 2374 |
+
"step_time": 24.749777329619974
|
| 2375 |
+
},
|
| 2376 |
+
{
|
| 2377 |
+
"clip_ratio/high_max": 0.0016891301929717883,
|
| 2378 |
+
"clip_ratio/high_mean": 0.00043829998685396276,
|
| 2379 |
+
"clip_ratio/low_mean": 0.00016886290541151538,
|
| 2380 |
+
"clip_ratio/low_min": 0.0,
|
| 2381 |
+
"clip_ratio/region_mean": 0.0006071628922654781,
|
| 2382 |
+
"epoch": 0.98125,
|
| 2383 |
+
"grad_norm": 3.401791637807875,
|
| 2384 |
+
"kl": 0.0222931137541309,
|
| 2385 |
+
"learning_rate": 9.240278374995636e-10,
|
| 2386 |
+
"loss": 0.0003452016389928758,
|
| 2387 |
+
"step": 157,
|
| 2388 |
+
"step_time": 24.73414495307952
|
| 2389 |
+
},
|
| 2390 |
+
{
|
| 2391 |
+
"clip_ratio/high_max": 0.0014769778645131737,
|
| 2392 |
+
"clip_ratio/high_mean": 0.0003692444661282934,
|
| 2393 |
+
"clip_ratio/low_mean": 0.0005605397163890302,
|
| 2394 |
+
"clip_ratio/low_min": 0.0,
|
| 2395 |
+
"clip_ratio/region_mean": 0.0009297841679654084,
|
| 2396 |
+
"epoch": 0.9875,
|
| 2397 |
+
"grad_norm": 3.425262099278379,
|
| 2398 |
+
"kl": 0.013876177137717605,
|
| 2399 |
+
"learning_rate": 4.1074934355384006e-10,
|
| 2400 |
+
"loss": 0.00027002766728401184,
|
| 2401 |
+
"step": 158,
|
| 2402 |
+
"step_time": 25.09980893228203
|
| 2403 |
+
},
|
| 2404 |
+
{
|
| 2405 |
+
"clip_ratio/high_max": 0.0024547354551032186,
|
| 2406 |
+
"clip_ratio/high_mean": 0.0006786526791984215,
|
| 2407 |
+
"clip_ratio/low_mean": 0.0003988959506386891,
|
| 2408 |
+
"clip_ratio/low_min": 0.0,
|
| 2409 |
+
"clip_ratio/region_mean": 0.0010775486298371106,
|
| 2410 |
+
"epoch": 0.99375,
|
| 2411 |
+
"grad_norm": 4.031048370091605,
|
| 2412 |
+
"kl": 0.036721623269841075,
|
| 2413 |
+
"learning_rate": 1.0269788274358537e-10,
|
| 2414 |
+
"loss": 0.000984895508736372,
|
| 2415 |
+
"step": 159,
|
| 2416 |
+
"step_time": 26.58611388411373
|
| 2417 |
+
},
|
| 2418 |
+
{
|
| 2419 |
+
"clip_ratio/high_max": 0.0015673136222176254,
|
| 2420 |
+
"clip_ratio/high_mean": 0.00039182840555440634,
|
| 2421 |
+
"clip_ratio/low_mean": 7.859133620513603e-05,
|
| 2422 |
+
"clip_ratio/low_min": 0.0,
|
| 2423 |
+
"clip_ratio/region_mean": 0.00047041973994055297,
|
| 2424 |
+
"epoch": 1.0,
|
| 2425 |
+
"grad_norm": 2.757834844247163,
|
| 2426 |
+
"kl": 0.03755952534265816,
|
| 2427 |
+
"learning_rate": 0.0,
|
| 2428 |
+
"loss": 0.00028778373962268233,
|
| 2429 |
+
"step": 160,
|
| 2430 |
+
"step_time": 25.10634110402316
|
| 2431 |
+
}
|
| 2432 |
+
],
|
| 2433 |
+
"logging_steps": 1,
|
| 2434 |
+
"max_steps": 160,
|
| 2435 |
+
"num_input_tokens_seen": 0,
|
| 2436 |
+
"num_train_epochs": 1,
|
| 2437 |
+
"save_steps": 80,
|
| 2438 |
+
"stateful_callbacks": {
|
| 2439 |
+
"TrainerControl": {
|
| 2440 |
+
"args": {
|
| 2441 |
+
"should_epoch_stop": false,
|
| 2442 |
+
"should_evaluate": false,
|
| 2443 |
+
"should_log": false,
|
| 2444 |
+
"should_save": true,
|
| 2445 |
+
"should_training_stop": true
|
| 2446 |
+
},
|
| 2447 |
+
"attributes": {}
|
| 2448 |
+
}
|
| 2449 |
+
},
|
| 2450 |
+
"total_flos": 0.0,
|
| 2451 |
+
"train_batch_size": 1,
|
| 2452 |
+
"trial_name": null,
|
| 2453 |
+
"trial_params": null
|
| 2454 |
+
}
|
a4_fixed_images/round_2/chat_template.jinja
ADDED
|
@@ -0,0 +1,154 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{%- set image_count = namespace(value=0) %}
|
| 2 |
+
{%- set video_count = namespace(value=0) %}
|
| 3 |
+
{%- macro render_content(content, do_vision_count, is_system_content=false) %}
|
| 4 |
+
{%- if content is string %}
|
| 5 |
+
{{- content }}
|
| 6 |
+
{%- elif content is iterable and content is not mapping %}
|
| 7 |
+
{%- for item in content %}
|
| 8 |
+
{%- if 'image' in item or 'image_url' in item or item.type == 'image' %}
|
| 9 |
+
{%- if is_system_content %}
|
| 10 |
+
{{- raise_exception('System message cannot contain images.') }}
|
| 11 |
+
{%- endif %}
|
| 12 |
+
{%- if do_vision_count %}
|
| 13 |
+
{%- set image_count.value = image_count.value + 1 %}
|
| 14 |
+
{%- endif %}
|
| 15 |
+
{%- if add_vision_id %}
|
| 16 |
+
{{- 'Picture ' ~ image_count.value ~ ': ' }}
|
| 17 |
+
{%- endif %}
|
| 18 |
+
{{- '<|vision_start|><|image_pad|><|vision_end|>' }}
|
| 19 |
+
{%- elif 'video' in item or item.type == 'video' %}
|
| 20 |
+
{%- if is_system_content %}
|
| 21 |
+
{{- raise_exception('System message cannot contain videos.') }}
|
| 22 |
+
{%- endif %}
|
| 23 |
+
{%- if do_vision_count %}
|
| 24 |
+
{%- set video_count.value = video_count.value + 1 %}
|
| 25 |
+
{%- endif %}
|
| 26 |
+
{%- if add_vision_id %}
|
| 27 |
+
{{- 'Video ' ~ video_count.value ~ ': ' }}
|
| 28 |
+
{%- endif %}
|
| 29 |
+
{{- '<|vision_start|><|video_pad|><|vision_end|>' }}
|
| 30 |
+
{%- elif 'text' in item %}
|
| 31 |
+
{{- item.text }}
|
| 32 |
+
{%- else %}
|
| 33 |
+
{{- raise_exception('Unexpected item type in content.') }}
|
| 34 |
+
{%- endif %}
|
| 35 |
+
{%- endfor %}
|
| 36 |
+
{%- elif content is none or content is undefined %}
|
| 37 |
+
{{- '' }}
|
| 38 |
+
{%- else %}
|
| 39 |
+
{{- raise_exception('Unexpected content type.') }}
|
| 40 |
+
{%- endif %}
|
| 41 |
+
{%- endmacro %}
|
| 42 |
+
{%- if not messages %}
|
| 43 |
+
{{- raise_exception('No messages provided.') }}
|
| 44 |
+
{%- endif %}
|
| 45 |
+
{%- if tools and tools is iterable and tools is not mapping %}
|
| 46 |
+
{{- '<|im_start|>system\n' }}
|
| 47 |
+
{{- "# Tools\n\nYou have access to the following functions:\n\n<tools>" }}
|
| 48 |
+
{%- for tool in tools %}
|
| 49 |
+
{{- "\n" }}
|
| 50 |
+
{{- tool | tojson }}
|
| 51 |
+
{%- endfor %}
|
| 52 |
+
{{- "\n</tools>" }}
|
| 53 |
+
{{- '\n\nIf you choose to call a function ONLY reply in the following format with NO suffix:\n\n<tool_call>\n<function=example_function_name>\n<parameter=example_parameter_1>\nvalue_1\n</parameter>\n<parameter=example_parameter_2>\nThis is the value for the second parameter\nthat can span\nmultiple lines\n</parameter>\n</function>\n</tool_call>\n\n<IMPORTANT>\nReminder:\n- Function calls MUST follow the specified format: an inner <function=...></function> block must be nested within <tool_call></tool_call> XML tags\n- Required parameters MUST be specified\n- You may provide optional reasoning for your function call in natural language BEFORE the function call, but NOT after\n- If there is no function call available, answer the question like normal with your current knowledge and do not tell the user about function calls\n</IMPORTANT>' }}
|
| 54 |
+
{%- if messages[0].role == 'system' %}
|
| 55 |
+
{%- set content = render_content(messages[0].content, false, true)|trim %}
|
| 56 |
+
{%- if content %}
|
| 57 |
+
{{- '\n\n' + content }}
|
| 58 |
+
{%- endif %}
|
| 59 |
+
{%- endif %}
|
| 60 |
+
{{- '<|im_end|>\n' }}
|
| 61 |
+
{%- else %}
|
| 62 |
+
{%- if messages[0].role == 'system' %}
|
| 63 |
+
{%- set content = render_content(messages[0].content, false, true)|trim %}
|
| 64 |
+
{{- '<|im_start|>system\n' + content + '<|im_end|>\n' }}
|
| 65 |
+
{%- endif %}
|
| 66 |
+
{%- endif %}
|
| 67 |
+
{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
|
| 68 |
+
{%- for message in messages[::-1] %}
|
| 69 |
+
{%- set index = (messages|length - 1) - loop.index0 %}
|
| 70 |
+
{%- if ns.multi_step_tool and message.role == "user" %}
|
| 71 |
+
{%- set content = render_content(message.content, false)|trim %}
|
| 72 |
+
{%- if not(content.startswith('<tool_response>') and content.endswith('</tool_response>')) %}
|
| 73 |
+
{%- set ns.multi_step_tool = false %}
|
| 74 |
+
{%- set ns.last_query_index = index %}
|
| 75 |
+
{%- endif %}
|
| 76 |
+
{%- endif %}
|
| 77 |
+
{%- endfor %}
|
| 78 |
+
{%- if ns.multi_step_tool %}
|
| 79 |
+
{{- raise_exception('No user query found in messages.') }}
|
| 80 |
+
{%- endif %}
|
| 81 |
+
{%- for message in messages %}
|
| 82 |
+
{%- set content = render_content(message.content, true)|trim %}
|
| 83 |
+
{%- if message.role == "system" %}
|
| 84 |
+
{%- if not loop.first %}
|
| 85 |
+
{{- raise_exception('System message must be at the beginning.') }}
|
| 86 |
+
{%- endif %}
|
| 87 |
+
{%- elif message.role == "user" %}
|
| 88 |
+
{{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
|
| 89 |
+
{%- elif message.role == "assistant" %}
|
| 90 |
+
{%- set reasoning_content = '' %}
|
| 91 |
+
{%- if message.reasoning_content is string %}
|
| 92 |
+
{%- set reasoning_content = message.reasoning_content %}
|
| 93 |
+
{%- else %}
|
| 94 |
+
{%- if '</think>' in content %}
|
| 95 |
+
{%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
|
| 96 |
+
{%- set content = content.split('</think>')[-1].lstrip('\n') %}
|
| 97 |
+
{%- endif %}
|
| 98 |
+
{%- endif %}
|
| 99 |
+
{%- set reasoning_content = reasoning_content|trim %}
|
| 100 |
+
{%- if loop.index0 > ns.last_query_index %}
|
| 101 |
+
{{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content + '\n</think>\n\n' + content }}
|
| 102 |
+
{%- else %}
|
| 103 |
+
{{- '<|im_start|>' + message.role + '\n' + content }}
|
| 104 |
+
{%- endif %}
|
| 105 |
+
{%- if message.tool_calls and message.tool_calls is iterable and message.tool_calls is not mapping %}
|
| 106 |
+
{%- for tool_call in message.tool_calls %}
|
| 107 |
+
{%- if tool_call.function is defined %}
|
| 108 |
+
{%- set tool_call = tool_call.function %}
|
| 109 |
+
{%- endif %}
|
| 110 |
+
{%- if loop.first %}
|
| 111 |
+
{%- if content|trim %}
|
| 112 |
+
{{- '\n\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
|
| 113 |
+
{%- else %}
|
| 114 |
+
{{- '<tool_call>\n<function=' + tool_call.name + '>\n' }}
|
| 115 |
+
{%- endif %}
|
| 116 |
+
{%- else %}
|
| 117 |
+
{{- '\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
|
| 118 |
+
{%- endif %}
|
| 119 |
+
{%- if tool_call.arguments is defined %}
|
| 120 |
+
{%- for args_name, args_value in tool_call.arguments|items %}
|
| 121 |
+
{{- '<parameter=' + args_name + '>\n' }}
|
| 122 |
+
{%- set args_value = args_value | tojson | safe if args_value is mapping or (args_value is sequence and args_value is not string) else args_value | string %}
|
| 123 |
+
{{- args_value }}
|
| 124 |
+
{{- '\n</parameter>\n' }}
|
| 125 |
+
{%- endfor %}
|
| 126 |
+
{%- endif %}
|
| 127 |
+
{{- '</function>\n</tool_call>' }}
|
| 128 |
+
{%- endfor %}
|
| 129 |
+
{%- endif %}
|
| 130 |
+
{{- '<|im_end|>\n' }}
|
| 131 |
+
{%- elif message.role == "tool" %}
|
| 132 |
+
{%- if loop.previtem and loop.previtem.role != "tool" %}
|
| 133 |
+
{{- '<|im_start|>user' }}
|
| 134 |
+
{%- endif %}
|
| 135 |
+
{{- '\n<tool_response>\n' }}
|
| 136 |
+
{{- content }}
|
| 137 |
+
{{- '\n</tool_response>' }}
|
| 138 |
+
{%- if not loop.last and loop.nextitem.role != "tool" %}
|
| 139 |
+
{{- '<|im_end|>\n' }}
|
| 140 |
+
{%- elif loop.last %}
|
| 141 |
+
{{- '<|im_end|>\n' }}
|
| 142 |
+
{%- endif %}
|
| 143 |
+
{%- else %}
|
| 144 |
+
{{- raise_exception('Unexpected message role.') }}
|
| 145 |
+
{%- endif %}
|
| 146 |
+
{%- endfor %}
|
| 147 |
+
{%- if add_generation_prompt %}
|
| 148 |
+
{{- '<|im_start|>assistant\n' }}
|
| 149 |
+
{%- if enable_thinking is defined and enable_thinking is false %}
|
| 150 |
+
{{- '<think>\n\n</think>\n\n' }}
|
| 151 |
+
{%- else %}
|
| 152 |
+
{{- '<think>\n' }}
|
| 153 |
+
{%- endif %}
|
| 154 |
+
{%- endif %}
|
a4_fixed_images/round_2/config.json
ADDED
|
@@ -0,0 +1,113 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"Qwen3_5ForConditionalGeneration"
|
| 4 |
+
],
|
| 5 |
+
"dtype": "bfloat16",
|
| 6 |
+
"eos_token_id": 248046,
|
| 7 |
+
"hidden_size": 4096,
|
| 8 |
+
"image_token_id": 248056,
|
| 9 |
+
"model_type": "qwen3_5",
|
| 10 |
+
"pad_token_id": 248044,
|
| 11 |
+
"text_config": {
|
| 12 |
+
"attention_bias": false,
|
| 13 |
+
"attention_dropout": 0.0,
|
| 14 |
+
"attn_output_gate": true,
|
| 15 |
+
"bos_token_id": null,
|
| 16 |
+
"dtype": "bfloat16",
|
| 17 |
+
"eos_token_id": 248044,
|
| 18 |
+
"full_attention_interval": 4,
|
| 19 |
+
"head_dim": 256,
|
| 20 |
+
"hidden_act": "silu",
|
| 21 |
+
"hidden_size": 4096,
|
| 22 |
+
"initializer_range": 0.02,
|
| 23 |
+
"intermediate_size": 12288,
|
| 24 |
+
"layer_types": [
|
| 25 |
+
"linear_attention",
|
| 26 |
+
"linear_attention",
|
| 27 |
+
"linear_attention",
|
| 28 |
+
"full_attention",
|
| 29 |
+
"linear_attention",
|
| 30 |
+
"linear_attention",
|
| 31 |
+
"linear_attention",
|
| 32 |
+
"full_attention",
|
| 33 |
+
"linear_attention",
|
| 34 |
+
"linear_attention",
|
| 35 |
+
"linear_attention",
|
| 36 |
+
"full_attention",
|
| 37 |
+
"linear_attention",
|
| 38 |
+
"linear_attention",
|
| 39 |
+
"linear_attention",
|
| 40 |
+
"full_attention",
|
| 41 |
+
"linear_attention",
|
| 42 |
+
"linear_attention",
|
| 43 |
+
"linear_attention",
|
| 44 |
+
"full_attention",
|
| 45 |
+
"linear_attention",
|
| 46 |
+
"linear_attention",
|
| 47 |
+
"linear_attention",
|
| 48 |
+
"full_attention",
|
| 49 |
+
"linear_attention",
|
| 50 |
+
"linear_attention",
|
| 51 |
+
"linear_attention",
|
| 52 |
+
"full_attention",
|
| 53 |
+
"linear_attention",
|
| 54 |
+
"linear_attention",
|
| 55 |
+
"linear_attention",
|
| 56 |
+
"full_attention"
|
| 57 |
+
],
|
| 58 |
+
"linear_conv_kernel_dim": 4,
|
| 59 |
+
"linear_key_head_dim": 128,
|
| 60 |
+
"linear_num_key_heads": 16,
|
| 61 |
+
"linear_num_value_heads": 32,
|
| 62 |
+
"linear_value_head_dim": 128,
|
| 63 |
+
"mamba_ssm_dtype": "float32",
|
| 64 |
+
"max_position_embeddings": 262144,
|
| 65 |
+
"mlp_only_layers": [],
|
| 66 |
+
"model_type": "qwen3_5_text",
|
| 67 |
+
"mtp_num_hidden_layers": 1,
|
| 68 |
+
"mtp_use_dedicated_embeddings": false,
|
| 69 |
+
"num_attention_heads": 16,
|
| 70 |
+
"num_hidden_layers": 32,
|
| 71 |
+
"num_key_value_heads": 4,
|
| 72 |
+
"pad_token_id": 248044,
|
| 73 |
+
"partial_rotary_factor": 0.25,
|
| 74 |
+
"rms_norm_eps": 1e-06,
|
| 75 |
+
"rope_parameters": {
|
| 76 |
+
"mrope_interleaved": true,
|
| 77 |
+
"mrope_section": [
|
| 78 |
+
11,
|
| 79 |
+
11,
|
| 80 |
+
10
|
| 81 |
+
],
|
| 82 |
+
"partial_rotary_factor": 0.25,
|
| 83 |
+
"rope_theta": 10000000,
|
| 84 |
+
"rope_type": "default"
|
| 85 |
+
},
|
| 86 |
+
"tie_word_embeddings": false,
|
| 87 |
+
"use_cache": false,
|
| 88 |
+
"vocab_size": 248320
|
| 89 |
+
},
|
| 90 |
+
"tie_word_embeddings": false,
|
| 91 |
+
"transformers_version": "5.10.4",
|
| 92 |
+
"use_cache": false,
|
| 93 |
+
"video_token_id": 248057,
|
| 94 |
+
"vision_config": {
|
| 95 |
+
"deepstack_visual_indexes": [],
|
| 96 |
+
"depth": 27,
|
| 97 |
+
"dtype": "bfloat16",
|
| 98 |
+
"hidden_act": "gelu_pytorch_tanh",
|
| 99 |
+
"hidden_size": 1152,
|
| 100 |
+
"in_channels": 3,
|
| 101 |
+
"initializer_range": 0.02,
|
| 102 |
+
"intermediate_size": 4304,
|
| 103 |
+
"model_type": "qwen3_5_vision",
|
| 104 |
+
"num_heads": 16,
|
| 105 |
+
"num_position_embeddings": 2304,
|
| 106 |
+
"out_hidden_size": 4096,
|
| 107 |
+
"patch_size": 16,
|
| 108 |
+
"spatial_merge_size": 2,
|
| 109 |
+
"temporal_patch_size": 2
|
| 110 |
+
},
|
| 111 |
+
"vision_end_token_id": 248054,
|
| 112 |
+
"vision_start_token_id": 248053
|
| 113 |
+
}
|
a4_fixed_images/round_2/generation_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"eos_token_id": [
|
| 4 |
+
248044,
|
| 5 |
+
248046
|
| 6 |
+
],
|
| 7 |
+
"transformers_version": "5.10.4",
|
| 8 |
+
"use_cache": true
|
| 9 |
+
}
|
a4_fixed_images/round_2/model.safetensors.index.json
ADDED
|
@@ -0,0 +1,768 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"metadata": {
|
| 3 |
+
"total_parameters": 1469680,
|
| 4 |
+
"total_size": 18819627488
|
| 5 |
+
},
|
| 6 |
+
"weight_map": {
|
| 7 |
+
"lm_head.weight": "model-00001-of-00004.safetensors",
|
| 8 |
+
"model.language_model.embed_tokens.weight": "model-00001-of-00004.safetensors",
|
| 9 |
+
"model.language_model.layers.0.input_layernorm.weight": "model-00001-of-00004.safetensors",
|
| 10 |
+
"model.language_model.layers.0.linear_attn.A_log": "model-00001-of-00004.safetensors",
|
| 11 |
+
"model.language_model.layers.0.linear_attn.conv1d.weight": "model-00001-of-00004.safetensors",
|
| 12 |
+
"model.language_model.layers.0.linear_attn.dt_bias": "model-00001-of-00004.safetensors",
|
| 13 |
+
"model.language_model.layers.0.linear_attn.in_proj_a.weight": "model-00001-of-00004.safetensors",
|
| 14 |
+
"model.language_model.layers.0.linear_attn.in_proj_b.weight": "model-00001-of-00004.safetensors",
|
| 15 |
+
"model.language_model.layers.0.linear_attn.in_proj_qkv.weight": "model-00001-of-00004.safetensors",
|
| 16 |
+
"model.language_model.layers.0.linear_attn.in_proj_z.weight": "model-00001-of-00004.safetensors",
|
| 17 |
+
"model.language_model.layers.0.linear_attn.norm.weight": "model-00001-of-00004.safetensors",
|
| 18 |
+
"model.language_model.layers.0.linear_attn.out_proj.weight": "model-00001-of-00004.safetensors",
|
| 19 |
+
"model.language_model.layers.0.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
|
| 20 |
+
"model.language_model.layers.0.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
|
| 21 |
+
"model.language_model.layers.0.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
|
| 22 |
+
"model.language_model.layers.0.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
|
| 23 |
+
"model.language_model.layers.1.input_layernorm.weight": "model-00001-of-00004.safetensors",
|
| 24 |
+
"model.language_model.layers.1.linear_attn.A_log": "model-00001-of-00004.safetensors",
|
| 25 |
+
"model.language_model.layers.1.linear_attn.conv1d.weight": "model-00001-of-00004.safetensors",
|
| 26 |
+
"model.language_model.layers.1.linear_attn.dt_bias": "model-00001-of-00004.safetensors",
|
| 27 |
+
"model.language_model.layers.1.linear_attn.in_proj_a.weight": "model-00001-of-00004.safetensors",
|
| 28 |
+
"model.language_model.layers.1.linear_attn.in_proj_b.weight": "model-00001-of-00004.safetensors",
|
| 29 |
+
"model.language_model.layers.1.linear_attn.in_proj_qkv.weight": "model-00001-of-00004.safetensors",
|
| 30 |
+
"model.language_model.layers.1.linear_attn.in_proj_z.weight": "model-00001-of-00004.safetensors",
|
| 31 |
+
"model.language_model.layers.1.linear_attn.norm.weight": "model-00001-of-00004.safetensors",
|
| 32 |
+
"model.language_model.layers.1.linear_attn.out_proj.weight": "model-00001-of-00004.safetensors",
|
| 33 |
+
"model.language_model.layers.1.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
|
| 34 |
+
"model.language_model.layers.1.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
|
| 35 |
+
"model.language_model.layers.1.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
|
| 36 |
+
"model.language_model.layers.1.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
|
| 37 |
+
"model.language_model.layers.10.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 38 |
+
"model.language_model.layers.10.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 39 |
+
"model.language_model.layers.10.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 40 |
+
"model.language_model.layers.10.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 41 |
+
"model.language_model.layers.10.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 42 |
+
"model.language_model.layers.10.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 43 |
+
"model.language_model.layers.10.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 44 |
+
"model.language_model.layers.10.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 45 |
+
"model.language_model.layers.10.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 46 |
+
"model.language_model.layers.10.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 47 |
+
"model.language_model.layers.10.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 48 |
+
"model.language_model.layers.10.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 49 |
+
"model.language_model.layers.10.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 50 |
+
"model.language_model.layers.10.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 51 |
+
"model.language_model.layers.11.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 52 |
+
"model.language_model.layers.11.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 53 |
+
"model.language_model.layers.11.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 54 |
+
"model.language_model.layers.11.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 55 |
+
"model.language_model.layers.11.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 56 |
+
"model.language_model.layers.11.self_attn.k_norm.weight": "model-00002-of-00004.safetensors",
|
| 57 |
+
"model.language_model.layers.11.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
|
| 58 |
+
"model.language_model.layers.11.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
|
| 59 |
+
"model.language_model.layers.11.self_attn.q_norm.weight": "model-00002-of-00004.safetensors",
|
| 60 |
+
"model.language_model.layers.11.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
|
| 61 |
+
"model.language_model.layers.11.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
|
| 62 |
+
"model.language_model.layers.12.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 63 |
+
"model.language_model.layers.12.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 64 |
+
"model.language_model.layers.12.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 65 |
+
"model.language_model.layers.12.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 66 |
+
"model.language_model.layers.12.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 67 |
+
"model.language_model.layers.12.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 68 |
+
"model.language_model.layers.12.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 69 |
+
"model.language_model.layers.12.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 70 |
+
"model.language_model.layers.12.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 71 |
+
"model.language_model.layers.12.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 72 |
+
"model.language_model.layers.12.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 73 |
+
"model.language_model.layers.12.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 74 |
+
"model.language_model.layers.12.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 75 |
+
"model.language_model.layers.12.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 76 |
+
"model.language_model.layers.13.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 77 |
+
"model.language_model.layers.13.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 78 |
+
"model.language_model.layers.13.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 79 |
+
"model.language_model.layers.13.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 80 |
+
"model.language_model.layers.13.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 81 |
+
"model.language_model.layers.13.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 82 |
+
"model.language_model.layers.13.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 83 |
+
"model.language_model.layers.13.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 84 |
+
"model.language_model.layers.13.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 85 |
+
"model.language_model.layers.13.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 86 |
+
"model.language_model.layers.13.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 87 |
+
"model.language_model.layers.13.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 88 |
+
"model.language_model.layers.13.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 89 |
+
"model.language_model.layers.13.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 90 |
+
"model.language_model.layers.14.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 91 |
+
"model.language_model.layers.14.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 92 |
+
"model.language_model.layers.14.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 93 |
+
"model.language_model.layers.14.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 94 |
+
"model.language_model.layers.14.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 95 |
+
"model.language_model.layers.14.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 96 |
+
"model.language_model.layers.14.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 97 |
+
"model.language_model.layers.14.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 98 |
+
"model.language_model.layers.14.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 99 |
+
"model.language_model.layers.14.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 100 |
+
"model.language_model.layers.14.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 101 |
+
"model.language_model.layers.14.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 102 |
+
"model.language_model.layers.14.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 103 |
+
"model.language_model.layers.14.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 104 |
+
"model.language_model.layers.15.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 105 |
+
"model.language_model.layers.15.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 106 |
+
"model.language_model.layers.15.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 107 |
+
"model.language_model.layers.15.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 108 |
+
"model.language_model.layers.15.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 109 |
+
"model.language_model.layers.15.self_attn.k_norm.weight": "model-00003-of-00004.safetensors",
|
| 110 |
+
"model.language_model.layers.15.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
|
| 111 |
+
"model.language_model.layers.15.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
|
| 112 |
+
"model.language_model.layers.15.self_attn.q_norm.weight": "model-00003-of-00004.safetensors",
|
| 113 |
+
"model.language_model.layers.15.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
|
| 114 |
+
"model.language_model.layers.15.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
|
| 115 |
+
"model.language_model.layers.16.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 116 |
+
"model.language_model.layers.16.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 117 |
+
"model.language_model.layers.16.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 118 |
+
"model.language_model.layers.16.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 119 |
+
"model.language_model.layers.16.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 120 |
+
"model.language_model.layers.16.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 121 |
+
"model.language_model.layers.16.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 122 |
+
"model.language_model.layers.16.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 123 |
+
"model.language_model.layers.16.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 124 |
+
"model.language_model.layers.16.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 125 |
+
"model.language_model.layers.16.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 126 |
+
"model.language_model.layers.16.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 127 |
+
"model.language_model.layers.16.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 128 |
+
"model.language_model.layers.16.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 129 |
+
"model.language_model.layers.17.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 130 |
+
"model.language_model.layers.17.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 131 |
+
"model.language_model.layers.17.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 132 |
+
"model.language_model.layers.17.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 133 |
+
"model.language_model.layers.17.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 134 |
+
"model.language_model.layers.17.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 135 |
+
"model.language_model.layers.17.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 136 |
+
"model.language_model.layers.17.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 137 |
+
"model.language_model.layers.17.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 138 |
+
"model.language_model.layers.17.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 139 |
+
"model.language_model.layers.17.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 140 |
+
"model.language_model.layers.17.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 141 |
+
"model.language_model.layers.17.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 142 |
+
"model.language_model.layers.17.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 143 |
+
"model.language_model.layers.18.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 144 |
+
"model.language_model.layers.18.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 145 |
+
"model.language_model.layers.18.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 146 |
+
"model.language_model.layers.18.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 147 |
+
"model.language_model.layers.18.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 148 |
+
"model.language_model.layers.18.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 149 |
+
"model.language_model.layers.18.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 150 |
+
"model.language_model.layers.18.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 151 |
+
"model.language_model.layers.18.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 152 |
+
"model.language_model.layers.18.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 153 |
+
"model.language_model.layers.18.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 154 |
+
"model.language_model.layers.18.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 155 |
+
"model.language_model.layers.18.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 156 |
+
"model.language_model.layers.18.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 157 |
+
"model.language_model.layers.19.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 158 |
+
"model.language_model.layers.19.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 159 |
+
"model.language_model.layers.19.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 160 |
+
"model.language_model.layers.19.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 161 |
+
"model.language_model.layers.19.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 162 |
+
"model.language_model.layers.19.self_attn.k_norm.weight": "model-00003-of-00004.safetensors",
|
| 163 |
+
"model.language_model.layers.19.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
|
| 164 |
+
"model.language_model.layers.19.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
|
| 165 |
+
"model.language_model.layers.19.self_attn.q_norm.weight": "model-00003-of-00004.safetensors",
|
| 166 |
+
"model.language_model.layers.19.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
|
| 167 |
+
"model.language_model.layers.19.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
|
| 168 |
+
"model.language_model.layers.2.input_layernorm.weight": "model-00001-of-00004.safetensors",
|
| 169 |
+
"model.language_model.layers.2.linear_attn.A_log": "model-00001-of-00004.safetensors",
|
| 170 |
+
"model.language_model.layers.2.linear_attn.conv1d.weight": "model-00001-of-00004.safetensors",
|
| 171 |
+
"model.language_model.layers.2.linear_attn.dt_bias": "model-00001-of-00004.safetensors",
|
| 172 |
+
"model.language_model.layers.2.linear_attn.in_proj_a.weight": "model-00001-of-00004.safetensors",
|
| 173 |
+
"model.language_model.layers.2.linear_attn.in_proj_b.weight": "model-00001-of-00004.safetensors",
|
| 174 |
+
"model.language_model.layers.2.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 175 |
+
"model.language_model.layers.2.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 176 |
+
"model.language_model.layers.2.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 177 |
+
"model.language_model.layers.2.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 178 |
+
"model.language_model.layers.2.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 179 |
+
"model.language_model.layers.2.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 180 |
+
"model.language_model.layers.2.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 181 |
+
"model.language_model.layers.2.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 182 |
+
"model.language_model.layers.20.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 183 |
+
"model.language_model.layers.20.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 184 |
+
"model.language_model.layers.20.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 185 |
+
"model.language_model.layers.20.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 186 |
+
"model.language_model.layers.20.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 187 |
+
"model.language_model.layers.20.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 188 |
+
"model.language_model.layers.20.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 189 |
+
"model.language_model.layers.20.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 190 |
+
"model.language_model.layers.20.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 191 |
+
"model.language_model.layers.20.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 192 |
+
"model.language_model.layers.20.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 193 |
+
"model.language_model.layers.20.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 194 |
+
"model.language_model.layers.20.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 195 |
+
"model.language_model.layers.20.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 196 |
+
"model.language_model.layers.21.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 197 |
+
"model.language_model.layers.21.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 198 |
+
"model.language_model.layers.21.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 199 |
+
"model.language_model.layers.21.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 200 |
+
"model.language_model.layers.21.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 201 |
+
"model.language_model.layers.21.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 202 |
+
"model.language_model.layers.21.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 203 |
+
"model.language_model.layers.21.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 204 |
+
"model.language_model.layers.21.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 205 |
+
"model.language_model.layers.21.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 206 |
+
"model.language_model.layers.21.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 207 |
+
"model.language_model.layers.21.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 208 |
+
"model.language_model.layers.21.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 209 |
+
"model.language_model.layers.21.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 210 |
+
"model.language_model.layers.22.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 211 |
+
"model.language_model.layers.22.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 212 |
+
"model.language_model.layers.22.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 213 |
+
"model.language_model.layers.22.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 214 |
+
"model.language_model.layers.22.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 215 |
+
"model.language_model.layers.22.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 216 |
+
"model.language_model.layers.22.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 217 |
+
"model.language_model.layers.22.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 218 |
+
"model.language_model.layers.22.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 219 |
+
"model.language_model.layers.22.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 220 |
+
"model.language_model.layers.22.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 221 |
+
"model.language_model.layers.22.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 222 |
+
"model.language_model.layers.22.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 223 |
+
"model.language_model.layers.22.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 224 |
+
"model.language_model.layers.23.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 225 |
+
"model.language_model.layers.23.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 226 |
+
"model.language_model.layers.23.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 227 |
+
"model.language_model.layers.23.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 228 |
+
"model.language_model.layers.23.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 229 |
+
"model.language_model.layers.23.self_attn.k_norm.weight": "model-00003-of-00004.safetensors",
|
| 230 |
+
"model.language_model.layers.23.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
|
| 231 |
+
"model.language_model.layers.23.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
|
| 232 |
+
"model.language_model.layers.23.self_attn.q_norm.weight": "model-00003-of-00004.safetensors",
|
| 233 |
+
"model.language_model.layers.23.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
|
| 234 |
+
"model.language_model.layers.23.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
|
| 235 |
+
"model.language_model.layers.24.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 236 |
+
"model.language_model.layers.24.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 237 |
+
"model.language_model.layers.24.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 238 |
+
"model.language_model.layers.24.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 239 |
+
"model.language_model.layers.24.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 240 |
+
"model.language_model.layers.24.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 241 |
+
"model.language_model.layers.24.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 242 |
+
"model.language_model.layers.24.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 243 |
+
"model.language_model.layers.24.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 244 |
+
"model.language_model.layers.24.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 245 |
+
"model.language_model.layers.24.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 246 |
+
"model.language_model.layers.24.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 247 |
+
"model.language_model.layers.24.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 248 |
+
"model.language_model.layers.24.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 249 |
+
"model.language_model.layers.25.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 250 |
+
"model.language_model.layers.25.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 251 |
+
"model.language_model.layers.25.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 252 |
+
"model.language_model.layers.25.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 253 |
+
"model.language_model.layers.25.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 254 |
+
"model.language_model.layers.25.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 255 |
+
"model.language_model.layers.25.linear_attn.in_proj_qkv.weight": "model-00004-of-00004.safetensors",
|
| 256 |
+
"model.language_model.layers.25.linear_attn.in_proj_z.weight": "model-00004-of-00004.safetensors",
|
| 257 |
+
"model.language_model.layers.25.linear_attn.norm.weight": "model-00004-of-00004.safetensors",
|
| 258 |
+
"model.language_model.layers.25.linear_attn.out_proj.weight": "model-00004-of-00004.safetensors",
|
| 259 |
+
"model.language_model.layers.25.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 260 |
+
"model.language_model.layers.25.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 261 |
+
"model.language_model.layers.25.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 262 |
+
"model.language_model.layers.25.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 263 |
+
"model.language_model.layers.26.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 264 |
+
"model.language_model.layers.26.linear_attn.A_log": "model-00004-of-00004.safetensors",
|
| 265 |
+
"model.language_model.layers.26.linear_attn.conv1d.weight": "model-00004-of-00004.safetensors",
|
| 266 |
+
"model.language_model.layers.26.linear_attn.dt_bias": "model-00004-of-00004.safetensors",
|
| 267 |
+
"model.language_model.layers.26.linear_attn.in_proj_a.weight": "model-00004-of-00004.safetensors",
|
| 268 |
+
"model.language_model.layers.26.linear_attn.in_proj_b.weight": "model-00004-of-00004.safetensors",
|
| 269 |
+
"model.language_model.layers.26.linear_attn.in_proj_qkv.weight": "model-00004-of-00004.safetensors",
|
| 270 |
+
"model.language_model.layers.26.linear_attn.in_proj_z.weight": "model-00004-of-00004.safetensors",
|
| 271 |
+
"model.language_model.layers.26.linear_attn.norm.weight": "model-00004-of-00004.safetensors",
|
| 272 |
+
"model.language_model.layers.26.linear_attn.out_proj.weight": "model-00004-of-00004.safetensors",
|
| 273 |
+
"model.language_model.layers.26.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 274 |
+
"model.language_model.layers.26.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 275 |
+
"model.language_model.layers.26.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 276 |
+
"model.language_model.layers.26.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 277 |
+
"model.language_model.layers.27.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 278 |
+
"model.language_model.layers.27.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 279 |
+
"model.language_model.layers.27.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 280 |
+
"model.language_model.layers.27.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 281 |
+
"model.language_model.layers.27.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 282 |
+
"model.language_model.layers.27.self_attn.k_norm.weight": "model-00004-of-00004.safetensors",
|
| 283 |
+
"model.language_model.layers.27.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
|
| 284 |
+
"model.language_model.layers.27.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
|
| 285 |
+
"model.language_model.layers.27.self_attn.q_norm.weight": "model-00004-of-00004.safetensors",
|
| 286 |
+
"model.language_model.layers.27.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
|
| 287 |
+
"model.language_model.layers.27.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
|
| 288 |
+
"model.language_model.layers.28.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 289 |
+
"model.language_model.layers.28.linear_attn.A_log": "model-00004-of-00004.safetensors",
|
| 290 |
+
"model.language_model.layers.28.linear_attn.conv1d.weight": "model-00004-of-00004.safetensors",
|
| 291 |
+
"model.language_model.layers.28.linear_attn.dt_bias": "model-00004-of-00004.safetensors",
|
| 292 |
+
"model.language_model.layers.28.linear_attn.in_proj_a.weight": "model-00004-of-00004.safetensors",
|
| 293 |
+
"model.language_model.layers.28.linear_attn.in_proj_b.weight": "model-00004-of-00004.safetensors",
|
| 294 |
+
"model.language_model.layers.28.linear_attn.in_proj_qkv.weight": "model-00004-of-00004.safetensors",
|
| 295 |
+
"model.language_model.layers.28.linear_attn.in_proj_z.weight": "model-00004-of-00004.safetensors",
|
| 296 |
+
"model.language_model.layers.28.linear_attn.norm.weight": "model-00004-of-00004.safetensors",
|
| 297 |
+
"model.language_model.layers.28.linear_attn.out_proj.weight": "model-00004-of-00004.safetensors",
|
| 298 |
+
"model.language_model.layers.28.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 299 |
+
"model.language_model.layers.28.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 300 |
+
"model.language_model.layers.28.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 301 |
+
"model.language_model.layers.28.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 302 |
+
"model.language_model.layers.29.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 303 |
+
"model.language_model.layers.29.linear_attn.A_log": "model-00004-of-00004.safetensors",
|
| 304 |
+
"model.language_model.layers.29.linear_attn.conv1d.weight": "model-00004-of-00004.safetensors",
|
| 305 |
+
"model.language_model.layers.29.linear_attn.dt_bias": "model-00004-of-00004.safetensors",
|
| 306 |
+
"model.language_model.layers.29.linear_attn.in_proj_a.weight": "model-00004-of-00004.safetensors",
|
| 307 |
+
"model.language_model.layers.29.linear_attn.in_proj_b.weight": "model-00004-of-00004.safetensors",
|
| 308 |
+
"model.language_model.layers.29.linear_attn.in_proj_qkv.weight": "model-00004-of-00004.safetensors",
|
| 309 |
+
"model.language_model.layers.29.linear_attn.in_proj_z.weight": "model-00004-of-00004.safetensors",
|
| 310 |
+
"model.language_model.layers.29.linear_attn.norm.weight": "model-00004-of-00004.safetensors",
|
| 311 |
+
"model.language_model.layers.29.linear_attn.out_proj.weight": "model-00004-of-00004.safetensors",
|
| 312 |
+
"model.language_model.layers.29.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 313 |
+
"model.language_model.layers.29.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 314 |
+
"model.language_model.layers.29.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 315 |
+
"model.language_model.layers.29.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 316 |
+
"model.language_model.layers.3.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 317 |
+
"model.language_model.layers.3.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 318 |
+
"model.language_model.layers.3.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 319 |
+
"model.language_model.layers.3.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 320 |
+
"model.language_model.layers.3.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 321 |
+
"model.language_model.layers.3.self_attn.k_norm.weight": "model-00002-of-00004.safetensors",
|
| 322 |
+
"model.language_model.layers.3.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
|
| 323 |
+
"model.language_model.layers.3.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
|
| 324 |
+
"model.language_model.layers.3.self_attn.q_norm.weight": "model-00002-of-00004.safetensors",
|
| 325 |
+
"model.language_model.layers.3.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
|
| 326 |
+
"model.language_model.layers.3.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
|
| 327 |
+
"model.language_model.layers.30.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 328 |
+
"model.language_model.layers.30.linear_attn.A_log": "model-00004-of-00004.safetensors",
|
| 329 |
+
"model.language_model.layers.30.linear_attn.conv1d.weight": "model-00004-of-00004.safetensors",
|
| 330 |
+
"model.language_model.layers.30.linear_attn.dt_bias": "model-00004-of-00004.safetensors",
|
| 331 |
+
"model.language_model.layers.30.linear_attn.in_proj_a.weight": "model-00004-of-00004.safetensors",
|
| 332 |
+
"model.language_model.layers.30.linear_attn.in_proj_b.weight": "model-00004-of-00004.safetensors",
|
| 333 |
+
"model.language_model.layers.30.linear_attn.in_proj_qkv.weight": "model-00004-of-00004.safetensors",
|
| 334 |
+
"model.language_model.layers.30.linear_attn.in_proj_z.weight": "model-00004-of-00004.safetensors",
|
| 335 |
+
"model.language_model.layers.30.linear_attn.norm.weight": "model-00004-of-00004.safetensors",
|
| 336 |
+
"model.language_model.layers.30.linear_attn.out_proj.weight": "model-00004-of-00004.safetensors",
|
| 337 |
+
"model.language_model.layers.30.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 338 |
+
"model.language_model.layers.30.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 339 |
+
"model.language_model.layers.30.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 340 |
+
"model.language_model.layers.30.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 341 |
+
"model.language_model.layers.31.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 342 |
+
"model.language_model.layers.31.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 343 |
+
"model.language_model.layers.31.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 344 |
+
"model.language_model.layers.31.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 345 |
+
"model.language_model.layers.31.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 346 |
+
"model.language_model.layers.31.self_attn.k_norm.weight": "model-00004-of-00004.safetensors",
|
| 347 |
+
"model.language_model.layers.31.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
|
| 348 |
+
"model.language_model.layers.31.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
|
| 349 |
+
"model.language_model.layers.31.self_attn.q_norm.weight": "model-00004-of-00004.safetensors",
|
| 350 |
+
"model.language_model.layers.31.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
|
| 351 |
+
"model.language_model.layers.31.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
|
| 352 |
+
"model.language_model.layers.4.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 353 |
+
"model.language_model.layers.4.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 354 |
+
"model.language_model.layers.4.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 355 |
+
"model.language_model.layers.4.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 356 |
+
"model.language_model.layers.4.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 357 |
+
"model.language_model.layers.4.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 358 |
+
"model.language_model.layers.4.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 359 |
+
"model.language_model.layers.4.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 360 |
+
"model.language_model.layers.4.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 361 |
+
"model.language_model.layers.4.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 362 |
+
"model.language_model.layers.4.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 363 |
+
"model.language_model.layers.4.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 364 |
+
"model.language_model.layers.4.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 365 |
+
"model.language_model.layers.4.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 366 |
+
"model.language_model.layers.5.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 367 |
+
"model.language_model.layers.5.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 368 |
+
"model.language_model.layers.5.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 369 |
+
"model.language_model.layers.5.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 370 |
+
"model.language_model.layers.5.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 371 |
+
"model.language_model.layers.5.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 372 |
+
"model.language_model.layers.5.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 373 |
+
"model.language_model.layers.5.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 374 |
+
"model.language_model.layers.5.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 375 |
+
"model.language_model.layers.5.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 376 |
+
"model.language_model.layers.5.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 377 |
+
"model.language_model.layers.5.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 378 |
+
"model.language_model.layers.5.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 379 |
+
"model.language_model.layers.5.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 380 |
+
"model.language_model.layers.6.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 381 |
+
"model.language_model.layers.6.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 382 |
+
"model.language_model.layers.6.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 383 |
+
"model.language_model.layers.6.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 384 |
+
"model.language_model.layers.6.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 385 |
+
"model.language_model.layers.6.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 386 |
+
"model.language_model.layers.6.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 387 |
+
"model.language_model.layers.6.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 388 |
+
"model.language_model.layers.6.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 389 |
+
"model.language_model.layers.6.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 390 |
+
"model.language_model.layers.6.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 391 |
+
"model.language_model.layers.6.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 392 |
+
"model.language_model.layers.6.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 393 |
+
"model.language_model.layers.6.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 394 |
+
"model.language_model.layers.7.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 395 |
+
"model.language_model.layers.7.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 396 |
+
"model.language_model.layers.7.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 397 |
+
"model.language_model.layers.7.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 398 |
+
"model.language_model.layers.7.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 399 |
+
"model.language_model.layers.7.self_attn.k_norm.weight": "model-00002-of-00004.safetensors",
|
| 400 |
+
"model.language_model.layers.7.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
|
| 401 |
+
"model.language_model.layers.7.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
|
| 402 |
+
"model.language_model.layers.7.self_attn.q_norm.weight": "model-00002-of-00004.safetensors",
|
| 403 |
+
"model.language_model.layers.7.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
|
| 404 |
+
"model.language_model.layers.7.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
|
| 405 |
+
"model.language_model.layers.8.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 406 |
+
"model.language_model.layers.8.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 407 |
+
"model.language_model.layers.8.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 408 |
+
"model.language_model.layers.8.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 409 |
+
"model.language_model.layers.8.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 410 |
+
"model.language_model.layers.8.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 411 |
+
"model.language_model.layers.8.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 412 |
+
"model.language_model.layers.8.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 413 |
+
"model.language_model.layers.8.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 414 |
+
"model.language_model.layers.8.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 415 |
+
"model.language_model.layers.8.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 416 |
+
"model.language_model.layers.8.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 417 |
+
"model.language_model.layers.8.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 418 |
+
"model.language_model.layers.8.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 419 |
+
"model.language_model.layers.9.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 420 |
+
"model.language_model.layers.9.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 421 |
+
"model.language_model.layers.9.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 422 |
+
"model.language_model.layers.9.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 423 |
+
"model.language_model.layers.9.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 424 |
+
"model.language_model.layers.9.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 425 |
+
"model.language_model.layers.9.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 426 |
+
"model.language_model.layers.9.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 427 |
+
"model.language_model.layers.9.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 428 |
+
"model.language_model.layers.9.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 429 |
+
"model.language_model.layers.9.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 430 |
+
"model.language_model.layers.9.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 431 |
+
"model.language_model.layers.9.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 432 |
+
"model.language_model.layers.9.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 433 |
+
"model.language_model.norm.weight": "model-00004-of-00004.safetensors",
|
| 434 |
+
"model.visual.blocks.0.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 435 |
+
"model.visual.blocks.0.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 436 |
+
"model.visual.blocks.0.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 437 |
+
"model.visual.blocks.0.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 438 |
+
"model.visual.blocks.0.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 439 |
+
"model.visual.blocks.0.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 440 |
+
"model.visual.blocks.0.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 441 |
+
"model.visual.blocks.0.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 442 |
+
"model.visual.blocks.0.norm1.bias": "model-00004-of-00004.safetensors",
|
| 443 |
+
"model.visual.blocks.0.norm1.weight": "model-00004-of-00004.safetensors",
|
| 444 |
+
"model.visual.blocks.0.norm2.bias": "model-00004-of-00004.safetensors",
|
| 445 |
+
"model.visual.blocks.0.norm2.weight": "model-00004-of-00004.safetensors",
|
| 446 |
+
"model.visual.blocks.1.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 447 |
+
"model.visual.blocks.1.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 448 |
+
"model.visual.blocks.1.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 449 |
+
"model.visual.blocks.1.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 450 |
+
"model.visual.blocks.1.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 451 |
+
"model.visual.blocks.1.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 452 |
+
"model.visual.blocks.1.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 453 |
+
"model.visual.blocks.1.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 454 |
+
"model.visual.blocks.1.norm1.bias": "model-00004-of-00004.safetensors",
|
| 455 |
+
"model.visual.blocks.1.norm1.weight": "model-00004-of-00004.safetensors",
|
| 456 |
+
"model.visual.blocks.1.norm2.bias": "model-00004-of-00004.safetensors",
|
| 457 |
+
"model.visual.blocks.1.norm2.weight": "model-00004-of-00004.safetensors",
|
| 458 |
+
"model.visual.blocks.10.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 459 |
+
"model.visual.blocks.10.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 460 |
+
"model.visual.blocks.10.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 461 |
+
"model.visual.blocks.10.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 462 |
+
"model.visual.blocks.10.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 463 |
+
"model.visual.blocks.10.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 464 |
+
"model.visual.blocks.10.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 465 |
+
"model.visual.blocks.10.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 466 |
+
"model.visual.blocks.10.norm1.bias": "model-00004-of-00004.safetensors",
|
| 467 |
+
"model.visual.blocks.10.norm1.weight": "model-00004-of-00004.safetensors",
|
| 468 |
+
"model.visual.blocks.10.norm2.bias": "model-00004-of-00004.safetensors",
|
| 469 |
+
"model.visual.blocks.10.norm2.weight": "model-00004-of-00004.safetensors",
|
| 470 |
+
"model.visual.blocks.11.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 471 |
+
"model.visual.blocks.11.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 472 |
+
"model.visual.blocks.11.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 473 |
+
"model.visual.blocks.11.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 474 |
+
"model.visual.blocks.11.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 475 |
+
"model.visual.blocks.11.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 476 |
+
"model.visual.blocks.11.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 477 |
+
"model.visual.blocks.11.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 478 |
+
"model.visual.blocks.11.norm1.bias": "model-00004-of-00004.safetensors",
|
| 479 |
+
"model.visual.blocks.11.norm1.weight": "model-00004-of-00004.safetensors",
|
| 480 |
+
"model.visual.blocks.11.norm2.bias": "model-00004-of-00004.safetensors",
|
| 481 |
+
"model.visual.blocks.11.norm2.weight": "model-00004-of-00004.safetensors",
|
| 482 |
+
"model.visual.blocks.12.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 483 |
+
"model.visual.blocks.12.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 484 |
+
"model.visual.blocks.12.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 485 |
+
"model.visual.blocks.12.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 486 |
+
"model.visual.blocks.12.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 487 |
+
"model.visual.blocks.12.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 488 |
+
"model.visual.blocks.12.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 489 |
+
"model.visual.blocks.12.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 490 |
+
"model.visual.blocks.12.norm1.bias": "model-00004-of-00004.safetensors",
|
| 491 |
+
"model.visual.blocks.12.norm1.weight": "model-00004-of-00004.safetensors",
|
| 492 |
+
"model.visual.blocks.12.norm2.bias": "model-00004-of-00004.safetensors",
|
| 493 |
+
"model.visual.blocks.12.norm2.weight": "model-00004-of-00004.safetensors",
|
| 494 |
+
"model.visual.blocks.13.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 495 |
+
"model.visual.blocks.13.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 496 |
+
"model.visual.blocks.13.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 497 |
+
"model.visual.blocks.13.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 498 |
+
"model.visual.blocks.13.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 499 |
+
"model.visual.blocks.13.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 500 |
+
"model.visual.blocks.13.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 501 |
+
"model.visual.blocks.13.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 502 |
+
"model.visual.blocks.13.norm1.bias": "model-00004-of-00004.safetensors",
|
| 503 |
+
"model.visual.blocks.13.norm1.weight": "model-00004-of-00004.safetensors",
|
| 504 |
+
"model.visual.blocks.13.norm2.bias": "model-00004-of-00004.safetensors",
|
| 505 |
+
"model.visual.blocks.13.norm2.weight": "model-00004-of-00004.safetensors",
|
| 506 |
+
"model.visual.blocks.14.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 507 |
+
"model.visual.blocks.14.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 508 |
+
"model.visual.blocks.14.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 509 |
+
"model.visual.blocks.14.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 510 |
+
"model.visual.blocks.14.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 511 |
+
"model.visual.blocks.14.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 512 |
+
"model.visual.blocks.14.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 513 |
+
"model.visual.blocks.14.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 514 |
+
"model.visual.blocks.14.norm1.bias": "model-00004-of-00004.safetensors",
|
| 515 |
+
"model.visual.blocks.14.norm1.weight": "model-00004-of-00004.safetensors",
|
| 516 |
+
"model.visual.blocks.14.norm2.bias": "model-00004-of-00004.safetensors",
|
| 517 |
+
"model.visual.blocks.14.norm2.weight": "model-00004-of-00004.safetensors",
|
| 518 |
+
"model.visual.blocks.15.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 519 |
+
"model.visual.blocks.15.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 520 |
+
"model.visual.blocks.15.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 521 |
+
"model.visual.blocks.15.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 522 |
+
"model.visual.blocks.15.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 523 |
+
"model.visual.blocks.15.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 524 |
+
"model.visual.blocks.15.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 525 |
+
"model.visual.blocks.15.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 526 |
+
"model.visual.blocks.15.norm1.bias": "model-00004-of-00004.safetensors",
|
| 527 |
+
"model.visual.blocks.15.norm1.weight": "model-00004-of-00004.safetensors",
|
| 528 |
+
"model.visual.blocks.15.norm2.bias": "model-00004-of-00004.safetensors",
|
| 529 |
+
"model.visual.blocks.15.norm2.weight": "model-00004-of-00004.safetensors",
|
| 530 |
+
"model.visual.blocks.16.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 531 |
+
"model.visual.blocks.16.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 532 |
+
"model.visual.blocks.16.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 533 |
+
"model.visual.blocks.16.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 534 |
+
"model.visual.blocks.16.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 535 |
+
"model.visual.blocks.16.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 536 |
+
"model.visual.blocks.16.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 537 |
+
"model.visual.blocks.16.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 538 |
+
"model.visual.blocks.16.norm1.bias": "model-00004-of-00004.safetensors",
|
| 539 |
+
"model.visual.blocks.16.norm1.weight": "model-00004-of-00004.safetensors",
|
| 540 |
+
"model.visual.blocks.16.norm2.bias": "model-00004-of-00004.safetensors",
|
| 541 |
+
"model.visual.blocks.16.norm2.weight": "model-00004-of-00004.safetensors",
|
| 542 |
+
"model.visual.blocks.17.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 543 |
+
"model.visual.blocks.17.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 544 |
+
"model.visual.blocks.17.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 545 |
+
"model.visual.blocks.17.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 546 |
+
"model.visual.blocks.17.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 547 |
+
"model.visual.blocks.17.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 548 |
+
"model.visual.blocks.17.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 549 |
+
"model.visual.blocks.17.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 550 |
+
"model.visual.blocks.17.norm1.bias": "model-00004-of-00004.safetensors",
|
| 551 |
+
"model.visual.blocks.17.norm1.weight": "model-00004-of-00004.safetensors",
|
| 552 |
+
"model.visual.blocks.17.norm2.bias": "model-00004-of-00004.safetensors",
|
| 553 |
+
"model.visual.blocks.17.norm2.weight": "model-00004-of-00004.safetensors",
|
| 554 |
+
"model.visual.blocks.18.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 555 |
+
"model.visual.blocks.18.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 556 |
+
"model.visual.blocks.18.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 557 |
+
"model.visual.blocks.18.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 558 |
+
"model.visual.blocks.18.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 559 |
+
"model.visual.blocks.18.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 560 |
+
"model.visual.blocks.18.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 561 |
+
"model.visual.blocks.18.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 562 |
+
"model.visual.blocks.18.norm1.bias": "model-00004-of-00004.safetensors",
|
| 563 |
+
"model.visual.blocks.18.norm1.weight": "model-00004-of-00004.safetensors",
|
| 564 |
+
"model.visual.blocks.18.norm2.bias": "model-00004-of-00004.safetensors",
|
| 565 |
+
"model.visual.blocks.18.norm2.weight": "model-00004-of-00004.safetensors",
|
| 566 |
+
"model.visual.blocks.19.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 567 |
+
"model.visual.blocks.19.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 568 |
+
"model.visual.blocks.19.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 569 |
+
"model.visual.blocks.19.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 570 |
+
"model.visual.blocks.19.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 571 |
+
"model.visual.blocks.19.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 572 |
+
"model.visual.blocks.19.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 573 |
+
"model.visual.blocks.19.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 574 |
+
"model.visual.blocks.19.norm1.bias": "model-00004-of-00004.safetensors",
|
| 575 |
+
"model.visual.blocks.19.norm1.weight": "model-00004-of-00004.safetensors",
|
| 576 |
+
"model.visual.blocks.19.norm2.bias": "model-00004-of-00004.safetensors",
|
| 577 |
+
"model.visual.blocks.19.norm2.weight": "model-00004-of-00004.safetensors",
|
| 578 |
+
"model.visual.blocks.2.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 579 |
+
"model.visual.blocks.2.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 580 |
+
"model.visual.blocks.2.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 581 |
+
"model.visual.blocks.2.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 582 |
+
"model.visual.blocks.2.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 583 |
+
"model.visual.blocks.2.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 584 |
+
"model.visual.blocks.2.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 585 |
+
"model.visual.blocks.2.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 586 |
+
"model.visual.blocks.2.norm1.bias": "model-00004-of-00004.safetensors",
|
| 587 |
+
"model.visual.blocks.2.norm1.weight": "model-00004-of-00004.safetensors",
|
| 588 |
+
"model.visual.blocks.2.norm2.bias": "model-00004-of-00004.safetensors",
|
| 589 |
+
"model.visual.blocks.2.norm2.weight": "model-00004-of-00004.safetensors",
|
| 590 |
+
"model.visual.blocks.20.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 591 |
+
"model.visual.blocks.20.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 592 |
+
"model.visual.blocks.20.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 593 |
+
"model.visual.blocks.20.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 594 |
+
"model.visual.blocks.20.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 595 |
+
"model.visual.blocks.20.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 596 |
+
"model.visual.blocks.20.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 597 |
+
"model.visual.blocks.20.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 598 |
+
"model.visual.blocks.20.norm1.bias": "model-00004-of-00004.safetensors",
|
| 599 |
+
"model.visual.blocks.20.norm1.weight": "model-00004-of-00004.safetensors",
|
| 600 |
+
"model.visual.blocks.20.norm2.bias": "model-00004-of-00004.safetensors",
|
| 601 |
+
"model.visual.blocks.20.norm2.weight": "model-00004-of-00004.safetensors",
|
| 602 |
+
"model.visual.blocks.21.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 603 |
+
"model.visual.blocks.21.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 604 |
+
"model.visual.blocks.21.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 605 |
+
"model.visual.blocks.21.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 606 |
+
"model.visual.blocks.21.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 607 |
+
"model.visual.blocks.21.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 608 |
+
"model.visual.blocks.21.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 609 |
+
"model.visual.blocks.21.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 610 |
+
"model.visual.blocks.21.norm1.bias": "model-00004-of-00004.safetensors",
|
| 611 |
+
"model.visual.blocks.21.norm1.weight": "model-00004-of-00004.safetensors",
|
| 612 |
+
"model.visual.blocks.21.norm2.bias": "model-00004-of-00004.safetensors",
|
| 613 |
+
"model.visual.blocks.21.norm2.weight": "model-00004-of-00004.safetensors",
|
| 614 |
+
"model.visual.blocks.22.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 615 |
+
"model.visual.blocks.22.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 616 |
+
"model.visual.blocks.22.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 617 |
+
"model.visual.blocks.22.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 618 |
+
"model.visual.blocks.22.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 619 |
+
"model.visual.blocks.22.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 620 |
+
"model.visual.blocks.22.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 621 |
+
"model.visual.blocks.22.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 622 |
+
"model.visual.blocks.22.norm1.bias": "model-00004-of-00004.safetensors",
|
| 623 |
+
"model.visual.blocks.22.norm1.weight": "model-00004-of-00004.safetensors",
|
| 624 |
+
"model.visual.blocks.22.norm2.bias": "model-00004-of-00004.safetensors",
|
| 625 |
+
"model.visual.blocks.22.norm2.weight": "model-00004-of-00004.safetensors",
|
| 626 |
+
"model.visual.blocks.23.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 627 |
+
"model.visual.blocks.23.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 628 |
+
"model.visual.blocks.23.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 629 |
+
"model.visual.blocks.23.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 630 |
+
"model.visual.blocks.23.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 631 |
+
"model.visual.blocks.23.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 632 |
+
"model.visual.blocks.23.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 633 |
+
"model.visual.blocks.23.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 634 |
+
"model.visual.blocks.23.norm1.bias": "model-00004-of-00004.safetensors",
|
| 635 |
+
"model.visual.blocks.23.norm1.weight": "model-00004-of-00004.safetensors",
|
| 636 |
+
"model.visual.blocks.23.norm2.bias": "model-00004-of-00004.safetensors",
|
| 637 |
+
"model.visual.blocks.23.norm2.weight": "model-00004-of-00004.safetensors",
|
| 638 |
+
"model.visual.blocks.24.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 639 |
+
"model.visual.blocks.24.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 640 |
+
"model.visual.blocks.24.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 641 |
+
"model.visual.blocks.24.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 642 |
+
"model.visual.blocks.24.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 643 |
+
"model.visual.blocks.24.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 644 |
+
"model.visual.blocks.24.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 645 |
+
"model.visual.blocks.24.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 646 |
+
"model.visual.blocks.24.norm1.bias": "model-00004-of-00004.safetensors",
|
| 647 |
+
"model.visual.blocks.24.norm1.weight": "model-00004-of-00004.safetensors",
|
| 648 |
+
"model.visual.blocks.24.norm2.bias": "model-00004-of-00004.safetensors",
|
| 649 |
+
"model.visual.blocks.24.norm2.weight": "model-00004-of-00004.safetensors",
|
| 650 |
+
"model.visual.blocks.25.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 651 |
+
"model.visual.blocks.25.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 652 |
+
"model.visual.blocks.25.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 653 |
+
"model.visual.blocks.25.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 654 |
+
"model.visual.blocks.25.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 655 |
+
"model.visual.blocks.25.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 656 |
+
"model.visual.blocks.25.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 657 |
+
"model.visual.blocks.25.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 658 |
+
"model.visual.blocks.25.norm1.bias": "model-00004-of-00004.safetensors",
|
| 659 |
+
"model.visual.blocks.25.norm1.weight": "model-00004-of-00004.safetensors",
|
| 660 |
+
"model.visual.blocks.25.norm2.bias": "model-00004-of-00004.safetensors",
|
| 661 |
+
"model.visual.blocks.25.norm2.weight": "model-00004-of-00004.safetensors",
|
| 662 |
+
"model.visual.blocks.26.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 663 |
+
"model.visual.blocks.26.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 664 |
+
"model.visual.blocks.26.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 665 |
+
"model.visual.blocks.26.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 666 |
+
"model.visual.blocks.26.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 667 |
+
"model.visual.blocks.26.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 668 |
+
"model.visual.blocks.26.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 669 |
+
"model.visual.blocks.26.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 670 |
+
"model.visual.blocks.26.norm1.bias": "model-00004-of-00004.safetensors",
|
| 671 |
+
"model.visual.blocks.26.norm1.weight": "model-00004-of-00004.safetensors",
|
| 672 |
+
"model.visual.blocks.26.norm2.bias": "model-00004-of-00004.safetensors",
|
| 673 |
+
"model.visual.blocks.26.norm2.weight": "model-00004-of-00004.safetensors",
|
| 674 |
+
"model.visual.blocks.3.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 675 |
+
"model.visual.blocks.3.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 676 |
+
"model.visual.blocks.3.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 677 |
+
"model.visual.blocks.3.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 678 |
+
"model.visual.blocks.3.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 679 |
+
"model.visual.blocks.3.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 680 |
+
"model.visual.blocks.3.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 681 |
+
"model.visual.blocks.3.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 682 |
+
"model.visual.blocks.3.norm1.bias": "model-00004-of-00004.safetensors",
|
| 683 |
+
"model.visual.blocks.3.norm1.weight": "model-00004-of-00004.safetensors",
|
| 684 |
+
"model.visual.blocks.3.norm2.bias": "model-00004-of-00004.safetensors",
|
| 685 |
+
"model.visual.blocks.3.norm2.weight": "model-00004-of-00004.safetensors",
|
| 686 |
+
"model.visual.blocks.4.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 687 |
+
"model.visual.blocks.4.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 688 |
+
"model.visual.blocks.4.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 689 |
+
"model.visual.blocks.4.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 690 |
+
"model.visual.blocks.4.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 691 |
+
"model.visual.blocks.4.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 692 |
+
"model.visual.blocks.4.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 693 |
+
"model.visual.blocks.4.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 694 |
+
"model.visual.blocks.4.norm1.bias": "model-00004-of-00004.safetensors",
|
| 695 |
+
"model.visual.blocks.4.norm1.weight": "model-00004-of-00004.safetensors",
|
| 696 |
+
"model.visual.blocks.4.norm2.bias": "model-00004-of-00004.safetensors",
|
| 697 |
+
"model.visual.blocks.4.norm2.weight": "model-00004-of-00004.safetensors",
|
| 698 |
+
"model.visual.blocks.5.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 699 |
+
"model.visual.blocks.5.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 700 |
+
"model.visual.blocks.5.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 701 |
+
"model.visual.blocks.5.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 702 |
+
"model.visual.blocks.5.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 703 |
+
"model.visual.blocks.5.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 704 |
+
"model.visual.blocks.5.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 705 |
+
"model.visual.blocks.5.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 706 |
+
"model.visual.blocks.5.norm1.bias": "model-00004-of-00004.safetensors",
|
| 707 |
+
"model.visual.blocks.5.norm1.weight": "model-00004-of-00004.safetensors",
|
| 708 |
+
"model.visual.blocks.5.norm2.bias": "model-00004-of-00004.safetensors",
|
| 709 |
+
"model.visual.blocks.5.norm2.weight": "model-00004-of-00004.safetensors",
|
| 710 |
+
"model.visual.blocks.6.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 711 |
+
"model.visual.blocks.6.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 712 |
+
"model.visual.blocks.6.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 713 |
+
"model.visual.blocks.6.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 714 |
+
"model.visual.blocks.6.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 715 |
+
"model.visual.blocks.6.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 716 |
+
"model.visual.blocks.6.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 717 |
+
"model.visual.blocks.6.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 718 |
+
"model.visual.blocks.6.norm1.bias": "model-00004-of-00004.safetensors",
|
| 719 |
+
"model.visual.blocks.6.norm1.weight": "model-00004-of-00004.safetensors",
|
| 720 |
+
"model.visual.blocks.6.norm2.bias": "model-00004-of-00004.safetensors",
|
| 721 |
+
"model.visual.blocks.6.norm2.weight": "model-00004-of-00004.safetensors",
|
| 722 |
+
"model.visual.blocks.7.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 723 |
+
"model.visual.blocks.7.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 724 |
+
"model.visual.blocks.7.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 725 |
+
"model.visual.blocks.7.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 726 |
+
"model.visual.blocks.7.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 727 |
+
"model.visual.blocks.7.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 728 |
+
"model.visual.blocks.7.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 729 |
+
"model.visual.blocks.7.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 730 |
+
"model.visual.blocks.7.norm1.bias": "model-00004-of-00004.safetensors",
|
| 731 |
+
"model.visual.blocks.7.norm1.weight": "model-00004-of-00004.safetensors",
|
| 732 |
+
"model.visual.blocks.7.norm2.bias": "model-00004-of-00004.safetensors",
|
| 733 |
+
"model.visual.blocks.7.norm2.weight": "model-00004-of-00004.safetensors",
|
| 734 |
+
"model.visual.blocks.8.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 735 |
+
"model.visual.blocks.8.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 736 |
+
"model.visual.blocks.8.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 737 |
+
"model.visual.blocks.8.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 738 |
+
"model.visual.blocks.8.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 739 |
+
"model.visual.blocks.8.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 740 |
+
"model.visual.blocks.8.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 741 |
+
"model.visual.blocks.8.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 742 |
+
"model.visual.blocks.8.norm1.bias": "model-00004-of-00004.safetensors",
|
| 743 |
+
"model.visual.blocks.8.norm1.weight": "model-00004-of-00004.safetensors",
|
| 744 |
+
"model.visual.blocks.8.norm2.bias": "model-00004-of-00004.safetensors",
|
| 745 |
+
"model.visual.blocks.8.norm2.weight": "model-00004-of-00004.safetensors",
|
| 746 |
+
"model.visual.blocks.9.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 747 |
+
"model.visual.blocks.9.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 748 |
+
"model.visual.blocks.9.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 749 |
+
"model.visual.blocks.9.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 750 |
+
"model.visual.blocks.9.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 751 |
+
"model.visual.blocks.9.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 752 |
+
"model.visual.blocks.9.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 753 |
+
"model.visual.blocks.9.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 754 |
+
"model.visual.blocks.9.norm1.bias": "model-00004-of-00004.safetensors",
|
| 755 |
+
"model.visual.blocks.9.norm1.weight": "model-00004-of-00004.safetensors",
|
| 756 |
+
"model.visual.blocks.9.norm2.bias": "model-00004-of-00004.safetensors",
|
| 757 |
+
"model.visual.blocks.9.norm2.weight": "model-00004-of-00004.safetensors",
|
| 758 |
+
"model.visual.merger.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 759 |
+
"model.visual.merger.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 760 |
+
"model.visual.merger.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 761 |
+
"model.visual.merger.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 762 |
+
"model.visual.merger.norm.bias": "model-00004-of-00004.safetensors",
|
| 763 |
+
"model.visual.merger.norm.weight": "model-00004-of-00004.safetensors",
|
| 764 |
+
"model.visual.patch_embed.proj.bias": "model-00004-of-00004.safetensors",
|
| 765 |
+
"model.visual.patch_embed.proj.weight": "model-00004-of-00004.safetensors",
|
| 766 |
+
"model.visual.pos_embed.weight": "model-00004-of-00004.safetensors"
|
| 767 |
+
}
|
| 768 |
+
}
|
a4_fixed_images/round_2/preprocessor_config.json
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"size": {
|
| 3 |
+
"longest_edge": 16777216,
|
| 4 |
+
"shortest_edge": 65536
|
| 5 |
+
},
|
| 6 |
+
"patch_size": 16,
|
| 7 |
+
"temporal_patch_size": 2,
|
| 8 |
+
"merge_size": 2,
|
| 9 |
+
"image_mean": [
|
| 10 |
+
0.5,
|
| 11 |
+
0.5,
|
| 12 |
+
0.5
|
| 13 |
+
],
|
| 14 |
+
"image_std": [
|
| 15 |
+
0.5,
|
| 16 |
+
0.5,
|
| 17 |
+
0.5
|
| 18 |
+
],
|
| 19 |
+
"processor_class": "Qwen3VLProcessor",
|
| 20 |
+
"image_processor_type": "Qwen2VLImageProcessorFast"
|
| 21 |
+
}
|
a4_fixed_images/round_2/processor_config.json
ADDED
|
@@ -0,0 +1,60 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"image_processor": {
|
| 3 |
+
"do_convert_rgb": true,
|
| 4 |
+
"do_normalize": true,
|
| 5 |
+
"do_rescale": true,
|
| 6 |
+
"do_resize": true,
|
| 7 |
+
"image_mean": [
|
| 8 |
+
0.5,
|
| 9 |
+
0.5,
|
| 10 |
+
0.5
|
| 11 |
+
],
|
| 12 |
+
"image_processor_type": "Qwen2VLImageProcessor",
|
| 13 |
+
"image_std": [
|
| 14 |
+
0.5,
|
| 15 |
+
0.5,
|
| 16 |
+
0.5
|
| 17 |
+
],
|
| 18 |
+
"merge_size": 2,
|
| 19 |
+
"patch_size": 16,
|
| 20 |
+
"resample": 3,
|
| 21 |
+
"rescale_factor": 0.00392156862745098,
|
| 22 |
+
"size": {
|
| 23 |
+
"longest_edge": 16777216,
|
| 24 |
+
"shortest_edge": 65536
|
| 25 |
+
},
|
| 26 |
+
"temporal_patch_size": 2
|
| 27 |
+
},
|
| 28 |
+
"processor_class": "Qwen3VLProcessor",
|
| 29 |
+
"video_processor": {
|
| 30 |
+
"do_convert_rgb": true,
|
| 31 |
+
"do_normalize": true,
|
| 32 |
+
"do_rescale": true,
|
| 33 |
+
"do_resize": true,
|
| 34 |
+
"do_sample_frames": true,
|
| 35 |
+
"fps": 2,
|
| 36 |
+
"image_mean": [
|
| 37 |
+
0.5,
|
| 38 |
+
0.5,
|
| 39 |
+
0.5
|
| 40 |
+
],
|
| 41 |
+
"image_std": [
|
| 42 |
+
0.5,
|
| 43 |
+
0.5,
|
| 44 |
+
0.5
|
| 45 |
+
],
|
| 46 |
+
"max_frames": 768,
|
| 47 |
+
"merge_size": 2,
|
| 48 |
+
"min_frames": 4,
|
| 49 |
+
"patch_size": 16,
|
| 50 |
+
"resample": 3,
|
| 51 |
+
"rescale_factor": 0.00392156862745098,
|
| 52 |
+
"return_metadata": false,
|
| 53 |
+
"size": {
|
| 54 |
+
"longest_edge": 25165824,
|
| 55 |
+
"shortest_edge": 4096
|
| 56 |
+
},
|
| 57 |
+
"temporal_patch_size": 2,
|
| 58 |
+
"video_processor_type": "Qwen3VLVideoProcessor"
|
| 59 |
+
}
|
| 60 |
+
}
|
a4_fixed_images/round_2/tokenizer_config.json
ADDED
|
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"audio_bos_token": "<|audio_start|>",
|
| 4 |
+
"audio_eos_token": "<|audio_end|>",
|
| 5 |
+
"audio_token": "<|audio_pad|>",
|
| 6 |
+
"backend": "tokenizers",
|
| 7 |
+
"bos_token": null,
|
| 8 |
+
"clean_up_tokenization_spaces": false,
|
| 9 |
+
"eos_token": "<|im_end|>",
|
| 10 |
+
"errors": "replace",
|
| 11 |
+
"image_token": "<|image_pad|>",
|
| 12 |
+
"is_local": true,
|
| 13 |
+
"local_files_only": true,
|
| 14 |
+
"model_max_length": 262144,
|
| 15 |
+
"model_specific_special_tokens": {
|
| 16 |
+
"audio_bos_token": "<|audio_start|>",
|
| 17 |
+
"audio_eos_token": "<|audio_end|>",
|
| 18 |
+
"audio_token": "<|audio_pad|>",
|
| 19 |
+
"image_token": "<|image_pad|>",
|
| 20 |
+
"video_token": "<|video_pad|>",
|
| 21 |
+
"vision_bos_token": "<|vision_start|>",
|
| 22 |
+
"vision_eos_token": "<|vision_end|>"
|
| 23 |
+
},
|
| 24 |
+
"pad_token": "<|endoftext|>",
|
| 25 |
+
"pretokenize_regex": "(?i:'s|'t|'re|'ve|'m|'ll|'d)|[^\\r\\n\\p{L}\\p{N}]?[\\p{L}\\p{M}]+|\\p{N}| ?[^\\s\\p{L}\\p{M}\\p{N}]+[\\r\\n]*|\\s*[\\r\\n]+|\\s+(?!\\S)|\\s+",
|
| 26 |
+
"processor_class": "Qwen3VLProcessor",
|
| 27 |
+
"split_special_tokens": false,
|
| 28 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 29 |
+
"unk_token": null,
|
| 30 |
+
"video_token": "<|video_pad|>",
|
| 31 |
+
"vision_bos_token": "<|vision_start|>",
|
| 32 |
+
"vision_eos_token": "<|vision_end|>"
|
| 33 |
+
}
|
a4_fixed_images/round_2/trainer_state.json
ADDED
|
@@ -0,0 +1,2454 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 1.0,
|
| 6 |
+
"eval_steps": 80.0,
|
| 7 |
+
"global_step": 160,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"clip_ratio/high_max": 0.0,
|
| 14 |
+
"clip_ratio/high_mean": 0.0,
|
| 15 |
+
"clip_ratio/low_mean": 0.0,
|
| 16 |
+
"clip_ratio/low_min": 0.0,
|
| 17 |
+
"clip_ratio/region_mean": 0.0,
|
| 18 |
+
"completions/clipped_ratio": 0.19140625,
|
| 19 |
+
"completions/max_length": 2048.0,
|
| 20 |
+
"completions/mean_length": 811.37109375,
|
| 21 |
+
"completions/min_length": 67.0,
|
| 22 |
+
"epoch": 0.00625,
|
| 23 |
+
"frac_reward_zero_std": 0.1875,
|
| 24 |
+
"grad_norm": 4.695847233095537,
|
| 25 |
+
"kl": 0.0,
|
| 26 |
+
"learning_rate": 2e-07,
|
| 27 |
+
"loss": -2.7939677238464355e-09,
|
| 28 |
+
"reward": 0.3828125,
|
| 29 |
+
"reward_std": 0.3799312710762024,
|
| 30 |
+
"rewards/MajorityAccuracy/mean": 0.3828125,
|
| 31 |
+
"rewards/MajorityAccuracy/std": 0.48702529072761536,
|
| 32 |
+
"step": 1,
|
| 33 |
+
"step_time": 204.9993000351824
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"clip_ratio/high_max": 0.0,
|
| 37 |
+
"clip_ratio/high_mean": 0.0,
|
| 38 |
+
"clip_ratio/low_mean": 0.0,
|
| 39 |
+
"clip_ratio/low_min": 0.0,
|
| 40 |
+
"clip_ratio/region_mean": 0.0,
|
| 41 |
+
"epoch": 0.0125,
|
| 42 |
+
"grad_norm": 3.250089562297734,
|
| 43 |
+
"kl": 0.0,
|
| 44 |
+
"learning_rate": 4e-07,
|
| 45 |
+
"loss": -1.862645149230957e-09,
|
| 46 |
+
"step": 2,
|
| 47 |
+
"step_time": 22.695769495796412
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"clip_ratio/high_max": 0.0007321182056330144,
|
| 51 |
+
"clip_ratio/high_mean": 0.0001830295514082536,
|
| 52 |
+
"clip_ratio/low_mean": 0.0007105598851921968,
|
| 53 |
+
"clip_ratio/low_min": 0.0,
|
| 54 |
+
"clip_ratio/region_mean": 0.000893589443876408,
|
| 55 |
+
"epoch": 0.01875,
|
| 56 |
+
"grad_norm": 3.7183506346963124,
|
| 57 |
+
"kl": 0.00018928675672214013,
|
| 58 |
+
"learning_rate": 6e-07,
|
| 59 |
+
"loss": -0.00017974525690078735,
|
| 60 |
+
"step": 3,
|
| 61 |
+
"step_time": 25.028152221348137
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"clip_ratio/high_max": 0.0033590660023037344,
|
| 65 |
+
"clip_ratio/high_mean": 0.0008570507998229004,
|
| 66 |
+
"clip_ratio/low_mean": 0.00016336305998265743,
|
| 67 |
+
"clip_ratio/low_min": 0.0,
|
| 68 |
+
"clip_ratio/region_mean": 0.0010204138598055579,
|
| 69 |
+
"epoch": 0.025,
|
| 70 |
+
"grad_norm": 4.98871297098248,
|
| 71 |
+
"kl": 0.00026786337184603326,
|
| 72 |
+
"learning_rate": 8e-07,
|
| 73 |
+
"loss": 0.0007005361840128899,
|
| 74 |
+
"step": 4,
|
| 75 |
+
"step_time": 24.52451429190114
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"clip_ratio/high_max": 0.00031984080851543695,
|
| 79 |
+
"clip_ratio/high_mean": 7.996020212885924e-05,
|
| 80 |
+
"clip_ratio/low_mean": 9.461164700041991e-05,
|
| 81 |
+
"clip_ratio/low_min": 0.0,
|
| 82 |
+
"clip_ratio/region_mean": 0.00017457184912927914,
|
| 83 |
+
"epoch": 0.03125,
|
| 84 |
+
"grad_norm": 2.130904855632033,
|
| 85 |
+
"kl": 0.0001346019407719723,
|
| 86 |
+
"learning_rate": 1e-06,
|
| 87 |
+
"loss": -0.0002031590265687555,
|
| 88 |
+
"step": 5,
|
| 89 |
+
"step_time": 25.13552548037842
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"clip_ratio/high_max": 0.0033922659640666097,
|
| 93 |
+
"clip_ratio/high_mean": 0.0010523226010263897,
|
| 94 |
+
"clip_ratio/low_mean": 0.0005023990997869987,
|
| 95 |
+
"clip_ratio/low_min": 0.0,
|
| 96 |
+
"clip_ratio/region_mean": 0.0015547217117273249,
|
| 97 |
+
"epoch": 0.0375,
|
| 98 |
+
"grad_norm": 3.421508152812596,
|
| 99 |
+
"kl": 0.00036621685467252973,
|
| 100 |
+
"learning_rate": 9.998973021172564e-07,
|
| 101 |
+
"loss": -6.486475467681885e-05,
|
| 102 |
+
"step": 6,
|
| 103 |
+
"step_time": 23.172815908212215
|
| 104 |
+
},
|
| 105 |
+
{
|
| 106 |
+
"clip_ratio/high_max": 0.003006511258718092,
|
| 107 |
+
"clip_ratio/high_mean": 0.0009609404241928132,
|
| 108 |
+
"clip_ratio/low_mean": 0.0008985795429907739,
|
| 109 |
+
"clip_ratio/low_min": 6.103515625e-05,
|
| 110 |
+
"clip_ratio/region_mean": 0.0018595199871924706,
|
| 111 |
+
"epoch": 0.04375,
|
| 112 |
+
"grad_norm": 2.1657094604600395,
|
| 113 |
+
"kl": 0.00024832982126099523,
|
| 114 |
+
"learning_rate": 9.99589250656446e-07,
|
| 115 |
+
"loss": -0.00013617239892482758,
|
| 116 |
+
"step": 7,
|
| 117 |
+
"step_time": 24.88068441906944
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"clip_ratio/high_max": 0.003329182043671608,
|
| 121 |
+
"clip_ratio/high_mean": 0.000832295510917902,
|
| 122 |
+
"clip_ratio/low_mean": 0.00035022383963223547,
|
| 123 |
+
"clip_ratio/low_min": 0.0,
|
| 124 |
+
"clip_ratio/region_mean": 0.0011825193505501375,
|
| 125 |
+
"epoch": 0.05,
|
| 126 |
+
"grad_norm": 3.4404457908493216,
|
| 127 |
+
"kl": 0.000299539466141141,
|
| 128 |
+
"learning_rate": 9.990759721625005e-07,
|
| 129 |
+
"loss": -0.00037156135658733547,
|
| 130 |
+
"step": 8,
|
| 131 |
+
"step_time": 24.741218995302916
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"clip_ratio/high_max": 0.0,
|
| 135 |
+
"clip_ratio/high_mean": 0.0,
|
| 136 |
+
"clip_ratio/low_mean": 0.0,
|
| 137 |
+
"clip_ratio/low_min": 0.0,
|
| 138 |
+
"clip_ratio/region_mean": 0.0,
|
| 139 |
+
"completions/clipped_ratio": 0.36328125,
|
| 140 |
+
"completions/max_length": 2048.0,
|
| 141 |
+
"completions/mean_length": 1107.8359375,
|
| 142 |
+
"completions/min_length": 47.0,
|
| 143 |
+
"epoch": 0.05625,
|
| 144 |
+
"frac_reward_zero_std": 0.375,
|
| 145 |
+
"grad_norm": 1.8750759745832997,
|
| 146 |
+
"kl": 0.00028178822321933694,
|
| 147 |
+
"learning_rate": 9.983576774858775e-07,
|
| 148 |
+
"loss": 2.8166105039417744e-06,
|
| 149 |
+
"reward": 0.35546875,
|
| 150 |
+
"reward_std": 0.2964368462562561,
|
| 151 |
+
"rewards/MajorityAccuracy/mean": 0.35546875,
|
| 152 |
+
"rewards/MajorityAccuracy/std": 0.4795927405357361,
|
| 153 |
+
"step": 9,
|
| 154 |
+
"step_time": 188.7468909942545
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"clip_ratio/high_max": 0.00031420186132891104,
|
| 158 |
+
"clip_ratio/high_mean": 7.855046533222776e-05,
|
| 159 |
+
"clip_ratio/low_mean": 0.0005635773359244922,
|
| 160 |
+
"clip_ratio/low_min": 0.0,
|
| 161 |
+
"clip_ratio/region_mean": 0.00064212780125672,
|
| 162 |
+
"epoch": 0.0625,
|
| 163 |
+
"grad_norm": 1.8719195258626709,
|
| 164 |
+
"kl": 0.00025997486409323756,
|
| 165 |
+
"learning_rate": 9.974346616959475e-07,
|
| 166 |
+
"loss": 3.3422162232454866e-05,
|
| 167 |
+
"step": 10,
|
| 168 |
+
"step_time": 25.313482995610684
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"clip_ratio/high_max": 0.006897328770719469,
|
| 172 |
+
"clip_ratio/high_mean": 0.0017243321926798671,
|
| 173 |
+
"clip_ratio/low_mean": 9.527438669465482e-05,
|
| 174 |
+
"clip_ratio/low_min": 0.0,
|
| 175 |
+
"clip_ratio/region_mean": 0.0018196066084783524,
|
| 176 |
+
"epoch": 0.06875,
|
| 177 |
+
"grad_norm": 5.529847657786009,
|
| 178 |
+
"kl": 0.0014605110336560756,
|
| 179 |
+
"learning_rate": 9.963073039597796e-07,
|
| 180 |
+
"loss": 0.00012450278154574335,
|
| 181 |
+
"step": 11,
|
| 182 |
+
"step_time": 24.310499145649374
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"clip_ratio/high_max": 0.0027692848234437406,
|
| 186 |
+
"clip_ratio/high_mean": 0.000757357629481703,
|
| 187 |
+
"clip_ratio/low_mean": 0.0005131799807713833,
|
| 188 |
+
"clip_ratio/low_min": 0.0,
|
| 189 |
+
"clip_ratio/region_mean": 0.0012705375811492559,
|
| 190 |
+
"epoch": 0.075,
|
| 191 |
+
"grad_norm": 3.167401306184357,
|
| 192 |
+
"kl": 0.0004811811668332666,
|
| 193 |
+
"learning_rate": 9.949760673863846e-07,
|
| 194 |
+
"loss": -0.00011729255493264645,
|
| 195 |
+
"step": 12,
|
| 196 |
+
"step_time": 24.787159980740398
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"clip_ratio/high_max": 0.0010050410564872436,
|
| 200 |
+
"clip_ratio/high_mean": 0.0002512602641218109,
|
| 201 |
+
"clip_ratio/low_mean": 0.0003743253728316631,
|
| 202 |
+
"clip_ratio/low_min": 0.0,
|
| 203 |
+
"clip_ratio/region_mean": 0.0006255856351344846,
|
| 204 |
+
"epoch": 0.08125,
|
| 205 |
+
"grad_norm": 2.42369443245261,
|
| 206 |
+
"kl": 0.0003137275925837457,
|
| 207 |
+
"learning_rate": 9.934414988364722e-07,
|
| 208 |
+
"loss": -0.00018403635476715863,
|
| 209 |
+
"step": 13,
|
| 210 |
+
"step_time": 25.317080626729876
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"clip_ratio/high_max": 0.0,
|
| 214 |
+
"clip_ratio/high_mean": 0.0,
|
| 215 |
+
"clip_ratio/low_mean": 0.0,
|
| 216 |
+
"clip_ratio/low_min": 0.0,
|
| 217 |
+
"clip_ratio/region_mean": 0.0,
|
| 218 |
+
"epoch": 0.0875,
|
| 219 |
+
"grad_norm": 0.009363274739211482,
|
| 220 |
+
"kl": 0.0007699248380959034,
|
| 221 |
+
"learning_rate": 9.917042286978063e-07,
|
| 222 |
+
"loss": 7.699379239056725e-06,
|
| 223 |
+
"step": 14,
|
| 224 |
+
"step_time": 25.456473756115884
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"clip_ratio/high_max": 0.005411284801084548,
|
| 228 |
+
"clip_ratio/high_mean": 0.0015470033249584958,
|
| 229 |
+
"clip_ratio/low_mean": 0.0028056614810338942,
|
| 230 |
+
"clip_ratio/low_min": 0.0,
|
| 231 |
+
"clip_ratio/region_mean": 0.004352664764155634,
|
| 232 |
+
"epoch": 0.09375,
|
| 233 |
+
"grad_norm": 4.331338527174864,
|
| 234 |
+
"kl": 0.001316427646088414,
|
| 235 |
+
"learning_rate": 9.897649706262473e-07,
|
| 236 |
+
"loss": -0.0004254467785358429,
|
| 237 |
+
"step": 15,
|
| 238 |
+
"step_time": 24.316636170260608
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"clip_ratio/high_max": 0.002973238573758863,
|
| 242 |
+
"clip_ratio/high_mean": 0.0007433096434397157,
|
| 243 |
+
"clip_ratio/low_mean": 0.002325320128875319,
|
| 244 |
+
"clip_ratio/low_min": 0.0,
|
| 245 |
+
"clip_ratio/region_mean": 0.0030686297541251406,
|
| 246 |
+
"epoch": 0.1,
|
| 247 |
+
"grad_norm": 1.7015851061167304,
|
| 248 |
+
"kl": 0.0008961877902038395,
|
| 249 |
+
"learning_rate": 9.87624521252587e-07,
|
| 250 |
+
"loss": 0.00015197505126707256,
|
| 251 |
+
"step": 16,
|
| 252 |
+
"step_time": 24.83251514006406
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"clip_ratio/high_max": 0.0,
|
| 256 |
+
"clip_ratio/high_mean": 0.0,
|
| 257 |
+
"clip_ratio/low_mean": 0.0,
|
| 258 |
+
"clip_ratio/low_min": 0.0,
|
| 259 |
+
"clip_ratio/region_mean": 0.0,
|
| 260 |
+
"completions/clipped_ratio": 0.22265625,
|
| 261 |
+
"completions/max_length": 2048.0,
|
| 262 |
+
"completions/mean_length": 810.59765625,
|
| 263 |
+
"completions/min_length": 70.0,
|
| 264 |
+
"epoch": 0.10625,
|
| 265 |
+
"frac_reward_zero_std": 0.25,
|
| 266 |
+
"grad_norm": 2.9001268690178845,
|
| 267 |
+
"kl": 0.0014100664266152307,
|
| 268 |
+
"learning_rate": 9.852837598553008e-07,
|
| 269 |
+
"loss": 1.4077360901865177e-05,
|
| 270 |
+
"reward": 0.41015625,
|
| 271 |
+
"reward_std": 0.33786025643348694,
|
| 272 |
+
"rewards/MajorityAccuracy/mean": 0.41015625,
|
| 273 |
+
"rewards/MajorityAccuracy/std": 0.49282535910606384,
|
| 274 |
+
"step": 17,
|
| 275 |
+
"step_time": 188.04850154649466
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"clip_ratio/high_max": 0.0030763528775423765,
|
| 279 |
+
"clip_ratio/high_mean": 0.0007690882193855941,
|
| 280 |
+
"clip_ratio/low_mean": 0.0005547774635488167,
|
| 281 |
+
"clip_ratio/low_min": 0.0,
|
| 282 |
+
"clip_ratio/region_mean": 0.0013238656829344109,
|
| 283 |
+
"epoch": 0.1125,
|
| 284 |
+
"grad_norm": 4.549244699765559,
|
| 285 |
+
"kl": 0.003908877115463838,
|
| 286 |
+
"learning_rate": 9.827436479993468e-07,
|
| 287 |
+
"loss": 0.0004244573356118053,
|
| 288 |
+
"step": 18,
|
| 289 |
+
"step_time": 24.49141151458025
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"clip_ratio/high_max": 0.004936734621878713,
|
| 293 |
+
"clip_ratio/high_mean": 0.0013835155550623313,
|
| 294 |
+
"clip_ratio/low_mean": 0.0,
|
| 295 |
+
"clip_ratio/low_min": 0.0,
|
| 296 |
+
"clip_ratio/region_mean": 0.0013835155550623313,
|
| 297 |
+
"epoch": 0.11875,
|
| 298 |
+
"grad_norm": 5.903586513388641,
|
| 299 |
+
"kl": 0.00327404064591974,
|
| 300 |
+
"learning_rate": 9.800052291411629e-07,
|
| 301 |
+
"loss": 0.00038774037966504693,
|
| 302 |
+
"step": 19,
|
| 303 |
+
"step_time": 26.090277417562902
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"clip_ratio/high_max": 0.011434328858740628,
|
| 307 |
+
"clip_ratio/high_mean": 0.0030632415728177875,
|
| 308 |
+
"clip_ratio/low_mean": 0.0012447981862351298,
|
| 309 |
+
"clip_ratio/low_min": 0.0,
|
| 310 |
+
"clip_ratio/region_mean": 0.004308039788156748,
|
| 311 |
+
"epoch": 0.125,
|
| 312 |
+
"grad_norm": 2.641389410783309,
|
| 313 |
+
"kl": 0.005112184124300256,
|
| 314 |
+
"learning_rate": 9.770696282000244e-07,
|
| 315 |
+
"loss": 9.221286745741963e-06,
|
| 316 |
+
"step": 20,
|
| 317 |
+
"step_time": 24.79021643055603
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"clip_ratio/high_max": 0.01045236708887387,
|
| 321 |
+
"clip_ratio/high_mean": 0.0037997371910023503,
|
| 322 |
+
"clip_ratio/low_mean": 0.004448927218618337,
|
| 323 |
+
"clip_ratio/low_min": 0.00018712575547397137,
|
| 324 |
+
"clip_ratio/region_mean": 0.008248664264101535,
|
| 325 |
+
"epoch": 0.13125,
|
| 326 |
+
"grad_norm": 3.28844334770424,
|
| 327 |
+
"kl": 0.006296496605500579,
|
| 328 |
+
"learning_rate": 9.739380510959364e-07,
|
| 329 |
+
"loss": 0.0007192716002464294,
|
| 330 |
+
"step": 21,
|
| 331 |
+
"step_time": 22.992431721650064
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"clip_ratio/high_max": 0.002427915926091373,
|
| 335 |
+
"clip_ratio/high_mean": 0.0007392821134999394,
|
| 336 |
+
"clip_ratio/low_mean": 0.0010173261325689964,
|
| 337 |
+
"clip_ratio/low_min": 0.0,
|
| 338 |
+
"clip_ratio/region_mean": 0.0017566082460689358,
|
| 339 |
+
"epoch": 0.1375,
|
| 340 |
+
"grad_norm": 1.566472734339869,
|
| 341 |
+
"kl": 0.0020368542027426884,
|
| 342 |
+
"learning_rate": 9.706117842542516e-07,
|
| 343 |
+
"loss": 0.0005367505946196616,
|
| 344 |
+
"step": 22,
|
| 345 |
+
"step_time": 24.65034456597641
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"clip_ratio/high_max": 0.00453313731122762,
|
| 349 |
+
"clip_ratio/high_mean": 0.0018452076328685507,
|
| 350 |
+
"clip_ratio/low_mean": 0.0013239416730357334,
|
| 351 |
+
"clip_ratio/low_min": 0.0,
|
| 352 |
+
"clip_ratio/region_mean": 0.0031691492622485384,
|
| 353 |
+
"epoch": 0.14375,
|
| 354 |
+
"grad_norm": 2.359652589422611,
|
| 355 |
+
"kl": 0.0067315028281882405,
|
| 356 |
+
"learning_rate": 9.670921940772186e-07,
|
| 357 |
+
"loss": 0.0005433768965303898,
|
| 358 |
+
"step": 23,
|
| 359 |
+
"step_time": 24.42431439459324
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"clip_ratio/high_max": 0.0070750270679127425,
|
| 363 |
+
"clip_ratio/high_mean": 0.0018996041180798784,
|
| 364 |
+
"clip_ratio/low_mean": 0.0005249377863947302,
|
| 365 |
+
"clip_ratio/low_min": 0.0,
|
| 366 |
+
"clip_ratio/region_mean": 0.0024245419044746086,
|
| 367 |
+
"epoch": 0.15,
|
| 368 |
+
"grad_norm": 1.4214340389681122,
|
| 369 |
+
"kl": 0.0018519359146011993,
|
| 370 |
+
"learning_rate": 9.633807263826744e-07,
|
| 371 |
+
"loss": 3.607954931794666e-05,
|
| 372 |
+
"step": 24,
|
| 373 |
+
"step_time": 25.456752769183367
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"clip_ratio/high_max": 0.0,
|
| 377 |
+
"clip_ratio/high_mean": 0.0,
|
| 378 |
+
"clip_ratio/low_mean": 0.0,
|
| 379 |
+
"clip_ratio/low_min": 0.0,
|
| 380 |
+
"clip_ratio/region_mean": 0.0,
|
| 381 |
+
"completions/clipped_ratio": 0.2109375,
|
| 382 |
+
"completions/max_length": 2048.0,
|
| 383 |
+
"completions/mean_length": 867.375,
|
| 384 |
+
"completions/min_length": 62.0,
|
| 385 |
+
"epoch": 0.15625,
|
| 386 |
+
"frac_reward_zero_std": 0.28125,
|
| 387 |
+
"grad_norm": 2.639118083929155,
|
| 388 |
+
"kl": 0.0029544061690103263,
|
| 389 |
+
"learning_rate": 9.594789058101153e-07,
|
| 390 |
+
"loss": 2.9526461730711162e-05,
|
| 391 |
+
"reward": 0.46875,
|
| 392 |
+
"reward_std": 0.32917070388793945,
|
| 393 |
+
"rewards/MajorityAccuracy/mean": 0.46875,
|
| 394 |
+
"rewards/MajorityAccuracy/std": 0.5,
|
| 395 |
+
"step": 25,
|
| 396 |
+
"step_time": 187.03084949404
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"clip_ratio/high_max": 0.0032697223650757223,
|
| 400 |
+
"clip_ratio/high_mean": 0.0010943491724901833,
|
| 401 |
+
"clip_ratio/low_mean": 0.0003995299121015705,
|
| 402 |
+
"clip_ratio/low_min": 0.0,
|
| 403 |
+
"clip_ratio/region_mean": 0.0014938790482119657,
|
| 404 |
+
"epoch": 0.1625,
|
| 405 |
+
"grad_norm": 3.8766513135890115,
|
| 406 |
+
"kl": 0.005847665102919564,
|
| 407 |
+
"learning_rate": 9.55388335194388e-07,
|
| 408 |
+
"loss": 0.0002941591665148735,
|
| 409 |
+
"step": 26,
|
| 410 |
+
"step_time": 23.2085574362427
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"clip_ratio/high_max": 0.0024437837855657563,
|
| 414 |
+
"clip_ratio/high_mean": 0.0006109459463914391,
|
| 415 |
+
"clip_ratio/low_mean": 0.0005739888511016034,
|
| 416 |
+
"clip_ratio/low_min": 0.0,
|
| 417 |
+
"clip_ratio/region_mean": 0.0011849348011310212,
|
| 418 |
+
"epoch": 0.16875,
|
| 419 |
+
"grad_norm": 4.414518876246181,
|
| 420 |
+
"kl": 0.003952323633711785,
|
| 421 |
+
"learning_rate": 9.511106949072587e-07,
|
| 422 |
+
"loss": 0.00023994781076908112,
|
| 423 |
+
"step": 27,
|
| 424 |
+
"step_time": 24.607075489126146
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"clip_ratio/high_max": 0.0056166284484788775,
|
| 428 |
+
"clip_ratio/high_mean": 0.0014994314988143742,
|
| 429 |
+
"clip_ratio/low_mean": 0.0012583689676830545,
|
| 430 |
+
"clip_ratio/low_min": 0.0,
|
| 431 |
+
"clip_ratio/region_mean": 0.0027578004373935983,
|
| 432 |
+
"epoch": 0.175,
|
| 433 |
+
"grad_norm": 2.4955510354241977,
|
| 434 |
+
"kl": 0.005823354033054784,
|
| 435 |
+
"learning_rate": 9.466477421671295e-07,
|
| 436 |
+
"loss": -0.00014097904204390943,
|
| 437 |
+
"step": 28,
|
| 438 |
+
"step_time": 23.855516084935516
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"clip_ratio/high_max": 0.0018382353009656072,
|
| 442 |
+
"clip_ratio/high_mean": 0.0004595588252414018,
|
| 443 |
+
"clip_ratio/low_mean": 8.844743751978967e-05,
|
| 444 |
+
"clip_ratio/low_min": 0.0,
|
| 445 |
+
"clip_ratio/region_mean": 0.0005480062627611915,
|
| 446 |
+
"epoch": 0.18125,
|
| 447 |
+
"grad_norm": 1.1774900011121756,
|
| 448 |
+
"kl": 0.00340640963986516,
|
| 449 |
+
"learning_rate": 9.420013103171891e-07,
|
| 450 |
+
"loss": -0.00010137036588275805,
|
| 451 |
+
"step": 29,
|
| 452 |
+
"step_time": 25.249828394968063
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"clip_ratio/high_max": 0.0014764554289286025,
|
| 456 |
+
"clip_ratio/high_mean": 0.0004541478574537905,
|
| 457 |
+
"clip_ratio/low_mean": 0.0004342738029663451,
|
| 458 |
+
"clip_ratio/low_min": 0.0,
|
| 459 |
+
"clip_ratio/region_mean": 0.0008884216658771038,
|
| 460 |
+
"epoch": 0.1875,
|
| 461 |
+
"grad_norm": 1.8072043472782933,
|
| 462 |
+
"kl": 0.0034895020071417093,
|
| 463 |
+
"learning_rate": 9.37173308072291e-07,
|
| 464 |
+
"loss": 0.00020149738702457398,
|
| 465 |
+
"step": 30,
|
| 466 |
+
"step_time": 25.337890869006515
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"clip_ratio/high_max": 0.006205308978678659,
|
| 470 |
+
"clip_ratio/high_mean": 0.0018616801316966303,
|
| 471 |
+
"clip_ratio/low_mean": 0.0005405775591498241,
|
| 472 |
+
"clip_ratio/low_min": 0.0,
|
| 473 |
+
"clip_ratio/region_mean": 0.002402257676294539,
|
| 474 |
+
"epoch": 0.19375,
|
| 475 |
+
"grad_norm": 1.2805238712548745,
|
| 476 |
+
"kl": 0.006296942010521889,
|
| 477 |
+
"learning_rate": 9.321657187348688e-07,
|
| 478 |
+
"loss": 4.792524123331532e-05,
|
| 479 |
+
"step": 31,
|
| 480 |
+
"step_time": 25.128453397657722
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"clip_ratio/high_max": 0.009001921163871884,
|
| 484 |
+
"clip_ratio/high_mean": 0.0040012859826674685,
|
| 485 |
+
"clip_ratio/low_mean": 0.0015654221751901787,
|
| 486 |
+
"clip_ratio/low_min": 0.0,
|
| 487 |
+
"clip_ratio/region_mean": 0.005566708336118609,
|
| 488 |
+
"epoch": 0.2,
|
| 489 |
+
"grad_norm": 3.155484679949791,
|
| 490 |
+
"kl": 0.005464277637656778,
|
| 491 |
+
"learning_rate": 9.269805993802128e-07,
|
| 492 |
+
"loss": -9.257346391677856e-07,
|
| 493 |
+
"step": 32,
|
| 494 |
+
"step_time": 22.808502685278654
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"clip_ratio/high_max": 0.0,
|
| 498 |
+
"clip_ratio/high_mean": 0.0,
|
| 499 |
+
"clip_ratio/low_mean": 0.0,
|
| 500 |
+
"clip_ratio/low_min": 0.0,
|
| 501 |
+
"clip_ratio/region_mean": 0.0,
|
| 502 |
+
"completions/clipped_ratio": 0.1875,
|
| 503 |
+
"completions/max_length": 2048.0,
|
| 504 |
+
"completions/mean_length": 868.03125,
|
| 505 |
+
"completions/min_length": 22.0,
|
| 506 |
+
"epoch": 0.20625,
|
| 507 |
+
"frac_reward_zero_std": 0.21875,
|
| 508 |
+
"grad_norm": 2.93743740724168,
|
| 509 |
+
"kl": 0.0035988678864669055,
|
| 510 |
+
"learning_rate": 9.216200800114411e-07,
|
| 511 |
+
"loss": 3.6024186556460336e-05,
|
| 512 |
+
"reward": 0.49609375,
|
| 513 |
+
"reward_std": 0.37453559041023254,
|
| 514 |
+
"rewards/MajorityAccuracy/mean": 0.49609375,
|
| 515 |
+
"rewards/MajorityAccuracy/std": 0.5009641051292419,
|
| 516 |
+
"step": 33,
|
| 517 |
+
"step_time": 182.29148081317544
|
| 518 |
+
},
|
| 519 |
+
{
|
| 520 |
+
"clip_ratio/high_max": 0.00164631029474549,
|
| 521 |
+
"clip_ratio/high_mean": 0.0005525330489035696,
|
| 522 |
+
"clip_ratio/low_mean": 0.0001341643146588467,
|
| 523 |
+
"clip_ratio/low_min": 0.0,
|
| 524 |
+
"clip_ratio/region_mean": 0.0006866973562864587,
|
| 525 |
+
"epoch": 0.2125,
|
| 526 |
+
"grad_norm": 4.00081718210801,
|
| 527 |
+
"kl": 0.005252757197013125,
|
| 528 |
+
"learning_rate": 9.160863626845119e-07,
|
| 529 |
+
"loss": 7.036514580249786e-05,
|
| 530 |
+
"step": 34,
|
| 531 |
+
"step_time": 24.649917823262513
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"clip_ratio/high_max": 0.0021446907267090864,
|
| 535 |
+
"clip_ratio/high_mean": 0.0006748751202394487,
|
| 536 |
+
"clip_ratio/low_mean": 0.0019597065256675705,
|
| 537 |
+
"clip_ratio/low_min": 0.0,
|
| 538 |
+
"clip_ratio/region_mean": 0.0026345816368120722,
|
| 539 |
+
"epoch": 0.21875,
|
| 540 |
+
"grad_norm": 9.208090035807414,
|
| 541 |
+
"kl": 0.014696626807563007,
|
| 542 |
+
"learning_rate": 9.103817206036382e-07,
|
| 543 |
+
"loss": 0.0008899364620447159,
|
| 544 |
+
"step": 35,
|
| 545 |
+
"step_time": 24.612306271214038
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"clip_ratio/high_max": 0.002291207143571228,
|
| 549 |
+
"clip_ratio/high_mean": 0.000572801785892807,
|
| 550 |
+
"clip_ratio/low_mean": 0.0008117158176901285,
|
| 551 |
+
"clip_ratio/low_min": 0.0,
|
| 552 |
+
"clip_ratio/region_mean": 0.0013845175963069778,
|
| 553 |
+
"epoch": 0.225,
|
| 554 |
+
"grad_norm": 2.1536893172676113,
|
| 555 |
+
"kl": 0.010512356704566628,
|
| 556 |
+
"learning_rate": 9.045084971874737e-07,
|
| 557 |
+
"loss": 0.0002936558739747852,
|
| 558 |
+
"step": 36,
|
| 559 |
+
"step_time": 24.321947866119444
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"clip_ratio/high_max": 0.003072796876949724,
|
| 563 |
+
"clip_ratio/high_mean": 0.0010432749022584176,
|
| 564 |
+
"clip_ratio/low_mean": 0.000468065838504117,
|
| 565 |
+
"clip_ratio/low_min": 0.0,
|
| 566 |
+
"clip_ratio/region_mean": 0.0015113407498574816,
|
| 567 |
+
"epoch": 0.23125,
|
| 568 |
+
"grad_norm": 1.672334320343945,
|
| 569 |
+
"kl": 0.009935911395587027,
|
| 570 |
+
"learning_rate": 8.984691051064574e-07,
|
| 571 |
+
"loss": 0.00023404788225889206,
|
| 572 |
+
"step": 37,
|
| 573 |
+
"step_time": 24.123631950467825
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"clip_ratio/high_max": 0.005803495034342632,
|
| 577 |
+
"clip_ratio/high_mean": 0.0017104373109759763,
|
| 578 |
+
"clip_ratio/low_mean": 0.001716340168059105,
|
| 579 |
+
"clip_ratio/low_min": 0.0,
|
| 580 |
+
"clip_ratio/region_mean": 0.003426777431741357,
|
| 581 |
+
"epoch": 0.2375,
|
| 582 |
+
"grad_norm": 1.6623348197113372,
|
| 583 |
+
"kl": 0.0071962117799557745,
|
| 584 |
+
"learning_rate": 8.922660252917087e-07,
|
| 585 |
+
"loss": 5.079137918073684e-05,
|
| 586 |
+
"step": 38,
|
| 587 |
+
"step_time": 23.770184409804642
|
| 588 |
+
},
|
| 589 |
+
{
|
| 590 |
+
"clip_ratio/high_max": 0.0017068649394786917,
|
| 591 |
+
"clip_ratio/high_mean": 0.0005194253844820196,
|
| 592 |
+
"clip_ratio/low_mean": 0.0005002229117962997,
|
| 593 |
+
"clip_ratio/low_min": 0.0,
|
| 594 |
+
"clip_ratio/region_mean": 0.0010196483053732663,
|
| 595 |
+
"epoch": 0.24375,
|
| 596 |
+
"grad_norm": 2.1122585751528478,
|
| 597 |
+
"kl": 0.004967384273186326,
|
| 598 |
+
"learning_rate": 8.859018059158809e-07,
|
| 599 |
+
"loss": 0.00044735620031133294,
|
| 600 |
+
"step": 39,
|
| 601 |
+
"step_time": 24.82885563839227
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"clip_ratio/high_max": 0.006567380507476628,
|
| 605 |
+
"clip_ratio/high_mean": 0.002394798764726147,
|
| 606 |
+
"clip_ratio/low_mean": 0.002628698402986629,
|
| 607 |
+
"clip_ratio/low_min": 0.0,
|
| 608 |
+
"clip_ratio/region_mean": 0.00502349715679884,
|
| 609 |
+
"epoch": 0.25,
|
| 610 |
+
"grad_norm": 3.5457104037247085,
|
| 611 |
+
"kl": 0.007067607366479933,
|
| 612 |
+
"learning_rate": 8.793790613463954e-07,
|
| 613 |
+
"loss": 0.0006516268476843834,
|
| 614 |
+
"step": 40,
|
| 615 |
+
"step_time": 25.275523873511702
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"clip_ratio/high_max": 0.0,
|
| 619 |
+
"clip_ratio/high_mean": 0.0,
|
| 620 |
+
"clip_ratio/low_mean": 0.0,
|
| 621 |
+
"clip_ratio/low_min": 0.0,
|
| 622 |
+
"clip_ratio/region_mean": 0.0,
|
| 623 |
+
"completions/clipped_ratio": 0.2421875,
|
| 624 |
+
"completions/max_length": 2048.0,
|
| 625 |
+
"completions/mean_length": 923.4609375,
|
| 626 |
+
"completions/min_length": 53.0,
|
| 627 |
+
"epoch": 0.25625,
|
| 628 |
+
"frac_reward_zero_std": 0.1875,
|
| 629 |
+
"grad_norm": 2.5442764232386703,
|
| 630 |
+
"kl": 0.00881995662348345,
|
| 631 |
+
"learning_rate": 8.727004710714798e-07,
|
| 632 |
+
"loss": 8.815968612907454e-05,
|
| 633 |
+
"reward": 0.42578125,
|
| 634 |
+
"reward_std": 0.3799288272857666,
|
| 635 |
+
"rewards/MajorityAccuracy/mean": 0.42578125,
|
| 636 |
+
"rewards/MajorityAccuracy/std": 0.49542948603630066,
|
| 637 |
+
"step": 41,
|
| 638 |
+
"step_time": 187.94554034573957
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"clip_ratio/high_max": 0.008006204865523614,
|
| 642 |
+
"clip_ratio/high_mean": 0.002803285195113858,
|
| 643 |
+
"clip_ratio/low_mean": 0.001350004416963202,
|
| 644 |
+
"clip_ratio/low_min": 0.0,
|
| 645 |
+
"clip_ratio/region_mean": 0.004153289541136473,
|
| 646 |
+
"epoch": 0.2625,
|
| 647 |
+
"grad_norm": 3.539274157669021,
|
| 648 |
+
"kl": 0.007835759955924004,
|
| 649 |
+
"learning_rate": 8.658687785994578e-07,
|
| 650 |
+
"loss": 0.0001817997545003891,
|
| 651 |
+
"step": 42,
|
| 652 |
+
"step_time": 23.699908271897584
|
| 653 |
+
},
|
| 654 |
+
{
|
| 655 |
+
"clip_ratio/high_max": 0.003921301322407089,
|
| 656 |
+
"clip_ratio/high_mean": 0.0010515320755075663,
|
| 657 |
+
"clip_ratio/low_mean": 0.0009233137534465641,
|
| 658 |
+
"clip_ratio/low_min": 0.0,
|
| 659 |
+
"clip_ratio/region_mean": 0.0019748457998503,
|
| 660 |
+
"epoch": 0.26875,
|
| 661 |
+
"grad_norm": 3.0116079527898707,
|
| 662 |
+
"kl": 0.00946639000903815,
|
| 663 |
+
"learning_rate": 8.588867903317394e-07,
|
| 664 |
+
"loss": 0.0004195755464024842,
|
| 665 |
+
"step": 43,
|
| 666 |
+
"step_time": 24.56235553883016
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"clip_ratio/high_max": 0.0007279056444531307,
|
| 670 |
+
"clip_ratio/high_mean": 0.00018197641111328267,
|
| 671 |
+
"clip_ratio/low_mean": 0.0012447108747437596,
|
| 672 |
+
"clip_ratio/low_min": 0.0,
|
| 673 |
+
"clip_ratio/region_mean": 0.0014266872822190635,
|
| 674 |
+
"epoch": 0.275,
|
| 675 |
+
"grad_norm": 2.163774622609758,
|
| 676 |
+
"kl": 0.010521094896830618,
|
| 677 |
+
"learning_rate": 8.51757374409974e-07,
|
| 678 |
+
"loss": -0.00018679485947359353,
|
| 679 |
+
"step": 44,
|
| 680 |
+
"step_time": 24.84438993409276
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"clip_ratio/high_max": 0.0033558645809534937,
|
| 684 |
+
"clip_ratio/high_mean": 0.0010430761103634723,
|
| 685 |
+
"clip_ratio/low_mean": 0.0008953890355769545,
|
| 686 |
+
"clip_ratio/low_min": 0.0,
|
| 687 |
+
"clip_ratio/region_mean": 0.0019384651677682996,
|
| 688 |
+
"epoch": 0.28125,
|
| 689 |
+
"grad_norm": 2.215785615034456,
|
| 690 |
+
"kl": 0.01043605338782072,
|
| 691 |
+
"learning_rate": 8.444834595378433e-07,
|
| 692 |
+
"loss": 0.00026373384753242135,
|
| 693 |
+
"step": 45,
|
| 694 |
+
"step_time": 25.92970998538658
|
| 695 |
+
},
|
| 696 |
+
{
|
| 697 |
+
"clip_ratio/high_max": 0.012492695066612214,
|
| 698 |
+
"clip_ratio/high_mean": 0.003614794826717116,
|
| 699 |
+
"clip_ratio/low_mean": 0.002276377912494354,
|
| 700 |
+
"clip_ratio/low_min": 0.0,
|
| 701 |
+
"clip_ratio/region_mean": 0.005891172710107639,
|
| 702 |
+
"epoch": 0.2875,
|
| 703 |
+
"grad_norm": 3.2339289589771414,
|
| 704 |
+
"kl": 0.014469430432654917,
|
| 705 |
+
"learning_rate": 8.370680337779736e-07,
|
| 706 |
+
"loss": 0.0013254610821604729,
|
| 707 |
+
"step": 46,
|
| 708 |
+
"step_time": 25.083384543657303
|
| 709 |
+
},
|
| 710 |
+
{
|
| 711 |
+
"clip_ratio/high_max": 0.006542220973642543,
|
| 712 |
+
"clip_ratio/high_mean": 0.0021601105108857155,
|
| 713 |
+
"clip_ratio/low_mean": 0.0010578286310192198,
|
| 714 |
+
"clip_ratio/low_min": 0.0,
|
| 715 |
+
"clip_ratio/region_mean": 0.003217939200112596,
|
| 716 |
+
"epoch": 0.29375,
|
| 717 |
+
"grad_norm": 2.438370087609297,
|
| 718 |
+
"kl": 0.00785081519279629,
|
| 719 |
+
"learning_rate": 8.295141433244659e-07,
|
| 720 |
+
"loss": 9.422376751899719e-05,
|
| 721 |
+
"step": 47,
|
| 722 |
+
"step_time": 24.57718371786177
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"clip_ratio/high_max": 0.003133799160423223,
|
| 726 |
+
"clip_ratio/high_mean": 0.0011595951164053986,
|
| 727 |
+
"clip_ratio/low_mean": 0.003114249266218394,
|
| 728 |
+
"clip_ratio/low_min": 0.0,
|
| 729 |
+
"clip_ratio/region_mean": 0.004273844373528846,
|
| 730 |
+
"epoch": 0.3,
|
| 731 |
+
"grad_norm": 2.37994495485011,
|
| 732 |
+
"kl": 0.014018401619978249,
|
| 733 |
+
"learning_rate": 8.218248912515442e-07,
|
| 734 |
+
"loss": 0.0002113285445375368,
|
| 735 |
+
"step": 48,
|
| 736 |
+
"step_time": 25.489909418392926
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"clip_ratio/high_max": 0.0,
|
| 740 |
+
"clip_ratio/high_mean": 0.0,
|
| 741 |
+
"clip_ratio/low_mean": 0.0,
|
| 742 |
+
"clip_ratio/low_min": 0.0,
|
| 743 |
+
"clip_ratio/region_mean": 0.0,
|
| 744 |
+
"completions/clipped_ratio": 0.234375,
|
| 745 |
+
"completions/max_length": 2048.0,
|
| 746 |
+
"completions/mean_length": 908.296875,
|
| 747 |
+
"completions/min_length": 64.0,
|
| 748 |
+
"epoch": 0.30625,
|
| 749 |
+
"frac_reward_zero_std": 0.21875,
|
| 750 |
+
"grad_norm": 9.250382506412379,
|
| 751 |
+
"kl": 0.011843732558190823,
|
| 752 |
+
"learning_rate": 8.140034362388397e-07,
|
| 753 |
+
"loss": 0.00011855550110340118,
|
| 754 |
+
"reward": 0.38671875,
|
| 755 |
+
"reward_std": 0.3553343117237091,
|
| 756 |
+
"rewards/MajorityAccuracy/mean": 0.38671875,
|
| 757 |
+
"rewards/MajorityAccuracy/std": 0.4879522919654846,
|
| 758 |
+
"step": 49,
|
| 759 |
+
"step_time": 187.01735600456595
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"clip_ratio/high_max": 0.0051809483775286935,
|
| 763 |
+
"clip_ratio/high_mean": 0.001341579152722261,
|
| 764 |
+
"clip_ratio/low_mean": 0.0005372997366066556,
|
| 765 |
+
"clip_ratio/low_min": 0.0,
|
| 766 |
+
"clip_ratio/region_mean": 0.001878878860225086,
|
| 767 |
+
"epoch": 0.3125,
|
| 768 |
+
"grad_norm": 4.410010675722964,
|
| 769 |
+
"kl": 0.017517635482363403,
|
| 770 |
+
"learning_rate": 8.060529912738314e-07,
|
| 771 |
+
"loss": 0.000716296024620533,
|
| 772 |
+
"step": 50,
|
| 773 |
+
"step_time": 24.88298148382455
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"clip_ratio/high_max": 0.003026965045137331,
|
| 777 |
+
"clip_ratio/high_mean": 0.0007842500635888427,
|
| 778 |
+
"clip_ratio/low_mean": 0.00045011717884335667,
|
| 779 |
+
"clip_ratio/low_min": 0.0,
|
| 780 |
+
"clip_ratio/region_mean": 0.0012343672351562418,
|
| 781 |
+
"epoch": 0.31875,
|
| 782 |
+
"grad_norm": 1.9662282985760757,
|
| 783 |
+
"kl": 0.010723810293711722,
|
| 784 |
+
"learning_rate": 7.979768223319785e-07,
|
| 785 |
+
"loss": -2.2221356630325317e-06,
|
| 786 |
+
"step": 51,
|
| 787 |
+
"step_time": 24.898542575072497
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"clip_ratio/high_max": 0.0018173230200773105,
|
| 791 |
+
"clip_ratio/high_mean": 0.0004543307550193276,
|
| 792 |
+
"clip_ratio/low_mean": 0.0007725453615421429,
|
| 793 |
+
"clip_ratio/low_min": 0.0,
|
| 794 |
+
"clip_ratio/region_mean": 0.00122687608745764,
|
| 795 |
+
"epoch": 0.325,
|
| 796 |
+
"grad_norm": 3.340761295684519,
|
| 797 |
+
"kl": 0.028362084180116653,
|
| 798 |
+
"learning_rate": 7.897782470350849e-07,
|
| 799 |
+
"loss": 0.00024911481887102127,
|
| 800 |
+
"step": 52,
|
| 801 |
+
"step_time": 24.108296901453286
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"clip_ratio/high_max": 0.002815430168993771,
|
| 805 |
+
"clip_ratio/high_mean": 0.0007038575422484428,
|
| 806 |
+
"clip_ratio/low_mean": 0.0006311663164524361,
|
| 807 |
+
"clip_ratio/low_min": 0.0,
|
| 808 |
+
"clip_ratio/region_mean": 0.0013350238514249213,
|
| 809 |
+
"epoch": 0.33125,
|
| 810 |
+
"grad_norm": 2.5887025036922546,
|
| 811 |
+
"kl": 0.014379832660779357,
|
| 812 |
+
"learning_rate": 7.814606332884488e-07,
|
| 813 |
+
"loss": 0.00029783882200717926,
|
| 814 |
+
"step": 53,
|
| 815 |
+
"step_time": 24.50713274301961
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"clip_ratio/high_max": 0.0017486806609667838,
|
| 819 |
+
"clip_ratio/high_mean": 0.0004912359072477557,
|
| 820 |
+
"clip_ratio/low_mean": 0.0012832818247261457,
|
| 821 |
+
"clip_ratio/low_min": 0.0,
|
| 822 |
+
"clip_ratio/region_mean": 0.0017745177465258166,
|
| 823 |
+
"epoch": 0.3375,
|
| 824 |
+
"grad_norm": 1.8276469537007265,
|
| 825 |
+
"kl": 0.011540349223650992,
|
| 826 |
+
"learning_rate": 7.730273978973552e-07,
|
| 827 |
+
"loss": 0.0004655093071050942,
|
| 828 |
+
"step": 54,
|
| 829 |
+
"step_time": 25.245238327421248
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"clip_ratio/high_max": 0.0034583521774038672,
|
| 833 |
+
"clip_ratio/high_mean": 0.001067510136635974,
|
| 834 |
+
"clip_ratio/low_mean": 0.001687644558842294,
|
| 835 |
+
"clip_ratio/low_min": 0.0,
|
| 836 |
+
"clip_ratio/region_mean": 0.0027551546809263527,
|
| 837 |
+
"epoch": 0.34375,
|
| 838 |
+
"grad_norm": 1.981583925681748,
|
| 839 |
+
"kl": 0.010874011437408626,
|
| 840 |
+
"learning_rate": 7.644820051634812e-07,
|
| 841 |
+
"loss": 0.00043419661233201623,
|
| 842 |
+
"step": 55,
|
| 843 |
+
"step_time": 25.42656168807298
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"clip_ratio/high_max": 0.0007907900289865211,
|
| 847 |
+
"clip_ratio/high_mean": 0.00019769750724663027,
|
| 848 |
+
"clip_ratio/low_mean": 0.00019139484720653854,
|
| 849 |
+
"clip_ratio/low_min": 0.0,
|
| 850 |
+
"clip_ratio/region_mean": 0.0003890923544531688,
|
| 851 |
+
"epoch": 0.35,
|
| 852 |
+
"grad_norm": 0.7837489837145147,
|
| 853 |
+
"kl": 0.010991487186402082,
|
| 854 |
+
"learning_rate": 7.558279654617912e-07,
|
| 855 |
+
"loss": 0.0001362740877084434,
|
| 856 |
+
"step": 56,
|
| 857 |
+
"step_time": 24.66771325469017
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"clip_ratio/high_max": 0.0,
|
| 861 |
+
"clip_ratio/high_mean": 0.0,
|
| 862 |
+
"clip_ratio/low_mean": 0.0,
|
| 863 |
+
"clip_ratio/low_min": 0.0,
|
| 864 |
+
"clip_ratio/region_mean": 0.0,
|
| 865 |
+
"completions/clipped_ratio": 0.23828125,
|
| 866 |
+
"completions/max_length": 2048.0,
|
| 867 |
+
"completions/mean_length": 932.921875,
|
| 868 |
+
"completions/min_length": 82.0,
|
| 869 |
+
"epoch": 0.35625,
|
| 870 |
+
"frac_reward_zero_std": 0.34375,
|
| 871 |
+
"grad_norm": 2.2583108823177933,
|
| 872 |
+
"kl": 0.019514249986968935,
|
| 873 |
+
"learning_rate": 7.470688337985029e-07,
|
| 874 |
+
"loss": 0.00019517933833412826,
|
| 875 |
+
"reward": 0.390625,
|
| 876 |
+
"reward_std": 0.30023884773254395,
|
| 877 |
+
"rewards/MajorityAccuracy/mean": 0.390625,
|
| 878 |
+
"rewards/MajorityAccuracy/std": 0.48884618282318115,
|
| 879 |
+
"step": 57,
|
| 880 |
+
"step_time": 188.06912696454674
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"clip_ratio/high_max": 0.0004370629321783781,
|
| 884 |
+
"clip_ratio/high_mean": 0.00010926573304459453,
|
| 885 |
+
"clip_ratio/low_mean": 7.62939453125e-05,
|
| 886 |
+
"clip_ratio/low_min": 0.0,
|
| 887 |
+
"clip_ratio/region_mean": 0.00018555967835709453,
|
| 888 |
+
"epoch": 0.3625,
|
| 889 |
+
"grad_norm": 5.7226994266443105,
|
| 890 |
+
"kl": 0.009066700702533126,
|
| 891 |
+
"learning_rate": 7.382082083507225e-07,
|
| 892 |
+
"loss": 0.0005095517262816429,
|
| 893 |
+
"step": 58,
|
| 894 |
+
"step_time": 25.29356095287949
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"clip_ratio/high_max": 0.004388058558106422,
|
| 898 |
+
"clip_ratio/high_mean": 0.0011852914831251837,
|
| 899 |
+
"clip_ratio/low_mean": 0.0,
|
| 900 |
+
"clip_ratio/low_min": 0.0,
|
| 901 |
+
"clip_ratio/region_mean": 0.0011852914831251837,
|
| 902 |
+
"epoch": 0.36875,
|
| 903 |
+
"grad_norm": 1.9371388982685351,
|
| 904 |
+
"kl": 0.015010003698989749,
|
| 905 |
+
"learning_rate": 7.292497289883432e-07,
|
| 906 |
+
"loss": 0.00030176318250596523,
|
| 907 |
+
"step": 59,
|
| 908 |
+
"step_time": 23.9305366454646
|
| 909 |
+
},
|
| 910 |
+
{
|
| 911 |
+
"clip_ratio/high_max": 0.0017632921881158836,
|
| 912 |
+
"clip_ratio/high_mean": 0.00047156580149021465,
|
| 913 |
+
"clip_ratio/low_mean": 0.0004072617539350176,
|
| 914 |
+
"clip_ratio/low_min": 0.0,
|
| 915 |
+
"clip_ratio/region_mean": 0.0008788275554252323,
|
| 916 |
+
"epoch": 0.375,
|
| 917 |
+
"grad_norm": 1.2819571126051028,
|
| 918 |
+
"kl": 0.00665118609322235,
|
| 919 |
+
"learning_rate": 7.201970757788171e-07,
|
| 920 |
+
"loss": -0.00011131103383377194,
|
| 921 |
+
"step": 60,
|
| 922 |
+
"step_time": 25.483606967143714
|
| 923 |
+
},
|
| 924 |
+
{
|
| 925 |
+
"clip_ratio/high_max": 0.0016244949074462056,
|
| 926 |
+
"clip_ratio/high_mean": 0.0004061237268615514,
|
| 927 |
+
"clip_ratio/low_mean": 0.0,
|
| 928 |
+
"clip_ratio/low_min": 0.0,
|
| 929 |
+
"clip_ratio/region_mean": 0.0004061237268615514,
|
| 930 |
+
"epoch": 0.38125,
|
| 931 |
+
"grad_norm": 1.3288414107769804,
|
| 932 |
+
"kl": 0.022712247911840677,
|
| 933 |
+
"learning_rate": 7.110539674754159e-07,
|
| 934 |
+
"loss": 0.00015298827202059329,
|
| 935 |
+
"step": 61,
|
| 936 |
+
"step_time": 24.314639389049262
|
| 937 |
+
},
|
| 938 |
+
{
|
| 939 |
+
"clip_ratio/high_max": 0.000841613145894371,
|
| 940 |
+
"clip_ratio/high_mean": 0.00025617965002311394,
|
| 941 |
+
"clip_ratio/low_mean": 0.0005493722856044769,
|
| 942 |
+
"clip_ratio/low_min": 0.0,
|
| 943 |
+
"clip_ratio/region_mean": 0.0008055519356275909,
|
| 944 |
+
"epoch": 0.3875,
|
| 945 |
+
"grad_norm": 1.5642735809941357,
|
| 946 |
+
"kl": 0.01005728857126087,
|
| 947 |
+
"learning_rate": 7.018241599895973e-07,
|
| 948 |
+
"loss": 0.0002405757550150156,
|
| 949 |
+
"step": 62,
|
| 950 |
+
"step_time": 25.182940600439906
|
| 951 |
+
},
|
| 952 |
+
{
|
| 953 |
+
"clip_ratio/high_max": 0.0048485293227713555,
|
| 954 |
+
"clip_ratio/high_mean": 0.0016093442900455557,
|
| 955 |
+
"clip_ratio/low_mean": 0.0011753323415177874,
|
| 956 |
+
"clip_ratio/low_min": 0.0,
|
| 957 |
+
"clip_ratio/region_mean": 0.002784676631563343,
|
| 958 |
+
"epoch": 0.39375,
|
| 959 |
+
"grad_norm": 3.83066845165236,
|
| 960 |
+
"kl": 0.02016834425739944,
|
| 961 |
+
"learning_rate": 6.925114448481088e-07,
|
| 962 |
+
"loss": 0.00044747069478034973,
|
| 963 |
+
"step": 63,
|
| 964 |
+
"step_time": 23.89273976860568
|
| 965 |
+
},
|
| 966 |
+
{
|
| 967 |
+
"clip_ratio/high_max": 0.009516188700217754,
|
| 968 |
+
"clip_ratio/high_mean": 0.0027301256341161206,
|
| 969 |
+
"clip_ratio/low_mean": 0.001683518145000562,
|
| 970 |
+
"clip_ratio/low_min": 0.0,
|
| 971 |
+
"clip_ratio/region_mean": 0.004413643735460937,
|
| 972 |
+
"epoch": 0.4,
|
| 973 |
+
"grad_norm": 2.345097908461299,
|
| 974 |
+
"kl": 0.011708203121088445,
|
| 975 |
+
"learning_rate": 6.831196476354614e-07,
|
| 976 |
+
"loss": 6.761201075278223e-05,
|
| 977 |
+
"step": 64,
|
| 978 |
+
"step_time": 25.246708754450083
|
| 979 |
+
},
|
| 980 |
+
{
|
| 981 |
+
"clip_ratio/high_max": 0.0,
|
| 982 |
+
"clip_ratio/high_mean": 0.0,
|
| 983 |
+
"clip_ratio/low_mean": 0.0,
|
| 984 |
+
"clip_ratio/low_min": 0.0,
|
| 985 |
+
"clip_ratio/region_mean": 0.0,
|
| 986 |
+
"completions/clipped_ratio": 0.203125,
|
| 987 |
+
"completions/max_length": 2048.0,
|
| 988 |
+
"completions/mean_length": 823.62109375,
|
| 989 |
+
"completions/min_length": 57.0,
|
| 990 |
+
"epoch": 0.40625,
|
| 991 |
+
"frac_reward_zero_std": 0.25,
|
| 992 |
+
"grad_norm": 3.3312885204356695,
|
| 993 |
+
"kl": 0.046999497804790735,
|
| 994 |
+
"learning_rate": 6.7365262642241e-07,
|
| 995 |
+
"loss": 0.0004701075959019363,
|
| 996 |
+
"reward": 0.42578125,
|
| 997 |
+
"reward_std": 0.32996463775634766,
|
| 998 |
+
"rewards/MajorityAccuracy/mean": 0.42578125,
|
| 999 |
+
"rewards/MajorityAccuracy/std": 0.49542948603630066,
|
| 1000 |
+
"step": 65,
|
| 1001 |
+
"step_time": 179.24735008971766
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"clip_ratio/high_max": 0.0004918264457955956,
|
| 1005 |
+
"clip_ratio/high_mean": 0.0001229566114488989,
|
| 1006 |
+
"clip_ratio/low_mean": 7.961783558130264e-05,
|
| 1007 |
+
"clip_ratio/low_min": 0.0,
|
| 1008 |
+
"clip_ratio/region_mean": 0.00020257444703020155,
|
| 1009 |
+
"epoch": 0.4125,
|
| 1010 |
+
"grad_norm": 3.0924235435049403,
|
| 1011 |
+
"kl": 0.01622297114226967,
|
| 1012 |
+
"learning_rate": 6.641142701810931e-07,
|
| 1013 |
+
"loss": 0.0004255983221810311,
|
| 1014 |
+
"step": 66,
|
| 1015 |
+
"step_time": 23.03433266095817
|
| 1016 |
+
},
|
| 1017 |
+
{
|
| 1018 |
+
"clip_ratio/high_max": 0.0005675968568539247,
|
| 1019 |
+
"clip_ratio/high_mean": 0.00014189921421348117,
|
| 1020 |
+
"clip_ratio/low_mean": 0.00023956904260558076,
|
| 1021 |
+
"clip_ratio/low_min": 0.0,
|
| 1022 |
+
"clip_ratio/region_mean": 0.00038146825318108313,
|
| 1023 |
+
"epoch": 0.41875,
|
| 1024 |
+
"grad_norm": 1.2093541380819843,
|
| 1025 |
+
"kl": 0.004841322137508541,
|
| 1026 |
+
"learning_rate": 6.545084971874736e-07,
|
| 1027 |
+
"loss": 0.00017882096290122718,
|
| 1028 |
+
"step": 67,
|
| 1029 |
+
"step_time": 24.990839080885053
|
| 1030 |
+
},
|
| 1031 |
+
{
|
| 1032 |
+
"clip_ratio/high_max": 0.0040726817678660154,
|
| 1033 |
+
"clip_ratio/high_mean": 0.0010181704419665039,
|
| 1034 |
+
"clip_ratio/low_mean": 0.0007858698918425944,
|
| 1035 |
+
"clip_ratio/low_min": 0.0,
|
| 1036 |
+
"clip_ratio/region_mean": 0.0018040403811028227,
|
| 1037 |
+
"epoch": 0.425,
|
| 1038 |
+
"grad_norm": 3.3235770105933833,
|
| 1039 |
+
"kl": 0.02504934696480632,
|
| 1040 |
+
"learning_rate": 6.448392534117461e-07,
|
| 1041 |
+
"loss": 0.00035140561521984637,
|
| 1042 |
+
"step": 68,
|
| 1043 |
+
"step_time": 22.304542739409953
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"clip_ratio/high_max": 0.0017348469627904706,
|
| 1047 |
+
"clip_ratio/high_mean": 0.0005006640840292675,
|
| 1048 |
+
"clip_ratio/low_mean": 0.0001936342814587988,
|
| 1049 |
+
"clip_ratio/low_min": 0.0,
|
| 1050 |
+
"clip_ratio/region_mean": 0.0006942983709450345,
|
| 1051 |
+
"epoch": 0.43125,
|
| 1052 |
+
"grad_norm": 1.3445033366191792,
|
| 1053 |
+
"kl": 0.01204421219881624,
|
| 1054 |
+
"learning_rate": 6.351105108973644e-07,
|
| 1055 |
+
"loss": 9.247203706763685e-06,
|
| 1056 |
+
"step": 69,
|
| 1057 |
+
"step_time": 25.187222557142377
|
| 1058 |
+
},
|
| 1059 |
+
{
|
| 1060 |
+
"clip_ratio/high_max": 0.0017655761912465096,
|
| 1061 |
+
"clip_ratio/high_mean": 0.0004413940478116274,
|
| 1062 |
+
"clip_ratio/low_mean": 0.00015376312512671575,
|
| 1063 |
+
"clip_ratio/low_min": 0.0,
|
| 1064 |
+
"clip_ratio/region_mean": 0.0005951571765763219,
|
| 1065 |
+
"epoch": 0.4375,
|
| 1066 |
+
"grad_norm": 1.905036812316653,
|
| 1067 |
+
"kl": 0.008500812691636384,
|
| 1068 |
+
"learning_rate": 6.253262661293602e-07,
|
| 1069 |
+
"loss": 9.516250429442152e-05,
|
| 1070 |
+
"step": 70,
|
| 1071 |
+
"step_time": 25.302472412586212
|
| 1072 |
+
},
|
| 1073 |
+
{
|
| 1074 |
+
"clip_ratio/high_max": 0.009644393838243559,
|
| 1075 |
+
"clip_ratio/high_mean": 0.003316150170576293,
|
| 1076 |
+
"clip_ratio/low_mean": 0.0023280788300326094,
|
| 1077 |
+
"clip_ratio/low_min": 0.0,
|
| 1078 |
+
"clip_ratio/region_mean": 0.00564422897878103,
|
| 1079 |
+
"epoch": 0.44375,
|
| 1080 |
+
"grad_norm": 3.879726354553133,
|
| 1081 |
+
"kl": 0.018917198292911053,
|
| 1082 |
+
"learning_rate": 6.154905383926216e-07,
|
| 1083 |
+
"loss": 0.00012452993541955948,
|
| 1084 |
+
"step": 71,
|
| 1085 |
+
"step_time": 23.1625173012726
|
| 1086 |
+
},
|
| 1087 |
+
{
|
| 1088 |
+
"clip_ratio/high_max": 0.012139661121182144,
|
| 1089 |
+
"clip_ratio/high_mean": 0.003906302415998653,
|
| 1090 |
+
"clip_ratio/low_mean": 0.0017974512556975242,
|
| 1091 |
+
"clip_ratio/low_min": 0.0,
|
| 1092 |
+
"clip_ratio/region_mean": 0.005703753646230325,
|
| 1093 |
+
"epoch": 0.45,
|
| 1094 |
+
"grad_norm": 2.984797482611713,
|
| 1095 |
+
"kl": 0.02129211788997054,
|
| 1096 |
+
"learning_rate": 6.056073681208037e-07,
|
| 1097 |
+
"loss": -8.94470140337944e-05,
|
| 1098 |
+
"step": 72,
|
| 1099 |
+
"step_time": 23.69795270031318
|
| 1100 |
+
},
|
| 1101 |
+
{
|
| 1102 |
+
"clip_ratio/high_max": 0.0,
|
| 1103 |
+
"clip_ratio/high_mean": 0.0,
|
| 1104 |
+
"clip_ratio/low_mean": 0.0,
|
| 1105 |
+
"clip_ratio/low_min": 0.0,
|
| 1106 |
+
"clip_ratio/region_mean": 0.0,
|
| 1107 |
+
"completions/clipped_ratio": 0.2265625,
|
| 1108 |
+
"completions/max_length": 2048.0,
|
| 1109 |
+
"completions/mean_length": 930.39453125,
|
| 1110 |
+
"completions/min_length": 73.0,
|
| 1111 |
+
"epoch": 0.45625,
|
| 1112 |
+
"frac_reward_zero_std": 0.21875,
|
| 1113 |
+
"grad_norm": 2.380479777073025,
|
| 1114 |
+
"kl": 0.01702312717679888,
|
| 1115 |
+
"learning_rate": 5.956808152365532e-07,
|
| 1116 |
+
"loss": 0.0001702345907688141,
|
| 1117 |
+
"reward": 0.296875,
|
| 1118 |
+
"reward_std": 0.3456100821495056,
|
| 1119 |
+
"rewards/MajorityAccuracy/mean": 0.296875,
|
| 1120 |
+
"rewards/MajorityAccuracy/std": 0.45777595043182373,
|
| 1121 |
+
"step": 73,
|
| 1122 |
+
"step_time": 186.38389498461038
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"clip_ratio/high_max": 0.0012056369450874627,
|
| 1126 |
+
"clip_ratio/high_mean": 0.00038133251655381173,
|
| 1127 |
+
"clip_ratio/low_mean": 0.0013879365178581793,
|
| 1128 |
+
"clip_ratio/low_min": 0.0,
|
| 1129 |
+
"clip_ratio/region_mean": 0.001769269016222097,
|
| 1130 |
+
"epoch": 0.4625,
|
| 1131 |
+
"grad_norm": 2.4473061948382373,
|
| 1132 |
+
"kl": 0.021478898357599974,
|
| 1133 |
+
"learning_rate": 5.857149574837268e-07,
|
| 1134 |
+
"loss": 0.0004841946065425873,
|
| 1135 |
+
"step": 74,
|
| 1136 |
+
"step_time": 23.700391384307295
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"clip_ratio/high_max": 0.0019410357053857297,
|
| 1140 |
+
"clip_ratio/high_mean": 0.0006716696298099123,
|
| 1141 |
+
"clip_ratio/low_mean": 0.0001742969725455623,
|
| 1142 |
+
"clip_ratio/low_min": 0.0,
|
| 1143 |
+
"clip_ratio/region_mean": 0.000845966613269411,
|
| 1144 |
+
"epoch": 0.46875,
|
| 1145 |
+
"grad_norm": 1.56652958088449,
|
| 1146 |
+
"kl": 0.011567855137400329,
|
| 1147 |
+
"learning_rate": 5.757138887522883e-07,
|
| 1148 |
+
"loss": 0.0002490454353392124,
|
| 1149 |
+
"step": 75,
|
| 1150 |
+
"step_time": 24.715592140331864
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"clip_ratio/high_max": 0.0017489914025645703,
|
| 1154 |
+
"clip_ratio/high_mean": 0.0004649272086680867,
|
| 1155 |
+
"clip_ratio/low_mean": 0.0010643473651725799,
|
| 1156 |
+
"clip_ratio/low_min": 0.0,
|
| 1157 |
+
"clip_ratio/region_mean": 0.0015292745883925818,
|
| 1158 |
+
"epoch": 0.475,
|
| 1159 |
+
"grad_norm": 4.0806837749376665,
|
| 1160 |
+
"kl": 0.029659639345481992,
|
| 1161 |
+
"learning_rate": 5.656817173965732e-07,
|
| 1162 |
+
"loss": 0.00011851545423269272,
|
| 1163 |
+
"step": 76,
|
| 1164 |
+
"step_time": 23.6789616830647
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"clip_ratio/high_max": 0.0015081901146913879,
|
| 1168 |
+
"clip_ratio/high_mean": 0.00043168040247110184,
|
| 1169 |
+
"clip_ratio/low_mean": 0.00035525966086424887,
|
| 1170 |
+
"clip_ratio/low_min": 0.0,
|
| 1171 |
+
"clip_ratio/region_mean": 0.0007869400487834355,
|
| 1172 |
+
"epoch": 0.48125,
|
| 1173 |
+
"grad_norm": 1.1652938534908752,
|
| 1174 |
+
"kl": 0.009804924367927015,
|
| 1175 |
+
"learning_rate": 5.556225645476118e-07,
|
| 1176 |
+
"loss": 0.00018100741726811975,
|
| 1177 |
+
"step": 77,
|
| 1178 |
+
"step_time": 25.201857208274305
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"clip_ratio/high_max": 0.002189831000578124,
|
| 1182 |
+
"clip_ratio/high_mean": 0.0006067670128686586,
|
| 1183 |
+
"clip_ratio/low_mean": 0.001718199870083481,
|
| 1184 |
+
"clip_ratio/low_min": 0.0,
|
| 1185 |
+
"clip_ratio/region_mean": 0.0023249668593052775,
|
| 1186 |
+
"epoch": 0.4875,
|
| 1187 |
+
"grad_norm": 3.534351043775539,
|
| 1188 |
+
"kl": 0.026298281038179994,
|
| 1189 |
+
"learning_rate": 5.455405624202032e-07,
|
| 1190 |
+
"loss": 0.0006033129175193608,
|
| 1191 |
+
"step": 78,
|
| 1192 |
+
"step_time": 23.404446750879288
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"clip_ratio/high_max": 0.007792175922077149,
|
| 1196 |
+
"clip_ratio/high_mean": 0.0019480439805192873,
|
| 1197 |
+
"clip_ratio/low_mean": 0.0030414680604735622,
|
| 1198 |
+
"clip_ratio/low_min": 0.0,
|
| 1199 |
+
"clip_ratio/region_mean": 0.004989512002794072,
|
| 1200 |
+
"epoch": 0.49375,
|
| 1201 |
+
"grad_norm": 2.9463560427789877,
|
| 1202 |
+
"kl": 0.019358637975528836,
|
| 1203 |
+
"learning_rate": 5.354398526154365e-07,
|
| 1204 |
+
"loss": 0.0007662462303414941,
|
| 1205 |
+
"step": 79,
|
| 1206 |
+
"step_time": 24.908305579330772
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"clip_ratio/high_max": 0.0028167102864244953,
|
| 1210 |
+
"clip_ratio/high_mean": 0.0007041775716061238,
|
| 1211 |
+
"clip_ratio/low_mean": 0.003963658571592532,
|
| 1212 |
+
"clip_ratio/low_min": 0.0,
|
| 1213 |
+
"clip_ratio/region_mean": 0.004667836183216423,
|
| 1214 |
+
"epoch": 0.5,
|
| 1215 |
+
"grad_norm": 2.446566923796506,
|
| 1216 |
+
"kl": 0.02591185679193586,
|
| 1217 |
+
"learning_rate": 5.253245844193564e-07,
|
| 1218 |
+
"loss": 0.0012754308991134167,
|
| 1219 |
+
"step": 80,
|
| 1220 |
+
"step_time": 25.27956625726074
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"clip_ratio/high_max": 0.0,
|
| 1224 |
+
"clip_ratio/high_mean": 0.0,
|
| 1225 |
+
"clip_ratio/low_mean": 0.0,
|
| 1226 |
+
"clip_ratio/low_min": 0.0,
|
| 1227 |
+
"clip_ratio/region_mean": 0.0,
|
| 1228 |
+
"completions/clipped_ratio": 0.25390625,
|
| 1229 |
+
"completions/max_length": 2048.0,
|
| 1230 |
+
"completions/mean_length": 973.64453125,
|
| 1231 |
+
"completions/min_length": 85.0,
|
| 1232 |
+
"epoch": 0.50625,
|
| 1233 |
+
"frac_reward_zero_std": 0.21875,
|
| 1234 |
+
"grad_norm": 2.476654140447937,
|
| 1235 |
+
"kl": 0.01259756286162883,
|
| 1236 |
+
"learning_rate": 5.151989130984714e-07,
|
| 1237 |
+
"loss": 0.00012598838657140732,
|
| 1238 |
+
"reward": 0.31640625,
|
| 1239 |
+
"reward_std": 0.36782193183898926,
|
| 1240 |
+
"rewards/MajorityAccuracy/mean": 0.31640625,
|
| 1241 |
+
"rewards/MajorityAccuracy/std": 0.4659844934940338,
|
| 1242 |
+
"step": 81,
|
| 1243 |
+
"step_time": 191.42544247116894
|
| 1244 |
+
},
|
| 1245 |
+
{
|
| 1246 |
+
"clip_ratio/high_max": 0.0007359016381087713,
|
| 1247 |
+
"clip_ratio/high_mean": 0.00020356788445496932,
|
| 1248 |
+
"clip_ratio/low_mean": 0.00029134953365428373,
|
| 1249 |
+
"clip_ratio/low_min": 0.0,
|
| 1250 |
+
"clip_ratio/region_mean": 0.0004949174217472319,
|
| 1251 |
+
"epoch": 0.5125,
|
| 1252 |
+
"grad_norm": 1.7058450036502937,
|
| 1253 |
+
"kl": 0.011631024710368365,
|
| 1254 |
+
"learning_rate": 5.050669981928055e-07,
|
| 1255 |
+
"loss": 0.00040643912507221103,
|
| 1256 |
+
"step": 82,
|
| 1257 |
+
"step_time": 25.59709671791643
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"clip_ratio/high_max": 0.0015254526952048764,
|
| 1261 |
+
"clip_ratio/high_mean": 0.0003813631738012191,
|
| 1262 |
+
"clip_ratio/low_mean": 4.57763671875e-05,
|
| 1263 |
+
"clip_ratio/low_min": 0.0,
|
| 1264 |
+
"clip_ratio/region_mean": 0.0004271395409887191,
|
| 1265 |
+
"epoch": 0.51875,
|
| 1266 |
+
"grad_norm": 2.930058914292042,
|
| 1267 |
+
"kl": 0.01623969106003642,
|
| 1268 |
+
"learning_rate": 4.949330018071946e-07,
|
| 1269 |
+
"loss": -5.0153237680206075e-05,
|
| 1270 |
+
"step": 83,
|
| 1271 |
+
"step_time": 24.98633300792426
|
| 1272 |
+
},
|
| 1273 |
+
{
|
| 1274 |
+
"clip_ratio/high_max": 0.0031060558394528925,
|
| 1275 |
+
"clip_ratio/high_mean": 0.000871884582011262,
|
| 1276 |
+
"clip_ratio/low_mean": 0.000713711509888526,
|
| 1277 |
+
"clip_ratio/low_min": 0.0,
|
| 1278 |
+
"clip_ratio/region_mean": 0.0015855961246415973,
|
| 1279 |
+
"epoch": 0.525,
|
| 1280 |
+
"grad_norm": 2.5502376512807126,
|
| 1281 |
+
"kl": 0.015290183364413679,
|
| 1282 |
+
"learning_rate": 4.848010869015287e-07,
|
| 1283 |
+
"loss": 0.000446440652012825,
|
| 1284 |
+
"step": 84,
|
| 1285 |
+
"step_time": 24.494921637699008
|
| 1286 |
+
},
|
| 1287 |
+
{
|
| 1288 |
+
"clip_ratio/high_max": 0.0034289442701265216,
|
| 1289 |
+
"clip_ratio/high_mean": 0.0010469080516486429,
|
| 1290 |
+
"clip_ratio/low_mean": 0.001035158449667506,
|
| 1291 |
+
"clip_ratio/low_min": 0.0,
|
| 1292 |
+
"clip_ratio/region_mean": 0.0020820665085921064,
|
| 1293 |
+
"epoch": 0.53125,
|
| 1294 |
+
"grad_norm": 3.2957675052366593,
|
| 1295 |
+
"kl": 0.02100019808858633,
|
| 1296 |
+
"learning_rate": 4.746754155806437e-07,
|
| 1297 |
+
"loss": -0.00013266864698380232,
|
| 1298 |
+
"step": 85,
|
| 1299 |
+
"step_time": 24.849586399272084
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"clip_ratio/high_max": 0.004159582138527185,
|
| 1303 |
+
"clip_ratio/high_mean": 0.0012507890351116657,
|
| 1304 |
+
"clip_ratio/low_mean": 0.0006684439886157634,
|
| 1305 |
+
"clip_ratio/low_min": 0.0,
|
| 1306 |
+
"clip_ratio/region_mean": 0.0019192330073565245,
|
| 1307 |
+
"epoch": 0.5375,
|
| 1308 |
+
"grad_norm": 2.7139091182137176,
|
| 1309 |
+
"kl": 0.017279300140216947,
|
| 1310 |
+
"learning_rate": 4.645601473845635e-07,
|
| 1311 |
+
"loss": 0.0006848186603747308,
|
| 1312 |
+
"step": 86,
|
| 1313 |
+
"step_time": 24.81594144180417
|
| 1314 |
+
},
|
| 1315 |
+
{
|
| 1316 |
+
"clip_ratio/high_max": 0.002681098907487467,
|
| 1317 |
+
"clip_ratio/high_mean": 0.0007314892063732259,
|
| 1318 |
+
"clip_ratio/low_mean": 0.0005771465657744557,
|
| 1319 |
+
"clip_ratio/low_min": 0.0,
|
| 1320 |
+
"clip_ratio/region_mean": 0.0013086357721476816,
|
| 1321 |
+
"epoch": 0.54375,
|
| 1322 |
+
"grad_norm": 1.4686409436880599,
|
| 1323 |
+
"kl": 0.024159747874364257,
|
| 1324 |
+
"learning_rate": 4.544594375797968e-07,
|
| 1325 |
+
"loss": 0.0005016487557440996,
|
| 1326 |
+
"step": 87,
|
| 1327 |
+
"step_time": 24.65445412043482
|
| 1328 |
+
},
|
| 1329 |
+
{
|
| 1330 |
+
"clip_ratio/high_max": 0.00242663589597214,
|
| 1331 |
+
"clip_ratio/high_mean": 0.000606658973993035,
|
| 1332 |
+
"clip_ratio/low_mean": 0.0022397852153517306,
|
| 1333 |
+
"clip_ratio/low_min": 0.0,
|
| 1334 |
+
"clip_ratio/region_mean": 0.0028464442439144477,
|
| 1335 |
+
"epoch": 0.55,
|
| 1336 |
+
"grad_norm": 2.592605248078999,
|
| 1337 |
+
"kl": 0.04230387136340141,
|
| 1338 |
+
"learning_rate": 4.443774354523882e-07,
|
| 1339 |
+
"loss": 0.0004846195224672556,
|
| 1340 |
+
"step": 88,
|
| 1341 |
+
"step_time": 25.043416779953986
|
| 1342 |
+
},
|
| 1343 |
+
{
|
| 1344 |
+
"clip_ratio/high_max": 0.0,
|
| 1345 |
+
"clip_ratio/high_mean": 0.0,
|
| 1346 |
+
"clip_ratio/low_mean": 0.0,
|
| 1347 |
+
"clip_ratio/low_min": 0.0,
|
| 1348 |
+
"clip_ratio/region_mean": 0.0,
|
| 1349 |
+
"completions/clipped_ratio": 0.37890625,
|
| 1350 |
+
"completions/max_length": 2048.0,
|
| 1351 |
+
"completions/mean_length": 1139.76171875,
|
| 1352 |
+
"completions/min_length": 66.0,
|
| 1353 |
+
"epoch": 0.55625,
|
| 1354 |
+
"frac_reward_zero_std": 0.1875,
|
| 1355 |
+
"grad_norm": 4.489858064580488,
|
| 1356 |
+
"kl": 0.03631933173164725,
|
| 1357 |
+
"learning_rate": 4.3431828260342675e-07,
|
| 1358 |
+
"loss": 0.00036364980041980743,
|
| 1359 |
+
"reward": 0.2734375,
|
| 1360 |
+
"reward_std": 0.3452342748641968,
|
| 1361 |
+
"rewards/MajorityAccuracy/mean": 0.2734375,
|
| 1362 |
+
"rewards/MajorityAccuracy/std": 0.44659659266471863,
|
| 1363 |
+
"step": 89,
|
| 1364 |
+
"step_time": 196.17982477601618
|
| 1365 |
+
},
|
| 1366 |
+
{
|
| 1367 |
+
"clip_ratio/high_max": 0.0036923085353919305,
|
| 1368 |
+
"clip_ratio/high_mean": 0.0009230771338479826,
|
| 1369 |
+
"clip_ratio/low_mean": 0.0007631553689861903,
|
| 1370 |
+
"clip_ratio/low_min": 0.0,
|
| 1371 |
+
"clip_ratio/region_mean": 0.0016862324773683213,
|
| 1372 |
+
"epoch": 0.5625,
|
| 1373 |
+
"grad_norm": 3.6702720741729733,
|
| 1374 |
+
"kl": 0.03623533179052174,
|
| 1375 |
+
"learning_rate": 4.242861112477118e-07,
|
| 1376 |
+
"loss": 0.0004895199090242386,
|
| 1377 |
+
"step": 90,
|
| 1378 |
+
"step_time": 24.441273680888116
|
| 1379 |
+
},
|
| 1380 |
+
{
|
| 1381 |
+
"clip_ratio/high_max": 0.00044057227205485106,
|
| 1382 |
+
"clip_ratio/high_mean": 0.00013330837828107178,
|
| 1383 |
+
"clip_ratio/low_mean": 0.0011051878682337701,
|
| 1384 |
+
"clip_ratio/low_min": 0.0,
|
| 1385 |
+
"clip_ratio/region_mean": 0.0012384962537907995,
|
| 1386 |
+
"epoch": 0.56875,
|
| 1387 |
+
"grad_norm": 2.1793644702138995,
|
| 1388 |
+
"kl": 0.010921389097347856,
|
| 1389 |
+
"learning_rate": 4.1428504251627325e-07,
|
| 1390 |
+
"loss": 0.0006457867566496134,
|
| 1391 |
+
"step": 91,
|
| 1392 |
+
"step_time": 25.30494431266561
|
| 1393 |
+
},
|
| 1394 |
+
{
|
| 1395 |
+
"clip_ratio/high_max": 0.0035454335375106893,
|
| 1396 |
+
"clip_ratio/high_mean": 0.0009643899647926446,
|
| 1397 |
+
"clip_ratio/low_mean": 0.00029465206898748875,
|
| 1398 |
+
"clip_ratio/low_min": 0.0,
|
| 1399 |
+
"clip_ratio/region_mean": 0.0012590420374181122,
|
| 1400 |
+
"epoch": 0.575,
|
| 1401 |
+
"grad_norm": 3.1710543924745522,
|
| 1402 |
+
"kl": 0.013836979516781867,
|
| 1403 |
+
"learning_rate": 4.0431918476344685e-07,
|
| 1404 |
+
"loss": 0.00018812026246450841,
|
| 1405 |
+
"step": 92,
|
| 1406 |
+
"step_time": 25.553095526061952
|
| 1407 |
+
},
|
| 1408 |
+
{
|
| 1409 |
+
"clip_ratio/high_max": 0.000989555730484426,
|
| 1410 |
+
"clip_ratio/high_mean": 0.0002473889326211065,
|
| 1411 |
+
"clip_ratio/low_mean": 0.0009219378771376796,
|
| 1412 |
+
"clip_ratio/low_min": 0.0,
|
| 1413 |
+
"clip_ratio/region_mean": 0.0011693268097587861,
|
| 1414 |
+
"epoch": 0.58125,
|
| 1415 |
+
"grad_norm": 1.9597912224545626,
|
| 1416 |
+
"kl": 0.013338020769879222,
|
| 1417 |
+
"learning_rate": 3.943926318791963e-07,
|
| 1418 |
+
"loss": -7.024944352451712e-05,
|
| 1419 |
+
"step": 93,
|
| 1420 |
+
"step_time": 25.42898074630648
|
| 1421 |
+
},
|
| 1422 |
+
{
|
| 1423 |
+
"clip_ratio/high_max": 0.0029973266064189374,
|
| 1424 |
+
"clip_ratio/high_mean": 0.0007668976977583952,
|
| 1425 |
+
"clip_ratio/low_mean": 0.0003723509071278386,
|
| 1426 |
+
"clip_ratio/low_min": 0.0,
|
| 1427 |
+
"clip_ratio/region_mean": 0.0011392486121621914,
|
| 1428 |
+
"epoch": 0.5875,
|
| 1429 |
+
"grad_norm": 2.0787223840068996,
|
| 1430 |
+
"kl": 0.009607368556316942,
|
| 1431 |
+
"learning_rate": 3.845094616073783e-07,
|
| 1432 |
+
"loss": 7.660418486921117e-05,
|
| 1433 |
+
"step": 94,
|
| 1434 |
+
"step_time": 25.229177705477923
|
| 1435 |
+
},
|
| 1436 |
+
{
|
| 1437 |
+
"clip_ratio/high_max": 0.01047526887850836,
|
| 1438 |
+
"clip_ratio/high_mean": 0.0032666080369381234,
|
| 1439 |
+
"clip_ratio/low_mean": 0.0029908504657214507,
|
| 1440 |
+
"clip_ratio/low_min": 0.0,
|
| 1441 |
+
"clip_ratio/region_mean": 0.006257458444451913,
|
| 1442 |
+
"epoch": 0.59375,
|
| 1443 |
+
"grad_norm": 4.374979672981791,
|
| 1444 |
+
"kl": 0.052292600739747286,
|
| 1445 |
+
"learning_rate": 3.7467373387063964e-07,
|
| 1446 |
+
"loss": 0.0013809623196721077,
|
| 1447 |
+
"step": 95,
|
| 1448 |
+
"step_time": 23.811121723148972
|
| 1449 |
+
},
|
| 1450 |
+
{
|
| 1451 |
+
"clip_ratio/high_max": 0.0021497112029464915,
|
| 1452 |
+
"clip_ratio/high_mean": 0.0005374278007366229,
|
| 1453 |
+
"clip_ratio/low_mean": 0.0005313278015819378,
|
| 1454 |
+
"clip_ratio/low_min": 0.0,
|
| 1455 |
+
"clip_ratio/region_mean": 0.001068755598680582,
|
| 1456 |
+
"epoch": 0.6,
|
| 1457 |
+
"grad_norm": 3.5433266169150297,
|
| 1458 |
+
"kl": 0.007281638216227293,
|
| 1459 |
+
"learning_rate": 3.648894891026358e-07,
|
| 1460 |
+
"loss": 4.350590461399406e-05,
|
| 1461 |
+
"step": 96,
|
| 1462 |
+
"step_time": 25.30344247072935
|
| 1463 |
+
},
|
| 1464 |
+
{
|
| 1465 |
+
"clip_ratio/high_max": 0.0,
|
| 1466 |
+
"clip_ratio/high_mean": 0.0,
|
| 1467 |
+
"clip_ratio/low_mean": 0.0,
|
| 1468 |
+
"clip_ratio/low_min": 0.0,
|
| 1469 |
+
"clip_ratio/region_mean": 0.0,
|
| 1470 |
+
"completions/clipped_ratio": 0.2265625,
|
| 1471 |
+
"completions/max_length": 2048.0,
|
| 1472 |
+
"completions/mean_length": 849.515625,
|
| 1473 |
+
"completions/min_length": 72.0,
|
| 1474 |
+
"epoch": 0.60625,
|
| 1475 |
+
"frac_reward_zero_std": 0.15625,
|
| 1476 |
+
"grad_norm": 1.6268355692919343,
|
| 1477 |
+
"kl": 0.015171386185102165,
|
| 1478 |
+
"learning_rate": 3.5516074658825397e-07,
|
| 1479 |
+
"loss": 0.0001517086784588173,
|
| 1480 |
+
"reward": 0.40625,
|
| 1481 |
+
"reward_std": 0.3745419383049011,
|
| 1482 |
+
"rewards/MajorityAccuracy/mean": 0.40625,
|
| 1483 |
+
"rewards/MajorityAccuracy/std": 0.49209436774253845,
|
| 1484 |
+
"step": 97,
|
| 1485 |
+
"step_time": 188.97810591384768
|
| 1486 |
+
},
|
| 1487 |
+
{
|
| 1488 |
+
"clip_ratio/high_max": 0.0010206142615061253,
|
| 1489 |
+
"clip_ratio/high_mean": 0.00025515356537653133,
|
| 1490 |
+
"clip_ratio/low_mean": 0.000977925374172628,
|
| 1491 |
+
"clip_ratio/low_min": 0.0,
|
| 1492 |
+
"clip_ratio/region_mean": 0.0012330789395491593,
|
| 1493 |
+
"epoch": 0.6125,
|
| 1494 |
+
"grad_norm": 4.136171257309263,
|
| 1495 |
+
"kl": 0.03110526269301772,
|
| 1496 |
+
"learning_rate": 3.454915028125263e-07,
|
| 1497 |
+
"loss": 0.000384659506380558,
|
| 1498 |
+
"step": 98,
|
| 1499 |
+
"step_time": 23.308477770537138
|
| 1500 |
+
},
|
| 1501 |
+
{
|
| 1502 |
+
"clip_ratio/high_max": 0.0034125668462365866,
|
| 1503 |
+
"clip_ratio/high_mean": 0.0010176706564379856,
|
| 1504 |
+
"clip_ratio/low_mean": 0.00015792158956173807,
|
| 1505 |
+
"clip_ratio/low_min": 0.0,
|
| 1506 |
+
"clip_ratio/region_mean": 0.001175592238723766,
|
| 1507 |
+
"epoch": 0.61875,
|
| 1508 |
+
"grad_norm": 2.017607044206946,
|
| 1509 |
+
"kl": 0.021671083755791187,
|
| 1510 |
+
"learning_rate": 3.3588572981890684e-07,
|
| 1511 |
+
"loss": 0.0004600776301231235,
|
| 1512 |
+
"step": 99,
|
| 1513 |
+
"step_time": 24.785724753979594
|
| 1514 |
+
},
|
| 1515 |
+
{
|
| 1516 |
+
"clip_ratio/high_max": 0.0014582278672605753,
|
| 1517 |
+
"clip_ratio/high_mean": 0.0004076010372955352,
|
| 1518 |
+
"clip_ratio/low_mean": 0.0018649821868166327,
|
| 1519 |
+
"clip_ratio/low_min": 6.103515625e-05,
|
| 1520 |
+
"clip_ratio/region_mean": 0.002272583224112168,
|
| 1521 |
+
"epoch": 0.625,
|
| 1522 |
+
"grad_norm": 3.719426428035184,
|
| 1523 |
+
"kl": 0.024606818333268166,
|
| 1524 |
+
"learning_rate": 3.263473735775899e-07,
|
| 1525 |
+
"loss": 0.000506189651787281,
|
| 1526 |
+
"step": 100,
|
| 1527 |
+
"step_time": 25.290831822901964
|
| 1528 |
+
},
|
| 1529 |
+
{
|
| 1530 |
+
"clip_ratio/high_max": 0.005143302551005036,
|
| 1531 |
+
"clip_ratio/high_mean": 0.0018211897986475378,
|
| 1532 |
+
"clip_ratio/low_mean": 0.0004312266537453979,
|
| 1533 |
+
"clip_ratio/low_min": 0.0,
|
| 1534 |
+
"clip_ratio/region_mean": 0.0022524164523929358,
|
| 1535 |
+
"epoch": 0.63125,
|
| 1536 |
+
"grad_norm": 3.6404990725091815,
|
| 1537 |
+
"kl": 0.01927510998211801,
|
| 1538 |
+
"learning_rate": 3.1688035236453865e-07,
|
| 1539 |
+
"loss": -0.00036476433160714805,
|
| 1540 |
+
"step": 101,
|
| 1541 |
+
"step_time": 25.03696014592424
|
| 1542 |
+
},
|
| 1543 |
+
{
|
| 1544 |
+
"clip_ratio/high_max": 0.002686045292648487,
|
| 1545 |
+
"clip_ratio/high_mean": 0.0008577359003538731,
|
| 1546 |
+
"clip_ratio/low_mean": 0.0005826876804349013,
|
| 1547 |
+
"clip_ratio/low_min": 0.0,
|
| 1548 |
+
"clip_ratio/region_mean": 0.001440423628082499,
|
| 1549 |
+
"epoch": 0.6375,
|
| 1550 |
+
"grad_norm": 1.8129779702501607,
|
| 1551 |
+
"kl": 0.013093323912471533,
|
| 1552 |
+
"learning_rate": 3.0748855515189096e-07,
|
| 1553 |
+
"loss": 6.10528513789177e-05,
|
| 1554 |
+
"step": 102,
|
| 1555 |
+
"step_time": 24.54435709444806
|
| 1556 |
+
},
|
| 1557 |
+
{
|
| 1558 |
+
"clip_ratio/high_max": 0.00701545673655346,
|
| 1559 |
+
"clip_ratio/high_mean": 0.002034133256529458,
|
| 1560 |
+
"clip_ratio/low_mean": 0.0017133841465692967,
|
| 1561 |
+
"clip_ratio/low_min": 0.0,
|
| 1562 |
+
"clip_ratio/region_mean": 0.0037475174758583307,
|
| 1563 |
+
"epoch": 0.64375,
|
| 1564 |
+
"grad_norm": 2.7942219697311192,
|
| 1565 |
+
"kl": 0.014251778367906809,
|
| 1566 |
+
"learning_rate": 2.981758400104028e-07,
|
| 1567 |
+
"loss": 0.0010412383126094937,
|
| 1568 |
+
"step": 103,
|
| 1569 |
+
"step_time": 25.128667521756142
|
| 1570 |
+
},
|
| 1571 |
+
{
|
| 1572 |
+
"clip_ratio/high_max": 0.005370853847125545,
|
| 1573 |
+
"clip_ratio/high_mean": 0.0019386288622627035,
|
| 1574 |
+
"clip_ratio/low_mean": 0.0010481294666533358,
|
| 1575 |
+
"clip_ratio/low_min": 0.0,
|
| 1576 |
+
"clip_ratio/region_mean": 0.0029867583652958274,
|
| 1577 |
+
"epoch": 0.65,
|
| 1578 |
+
"grad_norm": 3.0907025815216227,
|
| 1579 |
+
"kl": 0.016332319122739136,
|
| 1580 |
+
"learning_rate": 2.8894603252458403e-07,
|
| 1581 |
+
"loss": 6.217323243618011e-05,
|
| 1582 |
+
"step": 104,
|
| 1583 |
+
"step_time": 23.318413876462728
|
| 1584 |
+
},
|
| 1585 |
+
{
|
| 1586 |
+
"clip_ratio/high_max": 0.0,
|
| 1587 |
+
"clip_ratio/high_mean": 0.0,
|
| 1588 |
+
"clip_ratio/low_mean": 0.0,
|
| 1589 |
+
"clip_ratio/low_min": 0.0,
|
| 1590 |
+
"clip_ratio/region_mean": 0.0,
|
| 1591 |
+
"completions/clipped_ratio": 0.31640625,
|
| 1592 |
+
"completions/max_length": 2048.0,
|
| 1593 |
+
"completions/mean_length": 1120.609375,
|
| 1594 |
+
"completions/min_length": 76.0,
|
| 1595 |
+
"epoch": 0.65625,
|
| 1596 |
+
"frac_reward_zero_std": 0.5,
|
| 1597 |
+
"grad_norm": 2.0464087102585364,
|
| 1598 |
+
"kl": 0.01416267710737884,
|
| 1599 |
+
"learning_rate": 2.7980292422118277e-07,
|
| 1600 |
+
"loss": 0.00014167718472890556,
|
| 1601 |
+
"reward": 0.27734375,
|
| 1602 |
+
"reward_std": 0.21489110589027405,
|
| 1603 |
+
"rewards/MajorityAccuracy/mean": 0.27734375,
|
| 1604 |
+
"rewards/MajorityAccuracy/std": 0.4485645592212677,
|
| 1605 |
+
"step": 105,
|
| 1606 |
+
"step_time": 194.70132768852636
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"clip_ratio/high_max": 0.00022563176753465086,
|
| 1610 |
+
"clip_ratio/high_mean": 5.6407941883662716e-05,
|
| 1611 |
+
"clip_ratio/low_mean": 0.0,
|
| 1612 |
+
"clip_ratio/low_min": 0.0,
|
| 1613 |
+
"clip_ratio/region_mean": 5.6407941883662716e-05,
|
| 1614 |
+
"epoch": 0.6625,
|
| 1615 |
+
"grad_norm": 1.0674959904943593,
|
| 1616 |
+
"kl": 0.010573105071671307,
|
| 1617 |
+
"learning_rate": 2.70750271011657e-07,
|
| 1618 |
+
"loss": 0.00022347804042510688,
|
| 1619 |
+
"step": 106,
|
| 1620 |
+
"step_time": 23.597188628278673
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"clip_ratio/high_max": 0.003119844652246684,
|
| 1624 |
+
"clip_ratio/high_mean": 0.000779961163061671,
|
| 1625 |
+
"clip_ratio/low_mean": 0.0008782550576142967,
|
| 1626 |
+
"clip_ratio/low_min": 0.0,
|
| 1627 |
+
"clip_ratio/region_mean": 0.0016582162061240524,
|
| 1628 |
+
"epoch": 0.66875,
|
| 1629 |
+
"grad_norm": 4.496728847799243,
|
| 1630 |
+
"kl": 0.016367516247555614,
|
| 1631 |
+
"learning_rate": 2.6179179164927754e-07,
|
| 1632 |
+
"loss": 0.0001059561618603766,
|
| 1633 |
+
"step": 107,
|
| 1634 |
+
"step_time": 25.071860424708575
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"clip_ratio/high_max": 0.0016972935118246824,
|
| 1638 |
+
"clip_ratio/high_mean": 0.0004243233779561706,
|
| 1639 |
+
"clip_ratio/low_mean": 0.0011119607297587208,
|
| 1640 |
+
"clip_ratio/low_min": 0.0,
|
| 1641 |
+
"clip_ratio/region_mean": 0.0015362841040769126,
|
| 1642 |
+
"epoch": 0.675,
|
| 1643 |
+
"grad_norm": 2.472201468432442,
|
| 1644 |
+
"kl": 0.016185110085643828,
|
| 1645 |
+
"learning_rate": 2.529311662014972e-07,
|
| 1646 |
+
"loss": 0.0002276180894114077,
|
| 1647 |
+
"step": 108,
|
| 1648 |
+
"step_time": 25.17788119846955
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"clip_ratio/high_max": 0.0015682493831263855,
|
| 1652 |
+
"clip_ratio/high_mean": 0.0003920623457815964,
|
| 1653 |
+
"clip_ratio/low_mean": 1.52587890625e-05,
|
| 1654 |
+
"clip_ratio/low_min": 0.0,
|
| 1655 |
+
"clip_ratio/region_mean": 0.0004073211348440964,
|
| 1656 |
+
"epoch": 0.68125,
|
| 1657 |
+
"grad_norm": 2.4324401849920534,
|
| 1658 |
+
"kl": 0.03003188781440258,
|
| 1659 |
+
"learning_rate": 2.441720345382089e-07,
|
| 1660 |
+
"loss": 0.0002726036182139069,
|
| 1661 |
+
"step": 109,
|
| 1662 |
+
"step_time": 24.96178495604545
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"clip_ratio/high_max": 0.0,
|
| 1666 |
+
"clip_ratio/high_mean": 0.0,
|
| 1667 |
+
"clip_ratio/low_mean": 0.00046971591655164957,
|
| 1668 |
+
"clip_ratio/low_min": 0.0,
|
| 1669 |
+
"clip_ratio/region_mean": 0.00046971591655164957,
|
| 1670 |
+
"epoch": 0.6875,
|
| 1671 |
+
"grad_norm": 3.464969024362107,
|
| 1672 |
+
"kl": 0.03480562660843134,
|
| 1673 |
+
"learning_rate": 2.355179948365189e-07,
|
| 1674 |
+
"loss": 0.0006420800928026438,
|
| 1675 |
+
"step": 110,
|
| 1676 |
+
"step_time": 24.8954763491638
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"clip_ratio/high_max": 0.0005875440547242761,
|
| 1680 |
+
"clip_ratio/high_mean": 0.00014688601368106902,
|
| 1681 |
+
"clip_ratio/low_mean": 0.0002185720168199623,
|
| 1682 |
+
"clip_ratio/low_min": 0.0,
|
| 1683 |
+
"clip_ratio/region_mean": 0.0003654580305010313,
|
| 1684 |
+
"epoch": 0.69375,
|
| 1685 |
+
"grad_norm": 1.2144741425852867,
|
| 1686 |
+
"kl": 0.01694681867957115,
|
| 1687 |
+
"learning_rate": 2.26972602102645e-07,
|
| 1688 |
+
"loss": 0.000307480018818751,
|
| 1689 |
+
"step": 111,
|
| 1690 |
+
"step_time": 25.106293017975986
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"clip_ratio/high_max": 0.0021410137705970556,
|
| 1694 |
+
"clip_ratio/high_mean": 0.0005352534426492639,
|
| 1695 |
+
"clip_ratio/low_mean": 0.0013017337041674182,
|
| 1696 |
+
"clip_ratio/low_min": 0.0,
|
| 1697 |
+
"clip_ratio/region_mean": 0.0018369871540926397,
|
| 1698 |
+
"epoch": 0.7,
|
| 1699 |
+
"grad_norm": 1.8851320953054007,
|
| 1700 |
+
"kl": 0.019599810824729502,
|
| 1701 |
+
"learning_rate": 2.1853936671155127e-07,
|
| 1702 |
+
"loss": 0.00045443602721206844,
|
| 1703 |
+
"step": 112,
|
| 1704 |
+
"step_time": 25.279981053434312
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"clip_ratio/high_max": 0.0,
|
| 1708 |
+
"clip_ratio/high_mean": 0.0,
|
| 1709 |
+
"clip_ratio/low_mean": 0.0,
|
| 1710 |
+
"clip_ratio/low_min": 0.0,
|
| 1711 |
+
"clip_ratio/region_mean": 0.0,
|
| 1712 |
+
"completions/clipped_ratio": 0.32421875,
|
| 1713 |
+
"completions/max_length": 2048.0,
|
| 1714 |
+
"completions/mean_length": 1125.60546875,
|
| 1715 |
+
"completions/min_length": 96.0,
|
| 1716 |
+
"epoch": 0.70625,
|
| 1717 |
+
"frac_reward_zero_std": 0.28125,
|
| 1718 |
+
"grad_norm": 3.324088501872557,
|
| 1719 |
+
"kl": 0.021935274940915406,
|
| 1720 |
+
"learning_rate": 2.1022175296491512e-07,
|
| 1721 |
+
"loss": 0.0002193031832575798,
|
| 1722 |
+
"reward": 0.34375,
|
| 1723 |
+
"reward_std": 0.3235129714012146,
|
| 1724 |
+
"rewards/MajorityAccuracy/mean": 0.34375,
|
| 1725 |
+
"rewards/MajorityAccuracy/std": 0.47588926553726196,
|
| 1726 |
+
"step": 113,
|
| 1727 |
+
"step_time": 188.92285200301558
|
| 1728 |
+
},
|
| 1729 |
+
{
|
| 1730 |
+
"clip_ratio/high_max": 0.0005867388972546905,
|
| 1731 |
+
"clip_ratio/high_mean": 0.00014668472431367263,
|
| 1732 |
+
"clip_ratio/low_mean": 0.00014661161549156532,
|
| 1733 |
+
"clip_ratio/low_min": 0.0,
|
| 1734 |
+
"clip_ratio/region_mean": 0.00029329633980523795,
|
| 1735 |
+
"epoch": 0.7125,
|
| 1736 |
+
"grad_norm": 1.1289341579431351,
|
| 1737 |
+
"kl": 0.012859382666647434,
|
| 1738 |
+
"learning_rate": 2.0202317766802152e-07,
|
| 1739 |
+
"loss": 0.0002838941290974617,
|
| 1740 |
+
"step": 114,
|
| 1741 |
+
"step_time": 25.45572679908946
|
| 1742 |
+
},
|
| 1743 |
+
{
|
| 1744 |
+
"clip_ratio/high_max": 0.0015790283068781719,
|
| 1745 |
+
"clip_ratio/high_mean": 0.0005302938043314498,
|
| 1746 |
+
"clip_ratio/low_mean": 0.0002336966026632581,
|
| 1747 |
+
"clip_ratio/low_min": 0.0,
|
| 1748 |
+
"clip_ratio/region_mean": 0.0007639903997187503,
|
| 1749 |
+
"epoch": 0.71875,
|
| 1750 |
+
"grad_norm": 1.9414565323929176,
|
| 1751 |
+
"kl": 0.018882243428379297,
|
| 1752 |
+
"learning_rate": 1.9394700872616853e-07,
|
| 1753 |
+
"loss": 0.0003244606778025627,
|
| 1754 |
+
"step": 115,
|
| 1755 |
+
"step_time": 24.96849175542593
|
| 1756 |
+
},
|
| 1757 |
+
{
|
| 1758 |
+
"clip_ratio/high_max": 0.00048135107499547303,
|
| 1759 |
+
"clip_ratio/high_mean": 0.0001624536671442911,
|
| 1760 |
+
"clip_ratio/low_mean": 0.00027533352113096043,
|
| 1761 |
+
"clip_ratio/low_min": 0.0,
|
| 1762 |
+
"clip_ratio/region_mean": 0.00043778718827525154,
|
| 1763 |
+
"epoch": 0.725,
|
| 1764 |
+
"grad_norm": 1.7729704516205298,
|
| 1765 |
+
"kl": 0.015695497975684702,
|
| 1766 |
+
"learning_rate": 1.8599656376116024e-07,
|
| 1767 |
+
"loss": 0.00017167555051855743,
|
| 1768 |
+
"step": 116,
|
| 1769 |
+
"step_time": 25.350853241980076
|
| 1770 |
+
},
|
| 1771 |
+
{
|
| 1772 |
+
"clip_ratio/high_max": 0.0004672897339332849,
|
| 1773 |
+
"clip_ratio/high_mean": 0.00011682243348332122,
|
| 1774 |
+
"clip_ratio/low_mean": 0.00022242100385483354,
|
| 1775 |
+
"clip_ratio/low_min": 0.0,
|
| 1776 |
+
"clip_ratio/region_mean": 0.00033924343733815476,
|
| 1777 |
+
"epoch": 0.73125,
|
| 1778 |
+
"grad_norm": 1.2099690288804807,
|
| 1779 |
+
"kl": 0.013668992556631565,
|
| 1780 |
+
"learning_rate": 1.7817510874845582e-07,
|
| 1781 |
+
"loss": 9.471029625274241e-05,
|
| 1782 |
+
"step": 117,
|
| 1783 |
+
"step_time": 25.610421131830662
|
| 1784 |
+
},
|
| 1785 |
+
{
|
| 1786 |
+
"clip_ratio/high_max": 0.0013080856297165155,
|
| 1787 |
+
"clip_ratio/high_mean": 0.0004271816578693688,
|
| 1788 |
+
"clip_ratio/low_mean": 0.00026685596094466746,
|
| 1789 |
+
"clip_ratio/low_min": 0.0,
|
| 1790 |
+
"clip_ratio/region_mean": 0.0006940376188140363,
|
| 1791 |
+
"epoch": 0.7375,
|
| 1792 |
+
"grad_norm": 3.9208308947727293,
|
| 1793 |
+
"kl": 0.012728268280625343,
|
| 1794 |
+
"learning_rate": 1.7048585667553412e-07,
|
| 1795 |
+
"loss": 0.00031210752786137164,
|
| 1796 |
+
"step": 118,
|
| 1797 |
+
"step_time": 25.214739948976785
|
| 1798 |
+
},
|
| 1799 |
+
{
|
| 1800 |
+
"clip_ratio/high_max": 0.0018060915826936252,
|
| 1801 |
+
"clip_ratio/high_mean": 0.00047789420386834536,
|
| 1802 |
+
"clip_ratio/low_mean": 0.00045801994201610796,
|
| 1803 |
+
"clip_ratio/low_min": 0.0,
|
| 1804 |
+
"clip_ratio/region_mean": 0.0009359141477034427,
|
| 1805 |
+
"epoch": 0.74375,
|
| 1806 |
+
"grad_norm": 2.8514267404951426,
|
| 1807 |
+
"kl": 0.019006389542482793,
|
| 1808 |
+
"learning_rate": 1.6293196622202632e-07,
|
| 1809 |
+
"loss": -9.609166590962559e-05,
|
| 1810 |
+
"step": 119,
|
| 1811 |
+
"step_time": 24.889595428481698
|
| 1812 |
+
},
|
| 1813 |
+
{
|
| 1814 |
+
"clip_ratio/high_max": 0.0022044377692509443,
|
| 1815 |
+
"clip_ratio/high_mean": 0.0006893837780808099,
|
| 1816 |
+
"clip_ratio/low_mean": 0.00039617672155145556,
|
| 1817 |
+
"clip_ratio/low_min": 0.0,
|
| 1818 |
+
"clip_ratio/region_mean": 0.0010855605032702442,
|
| 1819 |
+
"epoch": 0.75,
|
| 1820 |
+
"grad_norm": 1.9611571606038112,
|
| 1821 |
+
"kl": 0.017608478665351868,
|
| 1822 |
+
"learning_rate": 1.555165404621567e-07,
|
| 1823 |
+
"loss": -5.368860001908615e-05,
|
| 1824 |
+
"step": 120,
|
| 1825 |
+
"step_time": 25.327955707442015
|
| 1826 |
+
},
|
| 1827 |
+
{
|
| 1828 |
+
"clip_ratio/high_max": 0.0,
|
| 1829 |
+
"clip_ratio/high_mean": 0.0,
|
| 1830 |
+
"clip_ratio/low_mean": 0.0,
|
| 1831 |
+
"clip_ratio/low_min": 0.0,
|
| 1832 |
+
"clip_ratio/region_mean": 0.0,
|
| 1833 |
+
"completions/clipped_ratio": 0.2578125,
|
| 1834 |
+
"completions/max_length": 2048.0,
|
| 1835 |
+
"completions/mean_length": 914.515625,
|
| 1836 |
+
"completions/min_length": 67.0,
|
| 1837 |
+
"epoch": 0.75625,
|
| 1838 |
+
"frac_reward_zero_std": 0.25,
|
| 1839 |
+
"grad_norm": 1.1832296217461817,
|
| 1840 |
+
"kl": 0.02371334913186729,
|
| 1841 |
+
"learning_rate": 1.4824262559002592e-07,
|
| 1842 |
+
"loss": 0.00023716743453405797,
|
| 1843 |
+
"reward": 0.4375,
|
| 1844 |
+
"reward_std": 0.3521764874458313,
|
| 1845 |
+
"rewards/MajorityAccuracy/mean": 0.4375,
|
| 1846 |
+
"rewards/MajorityAccuracy/std": 0.49705013632774353,
|
| 1847 |
+
"step": 121,
|
| 1848 |
+
"step_time": 192.67440181737766
|
| 1849 |
+
},
|
| 1850 |
+
{
|
| 1851 |
+
"clip_ratio/high_max": 0.013383902543864679,
|
| 1852 |
+
"clip_ratio/high_mean": 0.003986481327956426,
|
| 1853 |
+
"clip_ratio/low_mean": 0.001989592390600592,
|
| 1854 |
+
"clip_ratio/low_min": 0.0,
|
| 1855 |
+
"clip_ratio/region_mean": 0.005976073662168346,
|
| 1856 |
+
"epoch": 0.7625,
|
| 1857 |
+
"grad_norm": 16.667830138887425,
|
| 1858 |
+
"kl": 0.016483186278492212,
|
| 1859 |
+
"learning_rate": 1.4111320966826057e-07,
|
| 1860 |
+
"loss": 0.0001158304512500763,
|
| 1861 |
+
"step": 122,
|
| 1862 |
+
"step_time": 24.45521473698318
|
| 1863 |
+
},
|
| 1864 |
+
{
|
| 1865 |
+
"clip_ratio/high_max": 0.0,
|
| 1866 |
+
"clip_ratio/high_mean": 0.0,
|
| 1867 |
+
"clip_ratio/low_mean": 0.0003633694432210177,
|
| 1868 |
+
"clip_ratio/low_min": 0.0,
|
| 1869 |
+
"clip_ratio/region_mean": 0.0003633694432210177,
|
| 1870 |
+
"epoch": 0.76875,
|
| 1871 |
+
"grad_norm": 11.237347518299325,
|
| 1872 |
+
"kl": 0.013756084255874157,
|
| 1873 |
+
"learning_rate": 1.3413122140054217e-07,
|
| 1874 |
+
"loss": -9.51429974520579e-05,
|
| 1875 |
+
"step": 123,
|
| 1876 |
+
"step_time": 25.0142598352395
|
| 1877 |
+
},
|
| 1878 |
+
{
|
| 1879 |
+
"clip_ratio/high_max": 0.002558524865889922,
|
| 1880 |
+
"clip_ratio/high_mean": 0.0007441462730639614,
|
| 1881 |
+
"clip_ratio/low_mean": 0.0003528347187966574,
|
| 1882 |
+
"clip_ratio/low_min": 0.0,
|
| 1883 |
+
"clip_ratio/region_mean": 0.0010969809954985976,
|
| 1884 |
+
"epoch": 0.775,
|
| 1885 |
+
"grad_norm": 3.2653012917541466,
|
| 1886 |
+
"kl": 0.03484604391269386,
|
| 1887 |
+
"learning_rate": 1.272995289285202e-07,
|
| 1888 |
+
"loss": 0.0008791193249635398,
|
| 1889 |
+
"step": 124,
|
| 1890 |
+
"step_time": 23.20116938324645
|
| 1891 |
+
},
|
| 1892 |
+
{
|
| 1893 |
+
"clip_ratio/high_max": 0.0005897687515243888,
|
| 1894 |
+
"clip_ratio/high_mean": 0.00018139113672077656,
|
| 1895 |
+
"clip_ratio/low_mean": 0.00045278549805516377,
|
| 1896 |
+
"clip_ratio/low_min": 0.0,
|
| 1897 |
+
"clip_ratio/region_mean": 0.0006341766347759403,
|
| 1898 |
+
"epoch": 0.78125,
|
| 1899 |
+
"grad_norm": 1.7705980189436925,
|
| 1900 |
+
"kl": 0.018418508814647794,
|
| 1901 |
+
"learning_rate": 1.2062093865360457e-07,
|
| 1902 |
+
"loss": 0.00032354824361391366,
|
| 1903 |
+
"step": 125,
|
| 1904 |
+
"step_time": 24.805662186816335
|
| 1905 |
+
},
|
| 1906 |
+
{
|
| 1907 |
+
"clip_ratio/high_max": 0.007714232837315649,
|
| 1908 |
+
"clip_ratio/high_mean": 0.002126476538251154,
|
| 1909 |
+
"clip_ratio/low_mean": 0.00014504976570606232,
|
| 1910 |
+
"clip_ratio/low_min": 0.0,
|
| 1911 |
+
"clip_ratio/region_mean": 0.002271526274853386,
|
| 1912 |
+
"epoch": 0.7875,
|
| 1913 |
+
"grad_norm": 5.2930688744507535,
|
| 1914 |
+
"kl": 0.02015839098021388,
|
| 1915 |
+
"learning_rate": 1.1409819408411897e-07,
|
| 1916 |
+
"loss": 0.0003884662874042988,
|
| 1917 |
+
"step": 126,
|
| 1918 |
+
"step_time": 26.27512728003785
|
| 1919 |
+
},
|
| 1920 |
+
{
|
| 1921 |
+
"clip_ratio/high_max": 0.003208797068509739,
|
| 1922 |
+
"clip_ratio/high_mean": 0.0009308808930654777,
|
| 1923 |
+
"clip_ratio/low_mean": 0.0002435068308841437,
|
| 1924 |
+
"clip_ratio/low_min": 0.0,
|
| 1925 |
+
"clip_ratio/region_mean": 0.0011743877257686108,
|
| 1926 |
+
"epoch": 0.79375,
|
| 1927 |
+
"grad_norm": 1.7506601635123595,
|
| 1928 |
+
"kl": 0.013811586308293045,
|
| 1929 |
+
"learning_rate": 1.0773397470829143e-07,
|
| 1930 |
+
"loss": 0.00014651224773842841,
|
| 1931 |
+
"step": 127,
|
| 1932 |
+
"step_time": 26.774720322806388
|
| 1933 |
+
},
|
| 1934 |
+
{
|
| 1935 |
+
"clip_ratio/high_max": 0.00284015602665022,
|
| 1936 |
+
"clip_ratio/high_mean": 0.000710039006662555,
|
| 1937 |
+
"clip_ratio/low_mean": 0.00033165604691021144,
|
| 1938 |
+
"clip_ratio/low_min": 0.0,
|
| 1939 |
+
"clip_ratio/region_mean": 0.0010416950535727665,
|
| 1940 |
+
"epoch": 0.8,
|
| 1941 |
+
"grad_norm": 2.215589189978948,
|
| 1942 |
+
"kl": 0.01569803769234568,
|
| 1943 |
+
"learning_rate": 1.0153089489354256e-07,
|
| 1944 |
+
"loss": 0.00023033079924061894,
|
| 1945 |
+
"step": 128,
|
| 1946 |
+
"step_time": 25.10148276994005
|
| 1947 |
+
},
|
| 1948 |
+
{
|
| 1949 |
+
"clip_ratio/high_max": 0.0,
|
| 1950 |
+
"clip_ratio/high_mean": 0.0,
|
| 1951 |
+
"clip_ratio/low_mean": 0.0,
|
| 1952 |
+
"clip_ratio/low_min": 0.0,
|
| 1953 |
+
"clip_ratio/region_mean": 0.0,
|
| 1954 |
+
"completions/clipped_ratio": 0.25,
|
| 1955 |
+
"completions/max_length": 2048.0,
|
| 1956 |
+
"completions/mean_length": 931.98046875,
|
| 1957 |
+
"completions/min_length": 60.0,
|
| 1958 |
+
"epoch": 0.80625,
|
| 1959 |
+
"frac_reward_zero_std": 0.15625,
|
| 1960 |
+
"grad_norm": 6.01265361126305,
|
| 1961 |
+
"kl": 0.033148851012811065,
|
| 1962 |
+
"learning_rate": 9.549150281252632e-08,
|
| 1963 |
+
"loss": 0.0003314875066280365,
|
| 1964 |
+
"reward": 0.38671875,
|
| 1965 |
+
"reward_std": 0.37731024622917175,
|
| 1966 |
+
"rewards/MajorityAccuracy/mean": 0.38671875,
|
| 1967 |
+
"rewards/MajorityAccuracy/std": 0.4879522919654846,
|
| 1968 |
+
"step": 129,
|
| 1969 |
+
"step_time": 187.73209842201322
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"clip_ratio/high_max": 0.003201800958777312,
|
| 1973 |
+
"clip_ratio/high_mean": 0.0008303831946250284,
|
| 1974 |
+
"clip_ratio/low_mean": 0.0001220703125,
|
| 1975 |
+
"clip_ratio/low_min": 0.0,
|
| 1976 |
+
"clip_ratio/region_mean": 0.0009524535071250284,
|
| 1977 |
+
"epoch": 0.8125,
|
| 1978 |
+
"grad_norm": 2.075126484852606,
|
| 1979 |
+
"kl": 0.016945037292316556,
|
| 1980 |
+
"learning_rate": 8.961827939636196e-08,
|
| 1981 |
+
"loss": 8.903970592655241e-05,
|
| 1982 |
+
"step": 130,
|
| 1983 |
+
"step_time": 24.388125146273524
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"clip_ratio/high_max": 0.0019266994204372168,
|
| 1987 |
+
"clip_ratio/high_mean": 0.0005529547051992267,
|
| 1988 |
+
"clip_ratio/low_mean": 0.0008237439142249059,
|
| 1989 |
+
"clip_ratio/low_min": 0.0,
|
| 1990 |
+
"clip_ratio/region_mean": 0.0013766986194241326,
|
| 1991 |
+
"epoch": 0.81875,
|
| 1992 |
+
"grad_norm": 2.2245475019804504,
|
| 1993 |
+
"kl": 0.01919903000816703,
|
| 1994 |
+
"learning_rate": 8.391363731548811e-08,
|
| 1995 |
+
"loss": 0.00015516611165367067,
|
| 1996 |
+
"step": 131,
|
| 1997 |
+
"step_time": 25.292857798282057
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"clip_ratio/high_max": 0.0008202170574804768,
|
| 2001 |
+
"clip_ratio/high_mean": 0.0002050542643701192,
|
| 2002 |
+
"clip_ratio/low_mean": 0.0006309753080131486,
|
| 2003 |
+
"clip_ratio/low_min": 0.0,
|
| 2004 |
+
"clip_ratio/region_mean": 0.0008360295760212466,
|
| 2005 |
+
"epoch": 0.825,
|
| 2006 |
+
"grad_norm": 2.9461909813513194,
|
| 2007 |
+
"kl": 0.016898607136681676,
|
| 2008 |
+
"learning_rate": 7.837991998855897e-08,
|
| 2009 |
+
"loss": 0.00017759203910827637,
|
| 2010 |
+
"step": 132,
|
| 2011 |
+
"step_time": 25.315239606425166
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"clip_ratio/high_max": 0.005009059619624168,
|
| 2015 |
+
"clip_ratio/high_mean": 0.0014646384333900642,
|
| 2016 |
+
"clip_ratio/low_mean": 0.00046519010356860235,
|
| 2017 |
+
"clip_ratio/low_min": 0.0,
|
| 2018 |
+
"clip_ratio/region_mean": 0.0019298285224067513,
|
| 2019 |
+
"epoch": 0.83125,
|
| 2020 |
+
"grad_norm": 2.885945474523698,
|
| 2021 |
+
"kl": 0.018045995151624084,
|
| 2022 |
+
"learning_rate": 7.301940061978722e-08,
|
| 2023 |
+
"loss": -0.00022008176892995834,
|
| 2024 |
+
"step": 133,
|
| 2025 |
+
"step_time": 25.632155671715736
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"clip_ratio/high_max": 0.0034001428866758943,
|
| 2029 |
+
"clip_ratio/high_mean": 0.001236935961060226,
|
| 2030 |
+
"clip_ratio/low_mean": 0.0005371409497456625,
|
| 2031 |
+
"clip_ratio/low_min": 0.0,
|
| 2032 |
+
"clip_ratio/region_mean": 0.0017740768962539732,
|
| 2033 |
+
"epoch": 0.8375,
|
| 2034 |
+
"grad_norm": 3.6588051144632745,
|
| 2035 |
+
"kl": 0.0326204642187804,
|
| 2036 |
+
"learning_rate": 6.783428126513125e-08,
|
| 2037 |
+
"loss": 0.00044670887291431427,
|
| 2038 |
+
"step": 134,
|
| 2039 |
+
"step_time": 22.626671710517257
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"clip_ratio/high_max": 0.0023464912082999945,
|
| 2043 |
+
"clip_ratio/high_mean": 0.0006658369966316968,
|
| 2044 |
+
"clip_ratio/low_mean": 0.0004367511319287587,
|
| 2045 |
+
"clip_ratio/low_min": 0.0,
|
| 2046 |
+
"clip_ratio/region_mean": 0.0011025881285604555,
|
| 2047 |
+
"epoch": 0.84375,
|
| 2048 |
+
"grad_norm": 5.318214142953765,
|
| 2049 |
+
"kl": 0.017755185486748815,
|
| 2050 |
+
"learning_rate": 6.282669192770895e-08,
|
| 2051 |
+
"loss": 0.0003831141220871359,
|
| 2052 |
+
"step": 135,
|
| 2053 |
+
"step_time": 24.888322774786502
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"clip_ratio/high_max": 0.0011934986760024913,
|
| 2057 |
+
"clip_ratio/high_mean": 0.0002983746690006228,
|
| 2058 |
+
"clip_ratio/low_mean": 0.00037409725337056443,
|
| 2059 |
+
"clip_ratio/low_min": 0.0,
|
| 2060 |
+
"clip_ratio/region_mean": 0.0006724719278281555,
|
| 2061 |
+
"epoch": 0.85,
|
| 2062 |
+
"grad_norm": 2.240024974027618,
|
| 2063 |
+
"kl": 0.020275881397537887,
|
| 2064 |
+
"learning_rate": 5.799868968281074e-08,
|
| 2065 |
+
"loss": 7.000700861681253e-05,
|
| 2066 |
+
"step": 136,
|
| 2067 |
+
"step_time": 24.982830572407693
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"clip_ratio/high_max": 0.0,
|
| 2071 |
+
"clip_ratio/high_mean": 0.0,
|
| 2072 |
+
"clip_ratio/low_mean": 0.0,
|
| 2073 |
+
"clip_ratio/low_min": 0.0,
|
| 2074 |
+
"clip_ratio/region_mean": 0.0,
|
| 2075 |
+
"completions/clipped_ratio": 0.140625,
|
| 2076 |
+
"completions/max_length": 2048.0,
|
| 2077 |
+
"completions/mean_length": 791.046875,
|
| 2078 |
+
"completions/min_length": 76.0,
|
| 2079 |
+
"epoch": 0.85625,
|
| 2080 |
+
"frac_reward_zero_std": 0.0625,
|
| 2081 |
+
"grad_norm": 2.5738398743456763,
|
| 2082 |
+
"kl": 0.02360692317597568,
|
| 2083 |
+
"learning_rate": 5.335225783287051e-08,
|
| 2084 |
+
"loss": 0.00023609865456819534,
|
| 2085 |
+
"reward": 0.4375,
|
| 2086 |
+
"reward_std": 0.42424172163009644,
|
| 2087 |
+
"rewards/MajorityAccuracy/mean": 0.4375,
|
| 2088 |
+
"rewards/MajorityAccuracy/std": 0.49705013632774353,
|
| 2089 |
+
"step": 137,
|
| 2090 |
+
"step_time": 184.66426088567823
|
| 2091 |
+
},
|
| 2092 |
+
{
|
| 2093 |
+
"clip_ratio/high_max": 0.0020615982357412577,
|
| 2094 |
+
"clip_ratio/high_mean": 0.0005153995589353144,
|
| 2095 |
+
"clip_ratio/low_mean": 0.0005537912657018751,
|
| 2096 |
+
"clip_ratio/low_min": 0.0,
|
| 2097 |
+
"clip_ratio/region_mean": 0.0010691908246371895,
|
| 2098 |
+
"epoch": 0.8625,
|
| 2099 |
+
"grad_norm": 3.2571096505196167,
|
| 2100 |
+
"kl": 0.028625125298276544,
|
| 2101 |
+
"learning_rate": 4.888930509274125e-08,
|
| 2102 |
+
"loss": 0.0002624266780912876,
|
| 2103 |
+
"step": 138,
|
| 2104 |
+
"step_time": 23.191153191495687
|
| 2105 |
+
},
|
| 2106 |
+
{
|
| 2107 |
+
"clip_ratio/high_max": 0.0018699133361224085,
|
| 2108 |
+
"clip_ratio/high_mean": 0.0005250812973827124,
|
| 2109 |
+
"clip_ratio/low_mean": 0.0004181619206065079,
|
| 2110 |
+
"clip_ratio/low_min": 0.0,
|
| 2111 |
+
"clip_ratio/region_mean": 0.0009432432143512415,
|
| 2112 |
+
"epoch": 0.86875,
|
| 2113 |
+
"grad_norm": 2.5763289469523234,
|
| 2114 |
+
"kl": 0.02415002160705626,
|
| 2115 |
+
"learning_rate": 4.4611664805611794e-08,
|
| 2116 |
+
"loss": 0.00030940212309360504,
|
| 2117 |
+
"step": 139,
|
| 2118 |
+
"step_time": 25.155863158404827
|
| 2119 |
+
},
|
| 2120 |
+
{
|
| 2121 |
+
"clip_ratio/high_max": 0.0029300033565959893,
|
| 2122 |
+
"clip_ratio/high_mean": 0.0008879734723450383,
|
| 2123 |
+
"clip_ratio/low_mean": 0.00013679582480108365,
|
| 2124 |
+
"clip_ratio/low_min": 0.0,
|
| 2125 |
+
"clip_ratio/region_mean": 0.0010247692935081432,
|
| 2126 |
+
"epoch": 0.875,
|
| 2127 |
+
"grad_norm": 3.526713670895025,
|
| 2128 |
+
"kl": 0.025422130012884736,
|
| 2129 |
+
"learning_rate": 4.0521094189884696e-08,
|
| 2130 |
+
"loss": 0.0006344560533761978,
|
| 2131 |
+
"step": 140,
|
| 2132 |
+
"step_time": 24.59485837025568
|
| 2133 |
+
},
|
| 2134 |
+
{
|
| 2135 |
+
"clip_ratio/high_max": 0.0034560053900349885,
|
| 2136 |
+
"clip_ratio/high_mean": 0.001196836652525235,
|
| 2137 |
+
"clip_ratio/low_mean": 0.0009217964088747976,
|
| 2138 |
+
"clip_ratio/low_min": 0.0,
|
| 2139 |
+
"clip_ratio/region_mean": 0.0021186330614000326,
|
| 2140 |
+
"epoch": 0.88125,
|
| 2141 |
+
"grad_norm": 3.112117962162615,
|
| 2142 |
+
"kl": 0.023439400014467537,
|
| 2143 |
+
"learning_rate": 3.6619273617325695e-08,
|
| 2144 |
+
"loss": -5.265418440103531e-05,
|
| 2145 |
+
"step": 141,
|
| 2146 |
+
"step_time": 23.673489401116967
|
| 2147 |
+
},
|
| 2148 |
+
{
|
| 2149 |
+
"clip_ratio/high_max": 0.002226529613835737,
|
| 2150 |
+
"clip_ratio/high_mean": 0.0006640207066084258,
|
| 2151 |
+
"clip_ratio/low_mean": 0.00014140272105578333,
|
| 2152 |
+
"clip_ratio/low_min": 0.0,
|
| 2153 |
+
"clip_ratio/region_mean": 0.0008054234276642092,
|
| 2154 |
+
"epoch": 0.8875,
|
| 2155 |
+
"grad_norm": 4.3659877954438535,
|
| 2156 |
+
"kl": 0.025215451139956713,
|
| 2157 |
+
"learning_rate": 3.2907805922781476e-08,
|
| 2158 |
+
"loss": 0.000200640904949978,
|
| 2159 |
+
"step": 142,
|
| 2160 |
+
"step_time": 24.450828310567886
|
| 2161 |
+
},
|
| 2162 |
+
{
|
| 2163 |
+
"clip_ratio/high_max": 0.0026134509971598163,
|
| 2164 |
+
"clip_ratio/high_mean": 0.0008425519954471383,
|
| 2165 |
+
"clip_ratio/low_mean": 0.0003400194545974955,
|
| 2166 |
+
"clip_ratio/low_min": 0.0,
|
| 2167 |
+
"clip_ratio/region_mean": 0.0011825714227597928,
|
| 2168 |
+
"epoch": 0.89375,
|
| 2169 |
+
"grad_norm": 2.175101971539894,
|
| 2170 |
+
"kl": 0.017353363684378564,
|
| 2171 |
+
"learning_rate": 2.9388215745748345e-08,
|
| 2172 |
+
"loss": 3.263354301452637e-05,
|
| 2173 |
+
"step": 143,
|
| 2174 |
+
"step_time": 25.046978532336652
|
| 2175 |
+
},
|
| 2176 |
+
{
|
| 2177 |
+
"clip_ratio/high_max": 0.002092474576784298,
|
| 2178 |
+
"clip_ratio/high_mean": 0.0005748983821831644,
|
| 2179 |
+
"clip_ratio/low_mean": 0.0004017939427285455,
|
| 2180 |
+
"clip_ratio/low_min": 0.0,
|
| 2181 |
+
"clip_ratio/region_mean": 0.0009766923285496887,
|
| 2182 |
+
"epoch": 0.9,
|
| 2183 |
+
"grad_norm": 4.72415433005437,
|
| 2184 |
+
"kl": 0.012911462690681219,
|
| 2185 |
+
"learning_rate": 2.6061948904063658e-08,
|
| 2186 |
+
"loss": 0.0003595380112528801,
|
| 2187 |
+
"step": 144,
|
| 2188 |
+
"step_time": 25.19282961450517
|
| 2189 |
+
},
|
| 2190 |
+
{
|
| 2191 |
+
"clip_ratio/high_max": 0.0,
|
| 2192 |
+
"clip_ratio/high_mean": 0.0,
|
| 2193 |
+
"clip_ratio/low_mean": 0.0,
|
| 2194 |
+
"clip_ratio/low_min": 0.0,
|
| 2195 |
+
"clip_ratio/region_mean": 0.0,
|
| 2196 |
+
"completions/clipped_ratio": 0.26171875,
|
| 2197 |
+
"completions/max_length": 2048.0,
|
| 2198 |
+
"completions/mean_length": 1077.34765625,
|
| 2199 |
+
"completions/min_length": 67.0,
|
| 2200 |
+
"epoch": 0.90625,
|
| 2201 |
+
"frac_reward_zero_std": 0.21875,
|
| 2202 |
+
"grad_norm": 1.460928877017638,
|
| 2203 |
+
"kl": 0.0178944889921695,
|
| 2204 |
+
"learning_rate": 2.293037179997559e-08,
|
| 2205 |
+
"loss": 0.00017897864745464176,
|
| 2206 |
+
"reward": 0.3671875,
|
| 2207 |
+
"reward_std": 0.3427231013774872,
|
| 2208 |
+
"rewards/MajorityAccuracy/mean": 0.3671875,
|
| 2209 |
+
"rewards/MajorityAccuracy/std": 0.48298245668411255,
|
| 2210 |
+
"step": 145,
|
| 2211 |
+
"step_time": 189.73759700544178
|
| 2212 |
+
},
|
| 2213 |
+
{
|
| 2214 |
+
"clip_ratio/high_max": 0.0006253352476051077,
|
| 2215 |
+
"clip_ratio/high_mean": 0.00015633381190127693,
|
| 2216 |
+
"clip_ratio/low_mean": 0.00027326684721629135,
|
| 2217 |
+
"clip_ratio/low_min": 0.0,
|
| 2218 |
+
"clip_ratio/region_mean": 0.0004296006627555471,
|
| 2219 |
+
"epoch": 0.9125,
|
| 2220 |
+
"grad_norm": 3.5534997810544504,
|
| 2221 |
+
"kl": 0.026194309582933784,
|
| 2222 |
+
"learning_rate": 1.9994770858837107e-08,
|
| 2223 |
+
"loss": 0.00020220410078763962,
|
| 2224 |
+
"step": 146,
|
| 2225 |
+
"step_time": 24.759347829502076
|
| 2226 |
+
},
|
| 2227 |
+
{
|
| 2228 |
+
"clip_ratio/high_max": 0.0010165712737943977,
|
| 2229 |
+
"clip_ratio/high_mean": 0.00029050116791040637,
|
| 2230 |
+
"clip_ratio/low_mean": 0.000530568542671972,
|
| 2231 |
+
"clip_ratio/low_min": 0.0,
|
| 2232 |
+
"clip_ratio/region_mean": 0.00082106973604823,
|
| 2233 |
+
"epoch": 0.91875,
|
| 2234 |
+
"grad_norm": 1.7257619584109125,
|
| 2235 |
+
"kl": 0.015043306280858815,
|
| 2236 |
+
"learning_rate": 1.725635200065323e-08,
|
| 2237 |
+
"loss": -5.8428384363651276e-05,
|
| 2238 |
+
"step": 147,
|
| 2239 |
+
"step_time": 25.382796177640557
|
| 2240 |
+
},
|
| 2241 |
+
{
|
| 2242 |
+
"clip_ratio/high_max": 0.0008412618772126734,
|
| 2243 |
+
"clip_ratio/high_mean": 0.00021031546930316836,
|
| 2244 |
+
"clip_ratio/low_mean": 0.0004531816448434256,
|
| 2245 |
+
"clip_ratio/low_min": 0.0,
|
| 2246 |
+
"clip_ratio/region_mean": 0.0006634971250605304,
|
| 2247 |
+
"epoch": 0.925,
|
| 2248 |
+
"grad_norm": 1.8792715010063343,
|
| 2249 |
+
"kl": 0.013475793646648526,
|
| 2250 |
+
"learning_rate": 1.4716240144699187e-08,
|
| 2251 |
+
"loss": 0.0005341917276382446,
|
| 2252 |
+
"step": 148,
|
| 2253 |
+
"step_time": 25.198696110397577
|
| 2254 |
+
},
|
| 2255 |
+
{
|
| 2256 |
+
"clip_ratio/high_max": 0.004076726530911401,
|
| 2257 |
+
"clip_ratio/high_mean": 0.0013044481311226264,
|
| 2258 |
+
"clip_ratio/low_mean": 0.00037760482337034773,
|
| 2259 |
+
"clip_ratio/low_min": 0.0,
|
| 2260 |
+
"clip_ratio/region_mean": 0.0016820529399410589,
|
| 2261 |
+
"epoch": 0.93125,
|
| 2262 |
+
"grad_norm": 2.45372262953109,
|
| 2263 |
+
"kl": 0.03270056447945535,
|
| 2264 |
+
"learning_rate": 1.2375478747413015e-08,
|
| 2265 |
+
"loss": 0.00013369042426347733,
|
| 2266 |
+
"step": 149,
|
| 2267 |
+
"step_time": 24.006096094381064
|
| 2268 |
+
},
|
| 2269 |
+
{
|
| 2270 |
+
"clip_ratio/high_max": 0.0020262412799638696,
|
| 2271 |
+
"clip_ratio/high_mean": 0.0005630702671624022,
|
| 2272 |
+
"clip_ratio/low_mean": 0.0002406193616479868,
|
| 2273 |
+
"clip_ratio/low_min": 0.0,
|
| 2274 |
+
"clip_ratio/region_mean": 0.000803689628810389,
|
| 2275 |
+
"epoch": 0.9375,
|
| 2276 |
+
"grad_norm": 3.055083569637597,
|
| 2277 |
+
"kl": 0.01785694540012628,
|
| 2278 |
+
"learning_rate": 1.0235029373752757e-08,
|
| 2279 |
+
"loss": 3.506905341055244e-05,
|
| 2280 |
+
"step": 150,
|
| 2281 |
+
"step_time": 24.315687018912286
|
| 2282 |
+
},
|
| 2283 |
+
{
|
| 2284 |
+
"clip_ratio/high_max": 0.001950558384123724,
|
| 2285 |
+
"clip_ratio/high_mean": 0.0005649066933983704,
|
| 2286 |
+
"clip_ratio/low_mean": 8.587630145484582e-05,
|
| 2287 |
+
"clip_ratio/low_min": 0.0,
|
| 2288 |
+
"clip_ratio/region_mean": 0.0006507829948532162,
|
| 2289 |
+
"epoch": 0.94375,
|
| 2290 |
+
"grad_norm": 1.5620479879181497,
|
| 2291 |
+
"kl": 0.025568603770807385,
|
| 2292 |
+
"learning_rate": 8.295771302193721e-09,
|
| 2293 |
+
"loss": 0.0002807211421895772,
|
| 2294 |
+
"step": 151,
|
| 2295 |
+
"step_time": 24.9022874138318
|
| 2296 |
+
},
|
| 2297 |
+
{
|
| 2298 |
+
"clip_ratio/high_max": 0.0014459290832746774,
|
| 2299 |
+
"clip_ratio/high_mean": 0.0003825971289188601,
|
| 2300 |
+
"clip_ratio/low_mean": 0.0004820258982363157,
|
| 2301 |
+
"clip_ratio/low_min": 0.0,
|
| 2302 |
+
"clip_ratio/region_mean": 0.0008646230344311334,
|
| 2303 |
+
"epoch": 0.95,
|
| 2304 |
+
"grad_norm": 2.828172352218068,
|
| 2305 |
+
"kl": 0.02107399539090693,
|
| 2306 |
+
"learning_rate": 6.558501163527963e-09,
|
| 2307 |
+
"loss": 0.00041194763616658747,
|
| 2308 |
+
"step": 152,
|
| 2309 |
+
"step_time": 25.201037157326937
|
| 2310 |
+
},
|
| 2311 |
+
{
|
| 2312 |
+
"clip_ratio/high_max": 0.0,
|
| 2313 |
+
"clip_ratio/high_mean": 0.0,
|
| 2314 |
+
"clip_ratio/low_mean": 0.0,
|
| 2315 |
+
"clip_ratio/low_min": 0.0,
|
| 2316 |
+
"clip_ratio/region_mean": 0.0,
|
| 2317 |
+
"completions/clipped_ratio": 0.1875,
|
| 2318 |
+
"completions/max_length": 2048.0,
|
| 2319 |
+
"completions/mean_length": 764.9296875,
|
| 2320 |
+
"completions/min_length": 49.0,
|
| 2321 |
+
"epoch": 0.95625,
|
| 2322 |
+
"frac_reward_zero_std": 0.09375,
|
| 2323 |
+
"grad_norm": 3.473222251100786,
|
| 2324 |
+
"kl": 0.026403479394502938,
|
| 2325 |
+
"learning_rate": 5.023932613615445e-09,
|
| 2326 |
+
"loss": 0.0002638772130012512,
|
| 2327 |
+
"reward": 0.41796875,
|
| 2328 |
+
"reward_std": 0.42332667112350464,
|
| 2329 |
+
"rewards/MajorityAccuracy/mean": 0.41796875,
|
| 2330 |
+
"rewards/MajorityAccuracy/std": 0.49419113993644714,
|
| 2331 |
+
"step": 153,
|
| 2332 |
+
"step_time": 185.43863809900358
|
| 2333 |
+
},
|
| 2334 |
+
{
|
| 2335 |
+
"clip_ratio/high_max": 0.0015977444127202034,
|
| 2336 |
+
"clip_ratio/high_mean": 0.00039943610318005085,
|
| 2337 |
+
"clip_ratio/low_mean": 7.660566006961744e-05,
|
| 2338 |
+
"clip_ratio/low_min": 0.0,
|
| 2339 |
+
"clip_ratio/region_mean": 0.0004760417632496683,
|
| 2340 |
+
"epoch": 0.9625,
|
| 2341 |
+
"grad_norm": 2.1301817027333985,
|
| 2342 |
+
"kl": 0.03772760182619095,
|
| 2343 |
+
"learning_rate": 3.6926960402202666e-09,
|
| 2344 |
+
"loss": 0.0007470763521268964,
|
| 2345 |
+
"step": 154,
|
| 2346 |
+
"step_time": 24.48218126175925
|
| 2347 |
+
},
|
| 2348 |
+
{
|
| 2349 |
+
"clip_ratio/high_max": 0.002994970098370686,
|
| 2350 |
+
"clip_ratio/high_mean": 0.000780244568886701,
|
| 2351 |
+
"clip_ratio/low_mean": 0.0001373291015625,
|
| 2352 |
+
"clip_ratio/low_min": 0.0,
|
| 2353 |
+
"clip_ratio/region_mean": 0.000917573670449201,
|
| 2354 |
+
"epoch": 0.96875,
|
| 2355 |
+
"grad_norm": 4.3343071568894835,
|
| 2356 |
+
"kl": 0.021125117782503366,
|
| 2357 |
+
"learning_rate": 2.5653383040524224e-09,
|
| 2358 |
+
"loss": 0.0006277831271290779,
|
| 2359 |
+
"step": 155,
|
| 2360 |
+
"step_time": 25.263464014511555
|
| 2361 |
+
},
|
| 2362 |
+
{
|
| 2363 |
+
"clip_ratio/high_max": 0.0020884526893496513,
|
| 2364 |
+
"clip_ratio/high_mean": 0.0005723576323362067,
|
| 2365 |
+
"clip_ratio/low_mean": 0.0001918495909194462,
|
| 2366 |
+
"clip_ratio/low_min": 0.0,
|
| 2367 |
+
"clip_ratio/region_mean": 0.0007642072232556529,
|
| 2368 |
+
"epoch": 0.975,
|
| 2369 |
+
"grad_norm": 5.536031250589736,
|
| 2370 |
+
"kl": 0.02767080278135836,
|
| 2371 |
+
"learning_rate": 1.642322514122385e-09,
|
| 2372 |
+
"loss": 0.0012405328452587128,
|
| 2373 |
+
"step": 156,
|
| 2374 |
+
"step_time": 23.3633666834794
|
| 2375 |
+
},
|
| 2376 |
+
{
|
| 2377 |
+
"clip_ratio/high_max": 0.0015340471436502412,
|
| 2378 |
+
"clip_ratio/high_mean": 0.000575312184082577,
|
| 2379 |
+
"clip_ratio/low_mean": 0.00023834615421947092,
|
| 2380 |
+
"clip_ratio/low_min": 0.0,
|
| 2381 |
+
"clip_ratio/region_mean": 0.0008136583383020479,
|
| 2382 |
+
"epoch": 0.98125,
|
| 2383 |
+
"grad_norm": 3.0703038034724597,
|
| 2384 |
+
"kl": 0.016762026585638523,
|
| 2385 |
+
"learning_rate": 9.240278374995636e-10,
|
| 2386 |
+
"loss": 0.0003749486058950424,
|
| 2387 |
+
"step": 157,
|
| 2388 |
+
"step_time": 23.99086816702038
|
| 2389 |
+
},
|
| 2390 |
+
{
|
| 2391 |
+
"clip_ratio/high_max": 0.0014322113202069886,
|
| 2392 |
+
"clip_ratio/high_mean": 0.00038696125011483673,
|
| 2393 |
+
"clip_ratio/low_mean": 0.00037078796958667226,
|
| 2394 |
+
"clip_ratio/low_min": 0.0,
|
| 2395 |
+
"clip_ratio/region_mean": 0.000757749228796456,
|
| 2396 |
+
"epoch": 0.9875,
|
| 2397 |
+
"grad_norm": 1.4549710754418894,
|
| 2398 |
+
"kl": 0.013573615928180516,
|
| 2399 |
+
"learning_rate": 4.1074934355384006e-10,
|
| 2400 |
+
"loss": 0.0002349354326725006,
|
| 2401 |
+
"step": 158,
|
| 2402 |
+
"step_time": 25.49250327050686
|
| 2403 |
+
},
|
| 2404 |
+
{
|
| 2405 |
+
"clip_ratio/high_max": 0.004464268567971885,
|
| 2406 |
+
"clip_ratio/high_mean": 0.0014934779101167805,
|
| 2407 |
+
"clip_ratio/low_mean": 0.0003955696302000433,
|
| 2408 |
+
"clip_ratio/low_min": 0.0,
|
| 2409 |
+
"clip_ratio/region_mean": 0.0018890475403168239,
|
| 2410 |
+
"epoch": 0.99375,
|
| 2411 |
+
"grad_norm": 4.23813750775604,
|
| 2412 |
+
"kl": 0.028215029276907444,
|
| 2413 |
+
"learning_rate": 1.0269788274358537e-10,
|
| 2414 |
+
"loss": 0.0007316209375858307,
|
| 2415 |
+
"step": 159,
|
| 2416 |
+
"step_time": 22.230896414257586
|
| 2417 |
+
},
|
| 2418 |
+
{
|
| 2419 |
+
"clip_ratio/high_max": 0.000729057879652828,
|
| 2420 |
+
"clip_ratio/high_mean": 0.000182264469913207,
|
| 2421 |
+
"clip_ratio/low_mean": 0.00017667864085524343,
|
| 2422 |
+
"clip_ratio/low_min": 0.0,
|
| 2423 |
+
"clip_ratio/region_mean": 0.0003589431107684504,
|
| 2424 |
+
"epoch": 1.0,
|
| 2425 |
+
"grad_norm": 3.6253402587882535,
|
| 2426 |
+
"kl": 0.03702863282524049,
|
| 2427 |
+
"learning_rate": 0.0,
|
| 2428 |
+
"loss": 1.4083052519708872e-06,
|
| 2429 |
+
"step": 160,
|
| 2430 |
+
"step_time": 24.020924455020577
|
| 2431 |
+
}
|
| 2432 |
+
],
|
| 2433 |
+
"logging_steps": 1,
|
| 2434 |
+
"max_steps": 160,
|
| 2435 |
+
"num_input_tokens_seen": 0,
|
| 2436 |
+
"num_train_epochs": 1,
|
| 2437 |
+
"save_steps": 80,
|
| 2438 |
+
"stateful_callbacks": {
|
| 2439 |
+
"TrainerControl": {
|
| 2440 |
+
"args": {
|
| 2441 |
+
"should_epoch_stop": false,
|
| 2442 |
+
"should_evaluate": false,
|
| 2443 |
+
"should_log": false,
|
| 2444 |
+
"should_save": true,
|
| 2445 |
+
"should_training_stop": true
|
| 2446 |
+
},
|
| 2447 |
+
"attributes": {}
|
| 2448 |
+
}
|
| 2449 |
+
},
|
| 2450 |
+
"total_flos": 0.0,
|
| 2451 |
+
"train_batch_size": 1,
|
| 2452 |
+
"trial_name": null,
|
| 2453 |
+
"trial_params": null
|
| 2454 |
+
}
|
c1_no_history/round_1/chat_template.jinja
ADDED
|
@@ -0,0 +1,154 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{%- set image_count = namespace(value=0) %}
|
| 2 |
+
{%- set video_count = namespace(value=0) %}
|
| 3 |
+
{%- macro render_content(content, do_vision_count, is_system_content=false) %}
|
| 4 |
+
{%- if content is string %}
|
| 5 |
+
{{- content }}
|
| 6 |
+
{%- elif content is iterable and content is not mapping %}
|
| 7 |
+
{%- for item in content %}
|
| 8 |
+
{%- if 'image' in item or 'image_url' in item or item.type == 'image' %}
|
| 9 |
+
{%- if is_system_content %}
|
| 10 |
+
{{- raise_exception('System message cannot contain images.') }}
|
| 11 |
+
{%- endif %}
|
| 12 |
+
{%- if do_vision_count %}
|
| 13 |
+
{%- set image_count.value = image_count.value + 1 %}
|
| 14 |
+
{%- endif %}
|
| 15 |
+
{%- if add_vision_id %}
|
| 16 |
+
{{- 'Picture ' ~ image_count.value ~ ': ' }}
|
| 17 |
+
{%- endif %}
|
| 18 |
+
{{- '<|vision_start|><|image_pad|><|vision_end|>' }}
|
| 19 |
+
{%- elif 'video' in item or item.type == 'video' %}
|
| 20 |
+
{%- if is_system_content %}
|
| 21 |
+
{{- raise_exception('System message cannot contain videos.') }}
|
| 22 |
+
{%- endif %}
|
| 23 |
+
{%- if do_vision_count %}
|
| 24 |
+
{%- set video_count.value = video_count.value + 1 %}
|
| 25 |
+
{%- endif %}
|
| 26 |
+
{%- if add_vision_id %}
|
| 27 |
+
{{- 'Video ' ~ video_count.value ~ ': ' }}
|
| 28 |
+
{%- endif %}
|
| 29 |
+
{{- '<|vision_start|><|video_pad|><|vision_end|>' }}
|
| 30 |
+
{%- elif 'text' in item %}
|
| 31 |
+
{{- item.text }}
|
| 32 |
+
{%- else %}
|
| 33 |
+
{{- raise_exception('Unexpected item type in content.') }}
|
| 34 |
+
{%- endif %}
|
| 35 |
+
{%- endfor %}
|
| 36 |
+
{%- elif content is none or content is undefined %}
|
| 37 |
+
{{- '' }}
|
| 38 |
+
{%- else %}
|
| 39 |
+
{{- raise_exception('Unexpected content type.') }}
|
| 40 |
+
{%- endif %}
|
| 41 |
+
{%- endmacro %}
|
| 42 |
+
{%- if not messages %}
|
| 43 |
+
{{- raise_exception('No messages provided.') }}
|
| 44 |
+
{%- endif %}
|
| 45 |
+
{%- if tools and tools is iterable and tools is not mapping %}
|
| 46 |
+
{{- '<|im_start|>system\n' }}
|
| 47 |
+
{{- "# Tools\n\nYou have access to the following functions:\n\n<tools>" }}
|
| 48 |
+
{%- for tool in tools %}
|
| 49 |
+
{{- "\n" }}
|
| 50 |
+
{{- tool | tojson }}
|
| 51 |
+
{%- endfor %}
|
| 52 |
+
{{- "\n</tools>" }}
|
| 53 |
+
{{- '\n\nIf you choose to call a function ONLY reply in the following format with NO suffix:\n\n<tool_call>\n<function=example_function_name>\n<parameter=example_parameter_1>\nvalue_1\n</parameter>\n<parameter=example_parameter_2>\nThis is the value for the second parameter\nthat can span\nmultiple lines\n</parameter>\n</function>\n</tool_call>\n\n<IMPORTANT>\nReminder:\n- Function calls MUST follow the specified format: an inner <function=...></function> block must be nested within <tool_call></tool_call> XML tags\n- Required parameters MUST be specified\n- You may provide optional reasoning for your function call in natural language BEFORE the function call, but NOT after\n- If there is no function call available, answer the question like normal with your current knowledge and do not tell the user about function calls\n</IMPORTANT>' }}
|
| 54 |
+
{%- if messages[0].role == 'system' %}
|
| 55 |
+
{%- set content = render_content(messages[0].content, false, true)|trim %}
|
| 56 |
+
{%- if content %}
|
| 57 |
+
{{- '\n\n' + content }}
|
| 58 |
+
{%- endif %}
|
| 59 |
+
{%- endif %}
|
| 60 |
+
{{- '<|im_end|>\n' }}
|
| 61 |
+
{%- else %}
|
| 62 |
+
{%- if messages[0].role == 'system' %}
|
| 63 |
+
{%- set content = render_content(messages[0].content, false, true)|trim %}
|
| 64 |
+
{{- '<|im_start|>system\n' + content + '<|im_end|>\n' }}
|
| 65 |
+
{%- endif %}
|
| 66 |
+
{%- endif %}
|
| 67 |
+
{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
|
| 68 |
+
{%- for message in messages[::-1] %}
|
| 69 |
+
{%- set index = (messages|length - 1) - loop.index0 %}
|
| 70 |
+
{%- if ns.multi_step_tool and message.role == "user" %}
|
| 71 |
+
{%- set content = render_content(message.content, false)|trim %}
|
| 72 |
+
{%- if not(content.startswith('<tool_response>') and content.endswith('</tool_response>')) %}
|
| 73 |
+
{%- set ns.multi_step_tool = false %}
|
| 74 |
+
{%- set ns.last_query_index = index %}
|
| 75 |
+
{%- endif %}
|
| 76 |
+
{%- endif %}
|
| 77 |
+
{%- endfor %}
|
| 78 |
+
{%- if ns.multi_step_tool %}
|
| 79 |
+
{{- raise_exception('No user query found in messages.') }}
|
| 80 |
+
{%- endif %}
|
| 81 |
+
{%- for message in messages %}
|
| 82 |
+
{%- set content = render_content(message.content, true)|trim %}
|
| 83 |
+
{%- if message.role == "system" %}
|
| 84 |
+
{%- if not loop.first %}
|
| 85 |
+
{{- raise_exception('System message must be at the beginning.') }}
|
| 86 |
+
{%- endif %}
|
| 87 |
+
{%- elif message.role == "user" %}
|
| 88 |
+
{{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
|
| 89 |
+
{%- elif message.role == "assistant" %}
|
| 90 |
+
{%- set reasoning_content = '' %}
|
| 91 |
+
{%- if message.reasoning_content is string %}
|
| 92 |
+
{%- set reasoning_content = message.reasoning_content %}
|
| 93 |
+
{%- else %}
|
| 94 |
+
{%- if '</think>' in content %}
|
| 95 |
+
{%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
|
| 96 |
+
{%- set content = content.split('</think>')[-1].lstrip('\n') %}
|
| 97 |
+
{%- endif %}
|
| 98 |
+
{%- endif %}
|
| 99 |
+
{%- set reasoning_content = reasoning_content|trim %}
|
| 100 |
+
{%- if loop.index0 > ns.last_query_index %}
|
| 101 |
+
{{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content + '\n</think>\n\n' + content }}
|
| 102 |
+
{%- else %}
|
| 103 |
+
{{- '<|im_start|>' + message.role + '\n' + content }}
|
| 104 |
+
{%- endif %}
|
| 105 |
+
{%- if message.tool_calls and message.tool_calls is iterable and message.tool_calls is not mapping %}
|
| 106 |
+
{%- for tool_call in message.tool_calls %}
|
| 107 |
+
{%- if tool_call.function is defined %}
|
| 108 |
+
{%- set tool_call = tool_call.function %}
|
| 109 |
+
{%- endif %}
|
| 110 |
+
{%- if loop.first %}
|
| 111 |
+
{%- if content|trim %}
|
| 112 |
+
{{- '\n\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
|
| 113 |
+
{%- else %}
|
| 114 |
+
{{- '<tool_call>\n<function=' + tool_call.name + '>\n' }}
|
| 115 |
+
{%- endif %}
|
| 116 |
+
{%- else %}
|
| 117 |
+
{{- '\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
|
| 118 |
+
{%- endif %}
|
| 119 |
+
{%- if tool_call.arguments is defined %}
|
| 120 |
+
{%- for args_name, args_value in tool_call.arguments|items %}
|
| 121 |
+
{{- '<parameter=' + args_name + '>\n' }}
|
| 122 |
+
{%- set args_value = args_value | tojson | safe if args_value is mapping or (args_value is sequence and args_value is not string) else args_value | string %}
|
| 123 |
+
{{- args_value }}
|
| 124 |
+
{{- '\n</parameter>\n' }}
|
| 125 |
+
{%- endfor %}
|
| 126 |
+
{%- endif %}
|
| 127 |
+
{{- '</function>\n</tool_call>' }}
|
| 128 |
+
{%- endfor %}
|
| 129 |
+
{%- endif %}
|
| 130 |
+
{{- '<|im_end|>\n' }}
|
| 131 |
+
{%- elif message.role == "tool" %}
|
| 132 |
+
{%- if loop.previtem and loop.previtem.role != "tool" %}
|
| 133 |
+
{{- '<|im_start|>user' }}
|
| 134 |
+
{%- endif %}
|
| 135 |
+
{{- '\n<tool_response>\n' }}
|
| 136 |
+
{{- content }}
|
| 137 |
+
{{- '\n</tool_response>' }}
|
| 138 |
+
{%- if not loop.last and loop.nextitem.role != "tool" %}
|
| 139 |
+
{{- '<|im_end|>\n' }}
|
| 140 |
+
{%- elif loop.last %}
|
| 141 |
+
{{- '<|im_end|>\n' }}
|
| 142 |
+
{%- endif %}
|
| 143 |
+
{%- else %}
|
| 144 |
+
{{- raise_exception('Unexpected message role.') }}
|
| 145 |
+
{%- endif %}
|
| 146 |
+
{%- endfor %}
|
| 147 |
+
{%- if add_generation_prompt %}
|
| 148 |
+
{{- '<|im_start|>assistant\n' }}
|
| 149 |
+
{%- if enable_thinking is defined and enable_thinking is false %}
|
| 150 |
+
{{- '<think>\n\n</think>\n\n' }}
|
| 151 |
+
{%- else %}
|
| 152 |
+
{{- '<think>\n' }}
|
| 153 |
+
{%- endif %}
|
| 154 |
+
{%- endif %}
|
c1_no_history/round_1/config.json
ADDED
|
@@ -0,0 +1,113 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"Qwen3_5ForConditionalGeneration"
|
| 4 |
+
],
|
| 5 |
+
"dtype": "bfloat16",
|
| 6 |
+
"eos_token_id": 248046,
|
| 7 |
+
"hidden_size": 4096,
|
| 8 |
+
"image_token_id": 248056,
|
| 9 |
+
"model_type": "qwen3_5",
|
| 10 |
+
"pad_token_id": 248044,
|
| 11 |
+
"text_config": {
|
| 12 |
+
"attention_bias": false,
|
| 13 |
+
"attention_dropout": 0.0,
|
| 14 |
+
"attn_output_gate": true,
|
| 15 |
+
"bos_token_id": null,
|
| 16 |
+
"dtype": "bfloat16",
|
| 17 |
+
"eos_token_id": 248044,
|
| 18 |
+
"full_attention_interval": 4,
|
| 19 |
+
"head_dim": 256,
|
| 20 |
+
"hidden_act": "silu",
|
| 21 |
+
"hidden_size": 4096,
|
| 22 |
+
"initializer_range": 0.02,
|
| 23 |
+
"intermediate_size": 12288,
|
| 24 |
+
"layer_types": [
|
| 25 |
+
"linear_attention",
|
| 26 |
+
"linear_attention",
|
| 27 |
+
"linear_attention",
|
| 28 |
+
"full_attention",
|
| 29 |
+
"linear_attention",
|
| 30 |
+
"linear_attention",
|
| 31 |
+
"linear_attention",
|
| 32 |
+
"full_attention",
|
| 33 |
+
"linear_attention",
|
| 34 |
+
"linear_attention",
|
| 35 |
+
"linear_attention",
|
| 36 |
+
"full_attention",
|
| 37 |
+
"linear_attention",
|
| 38 |
+
"linear_attention",
|
| 39 |
+
"linear_attention",
|
| 40 |
+
"full_attention",
|
| 41 |
+
"linear_attention",
|
| 42 |
+
"linear_attention",
|
| 43 |
+
"linear_attention",
|
| 44 |
+
"full_attention",
|
| 45 |
+
"linear_attention",
|
| 46 |
+
"linear_attention",
|
| 47 |
+
"linear_attention",
|
| 48 |
+
"full_attention",
|
| 49 |
+
"linear_attention",
|
| 50 |
+
"linear_attention",
|
| 51 |
+
"linear_attention",
|
| 52 |
+
"full_attention",
|
| 53 |
+
"linear_attention",
|
| 54 |
+
"linear_attention",
|
| 55 |
+
"linear_attention",
|
| 56 |
+
"full_attention"
|
| 57 |
+
],
|
| 58 |
+
"linear_conv_kernel_dim": 4,
|
| 59 |
+
"linear_key_head_dim": 128,
|
| 60 |
+
"linear_num_key_heads": 16,
|
| 61 |
+
"linear_num_value_heads": 32,
|
| 62 |
+
"linear_value_head_dim": 128,
|
| 63 |
+
"mamba_ssm_dtype": "float32",
|
| 64 |
+
"max_position_embeddings": 262144,
|
| 65 |
+
"mlp_only_layers": [],
|
| 66 |
+
"model_type": "qwen3_5_text",
|
| 67 |
+
"mtp_num_hidden_layers": 1,
|
| 68 |
+
"mtp_use_dedicated_embeddings": false,
|
| 69 |
+
"num_attention_heads": 16,
|
| 70 |
+
"num_hidden_layers": 32,
|
| 71 |
+
"num_key_value_heads": 4,
|
| 72 |
+
"pad_token_id": 248044,
|
| 73 |
+
"partial_rotary_factor": 0.25,
|
| 74 |
+
"rms_norm_eps": 1e-06,
|
| 75 |
+
"rope_parameters": {
|
| 76 |
+
"mrope_interleaved": true,
|
| 77 |
+
"mrope_section": [
|
| 78 |
+
11,
|
| 79 |
+
11,
|
| 80 |
+
10
|
| 81 |
+
],
|
| 82 |
+
"partial_rotary_factor": 0.25,
|
| 83 |
+
"rope_theta": 10000000,
|
| 84 |
+
"rope_type": "default"
|
| 85 |
+
},
|
| 86 |
+
"tie_word_embeddings": false,
|
| 87 |
+
"use_cache": false,
|
| 88 |
+
"vocab_size": 248320
|
| 89 |
+
},
|
| 90 |
+
"tie_word_embeddings": false,
|
| 91 |
+
"transformers_version": "5.10.4",
|
| 92 |
+
"use_cache": false,
|
| 93 |
+
"video_token_id": 248057,
|
| 94 |
+
"vision_config": {
|
| 95 |
+
"deepstack_visual_indexes": [],
|
| 96 |
+
"depth": 27,
|
| 97 |
+
"dtype": "bfloat16",
|
| 98 |
+
"hidden_act": "gelu_pytorch_tanh",
|
| 99 |
+
"hidden_size": 1152,
|
| 100 |
+
"in_channels": 3,
|
| 101 |
+
"initializer_range": 0.02,
|
| 102 |
+
"intermediate_size": 4304,
|
| 103 |
+
"model_type": "qwen3_5_vision",
|
| 104 |
+
"num_heads": 16,
|
| 105 |
+
"num_position_embeddings": 2304,
|
| 106 |
+
"out_hidden_size": 4096,
|
| 107 |
+
"patch_size": 16,
|
| 108 |
+
"spatial_merge_size": 2,
|
| 109 |
+
"temporal_patch_size": 2
|
| 110 |
+
},
|
| 111 |
+
"vision_end_token_id": 248054,
|
| 112 |
+
"vision_start_token_id": 248053
|
| 113 |
+
}
|
c1_no_history/round_1/generation_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"eos_token_id": [
|
| 4 |
+
248044,
|
| 5 |
+
248046
|
| 6 |
+
],
|
| 7 |
+
"transformers_version": "5.10.4",
|
| 8 |
+
"use_cache": true
|
| 9 |
+
}
|
c1_no_history/round_1/model.safetensors.index.json
ADDED
|
@@ -0,0 +1,768 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"metadata": {
|
| 3 |
+
"total_parameters": 1469680,
|
| 4 |
+
"total_size": 18819627488
|
| 5 |
+
},
|
| 6 |
+
"weight_map": {
|
| 7 |
+
"lm_head.weight": "model-00001-of-00004.safetensors",
|
| 8 |
+
"model.language_model.embed_tokens.weight": "model-00001-of-00004.safetensors",
|
| 9 |
+
"model.language_model.layers.0.input_layernorm.weight": "model-00001-of-00004.safetensors",
|
| 10 |
+
"model.language_model.layers.0.linear_attn.A_log": "model-00001-of-00004.safetensors",
|
| 11 |
+
"model.language_model.layers.0.linear_attn.conv1d.weight": "model-00001-of-00004.safetensors",
|
| 12 |
+
"model.language_model.layers.0.linear_attn.dt_bias": "model-00001-of-00004.safetensors",
|
| 13 |
+
"model.language_model.layers.0.linear_attn.in_proj_a.weight": "model-00001-of-00004.safetensors",
|
| 14 |
+
"model.language_model.layers.0.linear_attn.in_proj_b.weight": "model-00001-of-00004.safetensors",
|
| 15 |
+
"model.language_model.layers.0.linear_attn.in_proj_qkv.weight": "model-00001-of-00004.safetensors",
|
| 16 |
+
"model.language_model.layers.0.linear_attn.in_proj_z.weight": "model-00001-of-00004.safetensors",
|
| 17 |
+
"model.language_model.layers.0.linear_attn.norm.weight": "model-00001-of-00004.safetensors",
|
| 18 |
+
"model.language_model.layers.0.linear_attn.out_proj.weight": "model-00001-of-00004.safetensors",
|
| 19 |
+
"model.language_model.layers.0.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
|
| 20 |
+
"model.language_model.layers.0.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
|
| 21 |
+
"model.language_model.layers.0.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
|
| 22 |
+
"model.language_model.layers.0.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
|
| 23 |
+
"model.language_model.layers.1.input_layernorm.weight": "model-00001-of-00004.safetensors",
|
| 24 |
+
"model.language_model.layers.1.linear_attn.A_log": "model-00001-of-00004.safetensors",
|
| 25 |
+
"model.language_model.layers.1.linear_attn.conv1d.weight": "model-00001-of-00004.safetensors",
|
| 26 |
+
"model.language_model.layers.1.linear_attn.dt_bias": "model-00001-of-00004.safetensors",
|
| 27 |
+
"model.language_model.layers.1.linear_attn.in_proj_a.weight": "model-00001-of-00004.safetensors",
|
| 28 |
+
"model.language_model.layers.1.linear_attn.in_proj_b.weight": "model-00001-of-00004.safetensors",
|
| 29 |
+
"model.language_model.layers.1.linear_attn.in_proj_qkv.weight": "model-00001-of-00004.safetensors",
|
| 30 |
+
"model.language_model.layers.1.linear_attn.in_proj_z.weight": "model-00001-of-00004.safetensors",
|
| 31 |
+
"model.language_model.layers.1.linear_attn.norm.weight": "model-00001-of-00004.safetensors",
|
| 32 |
+
"model.language_model.layers.1.linear_attn.out_proj.weight": "model-00001-of-00004.safetensors",
|
| 33 |
+
"model.language_model.layers.1.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
|
| 34 |
+
"model.language_model.layers.1.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
|
| 35 |
+
"model.language_model.layers.1.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
|
| 36 |
+
"model.language_model.layers.1.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
|
| 37 |
+
"model.language_model.layers.10.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 38 |
+
"model.language_model.layers.10.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 39 |
+
"model.language_model.layers.10.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 40 |
+
"model.language_model.layers.10.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 41 |
+
"model.language_model.layers.10.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 42 |
+
"model.language_model.layers.10.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 43 |
+
"model.language_model.layers.10.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 44 |
+
"model.language_model.layers.10.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 45 |
+
"model.language_model.layers.10.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 46 |
+
"model.language_model.layers.10.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 47 |
+
"model.language_model.layers.10.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 48 |
+
"model.language_model.layers.10.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 49 |
+
"model.language_model.layers.10.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 50 |
+
"model.language_model.layers.10.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 51 |
+
"model.language_model.layers.11.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 52 |
+
"model.language_model.layers.11.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 53 |
+
"model.language_model.layers.11.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 54 |
+
"model.language_model.layers.11.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 55 |
+
"model.language_model.layers.11.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 56 |
+
"model.language_model.layers.11.self_attn.k_norm.weight": "model-00002-of-00004.safetensors",
|
| 57 |
+
"model.language_model.layers.11.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
|
| 58 |
+
"model.language_model.layers.11.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
|
| 59 |
+
"model.language_model.layers.11.self_attn.q_norm.weight": "model-00002-of-00004.safetensors",
|
| 60 |
+
"model.language_model.layers.11.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
|
| 61 |
+
"model.language_model.layers.11.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
|
| 62 |
+
"model.language_model.layers.12.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 63 |
+
"model.language_model.layers.12.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 64 |
+
"model.language_model.layers.12.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 65 |
+
"model.language_model.layers.12.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 66 |
+
"model.language_model.layers.12.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 67 |
+
"model.language_model.layers.12.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 68 |
+
"model.language_model.layers.12.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 69 |
+
"model.language_model.layers.12.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 70 |
+
"model.language_model.layers.12.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 71 |
+
"model.language_model.layers.12.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 72 |
+
"model.language_model.layers.12.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 73 |
+
"model.language_model.layers.12.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 74 |
+
"model.language_model.layers.12.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 75 |
+
"model.language_model.layers.12.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 76 |
+
"model.language_model.layers.13.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 77 |
+
"model.language_model.layers.13.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 78 |
+
"model.language_model.layers.13.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 79 |
+
"model.language_model.layers.13.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 80 |
+
"model.language_model.layers.13.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 81 |
+
"model.language_model.layers.13.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 82 |
+
"model.language_model.layers.13.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 83 |
+
"model.language_model.layers.13.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 84 |
+
"model.language_model.layers.13.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 85 |
+
"model.language_model.layers.13.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 86 |
+
"model.language_model.layers.13.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 87 |
+
"model.language_model.layers.13.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 88 |
+
"model.language_model.layers.13.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 89 |
+
"model.language_model.layers.13.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 90 |
+
"model.language_model.layers.14.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 91 |
+
"model.language_model.layers.14.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 92 |
+
"model.language_model.layers.14.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 93 |
+
"model.language_model.layers.14.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 94 |
+
"model.language_model.layers.14.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 95 |
+
"model.language_model.layers.14.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 96 |
+
"model.language_model.layers.14.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 97 |
+
"model.language_model.layers.14.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 98 |
+
"model.language_model.layers.14.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 99 |
+
"model.language_model.layers.14.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 100 |
+
"model.language_model.layers.14.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 101 |
+
"model.language_model.layers.14.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 102 |
+
"model.language_model.layers.14.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 103 |
+
"model.language_model.layers.14.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 104 |
+
"model.language_model.layers.15.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 105 |
+
"model.language_model.layers.15.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 106 |
+
"model.language_model.layers.15.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 107 |
+
"model.language_model.layers.15.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 108 |
+
"model.language_model.layers.15.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 109 |
+
"model.language_model.layers.15.self_attn.k_norm.weight": "model-00003-of-00004.safetensors",
|
| 110 |
+
"model.language_model.layers.15.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
|
| 111 |
+
"model.language_model.layers.15.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
|
| 112 |
+
"model.language_model.layers.15.self_attn.q_norm.weight": "model-00003-of-00004.safetensors",
|
| 113 |
+
"model.language_model.layers.15.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
|
| 114 |
+
"model.language_model.layers.15.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
|
| 115 |
+
"model.language_model.layers.16.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 116 |
+
"model.language_model.layers.16.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 117 |
+
"model.language_model.layers.16.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 118 |
+
"model.language_model.layers.16.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 119 |
+
"model.language_model.layers.16.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 120 |
+
"model.language_model.layers.16.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 121 |
+
"model.language_model.layers.16.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 122 |
+
"model.language_model.layers.16.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 123 |
+
"model.language_model.layers.16.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 124 |
+
"model.language_model.layers.16.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 125 |
+
"model.language_model.layers.16.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 126 |
+
"model.language_model.layers.16.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 127 |
+
"model.language_model.layers.16.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 128 |
+
"model.language_model.layers.16.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 129 |
+
"model.language_model.layers.17.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 130 |
+
"model.language_model.layers.17.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 131 |
+
"model.language_model.layers.17.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 132 |
+
"model.language_model.layers.17.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 133 |
+
"model.language_model.layers.17.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 134 |
+
"model.language_model.layers.17.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 135 |
+
"model.language_model.layers.17.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 136 |
+
"model.language_model.layers.17.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 137 |
+
"model.language_model.layers.17.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 138 |
+
"model.language_model.layers.17.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 139 |
+
"model.language_model.layers.17.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 140 |
+
"model.language_model.layers.17.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 141 |
+
"model.language_model.layers.17.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 142 |
+
"model.language_model.layers.17.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 143 |
+
"model.language_model.layers.18.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 144 |
+
"model.language_model.layers.18.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 145 |
+
"model.language_model.layers.18.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 146 |
+
"model.language_model.layers.18.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 147 |
+
"model.language_model.layers.18.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 148 |
+
"model.language_model.layers.18.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 149 |
+
"model.language_model.layers.18.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 150 |
+
"model.language_model.layers.18.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 151 |
+
"model.language_model.layers.18.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 152 |
+
"model.language_model.layers.18.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 153 |
+
"model.language_model.layers.18.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 154 |
+
"model.language_model.layers.18.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 155 |
+
"model.language_model.layers.18.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 156 |
+
"model.language_model.layers.18.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 157 |
+
"model.language_model.layers.19.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 158 |
+
"model.language_model.layers.19.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 159 |
+
"model.language_model.layers.19.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 160 |
+
"model.language_model.layers.19.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 161 |
+
"model.language_model.layers.19.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 162 |
+
"model.language_model.layers.19.self_attn.k_norm.weight": "model-00003-of-00004.safetensors",
|
| 163 |
+
"model.language_model.layers.19.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
|
| 164 |
+
"model.language_model.layers.19.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
|
| 165 |
+
"model.language_model.layers.19.self_attn.q_norm.weight": "model-00003-of-00004.safetensors",
|
| 166 |
+
"model.language_model.layers.19.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
|
| 167 |
+
"model.language_model.layers.19.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
|
| 168 |
+
"model.language_model.layers.2.input_layernorm.weight": "model-00001-of-00004.safetensors",
|
| 169 |
+
"model.language_model.layers.2.linear_attn.A_log": "model-00001-of-00004.safetensors",
|
| 170 |
+
"model.language_model.layers.2.linear_attn.conv1d.weight": "model-00001-of-00004.safetensors",
|
| 171 |
+
"model.language_model.layers.2.linear_attn.dt_bias": "model-00001-of-00004.safetensors",
|
| 172 |
+
"model.language_model.layers.2.linear_attn.in_proj_a.weight": "model-00001-of-00004.safetensors",
|
| 173 |
+
"model.language_model.layers.2.linear_attn.in_proj_b.weight": "model-00001-of-00004.safetensors",
|
| 174 |
+
"model.language_model.layers.2.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 175 |
+
"model.language_model.layers.2.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 176 |
+
"model.language_model.layers.2.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 177 |
+
"model.language_model.layers.2.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 178 |
+
"model.language_model.layers.2.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 179 |
+
"model.language_model.layers.2.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 180 |
+
"model.language_model.layers.2.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 181 |
+
"model.language_model.layers.2.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 182 |
+
"model.language_model.layers.20.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 183 |
+
"model.language_model.layers.20.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 184 |
+
"model.language_model.layers.20.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 185 |
+
"model.language_model.layers.20.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 186 |
+
"model.language_model.layers.20.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 187 |
+
"model.language_model.layers.20.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 188 |
+
"model.language_model.layers.20.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 189 |
+
"model.language_model.layers.20.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 190 |
+
"model.language_model.layers.20.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 191 |
+
"model.language_model.layers.20.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 192 |
+
"model.language_model.layers.20.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 193 |
+
"model.language_model.layers.20.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 194 |
+
"model.language_model.layers.20.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 195 |
+
"model.language_model.layers.20.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 196 |
+
"model.language_model.layers.21.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 197 |
+
"model.language_model.layers.21.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 198 |
+
"model.language_model.layers.21.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 199 |
+
"model.language_model.layers.21.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 200 |
+
"model.language_model.layers.21.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 201 |
+
"model.language_model.layers.21.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 202 |
+
"model.language_model.layers.21.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 203 |
+
"model.language_model.layers.21.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 204 |
+
"model.language_model.layers.21.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 205 |
+
"model.language_model.layers.21.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 206 |
+
"model.language_model.layers.21.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 207 |
+
"model.language_model.layers.21.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 208 |
+
"model.language_model.layers.21.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 209 |
+
"model.language_model.layers.21.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 210 |
+
"model.language_model.layers.22.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 211 |
+
"model.language_model.layers.22.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 212 |
+
"model.language_model.layers.22.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 213 |
+
"model.language_model.layers.22.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 214 |
+
"model.language_model.layers.22.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 215 |
+
"model.language_model.layers.22.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 216 |
+
"model.language_model.layers.22.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 217 |
+
"model.language_model.layers.22.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 218 |
+
"model.language_model.layers.22.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 219 |
+
"model.language_model.layers.22.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 220 |
+
"model.language_model.layers.22.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 221 |
+
"model.language_model.layers.22.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 222 |
+
"model.language_model.layers.22.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 223 |
+
"model.language_model.layers.22.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 224 |
+
"model.language_model.layers.23.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 225 |
+
"model.language_model.layers.23.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 226 |
+
"model.language_model.layers.23.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 227 |
+
"model.language_model.layers.23.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 228 |
+
"model.language_model.layers.23.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 229 |
+
"model.language_model.layers.23.self_attn.k_norm.weight": "model-00003-of-00004.safetensors",
|
| 230 |
+
"model.language_model.layers.23.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
|
| 231 |
+
"model.language_model.layers.23.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
|
| 232 |
+
"model.language_model.layers.23.self_attn.q_norm.weight": "model-00003-of-00004.safetensors",
|
| 233 |
+
"model.language_model.layers.23.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
|
| 234 |
+
"model.language_model.layers.23.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
|
| 235 |
+
"model.language_model.layers.24.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 236 |
+
"model.language_model.layers.24.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 237 |
+
"model.language_model.layers.24.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 238 |
+
"model.language_model.layers.24.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 239 |
+
"model.language_model.layers.24.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 240 |
+
"model.language_model.layers.24.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 241 |
+
"model.language_model.layers.24.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 242 |
+
"model.language_model.layers.24.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 243 |
+
"model.language_model.layers.24.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 244 |
+
"model.language_model.layers.24.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 245 |
+
"model.language_model.layers.24.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 246 |
+
"model.language_model.layers.24.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 247 |
+
"model.language_model.layers.24.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 248 |
+
"model.language_model.layers.24.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 249 |
+
"model.language_model.layers.25.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 250 |
+
"model.language_model.layers.25.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 251 |
+
"model.language_model.layers.25.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 252 |
+
"model.language_model.layers.25.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 253 |
+
"model.language_model.layers.25.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 254 |
+
"model.language_model.layers.25.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 255 |
+
"model.language_model.layers.25.linear_attn.in_proj_qkv.weight": "model-00004-of-00004.safetensors",
|
| 256 |
+
"model.language_model.layers.25.linear_attn.in_proj_z.weight": "model-00004-of-00004.safetensors",
|
| 257 |
+
"model.language_model.layers.25.linear_attn.norm.weight": "model-00004-of-00004.safetensors",
|
| 258 |
+
"model.language_model.layers.25.linear_attn.out_proj.weight": "model-00004-of-00004.safetensors",
|
| 259 |
+
"model.language_model.layers.25.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 260 |
+
"model.language_model.layers.25.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 261 |
+
"model.language_model.layers.25.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 262 |
+
"model.language_model.layers.25.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 263 |
+
"model.language_model.layers.26.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 264 |
+
"model.language_model.layers.26.linear_attn.A_log": "model-00004-of-00004.safetensors",
|
| 265 |
+
"model.language_model.layers.26.linear_attn.conv1d.weight": "model-00004-of-00004.safetensors",
|
| 266 |
+
"model.language_model.layers.26.linear_attn.dt_bias": "model-00004-of-00004.safetensors",
|
| 267 |
+
"model.language_model.layers.26.linear_attn.in_proj_a.weight": "model-00004-of-00004.safetensors",
|
| 268 |
+
"model.language_model.layers.26.linear_attn.in_proj_b.weight": "model-00004-of-00004.safetensors",
|
| 269 |
+
"model.language_model.layers.26.linear_attn.in_proj_qkv.weight": "model-00004-of-00004.safetensors",
|
| 270 |
+
"model.language_model.layers.26.linear_attn.in_proj_z.weight": "model-00004-of-00004.safetensors",
|
| 271 |
+
"model.language_model.layers.26.linear_attn.norm.weight": "model-00004-of-00004.safetensors",
|
| 272 |
+
"model.language_model.layers.26.linear_attn.out_proj.weight": "model-00004-of-00004.safetensors",
|
| 273 |
+
"model.language_model.layers.26.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 274 |
+
"model.language_model.layers.26.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 275 |
+
"model.language_model.layers.26.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 276 |
+
"model.language_model.layers.26.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 277 |
+
"model.language_model.layers.27.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 278 |
+
"model.language_model.layers.27.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 279 |
+
"model.language_model.layers.27.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 280 |
+
"model.language_model.layers.27.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 281 |
+
"model.language_model.layers.27.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 282 |
+
"model.language_model.layers.27.self_attn.k_norm.weight": "model-00004-of-00004.safetensors",
|
| 283 |
+
"model.language_model.layers.27.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
|
| 284 |
+
"model.language_model.layers.27.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
|
| 285 |
+
"model.language_model.layers.27.self_attn.q_norm.weight": "model-00004-of-00004.safetensors",
|
| 286 |
+
"model.language_model.layers.27.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
|
| 287 |
+
"model.language_model.layers.27.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
|
| 288 |
+
"model.language_model.layers.28.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 289 |
+
"model.language_model.layers.28.linear_attn.A_log": "model-00004-of-00004.safetensors",
|
| 290 |
+
"model.language_model.layers.28.linear_attn.conv1d.weight": "model-00004-of-00004.safetensors",
|
| 291 |
+
"model.language_model.layers.28.linear_attn.dt_bias": "model-00004-of-00004.safetensors",
|
| 292 |
+
"model.language_model.layers.28.linear_attn.in_proj_a.weight": "model-00004-of-00004.safetensors",
|
| 293 |
+
"model.language_model.layers.28.linear_attn.in_proj_b.weight": "model-00004-of-00004.safetensors",
|
| 294 |
+
"model.language_model.layers.28.linear_attn.in_proj_qkv.weight": "model-00004-of-00004.safetensors",
|
| 295 |
+
"model.language_model.layers.28.linear_attn.in_proj_z.weight": "model-00004-of-00004.safetensors",
|
| 296 |
+
"model.language_model.layers.28.linear_attn.norm.weight": "model-00004-of-00004.safetensors",
|
| 297 |
+
"model.language_model.layers.28.linear_attn.out_proj.weight": "model-00004-of-00004.safetensors",
|
| 298 |
+
"model.language_model.layers.28.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 299 |
+
"model.language_model.layers.28.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 300 |
+
"model.language_model.layers.28.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 301 |
+
"model.language_model.layers.28.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 302 |
+
"model.language_model.layers.29.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 303 |
+
"model.language_model.layers.29.linear_attn.A_log": "model-00004-of-00004.safetensors",
|
| 304 |
+
"model.language_model.layers.29.linear_attn.conv1d.weight": "model-00004-of-00004.safetensors",
|
| 305 |
+
"model.language_model.layers.29.linear_attn.dt_bias": "model-00004-of-00004.safetensors",
|
| 306 |
+
"model.language_model.layers.29.linear_attn.in_proj_a.weight": "model-00004-of-00004.safetensors",
|
| 307 |
+
"model.language_model.layers.29.linear_attn.in_proj_b.weight": "model-00004-of-00004.safetensors",
|
| 308 |
+
"model.language_model.layers.29.linear_attn.in_proj_qkv.weight": "model-00004-of-00004.safetensors",
|
| 309 |
+
"model.language_model.layers.29.linear_attn.in_proj_z.weight": "model-00004-of-00004.safetensors",
|
| 310 |
+
"model.language_model.layers.29.linear_attn.norm.weight": "model-00004-of-00004.safetensors",
|
| 311 |
+
"model.language_model.layers.29.linear_attn.out_proj.weight": "model-00004-of-00004.safetensors",
|
| 312 |
+
"model.language_model.layers.29.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 313 |
+
"model.language_model.layers.29.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 314 |
+
"model.language_model.layers.29.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 315 |
+
"model.language_model.layers.29.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 316 |
+
"model.language_model.layers.3.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 317 |
+
"model.language_model.layers.3.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 318 |
+
"model.language_model.layers.3.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 319 |
+
"model.language_model.layers.3.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 320 |
+
"model.language_model.layers.3.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 321 |
+
"model.language_model.layers.3.self_attn.k_norm.weight": "model-00002-of-00004.safetensors",
|
| 322 |
+
"model.language_model.layers.3.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
|
| 323 |
+
"model.language_model.layers.3.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
|
| 324 |
+
"model.language_model.layers.3.self_attn.q_norm.weight": "model-00002-of-00004.safetensors",
|
| 325 |
+
"model.language_model.layers.3.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
|
| 326 |
+
"model.language_model.layers.3.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
|
| 327 |
+
"model.language_model.layers.30.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 328 |
+
"model.language_model.layers.30.linear_attn.A_log": "model-00004-of-00004.safetensors",
|
| 329 |
+
"model.language_model.layers.30.linear_attn.conv1d.weight": "model-00004-of-00004.safetensors",
|
| 330 |
+
"model.language_model.layers.30.linear_attn.dt_bias": "model-00004-of-00004.safetensors",
|
| 331 |
+
"model.language_model.layers.30.linear_attn.in_proj_a.weight": "model-00004-of-00004.safetensors",
|
| 332 |
+
"model.language_model.layers.30.linear_attn.in_proj_b.weight": "model-00004-of-00004.safetensors",
|
| 333 |
+
"model.language_model.layers.30.linear_attn.in_proj_qkv.weight": "model-00004-of-00004.safetensors",
|
| 334 |
+
"model.language_model.layers.30.linear_attn.in_proj_z.weight": "model-00004-of-00004.safetensors",
|
| 335 |
+
"model.language_model.layers.30.linear_attn.norm.weight": "model-00004-of-00004.safetensors",
|
| 336 |
+
"model.language_model.layers.30.linear_attn.out_proj.weight": "model-00004-of-00004.safetensors",
|
| 337 |
+
"model.language_model.layers.30.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 338 |
+
"model.language_model.layers.30.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 339 |
+
"model.language_model.layers.30.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 340 |
+
"model.language_model.layers.30.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 341 |
+
"model.language_model.layers.31.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 342 |
+
"model.language_model.layers.31.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 343 |
+
"model.language_model.layers.31.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 344 |
+
"model.language_model.layers.31.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 345 |
+
"model.language_model.layers.31.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 346 |
+
"model.language_model.layers.31.self_attn.k_norm.weight": "model-00004-of-00004.safetensors",
|
| 347 |
+
"model.language_model.layers.31.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
|
| 348 |
+
"model.language_model.layers.31.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
|
| 349 |
+
"model.language_model.layers.31.self_attn.q_norm.weight": "model-00004-of-00004.safetensors",
|
| 350 |
+
"model.language_model.layers.31.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
|
| 351 |
+
"model.language_model.layers.31.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
|
| 352 |
+
"model.language_model.layers.4.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 353 |
+
"model.language_model.layers.4.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 354 |
+
"model.language_model.layers.4.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 355 |
+
"model.language_model.layers.4.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 356 |
+
"model.language_model.layers.4.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 357 |
+
"model.language_model.layers.4.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 358 |
+
"model.language_model.layers.4.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 359 |
+
"model.language_model.layers.4.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 360 |
+
"model.language_model.layers.4.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 361 |
+
"model.language_model.layers.4.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 362 |
+
"model.language_model.layers.4.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 363 |
+
"model.language_model.layers.4.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 364 |
+
"model.language_model.layers.4.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 365 |
+
"model.language_model.layers.4.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 366 |
+
"model.language_model.layers.5.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 367 |
+
"model.language_model.layers.5.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 368 |
+
"model.language_model.layers.5.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 369 |
+
"model.language_model.layers.5.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 370 |
+
"model.language_model.layers.5.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 371 |
+
"model.language_model.layers.5.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 372 |
+
"model.language_model.layers.5.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 373 |
+
"model.language_model.layers.5.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 374 |
+
"model.language_model.layers.5.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 375 |
+
"model.language_model.layers.5.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 376 |
+
"model.language_model.layers.5.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 377 |
+
"model.language_model.layers.5.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 378 |
+
"model.language_model.layers.5.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 379 |
+
"model.language_model.layers.5.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 380 |
+
"model.language_model.layers.6.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 381 |
+
"model.language_model.layers.6.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 382 |
+
"model.language_model.layers.6.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 383 |
+
"model.language_model.layers.6.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 384 |
+
"model.language_model.layers.6.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 385 |
+
"model.language_model.layers.6.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 386 |
+
"model.language_model.layers.6.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 387 |
+
"model.language_model.layers.6.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 388 |
+
"model.language_model.layers.6.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 389 |
+
"model.language_model.layers.6.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 390 |
+
"model.language_model.layers.6.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 391 |
+
"model.language_model.layers.6.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 392 |
+
"model.language_model.layers.6.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 393 |
+
"model.language_model.layers.6.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 394 |
+
"model.language_model.layers.7.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 395 |
+
"model.language_model.layers.7.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 396 |
+
"model.language_model.layers.7.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 397 |
+
"model.language_model.layers.7.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 398 |
+
"model.language_model.layers.7.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 399 |
+
"model.language_model.layers.7.self_attn.k_norm.weight": "model-00002-of-00004.safetensors",
|
| 400 |
+
"model.language_model.layers.7.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
|
| 401 |
+
"model.language_model.layers.7.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
|
| 402 |
+
"model.language_model.layers.7.self_attn.q_norm.weight": "model-00002-of-00004.safetensors",
|
| 403 |
+
"model.language_model.layers.7.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
|
| 404 |
+
"model.language_model.layers.7.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
|
| 405 |
+
"model.language_model.layers.8.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 406 |
+
"model.language_model.layers.8.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 407 |
+
"model.language_model.layers.8.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 408 |
+
"model.language_model.layers.8.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 409 |
+
"model.language_model.layers.8.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 410 |
+
"model.language_model.layers.8.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 411 |
+
"model.language_model.layers.8.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 412 |
+
"model.language_model.layers.8.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 413 |
+
"model.language_model.layers.8.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 414 |
+
"model.language_model.layers.8.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 415 |
+
"model.language_model.layers.8.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 416 |
+
"model.language_model.layers.8.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 417 |
+
"model.language_model.layers.8.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 418 |
+
"model.language_model.layers.8.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 419 |
+
"model.language_model.layers.9.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 420 |
+
"model.language_model.layers.9.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 421 |
+
"model.language_model.layers.9.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 422 |
+
"model.language_model.layers.9.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 423 |
+
"model.language_model.layers.9.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 424 |
+
"model.language_model.layers.9.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 425 |
+
"model.language_model.layers.9.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 426 |
+
"model.language_model.layers.9.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 427 |
+
"model.language_model.layers.9.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 428 |
+
"model.language_model.layers.9.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 429 |
+
"model.language_model.layers.9.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 430 |
+
"model.language_model.layers.9.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 431 |
+
"model.language_model.layers.9.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 432 |
+
"model.language_model.layers.9.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 433 |
+
"model.language_model.norm.weight": "model-00004-of-00004.safetensors",
|
| 434 |
+
"model.visual.blocks.0.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 435 |
+
"model.visual.blocks.0.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 436 |
+
"model.visual.blocks.0.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 437 |
+
"model.visual.blocks.0.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 438 |
+
"model.visual.blocks.0.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 439 |
+
"model.visual.blocks.0.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 440 |
+
"model.visual.blocks.0.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 441 |
+
"model.visual.blocks.0.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 442 |
+
"model.visual.blocks.0.norm1.bias": "model-00004-of-00004.safetensors",
|
| 443 |
+
"model.visual.blocks.0.norm1.weight": "model-00004-of-00004.safetensors",
|
| 444 |
+
"model.visual.blocks.0.norm2.bias": "model-00004-of-00004.safetensors",
|
| 445 |
+
"model.visual.blocks.0.norm2.weight": "model-00004-of-00004.safetensors",
|
| 446 |
+
"model.visual.blocks.1.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 447 |
+
"model.visual.blocks.1.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 448 |
+
"model.visual.blocks.1.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 449 |
+
"model.visual.blocks.1.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 450 |
+
"model.visual.blocks.1.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 451 |
+
"model.visual.blocks.1.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 452 |
+
"model.visual.blocks.1.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 453 |
+
"model.visual.blocks.1.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 454 |
+
"model.visual.blocks.1.norm1.bias": "model-00004-of-00004.safetensors",
|
| 455 |
+
"model.visual.blocks.1.norm1.weight": "model-00004-of-00004.safetensors",
|
| 456 |
+
"model.visual.blocks.1.norm2.bias": "model-00004-of-00004.safetensors",
|
| 457 |
+
"model.visual.blocks.1.norm2.weight": "model-00004-of-00004.safetensors",
|
| 458 |
+
"model.visual.blocks.10.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 459 |
+
"model.visual.blocks.10.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 460 |
+
"model.visual.blocks.10.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 461 |
+
"model.visual.blocks.10.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 462 |
+
"model.visual.blocks.10.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 463 |
+
"model.visual.blocks.10.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 464 |
+
"model.visual.blocks.10.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 465 |
+
"model.visual.blocks.10.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 466 |
+
"model.visual.blocks.10.norm1.bias": "model-00004-of-00004.safetensors",
|
| 467 |
+
"model.visual.blocks.10.norm1.weight": "model-00004-of-00004.safetensors",
|
| 468 |
+
"model.visual.blocks.10.norm2.bias": "model-00004-of-00004.safetensors",
|
| 469 |
+
"model.visual.blocks.10.norm2.weight": "model-00004-of-00004.safetensors",
|
| 470 |
+
"model.visual.blocks.11.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 471 |
+
"model.visual.blocks.11.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 472 |
+
"model.visual.blocks.11.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 473 |
+
"model.visual.blocks.11.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 474 |
+
"model.visual.blocks.11.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 475 |
+
"model.visual.blocks.11.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 476 |
+
"model.visual.blocks.11.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 477 |
+
"model.visual.blocks.11.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 478 |
+
"model.visual.blocks.11.norm1.bias": "model-00004-of-00004.safetensors",
|
| 479 |
+
"model.visual.blocks.11.norm1.weight": "model-00004-of-00004.safetensors",
|
| 480 |
+
"model.visual.blocks.11.norm2.bias": "model-00004-of-00004.safetensors",
|
| 481 |
+
"model.visual.blocks.11.norm2.weight": "model-00004-of-00004.safetensors",
|
| 482 |
+
"model.visual.blocks.12.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 483 |
+
"model.visual.blocks.12.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 484 |
+
"model.visual.blocks.12.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 485 |
+
"model.visual.blocks.12.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 486 |
+
"model.visual.blocks.12.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 487 |
+
"model.visual.blocks.12.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 488 |
+
"model.visual.blocks.12.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 489 |
+
"model.visual.blocks.12.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 490 |
+
"model.visual.blocks.12.norm1.bias": "model-00004-of-00004.safetensors",
|
| 491 |
+
"model.visual.blocks.12.norm1.weight": "model-00004-of-00004.safetensors",
|
| 492 |
+
"model.visual.blocks.12.norm2.bias": "model-00004-of-00004.safetensors",
|
| 493 |
+
"model.visual.blocks.12.norm2.weight": "model-00004-of-00004.safetensors",
|
| 494 |
+
"model.visual.blocks.13.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 495 |
+
"model.visual.blocks.13.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 496 |
+
"model.visual.blocks.13.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 497 |
+
"model.visual.blocks.13.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 498 |
+
"model.visual.blocks.13.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 499 |
+
"model.visual.blocks.13.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 500 |
+
"model.visual.blocks.13.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 501 |
+
"model.visual.blocks.13.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 502 |
+
"model.visual.blocks.13.norm1.bias": "model-00004-of-00004.safetensors",
|
| 503 |
+
"model.visual.blocks.13.norm1.weight": "model-00004-of-00004.safetensors",
|
| 504 |
+
"model.visual.blocks.13.norm2.bias": "model-00004-of-00004.safetensors",
|
| 505 |
+
"model.visual.blocks.13.norm2.weight": "model-00004-of-00004.safetensors",
|
| 506 |
+
"model.visual.blocks.14.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 507 |
+
"model.visual.blocks.14.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 508 |
+
"model.visual.blocks.14.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 509 |
+
"model.visual.blocks.14.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 510 |
+
"model.visual.blocks.14.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 511 |
+
"model.visual.blocks.14.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 512 |
+
"model.visual.blocks.14.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 513 |
+
"model.visual.blocks.14.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 514 |
+
"model.visual.blocks.14.norm1.bias": "model-00004-of-00004.safetensors",
|
| 515 |
+
"model.visual.blocks.14.norm1.weight": "model-00004-of-00004.safetensors",
|
| 516 |
+
"model.visual.blocks.14.norm2.bias": "model-00004-of-00004.safetensors",
|
| 517 |
+
"model.visual.blocks.14.norm2.weight": "model-00004-of-00004.safetensors",
|
| 518 |
+
"model.visual.blocks.15.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 519 |
+
"model.visual.blocks.15.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 520 |
+
"model.visual.blocks.15.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 521 |
+
"model.visual.blocks.15.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 522 |
+
"model.visual.blocks.15.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 523 |
+
"model.visual.blocks.15.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 524 |
+
"model.visual.blocks.15.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 525 |
+
"model.visual.blocks.15.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 526 |
+
"model.visual.blocks.15.norm1.bias": "model-00004-of-00004.safetensors",
|
| 527 |
+
"model.visual.blocks.15.norm1.weight": "model-00004-of-00004.safetensors",
|
| 528 |
+
"model.visual.blocks.15.norm2.bias": "model-00004-of-00004.safetensors",
|
| 529 |
+
"model.visual.blocks.15.norm2.weight": "model-00004-of-00004.safetensors",
|
| 530 |
+
"model.visual.blocks.16.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 531 |
+
"model.visual.blocks.16.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 532 |
+
"model.visual.blocks.16.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 533 |
+
"model.visual.blocks.16.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 534 |
+
"model.visual.blocks.16.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 535 |
+
"model.visual.blocks.16.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 536 |
+
"model.visual.blocks.16.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 537 |
+
"model.visual.blocks.16.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 538 |
+
"model.visual.blocks.16.norm1.bias": "model-00004-of-00004.safetensors",
|
| 539 |
+
"model.visual.blocks.16.norm1.weight": "model-00004-of-00004.safetensors",
|
| 540 |
+
"model.visual.blocks.16.norm2.bias": "model-00004-of-00004.safetensors",
|
| 541 |
+
"model.visual.blocks.16.norm2.weight": "model-00004-of-00004.safetensors",
|
| 542 |
+
"model.visual.blocks.17.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 543 |
+
"model.visual.blocks.17.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 544 |
+
"model.visual.blocks.17.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 545 |
+
"model.visual.blocks.17.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 546 |
+
"model.visual.blocks.17.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 547 |
+
"model.visual.blocks.17.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 548 |
+
"model.visual.blocks.17.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 549 |
+
"model.visual.blocks.17.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 550 |
+
"model.visual.blocks.17.norm1.bias": "model-00004-of-00004.safetensors",
|
| 551 |
+
"model.visual.blocks.17.norm1.weight": "model-00004-of-00004.safetensors",
|
| 552 |
+
"model.visual.blocks.17.norm2.bias": "model-00004-of-00004.safetensors",
|
| 553 |
+
"model.visual.blocks.17.norm2.weight": "model-00004-of-00004.safetensors",
|
| 554 |
+
"model.visual.blocks.18.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 555 |
+
"model.visual.blocks.18.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 556 |
+
"model.visual.blocks.18.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 557 |
+
"model.visual.blocks.18.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 558 |
+
"model.visual.blocks.18.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 559 |
+
"model.visual.blocks.18.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 560 |
+
"model.visual.blocks.18.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 561 |
+
"model.visual.blocks.18.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 562 |
+
"model.visual.blocks.18.norm1.bias": "model-00004-of-00004.safetensors",
|
| 563 |
+
"model.visual.blocks.18.norm1.weight": "model-00004-of-00004.safetensors",
|
| 564 |
+
"model.visual.blocks.18.norm2.bias": "model-00004-of-00004.safetensors",
|
| 565 |
+
"model.visual.blocks.18.norm2.weight": "model-00004-of-00004.safetensors",
|
| 566 |
+
"model.visual.blocks.19.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 567 |
+
"model.visual.blocks.19.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 568 |
+
"model.visual.blocks.19.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 569 |
+
"model.visual.blocks.19.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 570 |
+
"model.visual.blocks.19.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 571 |
+
"model.visual.blocks.19.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 572 |
+
"model.visual.blocks.19.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 573 |
+
"model.visual.blocks.19.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 574 |
+
"model.visual.blocks.19.norm1.bias": "model-00004-of-00004.safetensors",
|
| 575 |
+
"model.visual.blocks.19.norm1.weight": "model-00004-of-00004.safetensors",
|
| 576 |
+
"model.visual.blocks.19.norm2.bias": "model-00004-of-00004.safetensors",
|
| 577 |
+
"model.visual.blocks.19.norm2.weight": "model-00004-of-00004.safetensors",
|
| 578 |
+
"model.visual.blocks.2.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 579 |
+
"model.visual.blocks.2.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 580 |
+
"model.visual.blocks.2.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 581 |
+
"model.visual.blocks.2.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 582 |
+
"model.visual.blocks.2.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 583 |
+
"model.visual.blocks.2.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 584 |
+
"model.visual.blocks.2.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 585 |
+
"model.visual.blocks.2.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 586 |
+
"model.visual.blocks.2.norm1.bias": "model-00004-of-00004.safetensors",
|
| 587 |
+
"model.visual.blocks.2.norm1.weight": "model-00004-of-00004.safetensors",
|
| 588 |
+
"model.visual.blocks.2.norm2.bias": "model-00004-of-00004.safetensors",
|
| 589 |
+
"model.visual.blocks.2.norm2.weight": "model-00004-of-00004.safetensors",
|
| 590 |
+
"model.visual.blocks.20.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 591 |
+
"model.visual.blocks.20.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 592 |
+
"model.visual.blocks.20.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 593 |
+
"model.visual.blocks.20.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 594 |
+
"model.visual.blocks.20.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 595 |
+
"model.visual.blocks.20.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 596 |
+
"model.visual.blocks.20.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 597 |
+
"model.visual.blocks.20.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 598 |
+
"model.visual.blocks.20.norm1.bias": "model-00004-of-00004.safetensors",
|
| 599 |
+
"model.visual.blocks.20.norm1.weight": "model-00004-of-00004.safetensors",
|
| 600 |
+
"model.visual.blocks.20.norm2.bias": "model-00004-of-00004.safetensors",
|
| 601 |
+
"model.visual.blocks.20.norm2.weight": "model-00004-of-00004.safetensors",
|
| 602 |
+
"model.visual.blocks.21.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 603 |
+
"model.visual.blocks.21.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 604 |
+
"model.visual.blocks.21.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 605 |
+
"model.visual.blocks.21.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 606 |
+
"model.visual.blocks.21.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 607 |
+
"model.visual.blocks.21.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 608 |
+
"model.visual.blocks.21.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 609 |
+
"model.visual.blocks.21.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 610 |
+
"model.visual.blocks.21.norm1.bias": "model-00004-of-00004.safetensors",
|
| 611 |
+
"model.visual.blocks.21.norm1.weight": "model-00004-of-00004.safetensors",
|
| 612 |
+
"model.visual.blocks.21.norm2.bias": "model-00004-of-00004.safetensors",
|
| 613 |
+
"model.visual.blocks.21.norm2.weight": "model-00004-of-00004.safetensors",
|
| 614 |
+
"model.visual.blocks.22.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 615 |
+
"model.visual.blocks.22.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 616 |
+
"model.visual.blocks.22.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 617 |
+
"model.visual.blocks.22.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 618 |
+
"model.visual.blocks.22.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 619 |
+
"model.visual.blocks.22.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 620 |
+
"model.visual.blocks.22.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 621 |
+
"model.visual.blocks.22.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 622 |
+
"model.visual.blocks.22.norm1.bias": "model-00004-of-00004.safetensors",
|
| 623 |
+
"model.visual.blocks.22.norm1.weight": "model-00004-of-00004.safetensors",
|
| 624 |
+
"model.visual.blocks.22.norm2.bias": "model-00004-of-00004.safetensors",
|
| 625 |
+
"model.visual.blocks.22.norm2.weight": "model-00004-of-00004.safetensors",
|
| 626 |
+
"model.visual.blocks.23.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 627 |
+
"model.visual.blocks.23.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 628 |
+
"model.visual.blocks.23.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 629 |
+
"model.visual.blocks.23.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 630 |
+
"model.visual.blocks.23.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 631 |
+
"model.visual.blocks.23.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 632 |
+
"model.visual.blocks.23.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 633 |
+
"model.visual.blocks.23.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 634 |
+
"model.visual.blocks.23.norm1.bias": "model-00004-of-00004.safetensors",
|
| 635 |
+
"model.visual.blocks.23.norm1.weight": "model-00004-of-00004.safetensors",
|
| 636 |
+
"model.visual.blocks.23.norm2.bias": "model-00004-of-00004.safetensors",
|
| 637 |
+
"model.visual.blocks.23.norm2.weight": "model-00004-of-00004.safetensors",
|
| 638 |
+
"model.visual.blocks.24.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 639 |
+
"model.visual.blocks.24.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 640 |
+
"model.visual.blocks.24.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 641 |
+
"model.visual.blocks.24.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 642 |
+
"model.visual.blocks.24.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 643 |
+
"model.visual.blocks.24.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 644 |
+
"model.visual.blocks.24.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 645 |
+
"model.visual.blocks.24.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 646 |
+
"model.visual.blocks.24.norm1.bias": "model-00004-of-00004.safetensors",
|
| 647 |
+
"model.visual.blocks.24.norm1.weight": "model-00004-of-00004.safetensors",
|
| 648 |
+
"model.visual.blocks.24.norm2.bias": "model-00004-of-00004.safetensors",
|
| 649 |
+
"model.visual.blocks.24.norm2.weight": "model-00004-of-00004.safetensors",
|
| 650 |
+
"model.visual.blocks.25.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 651 |
+
"model.visual.blocks.25.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 652 |
+
"model.visual.blocks.25.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 653 |
+
"model.visual.blocks.25.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 654 |
+
"model.visual.blocks.25.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 655 |
+
"model.visual.blocks.25.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 656 |
+
"model.visual.blocks.25.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 657 |
+
"model.visual.blocks.25.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 658 |
+
"model.visual.blocks.25.norm1.bias": "model-00004-of-00004.safetensors",
|
| 659 |
+
"model.visual.blocks.25.norm1.weight": "model-00004-of-00004.safetensors",
|
| 660 |
+
"model.visual.blocks.25.norm2.bias": "model-00004-of-00004.safetensors",
|
| 661 |
+
"model.visual.blocks.25.norm2.weight": "model-00004-of-00004.safetensors",
|
| 662 |
+
"model.visual.blocks.26.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 663 |
+
"model.visual.blocks.26.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 664 |
+
"model.visual.blocks.26.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 665 |
+
"model.visual.blocks.26.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 666 |
+
"model.visual.blocks.26.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 667 |
+
"model.visual.blocks.26.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 668 |
+
"model.visual.blocks.26.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 669 |
+
"model.visual.blocks.26.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 670 |
+
"model.visual.blocks.26.norm1.bias": "model-00004-of-00004.safetensors",
|
| 671 |
+
"model.visual.blocks.26.norm1.weight": "model-00004-of-00004.safetensors",
|
| 672 |
+
"model.visual.blocks.26.norm2.bias": "model-00004-of-00004.safetensors",
|
| 673 |
+
"model.visual.blocks.26.norm2.weight": "model-00004-of-00004.safetensors",
|
| 674 |
+
"model.visual.blocks.3.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 675 |
+
"model.visual.blocks.3.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 676 |
+
"model.visual.blocks.3.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 677 |
+
"model.visual.blocks.3.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 678 |
+
"model.visual.blocks.3.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 679 |
+
"model.visual.blocks.3.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 680 |
+
"model.visual.blocks.3.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 681 |
+
"model.visual.blocks.3.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 682 |
+
"model.visual.blocks.3.norm1.bias": "model-00004-of-00004.safetensors",
|
| 683 |
+
"model.visual.blocks.3.norm1.weight": "model-00004-of-00004.safetensors",
|
| 684 |
+
"model.visual.blocks.3.norm2.bias": "model-00004-of-00004.safetensors",
|
| 685 |
+
"model.visual.blocks.3.norm2.weight": "model-00004-of-00004.safetensors",
|
| 686 |
+
"model.visual.blocks.4.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 687 |
+
"model.visual.blocks.4.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 688 |
+
"model.visual.blocks.4.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 689 |
+
"model.visual.blocks.4.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 690 |
+
"model.visual.blocks.4.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 691 |
+
"model.visual.blocks.4.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 692 |
+
"model.visual.blocks.4.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 693 |
+
"model.visual.blocks.4.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 694 |
+
"model.visual.blocks.4.norm1.bias": "model-00004-of-00004.safetensors",
|
| 695 |
+
"model.visual.blocks.4.norm1.weight": "model-00004-of-00004.safetensors",
|
| 696 |
+
"model.visual.blocks.4.norm2.bias": "model-00004-of-00004.safetensors",
|
| 697 |
+
"model.visual.blocks.4.norm2.weight": "model-00004-of-00004.safetensors",
|
| 698 |
+
"model.visual.blocks.5.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 699 |
+
"model.visual.blocks.5.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 700 |
+
"model.visual.blocks.5.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 701 |
+
"model.visual.blocks.5.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 702 |
+
"model.visual.blocks.5.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 703 |
+
"model.visual.blocks.5.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 704 |
+
"model.visual.blocks.5.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 705 |
+
"model.visual.blocks.5.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 706 |
+
"model.visual.blocks.5.norm1.bias": "model-00004-of-00004.safetensors",
|
| 707 |
+
"model.visual.blocks.5.norm1.weight": "model-00004-of-00004.safetensors",
|
| 708 |
+
"model.visual.blocks.5.norm2.bias": "model-00004-of-00004.safetensors",
|
| 709 |
+
"model.visual.blocks.5.norm2.weight": "model-00004-of-00004.safetensors",
|
| 710 |
+
"model.visual.blocks.6.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 711 |
+
"model.visual.blocks.6.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 712 |
+
"model.visual.blocks.6.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 713 |
+
"model.visual.blocks.6.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 714 |
+
"model.visual.blocks.6.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 715 |
+
"model.visual.blocks.6.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 716 |
+
"model.visual.blocks.6.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 717 |
+
"model.visual.blocks.6.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 718 |
+
"model.visual.blocks.6.norm1.bias": "model-00004-of-00004.safetensors",
|
| 719 |
+
"model.visual.blocks.6.norm1.weight": "model-00004-of-00004.safetensors",
|
| 720 |
+
"model.visual.blocks.6.norm2.bias": "model-00004-of-00004.safetensors",
|
| 721 |
+
"model.visual.blocks.6.norm2.weight": "model-00004-of-00004.safetensors",
|
| 722 |
+
"model.visual.blocks.7.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 723 |
+
"model.visual.blocks.7.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 724 |
+
"model.visual.blocks.7.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 725 |
+
"model.visual.blocks.7.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 726 |
+
"model.visual.blocks.7.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 727 |
+
"model.visual.blocks.7.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 728 |
+
"model.visual.blocks.7.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 729 |
+
"model.visual.blocks.7.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 730 |
+
"model.visual.blocks.7.norm1.bias": "model-00004-of-00004.safetensors",
|
| 731 |
+
"model.visual.blocks.7.norm1.weight": "model-00004-of-00004.safetensors",
|
| 732 |
+
"model.visual.blocks.7.norm2.bias": "model-00004-of-00004.safetensors",
|
| 733 |
+
"model.visual.blocks.7.norm2.weight": "model-00004-of-00004.safetensors",
|
| 734 |
+
"model.visual.blocks.8.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 735 |
+
"model.visual.blocks.8.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 736 |
+
"model.visual.blocks.8.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 737 |
+
"model.visual.blocks.8.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 738 |
+
"model.visual.blocks.8.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 739 |
+
"model.visual.blocks.8.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 740 |
+
"model.visual.blocks.8.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 741 |
+
"model.visual.blocks.8.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 742 |
+
"model.visual.blocks.8.norm1.bias": "model-00004-of-00004.safetensors",
|
| 743 |
+
"model.visual.blocks.8.norm1.weight": "model-00004-of-00004.safetensors",
|
| 744 |
+
"model.visual.blocks.8.norm2.bias": "model-00004-of-00004.safetensors",
|
| 745 |
+
"model.visual.blocks.8.norm2.weight": "model-00004-of-00004.safetensors",
|
| 746 |
+
"model.visual.blocks.9.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 747 |
+
"model.visual.blocks.9.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 748 |
+
"model.visual.blocks.9.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 749 |
+
"model.visual.blocks.9.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 750 |
+
"model.visual.blocks.9.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 751 |
+
"model.visual.blocks.9.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 752 |
+
"model.visual.blocks.9.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 753 |
+
"model.visual.blocks.9.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 754 |
+
"model.visual.blocks.9.norm1.bias": "model-00004-of-00004.safetensors",
|
| 755 |
+
"model.visual.blocks.9.norm1.weight": "model-00004-of-00004.safetensors",
|
| 756 |
+
"model.visual.blocks.9.norm2.bias": "model-00004-of-00004.safetensors",
|
| 757 |
+
"model.visual.blocks.9.norm2.weight": "model-00004-of-00004.safetensors",
|
| 758 |
+
"model.visual.merger.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 759 |
+
"model.visual.merger.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 760 |
+
"model.visual.merger.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 761 |
+
"model.visual.merger.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 762 |
+
"model.visual.merger.norm.bias": "model-00004-of-00004.safetensors",
|
| 763 |
+
"model.visual.merger.norm.weight": "model-00004-of-00004.safetensors",
|
| 764 |
+
"model.visual.patch_embed.proj.bias": "model-00004-of-00004.safetensors",
|
| 765 |
+
"model.visual.patch_embed.proj.weight": "model-00004-of-00004.safetensors",
|
| 766 |
+
"model.visual.pos_embed.weight": "model-00004-of-00004.safetensors"
|
| 767 |
+
}
|
| 768 |
+
}
|
c1_no_history/round_1/preprocessor_config.json
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"size": {
|
| 3 |
+
"longest_edge": 16777216,
|
| 4 |
+
"shortest_edge": 65536
|
| 5 |
+
},
|
| 6 |
+
"patch_size": 16,
|
| 7 |
+
"temporal_patch_size": 2,
|
| 8 |
+
"merge_size": 2,
|
| 9 |
+
"image_mean": [
|
| 10 |
+
0.5,
|
| 11 |
+
0.5,
|
| 12 |
+
0.5
|
| 13 |
+
],
|
| 14 |
+
"image_std": [
|
| 15 |
+
0.5,
|
| 16 |
+
0.5,
|
| 17 |
+
0.5
|
| 18 |
+
],
|
| 19 |
+
"processor_class": "Qwen3VLProcessor",
|
| 20 |
+
"image_processor_type": "Qwen2VLImageProcessorFast"
|
| 21 |
+
}
|
c1_no_history/round_1/processor_config.json
ADDED
|
@@ -0,0 +1,60 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"image_processor": {
|
| 3 |
+
"do_convert_rgb": true,
|
| 4 |
+
"do_normalize": true,
|
| 5 |
+
"do_rescale": true,
|
| 6 |
+
"do_resize": true,
|
| 7 |
+
"image_mean": [
|
| 8 |
+
0.5,
|
| 9 |
+
0.5,
|
| 10 |
+
0.5
|
| 11 |
+
],
|
| 12 |
+
"image_processor_type": "Qwen2VLImageProcessor",
|
| 13 |
+
"image_std": [
|
| 14 |
+
0.5,
|
| 15 |
+
0.5,
|
| 16 |
+
0.5
|
| 17 |
+
],
|
| 18 |
+
"merge_size": 2,
|
| 19 |
+
"patch_size": 16,
|
| 20 |
+
"resample": 3,
|
| 21 |
+
"rescale_factor": 0.00392156862745098,
|
| 22 |
+
"size": {
|
| 23 |
+
"longest_edge": 16777216,
|
| 24 |
+
"shortest_edge": 65536
|
| 25 |
+
},
|
| 26 |
+
"temporal_patch_size": 2
|
| 27 |
+
},
|
| 28 |
+
"processor_class": "Qwen3VLProcessor",
|
| 29 |
+
"video_processor": {
|
| 30 |
+
"do_convert_rgb": true,
|
| 31 |
+
"do_normalize": true,
|
| 32 |
+
"do_rescale": true,
|
| 33 |
+
"do_resize": true,
|
| 34 |
+
"do_sample_frames": true,
|
| 35 |
+
"fps": 2,
|
| 36 |
+
"image_mean": [
|
| 37 |
+
0.5,
|
| 38 |
+
0.5,
|
| 39 |
+
0.5
|
| 40 |
+
],
|
| 41 |
+
"image_std": [
|
| 42 |
+
0.5,
|
| 43 |
+
0.5,
|
| 44 |
+
0.5
|
| 45 |
+
],
|
| 46 |
+
"max_frames": 768,
|
| 47 |
+
"merge_size": 2,
|
| 48 |
+
"min_frames": 4,
|
| 49 |
+
"patch_size": 16,
|
| 50 |
+
"resample": 3,
|
| 51 |
+
"rescale_factor": 0.00392156862745098,
|
| 52 |
+
"return_metadata": false,
|
| 53 |
+
"size": {
|
| 54 |
+
"longest_edge": 25165824,
|
| 55 |
+
"shortest_edge": 4096
|
| 56 |
+
},
|
| 57 |
+
"temporal_patch_size": 2,
|
| 58 |
+
"video_processor_type": "Qwen3VLVideoProcessor"
|
| 59 |
+
}
|
| 60 |
+
}
|
c1_no_history/round_1/tokenizer_config.json
ADDED
|
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"audio_bos_token": "<|audio_start|>",
|
| 4 |
+
"audio_eos_token": "<|audio_end|>",
|
| 5 |
+
"audio_token": "<|audio_pad|>",
|
| 6 |
+
"backend": "tokenizers",
|
| 7 |
+
"bos_token": null,
|
| 8 |
+
"clean_up_tokenization_spaces": false,
|
| 9 |
+
"eos_token": "<|im_end|>",
|
| 10 |
+
"errors": "replace",
|
| 11 |
+
"image_token": "<|image_pad|>",
|
| 12 |
+
"is_local": true,
|
| 13 |
+
"local_files_only": true,
|
| 14 |
+
"model_max_length": 262144,
|
| 15 |
+
"model_specific_special_tokens": {
|
| 16 |
+
"audio_bos_token": "<|audio_start|>",
|
| 17 |
+
"audio_eos_token": "<|audio_end|>",
|
| 18 |
+
"audio_token": "<|audio_pad|>",
|
| 19 |
+
"image_token": "<|image_pad|>",
|
| 20 |
+
"video_token": "<|video_pad|>",
|
| 21 |
+
"vision_bos_token": "<|vision_start|>",
|
| 22 |
+
"vision_eos_token": "<|vision_end|>"
|
| 23 |
+
},
|
| 24 |
+
"pad_token": "<|endoftext|>",
|
| 25 |
+
"pretokenize_regex": "(?i:'s|'t|'re|'ve|'m|'ll|'d)|[^\\r\\n\\p{L}\\p{N}]?[\\p{L}\\p{M}]+|\\p{N}| ?[^\\s\\p{L}\\p{M}\\p{N}]+[\\r\\n]*|\\s*[\\r\\n]+|\\s+(?!\\S)|\\s+",
|
| 26 |
+
"processor_class": "Qwen3VLProcessor",
|
| 27 |
+
"split_special_tokens": false,
|
| 28 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 29 |
+
"unk_token": null,
|
| 30 |
+
"video_token": "<|video_pad|>",
|
| 31 |
+
"vision_bos_token": "<|vision_start|>",
|
| 32 |
+
"vision_eos_token": "<|vision_end|>"
|
| 33 |
+
}
|
c1_no_history/round_1/trainer_state.json
ADDED
|
@@ -0,0 +1,2454 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 1.0,
|
| 6 |
+
"eval_steps": 80.0,
|
| 7 |
+
"global_step": 160,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"clip_ratio/high_max": 0.0,
|
| 14 |
+
"clip_ratio/high_mean": 0.0,
|
| 15 |
+
"clip_ratio/low_mean": 0.0,
|
| 16 |
+
"clip_ratio/low_min": 0.0,
|
| 17 |
+
"clip_ratio/region_mean": 0.0,
|
| 18 |
+
"completions/clipped_ratio": 0.57421875,
|
| 19 |
+
"completions/max_length": 2048.0,
|
| 20 |
+
"completions/mean_length": 1464.69140625,
|
| 21 |
+
"completions/min_length": 95.0,
|
| 22 |
+
"epoch": 0.00625,
|
| 23 |
+
"frac_reward_zero_std": 0.375,
|
| 24 |
+
"grad_norm": 1.2399364462606743,
|
| 25 |
+
"kl": 0.0,
|
| 26 |
+
"learning_rate": 2e-07,
|
| 27 |
+
"loss": 3.725290298461914e-09,
|
| 28 |
+
"reward": 0.28515625,
|
| 29 |
+
"reward_std": 0.28235548734664917,
|
| 30 |
+
"rewards/MajorityAccuracy/mean": 0.28515625,
|
| 31 |
+
"rewards/MajorityAccuracy/std": 0.4523732364177704,
|
| 32 |
+
"step": 1,
|
| 33 |
+
"step_time": 591.966953407973
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"clip_ratio/high_max": 0.0,
|
| 37 |
+
"clip_ratio/high_mean": 0.0,
|
| 38 |
+
"clip_ratio/low_mean": 0.0,
|
| 39 |
+
"clip_ratio/low_min": 0.0,
|
| 40 |
+
"clip_ratio/region_mean": 0.0,
|
| 41 |
+
"epoch": 0.0125,
|
| 42 |
+
"grad_norm": 2.7465013611479128,
|
| 43 |
+
"kl": 0.0,
|
| 44 |
+
"learning_rate": 4e-07,
|
| 45 |
+
"loss": -1.1175870895385742e-08,
|
| 46 |
+
"step": 2,
|
| 47 |
+
"step_time": 75.55144402757287
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"clip_ratio/high_max": 0.002859351414372213,
|
| 51 |
+
"clip_ratio/high_mean": 0.0007148378535930533,
|
| 52 |
+
"clip_ratio/low_mean": 0.00039386800926877186,
|
| 53 |
+
"clip_ratio/low_min": 0.0,
|
| 54 |
+
"clip_ratio/region_mean": 0.001108705866499804,
|
| 55 |
+
"epoch": 0.01875,
|
| 56 |
+
"grad_norm": 3.370540285978387,
|
| 57 |
+
"kl": 0.0002360330654482823,
|
| 58 |
+
"learning_rate": 6e-07,
|
| 59 |
+
"loss": 0.00017203204333782196,
|
| 60 |
+
"step": 3,
|
| 61 |
+
"step_time": 75.68978300318122
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"clip_ratio/high_max": 0.0005713289356208406,
|
| 65 |
+
"clip_ratio/high_mean": 0.00016763382336648647,
|
| 66 |
+
"clip_ratio/low_mean": 0.00013663022582477424,
|
| 67 |
+
"clip_ratio/low_min": 0.0,
|
| 68 |
+
"clip_ratio/region_mean": 0.0003042640473722713,
|
| 69 |
+
"epoch": 0.025,
|
| 70 |
+
"grad_norm": 1.3988420596002984,
|
| 71 |
+
"kl": 0.000188751231689821,
|
| 72 |
+
"learning_rate": 8e-07,
|
| 73 |
+
"loss": -1.964731200132519e-05,
|
| 74 |
+
"step": 4,
|
| 75 |
+
"step_time": 75.72735524177551
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"clip_ratio/high_max": 0.0005696416774298996,
|
| 79 |
+
"clip_ratio/high_mean": 0.00019831381359836087,
|
| 80 |
+
"clip_ratio/low_mean": 7.62939453125e-05,
|
| 81 |
+
"clip_ratio/low_min": 0.0,
|
| 82 |
+
"clip_ratio/region_mean": 0.00027460775891086087,
|
| 83 |
+
"epoch": 0.03125,
|
| 84 |
+
"grad_norm": 1.0278770898791965,
|
| 85 |
+
"kl": 0.00018560621901997365,
|
| 86 |
+
"learning_rate": 1e-06,
|
| 87 |
+
"loss": 8.074815559666604e-05,
|
| 88 |
+
"step": 5,
|
| 89 |
+
"step_time": 75.57767974585295
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"clip_ratio/high_max": 0.00018939393339678645,
|
| 93 |
+
"clip_ratio/high_mean": 4.734848334919661e-05,
|
| 94 |
+
"clip_ratio/low_mean": 0.00011041947254852857,
|
| 95 |
+
"clip_ratio/low_min": 0.0,
|
| 96 |
+
"clip_ratio/region_mean": 0.0001577679558977252,
|
| 97 |
+
"epoch": 0.0375,
|
| 98 |
+
"grad_norm": 0.7200135276848976,
|
| 99 |
+
"kl": 0.0001778916866896907,
|
| 100 |
+
"learning_rate": 9.998973021172564e-07,
|
| 101 |
+
"loss": 3.497612487990409e-05,
|
| 102 |
+
"step": 6,
|
| 103 |
+
"step_time": 74.54965971782804
|
| 104 |
+
},
|
| 105 |
+
{
|
| 106 |
+
"clip_ratio/high_max": 0.005128531236550771,
|
| 107 |
+
"clip_ratio/high_mean": 0.0015206824427878018,
|
| 108 |
+
"clip_ratio/low_mean": 0.0007454388542100787,
|
| 109 |
+
"clip_ratio/low_min": 0.0,
|
| 110 |
+
"clip_ratio/region_mean": 0.0022661213151877746,
|
| 111 |
+
"epoch": 0.04375,
|
| 112 |
+
"grad_norm": 3.1150883150510786,
|
| 113 |
+
"kl": 0.0005040397882112302,
|
| 114 |
+
"learning_rate": 9.99589250656446e-07,
|
| 115 |
+
"loss": 6.40491780359298e-05,
|
| 116 |
+
"step": 7,
|
| 117 |
+
"step_time": 75.48397311195731
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"clip_ratio/high_max": 0.0012702984822681174,
|
| 121 |
+
"clip_ratio/high_mean": 0.00033908184923348017,
|
| 122 |
+
"clip_ratio/low_mean": 0.0003645766410045326,
|
| 123 |
+
"clip_ratio/low_min": 0.0,
|
| 124 |
+
"clip_ratio/region_mean": 0.0007036584793240763,
|
| 125 |
+
"epoch": 0.05,
|
| 126 |
+
"grad_norm": 1.5355274633199465,
|
| 127 |
+
"kl": 0.00022402624017558992,
|
| 128 |
+
"learning_rate": 9.990759721625005e-07,
|
| 129 |
+
"loss": 4.699327109847218e-05,
|
| 130 |
+
"step": 8,
|
| 131 |
+
"step_time": 74.2727967388928
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"clip_ratio/high_max": 0.0,
|
| 135 |
+
"clip_ratio/high_mean": 0.0,
|
| 136 |
+
"clip_ratio/low_mean": 0.0,
|
| 137 |
+
"clip_ratio/low_min": 0.0,
|
| 138 |
+
"clip_ratio/region_mean": 0.0,
|
| 139 |
+
"completions/clipped_ratio": 0.640625,
|
| 140 |
+
"completions/max_length": 2048.0,
|
| 141 |
+
"completions/mean_length": 1590.27734375,
|
| 142 |
+
"completions/min_length": 8.0,
|
| 143 |
+
"epoch": 0.05625,
|
| 144 |
+
"frac_reward_zero_std": 0.53125,
|
| 145 |
+
"grad_norm": 1.9408762512249234,
|
| 146 |
+
"kl": 0.00039431219920516014,
|
| 147 |
+
"learning_rate": 9.983576774858775e-07,
|
| 148 |
+
"loss": 3.95067036151886e-06,
|
| 149 |
+
"reward": 0.26171875,
|
| 150 |
+
"reward_std": 0.20832079648971558,
|
| 151 |
+
"rewards/MajorityAccuracy/mean": 0.26171875,
|
| 152 |
+
"rewards/MajorityAccuracy/std": 0.4404313564300537,
|
| 153 |
+
"step": 9,
|
| 154 |
+
"step_time": 524.8124241456389
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"clip_ratio/high_max": 0.0,
|
| 158 |
+
"clip_ratio/high_mean": 0.0,
|
| 159 |
+
"clip_ratio/low_mean": 0.0,
|
| 160 |
+
"clip_ratio/low_min": 0.0,
|
| 161 |
+
"clip_ratio/region_mean": 0.0,
|
| 162 |
+
"epoch": 0.0625,
|
| 163 |
+
"grad_norm": 0.005279622830226628,
|
| 164 |
+
"kl": 0.0005260705947875977,
|
| 165 |
+
"learning_rate": 9.974346616959475e-07,
|
| 166 |
+
"loss": 5.264329956844449e-06,
|
| 167 |
+
"step": 10,
|
| 168 |
+
"step_time": 75.69239982962608
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"clip_ratio/high_max": 0.0015005267050582916,
|
| 172 |
+
"clip_ratio/high_mean": 0.00043054012348875403,
|
| 173 |
+
"clip_ratio/low_mean": 0.0008410179434577003,
|
| 174 |
+
"clip_ratio/low_min": 0.0,
|
| 175 |
+
"clip_ratio/region_mean": 0.0012715580596704967,
|
| 176 |
+
"epoch": 0.06875,
|
| 177 |
+
"grad_norm": 1.5397558988398616,
|
| 178 |
+
"kl": 0.000536054987605894,
|
| 179 |
+
"learning_rate": 9.963073039597796e-07,
|
| 180 |
+
"loss": 6.436955300159752e-05,
|
| 181 |
+
"step": 11,
|
| 182 |
+
"step_time": 75.62876082211733
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"clip_ratio/high_max": 0.0003669809229904786,
|
| 186 |
+
"clip_ratio/high_mean": 9.174523074761964e-05,
|
| 187 |
+
"clip_ratio/low_mean": 0.0007668163743801415,
|
| 188 |
+
"clip_ratio/low_min": 0.0,
|
| 189 |
+
"clip_ratio/region_mean": 0.0008585616014897823,
|
| 190 |
+
"epoch": 0.075,
|
| 191 |
+
"grad_norm": 4.0518895484255015,
|
| 192 |
+
"kl": 0.0010140740778297186,
|
| 193 |
+
"learning_rate": 9.949760673863846e-07,
|
| 194 |
+
"loss": 0.000598676735535264,
|
| 195 |
+
"step": 12,
|
| 196 |
+
"step_time": 73.49617353454232
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"clip_ratio/high_max": 0.0021458656701724976,
|
| 200 |
+
"clip_ratio/high_mean": 0.0005851803653058596,
|
| 201 |
+
"clip_ratio/low_mean": 0.0015239890599332284,
|
| 202 |
+
"clip_ratio/low_min": 0.0,
|
| 203 |
+
"clip_ratio/region_mean": 0.002109169421601109,
|
| 204 |
+
"epoch": 0.08125,
|
| 205 |
+
"grad_norm": 1.315436735900677,
|
| 206 |
+
"kl": 0.0012970006500836462,
|
| 207 |
+
"learning_rate": 9.934414988364722e-07,
|
| 208 |
+
"loss": 8.979473204817623e-05,
|
| 209 |
+
"step": 13,
|
| 210 |
+
"step_time": 75.8074538744986
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"clip_ratio/high_max": 0.0,
|
| 214 |
+
"clip_ratio/high_mean": 0.0,
|
| 215 |
+
"clip_ratio/low_mean": 0.0,
|
| 216 |
+
"clip_ratio/low_min": 0.0,
|
| 217 |
+
"clip_ratio/region_mean": 0.0,
|
| 218 |
+
"epoch": 0.0875,
|
| 219 |
+
"grad_norm": 0.013813342405294122,
|
| 220 |
+
"kl": 0.0010006478769355454,
|
| 221 |
+
"learning_rate": 9.917042286978063e-07,
|
| 222 |
+
"loss": 1.0007892342400737e-05,
|
| 223 |
+
"step": 14,
|
| 224 |
+
"step_time": 75.56084407120943
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"clip_ratio/high_max": 0.0023410479188896716,
|
| 228 |
+
"clip_ratio/high_mean": 0.0005852619797224179,
|
| 229 |
+
"clip_ratio/low_mean": 0.004099041507288348,
|
| 230 |
+
"clip_ratio/low_min": 0.0,
|
| 231 |
+
"clip_ratio/region_mean": 0.004684303465182893,
|
| 232 |
+
"epoch": 0.09375,
|
| 233 |
+
"grad_norm": 11.146355872065172,
|
| 234 |
+
"kl": 0.0013000990948057733,
|
| 235 |
+
"learning_rate": 9.897649706262473e-07,
|
| 236 |
+
"loss": 0.0003401850117370486,
|
| 237 |
+
"step": 15,
|
| 238 |
+
"step_time": 75.5984579026699
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"clip_ratio/high_max": 0.0029852829757146537,
|
| 242 |
+
"clip_ratio/high_mean": 0.0008241344912676141,
|
| 243 |
+
"clip_ratio/low_mean": 0.0014801025390625,
|
| 244 |
+
"clip_ratio/low_min": 0.0,
|
| 245 |
+
"clip_ratio/region_mean": 0.002304237030330114,
|
| 246 |
+
"epoch": 0.1,
|
| 247 |
+
"grad_norm": 1.3631621984566498,
|
| 248 |
+
"kl": 0.0012737210636259988,
|
| 249 |
+
"learning_rate": 9.87624521252587e-07,
|
| 250 |
+
"loss": -0.0001407551026204601,
|
| 251 |
+
"step": 16,
|
| 252 |
+
"step_time": 75.62825842946768
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"clip_ratio/high_max": 0.0,
|
| 256 |
+
"clip_ratio/high_mean": 0.0,
|
| 257 |
+
"clip_ratio/low_mean": 0.0,
|
| 258 |
+
"clip_ratio/low_min": 0.0,
|
| 259 |
+
"clip_ratio/region_mean": 0.0,
|
| 260 |
+
"completions/clipped_ratio": 0.5234375,
|
| 261 |
+
"completions/max_length": 2048.0,
|
| 262 |
+
"completions/mean_length": 1496.90234375,
|
| 263 |
+
"completions/min_length": 123.0,
|
| 264 |
+
"epoch": 0.10625,
|
| 265 |
+
"frac_reward_zero_std": 0.34375,
|
| 266 |
+
"grad_norm": 1.8554656983456794,
|
| 267 |
+
"kl": 0.001679333989159204,
|
| 268 |
+
"learning_rate": 9.852837598553008e-07,
|
| 269 |
+
"loss": 1.6797224816400558e-05,
|
| 270 |
+
"reward": 0.35546875,
|
| 271 |
+
"reward_std": 0.3041841983795166,
|
| 272 |
+
"rewards/MajorityAccuracy/mean": 0.35546875,
|
| 273 |
+
"rewards/MajorityAccuracy/std": 0.4795927405357361,
|
| 274 |
+
"step": 17,
|
| 275 |
+
"step_time": 521.3555744327605
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"clip_ratio/high_max": 0.00019771781808231026,
|
| 279 |
+
"clip_ratio/high_mean": 4.9429454520577565e-05,
|
| 280 |
+
"clip_ratio/low_mean": 0.000244140625,
|
| 281 |
+
"clip_ratio/low_min": 0.0,
|
| 282 |
+
"clip_ratio/region_mean": 0.00029357008315855637,
|
| 283 |
+
"epoch": 0.1125,
|
| 284 |
+
"grad_norm": 1.3291591784145218,
|
| 285 |
+
"kl": 0.002250333927804604,
|
| 286 |
+
"learning_rate": 9.827436479993468e-07,
|
| 287 |
+
"loss": 0.00010031653800979257,
|
| 288 |
+
"step": 18,
|
| 289 |
+
"step_time": 75.43238198384643
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"clip_ratio/high_max": 0.00137993957469007,
|
| 293 |
+
"clip_ratio/high_mean": 0.0004913946704618866,
|
| 294 |
+
"clip_ratio/low_mean": 0.000579833984375,
|
| 295 |
+
"clip_ratio/low_min": 0.0,
|
| 296 |
+
"clip_ratio/region_mean": 0.0010712286457419395,
|
| 297 |
+
"epoch": 0.11875,
|
| 298 |
+
"grad_norm": 3.5537448131974254,
|
| 299 |
+
"kl": 0.0014982544671511278,
|
| 300 |
+
"learning_rate": 9.800052291411629e-07,
|
| 301 |
+
"loss": 0.00011709332466125488,
|
| 302 |
+
"step": 19,
|
| 303 |
+
"step_time": 75.50367458537221
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"clip_ratio/high_max": 0.001593625325767789,
|
| 307 |
+
"clip_ratio/high_mean": 0.00039840633144194726,
|
| 308 |
+
"clip_ratio/low_mean": 0.0003543051097949501,
|
| 309 |
+
"clip_ratio/low_min": 0.0,
|
| 310 |
+
"clip_ratio/region_mean": 0.0007527114430558868,
|
| 311 |
+
"epoch": 0.125,
|
| 312 |
+
"grad_norm": 1.2719539516633678,
|
| 313 |
+
"kl": 0.002109079243382439,
|
| 314 |
+
"learning_rate": 9.770696282000244e-07,
|
| 315 |
+
"loss": 0.00018528600048739463,
|
| 316 |
+
"step": 20,
|
| 317 |
+
"step_time": 73.99633759260178
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"clip_ratio/high_max": 0.0056748297065496445,
|
| 321 |
+
"clip_ratio/high_mean": 0.0015474475221708417,
|
| 322 |
+
"clip_ratio/low_mean": 0.0006358166283462197,
|
| 323 |
+
"clip_ratio/low_min": 0.0,
|
| 324 |
+
"clip_ratio/region_mean": 0.002183264135965146,
|
| 325 |
+
"epoch": 0.13125,
|
| 326 |
+
"grad_norm": 1.2848392156933561,
|
| 327 |
+
"kl": 0.0027680258644977584,
|
| 328 |
+
"learning_rate": 9.739380510959364e-07,
|
| 329 |
+
"loss": 0.00012694425822701305,
|
| 330 |
+
"step": 21,
|
| 331 |
+
"step_time": 75.34346482530236
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"clip_ratio/high_max": 0.0039784511027392,
|
| 335 |
+
"clip_ratio/high_mean": 0.0009946127756848,
|
| 336 |
+
"clip_ratio/low_mean": 0.002669235836947337,
|
| 337 |
+
"clip_ratio/low_min": 0.0,
|
| 338 |
+
"clip_ratio/region_mean": 0.003663848612632137,
|
| 339 |
+
"epoch": 0.1375,
|
| 340 |
+
"grad_norm": 2.874897879658082,
|
| 341 |
+
"kl": 0.0033136552665382624,
|
| 342 |
+
"learning_rate": 9.706117842542516e-07,
|
| 343 |
+
"loss": 0.00021891400683671236,
|
| 344 |
+
"step": 22,
|
| 345 |
+
"step_time": 73.82131819799542
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"clip_ratio/high_max": 0.003953218256356195,
|
| 349 |
+
"clip_ratio/high_mean": 0.0015122052645892836,
|
| 350 |
+
"clip_ratio/low_mean": 0.00103759765625,
|
| 351 |
+
"clip_ratio/low_min": 0.0,
|
| 352 |
+
"clip_ratio/region_mean": 0.0025498028990114108,
|
| 353 |
+
"epoch": 0.14375,
|
| 354 |
+
"grad_norm": 1.0757543199661113,
|
| 355 |
+
"kl": 0.0022764830209780484,
|
| 356 |
+
"learning_rate": 9.670921940772186e-07,
|
| 357 |
+
"loss": 0.00029358870233409107,
|
| 358 |
+
"step": 23,
|
| 359 |
+
"step_time": 75.35770279169083
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"clip_ratio/high_max": 0.00441563525237143,
|
| 363 |
+
"clip_ratio/high_mean": 0.0013943721714895219,
|
| 364 |
+
"clip_ratio/low_mean": 0.0011307362947263755,
|
| 365 |
+
"clip_ratio/low_min": 0.0,
|
| 366 |
+
"clip_ratio/region_mean": 0.0025251084662158974,
|
| 367 |
+
"epoch": 0.15,
|
| 368 |
+
"grad_norm": 3.6329321603278695,
|
| 369 |
+
"kl": 0.003897866321494803,
|
| 370 |
+
"learning_rate": 9.633807263826744e-07,
|
| 371 |
+
"loss": 0.0005067691090516746,
|
| 372 |
+
"step": 24,
|
| 373 |
+
"step_time": 73.82892765849829
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"clip_ratio/high_max": 0.0,
|
| 377 |
+
"clip_ratio/high_mean": 0.0,
|
| 378 |
+
"clip_ratio/low_mean": 0.0,
|
| 379 |
+
"clip_ratio/low_min": 0.0,
|
| 380 |
+
"clip_ratio/region_mean": 0.0,
|
| 381 |
+
"completions/clipped_ratio": 0.5390625,
|
| 382 |
+
"completions/max_length": 2048.0,
|
| 383 |
+
"completions/mean_length": 1486.6171875,
|
| 384 |
+
"completions/min_length": 66.0,
|
| 385 |
+
"epoch": 0.15625,
|
| 386 |
+
"frac_reward_zero_std": 0.25,
|
| 387 |
+
"grad_norm": 2.323576138343025,
|
| 388 |
+
"kl": 0.0037999413325451314,
|
| 389 |
+
"learning_rate": 9.594789058101153e-07,
|
| 390 |
+
"loss": 3.8001686334609985e-05,
|
| 391 |
+
"reward": 0.29296875,
|
| 392 |
+
"reward_std": 0.34704914689064026,
|
| 393 |
+
"rewards/MajorityAccuracy/mean": 0.29296875,
|
| 394 |
+
"rewards/MajorityAccuracy/std": 0.45601576566696167,
|
| 395 |
+
"step": 25,
|
| 396 |
+
"step_time": 524.2894837036729
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"clip_ratio/high_max": 0.0007718996494077146,
|
| 400 |
+
"clip_ratio/high_mean": 0.000208369005122222,
|
| 401 |
+
"clip_ratio/low_mean": 0.0003662109375,
|
| 402 |
+
"clip_ratio/low_min": 0.0,
|
| 403 |
+
"clip_ratio/region_mean": 0.000574579942622222,
|
| 404 |
+
"epoch": 0.1625,
|
| 405 |
+
"grad_norm": 1.6553455817958194,
|
| 406 |
+
"kl": 0.0016797514108475298,
|
| 407 |
+
"learning_rate": 9.55388335194388e-07,
|
| 408 |
+
"loss": -0.0001247953623533249,
|
| 409 |
+
"step": 26,
|
| 410 |
+
"step_time": 74.6104701012373
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"clip_ratio/high_max": 0.001399816683260724,
|
| 414 |
+
"clip_ratio/high_mean": 0.0005214701050135773,
|
| 415 |
+
"clip_ratio/low_mean": 0.000274658203125,
|
| 416 |
+
"clip_ratio/low_min": 0.0,
|
| 417 |
+
"clip_ratio/region_mean": 0.0007961283117765561,
|
| 418 |
+
"epoch": 0.16875,
|
| 419 |
+
"grad_norm": 1.3733399126853045,
|
| 420 |
+
"kl": 0.0033611570834182203,
|
| 421 |
+
"learning_rate": 9.511106949072587e-07,
|
| 422 |
+
"loss": -0.00013023750216234475,
|
| 423 |
+
"step": 27,
|
| 424 |
+
"step_time": 74.61883232370019
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"clip_ratio/high_max": 8.638562576379627e-05,
|
| 428 |
+
"clip_ratio/high_mean": 2.1596406440949067e-05,
|
| 429 |
+
"clip_ratio/low_mean": 9.1552734375e-05,
|
| 430 |
+
"clip_ratio/low_min": 0.0,
|
| 431 |
+
"clip_ratio/region_mean": 0.00011314914081594907,
|
| 432 |
+
"epoch": 0.175,
|
| 433 |
+
"grad_norm": 0.7910584700721907,
|
| 434 |
+
"kl": 0.002703009929973632,
|
| 435 |
+
"learning_rate": 9.466477421671295e-07,
|
| 436 |
+
"loss": 0.00011785422248067334,
|
| 437 |
+
"step": 28,
|
| 438 |
+
"step_time": 75.23865249752998
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"clip_ratio/high_max": 0.0034115930611733347,
|
| 442 |
+
"clip_ratio/high_mean": 0.0008528982652933337,
|
| 443 |
+
"clip_ratio/low_mean": 0.0008532625361112878,
|
| 444 |
+
"clip_ratio/low_min": 0.0,
|
| 445 |
+
"clip_ratio/region_mean": 0.0017061607650248334,
|
| 446 |
+
"epoch": 0.18125,
|
| 447 |
+
"grad_norm": 1.8060285231328368,
|
| 448 |
+
"kl": 0.004940098471706733,
|
| 449 |
+
"learning_rate": 9.420013103171891e-07,
|
| 450 |
+
"loss": 1.1617245036177337e-05,
|
| 451 |
+
"step": 29,
|
| 452 |
+
"step_time": 75.57006607204676
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"clip_ratio/high_max": 0.00441058368596714,
|
| 456 |
+
"clip_ratio/high_mean": 0.0014366193463501986,
|
| 457 |
+
"clip_ratio/low_mean": 0.0008966944296844304,
|
| 458 |
+
"clip_ratio/low_min": 0.0,
|
| 459 |
+
"clip_ratio/region_mean": 0.0023333137651206926,
|
| 460 |
+
"epoch": 0.1875,
|
| 461 |
+
"grad_norm": 2.087026768816426,
|
| 462 |
+
"kl": 0.0044833502324763685,
|
| 463 |
+
"learning_rate": 9.37173308072291e-07,
|
| 464 |
+
"loss": 0.000512868573423475,
|
| 465 |
+
"step": 30,
|
| 466 |
+
"step_time": 75.53689527511597
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"clip_ratio/high_max": 0.007819346967153251,
|
| 470 |
+
"clip_ratio/high_mean": 0.0030232854769565165,
|
| 471 |
+
"clip_ratio/low_mean": 0.002723286372201983,
|
| 472 |
+
"clip_ratio/low_min": 0.0,
|
| 473 |
+
"clip_ratio/region_mean": 0.005746571769122966,
|
| 474 |
+
"epoch": 0.19375,
|
| 475 |
+
"grad_norm": 5.309644810306232,
|
| 476 |
+
"kl": 0.008327675808686763,
|
| 477 |
+
"learning_rate": 9.321657187348688e-07,
|
| 478 |
+
"loss": 0.0014059533132240176,
|
| 479 |
+
"step": 31,
|
| 480 |
+
"step_time": 75.66492963209748
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"clip_ratio/high_max": 0.007995587482582778,
|
| 484 |
+
"clip_ratio/high_mean": 0.0027412632334744558,
|
| 485 |
+
"clip_ratio/low_mean": 0.002029010094702244,
|
| 486 |
+
"clip_ratio/low_min": 0.000244140625,
|
| 487 |
+
"clip_ratio/region_mean": 0.004770273284520954,
|
| 488 |
+
"epoch": 0.2,
|
| 489 |
+
"grad_norm": 2.61422085008822,
|
| 490 |
+
"kl": 0.01011492992984131,
|
| 491 |
+
"learning_rate": 9.269805993802128e-07,
|
| 492 |
+
"loss": 0.0010216329246759415,
|
| 493 |
+
"step": 32,
|
| 494 |
+
"step_time": 73.56156662479043
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"clip_ratio/high_max": 0.0,
|
| 498 |
+
"clip_ratio/high_mean": 0.0,
|
| 499 |
+
"clip_ratio/low_mean": 0.0,
|
| 500 |
+
"clip_ratio/low_min": 0.0,
|
| 501 |
+
"clip_ratio/region_mean": 0.0,
|
| 502 |
+
"completions/clipped_ratio": 0.47265625,
|
| 503 |
+
"completions/max_length": 2048.0,
|
| 504 |
+
"completions/mean_length": 1432.97265625,
|
| 505 |
+
"completions/min_length": 89.0,
|
| 506 |
+
"epoch": 0.20625,
|
| 507 |
+
"frac_reward_zero_std": 0.3125,
|
| 508 |
+
"grad_norm": 3.143397562275028,
|
| 509 |
+
"kl": 0.0041240869322791696,
|
| 510 |
+
"learning_rate": 9.216200800114411e-07,
|
| 511 |
+
"loss": 4.125750274397433e-05,
|
| 512 |
+
"reward": 0.36328125,
|
| 513 |
+
"reward_std": 0.3128761649131775,
|
| 514 |
+
"rewards/MajorityAccuracy/mean": 0.36328125,
|
| 515 |
+
"rewards/MajorityAccuracy/std": 0.48188698291778564,
|
| 516 |
+
"step": 33,
|
| 517 |
+
"step_time": 522.7592743933201
|
| 518 |
+
},
|
| 519 |
+
{
|
| 520 |
+
"clip_ratio/high_max": 0.0010458010729053058,
|
| 521 |
+
"clip_ratio/high_mean": 0.00027750059962272644,
|
| 522 |
+
"clip_ratio/low_mean": 0.00018310546875,
|
| 523 |
+
"clip_ratio/low_min": 0.0,
|
| 524 |
+
"clip_ratio/region_mean": 0.00046060606473474763,
|
| 525 |
+
"epoch": 0.2125,
|
| 526 |
+
"grad_norm": 1.0489253198033208,
|
| 527 |
+
"kl": 0.0039875368238426745,
|
| 528 |
+
"learning_rate": 9.160863626845119e-07,
|
| 529 |
+
"loss": 7.475444726878777e-05,
|
| 530 |
+
"step": 34,
|
| 531 |
+
"step_time": 75.35879619047046
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"clip_ratio/high_max": 0.006793855456635356,
|
| 535 |
+
"clip_ratio/high_mean": 0.001769165217410773,
|
| 536 |
+
"clip_ratio/low_mean": 0.00035522287362255156,
|
| 537 |
+
"clip_ratio/low_min": 0.0,
|
| 538 |
+
"clip_ratio/region_mean": 0.002124388120137155,
|
| 539 |
+
"epoch": 0.21875,
|
| 540 |
+
"grad_norm": 2.3565914105442913,
|
| 541 |
+
"kl": 0.006722828664351255,
|
| 542 |
+
"learning_rate": 9.103817206036382e-07,
|
| 543 |
+
"loss": -0.0004410488763824105,
|
| 544 |
+
"step": 35,
|
| 545 |
+
"step_time": 73.64566445350647
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"clip_ratio/high_max": 0.0038139743846841156,
|
| 549 |
+
"clip_ratio/high_mean": 0.0012813252687919885,
|
| 550 |
+
"clip_ratio/low_mean": 0.0009471004232182167,
|
| 551 |
+
"clip_ratio/low_min": 0.0,
|
| 552 |
+
"clip_ratio/region_mean": 0.0022284256920102052,
|
| 553 |
+
"epoch": 0.225,
|
| 554 |
+
"grad_norm": 2.854614577340349,
|
| 555 |
+
"kl": 0.0049351580964867026,
|
| 556 |
+
"learning_rate": 9.045084971874737e-07,
|
| 557 |
+
"loss": 0.00012044515460729599,
|
| 558 |
+
"step": 36,
|
| 559 |
+
"step_time": 74.80945963412523
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"clip_ratio/high_max": 0.004141119570704177,
|
| 563 |
+
"clip_ratio/high_mean": 0.0010352798926760443,
|
| 564 |
+
"clip_ratio/low_mean": 0.00226214891881682,
|
| 565 |
+
"clip_ratio/low_min": 0.0,
|
| 566 |
+
"clip_ratio/region_mean": 0.003297428833320737,
|
| 567 |
+
"epoch": 0.23125,
|
| 568 |
+
"grad_norm": 2.1279023058114084,
|
| 569 |
+
"kl": 0.007933814078569412,
|
| 570 |
+
"learning_rate": 8.984691051064574e-07,
|
| 571 |
+
"loss": 0.00032359015312977135,
|
| 572 |
+
"step": 37,
|
| 573 |
+
"step_time": 75.54086620733142
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"clip_ratio/high_max": 0.0002189141814596951,
|
| 577 |
+
"clip_ratio/high_mean": 5.4728545364923775e-05,
|
| 578 |
+
"clip_ratio/low_mean": 0.0008649545779917389,
|
| 579 |
+
"clip_ratio/low_min": 0.0,
|
| 580 |
+
"clip_ratio/region_mean": 0.0009196831233566627,
|
| 581 |
+
"epoch": 0.2375,
|
| 582 |
+
"grad_norm": 0.6969537458477494,
|
| 583 |
+
"kl": 0.003762644366361201,
|
| 584 |
+
"learning_rate": 8.922660252917087e-07,
|
| 585 |
+
"loss": -5.0491144065745175e-05,
|
| 586 |
+
"step": 38,
|
| 587 |
+
"step_time": 75.61534526571631
|
| 588 |
+
},
|
| 589 |
+
{
|
| 590 |
+
"clip_ratio/high_max": 0.005213643045863137,
|
| 591 |
+
"clip_ratio/high_mean": 0.0019282817956991494,
|
| 592 |
+
"clip_ratio/low_mean": 0.0008544921875,
|
| 593 |
+
"clip_ratio/low_min": 0.0,
|
| 594 |
+
"clip_ratio/region_mean": 0.0027827739249914885,
|
| 595 |
+
"epoch": 0.24375,
|
| 596 |
+
"grad_norm": 1.565457482933576,
|
| 597 |
+
"kl": 0.006226467725355178,
|
| 598 |
+
"learning_rate": 8.859018059158809e-07,
|
| 599 |
+
"loss": 0.00023893453180789948,
|
| 600 |
+
"step": 39,
|
| 601 |
+
"step_time": 74.95028644055128
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"clip_ratio/high_max": 0.004562818707199767,
|
| 605 |
+
"clip_ratio/high_mean": 0.0017785916934371926,
|
| 606 |
+
"clip_ratio/low_mean": 0.0023568837495986372,
|
| 607 |
+
"clip_ratio/low_min": 6.103515625e-05,
|
| 608 |
+
"clip_ratio/region_mean": 0.004135475493967533,
|
| 609 |
+
"epoch": 0.25,
|
| 610 |
+
"grad_norm": 1.826916389014125,
|
| 611 |
+
"kl": 0.005191356758587062,
|
| 612 |
+
"learning_rate": 8.793790613463954e-07,
|
| 613 |
+
"loss": 0.00024130381643772125,
|
| 614 |
+
"step": 40,
|
| 615 |
+
"step_time": 75.46800241991878
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"clip_ratio/high_max": 0.0,
|
| 619 |
+
"clip_ratio/high_mean": 0.0,
|
| 620 |
+
"clip_ratio/low_mean": 0.0,
|
| 621 |
+
"clip_ratio/low_min": 0.0,
|
| 622 |
+
"clip_ratio/region_mean": 0.0,
|
| 623 |
+
"completions/clipped_ratio": 0.59375,
|
| 624 |
+
"completions/max_length": 2048.0,
|
| 625 |
+
"completions/mean_length": 1567.484375,
|
| 626 |
+
"completions/min_length": 212.0,
|
| 627 |
+
"epoch": 0.25625,
|
| 628 |
+
"frac_reward_zero_std": 0.375,
|
| 629 |
+
"grad_norm": 1.0466969221401228,
|
| 630 |
+
"kl": 0.00430725485784933,
|
| 631 |
+
"learning_rate": 8.727004710714798e-07,
|
| 632 |
+
"loss": 4.305015318095684e-05,
|
| 633 |
+
"reward": 0.28515625,
|
| 634 |
+
"reward_std": 0.27499276399612427,
|
| 635 |
+
"rewards/MajorityAccuracy/mean": 0.28515625,
|
| 636 |
+
"rewards/MajorityAccuracy/std": 0.4523732364177704,
|
| 637 |
+
"step": 41,
|
| 638 |
+
"step_time": 523.6532399393618
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"clip_ratio/high_max": 0.000766765973821748,
|
| 642 |
+
"clip_ratio/high_mean": 0.0002658959801919991,
|
| 643 |
+
"clip_ratio/low_mean": 0.000518798828125,
|
| 644 |
+
"clip_ratio/low_min": 0.0,
|
| 645 |
+
"clip_ratio/region_mean": 0.0007846948064980097,
|
| 646 |
+
"epoch": 0.2625,
|
| 647 |
+
"grad_norm": 1.5926133181675461,
|
| 648 |
+
"kl": 0.0054566432954743505,
|
| 649 |
+
"learning_rate": 8.658687785994578e-07,
|
| 650 |
+
"loss": -4.198402166366577e-06,
|
| 651 |
+
"step": 42,
|
| 652 |
+
"step_time": 74.46834290772676
|
| 653 |
+
},
|
| 654 |
+
{
|
| 655 |
+
"clip_ratio/high_max": 0.0005627803184324875,
|
| 656 |
+
"clip_ratio/high_mean": 0.0001406950796081219,
|
| 657 |
+
"clip_ratio/low_mean": 0.0004119873046875,
|
| 658 |
+
"clip_ratio/low_min": 0.0,
|
| 659 |
+
"clip_ratio/region_mean": 0.0005526823806576431,
|
| 660 |
+
"epoch": 0.26875,
|
| 661 |
+
"grad_norm": 1.1562895243654745,
|
| 662 |
+
"kl": 0.008215626236051321,
|
| 663 |
+
"learning_rate": 8.588867903317394e-07,
|
| 664 |
+
"loss": 8.126935426844284e-06,
|
| 665 |
+
"step": 43,
|
| 666 |
+
"step_time": 75.4747600145638
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"clip_ratio/high_max": 0.0,
|
| 670 |
+
"clip_ratio/high_mean": 0.0,
|
| 671 |
+
"clip_ratio/low_mean": 0.0,
|
| 672 |
+
"clip_ratio/low_min": 0.0,
|
| 673 |
+
"clip_ratio/region_mean": 0.0,
|
| 674 |
+
"epoch": 0.275,
|
| 675 |
+
"grad_norm": 0.019650576830844164,
|
| 676 |
+
"kl": 0.0063026980496943,
|
| 677 |
+
"learning_rate": 8.51757374409974e-07,
|
| 678 |
+
"loss": 6.310017488431185e-05,
|
| 679 |
+
"step": 44,
|
| 680 |
+
"step_time": 75.01477641239762
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"clip_ratio/high_max": 0.00804125118884258,
|
| 684 |
+
"clip_ratio/high_mean": 0.0023964928041095845,
|
| 685 |
+
"clip_ratio/low_mean": 0.0002943484578281641,
|
| 686 |
+
"clip_ratio/low_min": 0.0,
|
| 687 |
+
"clip_ratio/region_mean": 0.002690841291041579,
|
| 688 |
+
"epoch": 0.28125,
|
| 689 |
+
"grad_norm": 1.5956767370510891,
|
| 690 |
+
"kl": 0.012760324403643608,
|
| 691 |
+
"learning_rate": 8.444834595378433e-07,
|
| 692 |
+
"loss": 9.396413224749267e-05,
|
| 693 |
+
"step": 45,
|
| 694 |
+
"step_time": 73.58318983763456
|
| 695 |
+
},
|
| 696 |
+
{
|
| 697 |
+
"clip_ratio/high_max": 0.004093422838195693,
|
| 698 |
+
"clip_ratio/high_mean": 0.001312593418333563,
|
| 699 |
+
"clip_ratio/low_mean": 0.0012304132687859237,
|
| 700 |
+
"clip_ratio/low_min": 0.0,
|
| 701 |
+
"clip_ratio/region_mean": 0.0025430066307308152,
|
| 702 |
+
"epoch": 0.2875,
|
| 703 |
+
"grad_norm": 3.140158098225664,
|
| 704 |
+
"kl": 0.005673496052622795,
|
| 705 |
+
"learning_rate": 8.370680337779736e-07,
|
| 706 |
+
"loss": -2.602569293230772e-05,
|
| 707 |
+
"step": 46,
|
| 708 |
+
"step_time": 73.88425412774086
|
| 709 |
+
},
|
| 710 |
+
{
|
| 711 |
+
"clip_ratio/high_max": 0.005794720491394401,
|
| 712 |
+
"clip_ratio/high_mean": 0.0021390190086094663,
|
| 713 |
+
"clip_ratio/low_mean": 0.0013275146484375,
|
| 714 |
+
"clip_ratio/low_min": 0.0,
|
| 715 |
+
"clip_ratio/region_mean": 0.0034665336715988815,
|
| 716 |
+
"epoch": 0.29375,
|
| 717 |
+
"grad_norm": 1.7220128574964864,
|
| 718 |
+
"kl": 0.007395118765998632,
|
| 719 |
+
"learning_rate": 8.295141433244659e-07,
|
| 720 |
+
"loss": -9.272673923987895e-05,
|
| 721 |
+
"step": 47,
|
| 722 |
+
"step_time": 74.96287551522255
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"clip_ratio/high_max": 0.0027720580983441323,
|
| 726 |
+
"clip_ratio/high_mean": 0.000839258536871057,
|
| 727 |
+
"clip_ratio/low_mean": 0.003775283752474934,
|
| 728 |
+
"clip_ratio/low_min": 0.0010986328125,
|
| 729 |
+
"clip_ratio/region_mean": 0.0046145422966219485,
|
| 730 |
+
"epoch": 0.3,
|
| 731 |
+
"grad_norm": 2.70644677281796,
|
| 732 |
+
"kl": 0.006327775656245649,
|
| 733 |
+
"learning_rate": 8.218248912515442e-07,
|
| 734 |
+
"loss": 0.00027948617935180664,
|
| 735 |
+
"step": 48,
|
| 736 |
+
"step_time": 75.37967292964458
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"clip_ratio/high_max": 0.0,
|
| 740 |
+
"clip_ratio/high_mean": 0.0,
|
| 741 |
+
"clip_ratio/low_mean": 0.0,
|
| 742 |
+
"clip_ratio/low_min": 0.0,
|
| 743 |
+
"clip_ratio/region_mean": 0.0,
|
| 744 |
+
"completions/clipped_ratio": 0.328125,
|
| 745 |
+
"completions/max_length": 2048.0,
|
| 746 |
+
"completions/mean_length": 1141.3515625,
|
| 747 |
+
"completions/min_length": 72.0,
|
| 748 |
+
"epoch": 0.30625,
|
| 749 |
+
"frac_reward_zero_std": 0.28125,
|
| 750 |
+
"grad_norm": 2.391397698481597,
|
| 751 |
+
"kl": 0.009993361542001367,
|
| 752 |
+
"learning_rate": 8.140034362388397e-07,
|
| 753 |
+
"loss": 9.991973638534546e-05,
|
| 754 |
+
"reward": 0.3671875,
|
| 755 |
+
"reward_std": 0.33824339509010315,
|
| 756 |
+
"rewards/MajorityAccuracy/mean": 0.3671875,
|
| 757 |
+
"rewards/MajorityAccuracy/std": 0.48298245668411255,
|
| 758 |
+
"step": 49,
|
| 759 |
+
"step_time": 512.9214892312884
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"clip_ratio/high_max": 0.0020666105847340077,
|
| 763 |
+
"clip_ratio/high_mean": 0.0006830062302469742,
|
| 764 |
+
"clip_ratio/low_mean": 0.00012072090248693712,
|
| 765 |
+
"clip_ratio/low_min": 0.0,
|
| 766 |
+
"clip_ratio/region_mean": 0.0008037271363718901,
|
| 767 |
+
"epoch": 0.3125,
|
| 768 |
+
"grad_norm": 1.1311712092196637,
|
| 769 |
+
"kl": 0.007840007252525538,
|
| 770 |
+
"learning_rate": 8.060529912738314e-07,
|
| 771 |
+
"loss": 0.00019373028771951795,
|
| 772 |
+
"step": 50,
|
| 773 |
+
"step_time": 74.16181601211429
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"clip_ratio/high_max": 0.005059267321485095,
|
| 777 |
+
"clip_ratio/high_mean": 0.001389172251947457,
|
| 778 |
+
"clip_ratio/low_mean": 0.0003881874290527776,
|
| 779 |
+
"clip_ratio/low_min": 0.0,
|
| 780 |
+
"clip_ratio/region_mean": 0.0017773596846382134,
|
| 781 |
+
"epoch": 0.31875,
|
| 782 |
+
"grad_norm": 4.457680150518517,
|
| 783 |
+
"kl": 0.015099560958333313,
|
| 784 |
+
"learning_rate": 7.979768223319785e-07,
|
| 785 |
+
"loss": 0.0002656654396560043,
|
| 786 |
+
"step": 51,
|
| 787 |
+
"step_time": 75.25208542123437
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"clip_ratio/high_max": 0.015001942898379639,
|
| 791 |
+
"clip_ratio/high_mean": 0.004777847010700498,
|
| 792 |
+
"clip_ratio/low_mean": 0.0016292082436848432,
|
| 793 |
+
"clip_ratio/low_min": 0.0,
|
| 794 |
+
"clip_ratio/region_mean": 0.006407055363524705,
|
| 795 |
+
"epoch": 0.325,
|
| 796 |
+
"grad_norm": 6.939747517126915,
|
| 797 |
+
"kl": 0.02434373297728598,
|
| 798 |
+
"learning_rate": 7.897782470350849e-07,
|
| 799 |
+
"loss": 0.00047546811401844025,
|
| 800 |
+
"step": 52,
|
| 801 |
+
"step_time": 72.6264511346817
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"clip_ratio/high_max": 0.00048828125,
|
| 805 |
+
"clip_ratio/high_mean": 0.0001220703125,
|
| 806 |
+
"clip_ratio/low_mean": 0.0006708862638333812,
|
| 807 |
+
"clip_ratio/low_min": 0.0,
|
| 808 |
+
"clip_ratio/region_mean": 0.0007929565763333812,
|
| 809 |
+
"epoch": 0.33125,
|
| 810 |
+
"grad_norm": 1.4569601729475716,
|
| 811 |
+
"kl": 0.008915742160752416,
|
| 812 |
+
"learning_rate": 7.814606332884488e-07,
|
| 813 |
+
"loss": 0.00015942522441037,
|
| 814 |
+
"step": 53,
|
| 815 |
+
"step_time": 74.47659660130739
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"clip_ratio/high_max": 0.005292071495205164,
|
| 819 |
+
"clip_ratio/high_mean": 0.0015897179837338626,
|
| 820 |
+
"clip_ratio/low_mean": 0.0021800102840643376,
|
| 821 |
+
"clip_ratio/low_min": 0.0,
|
| 822 |
+
"clip_ratio/region_mean": 0.0037697282677982002,
|
| 823 |
+
"epoch": 0.3375,
|
| 824 |
+
"grad_norm": 3.196208892274776,
|
| 825 |
+
"kl": 0.012592139071784914,
|
| 826 |
+
"learning_rate": 7.730273978973552e-07,
|
| 827 |
+
"loss": 0.0006393707590177655,
|
| 828 |
+
"step": 54,
|
| 829 |
+
"step_time": 72.87352094054222
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"clip_ratio/high_max": 0.005318669660482556,
|
| 833 |
+
"clip_ratio/high_mean": 0.001962240508873947,
|
| 834 |
+
"clip_ratio/low_mean": 0.0033941361325560138,
|
| 835 |
+
"clip_ratio/low_min": 0.0,
|
| 836 |
+
"clip_ratio/region_mean": 0.005356376597774215,
|
| 837 |
+
"epoch": 0.34375,
|
| 838 |
+
"grad_norm": 2.3572025388454865,
|
| 839 |
+
"kl": 0.018138426588848233,
|
| 840 |
+
"learning_rate": 7.644820051634812e-07,
|
| 841 |
+
"loss": 0.00012287963181734085,
|
| 842 |
+
"step": 55,
|
| 843 |
+
"step_time": 73.92265523597598
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"clip_ratio/high_max": 0.0026121972186956555,
|
| 847 |
+
"clip_ratio/high_mean": 0.0007293432499864139,
|
| 848 |
+
"clip_ratio/low_mean": 0.0014072625344851986,
|
| 849 |
+
"clip_ratio/low_min": 0.0,
|
| 850 |
+
"clip_ratio/region_mean": 0.002136605777195655,
|
| 851 |
+
"epoch": 0.35,
|
| 852 |
+
"grad_norm": 1.9312080526730704,
|
| 853 |
+
"kl": 0.009327677777037024,
|
| 854 |
+
"learning_rate": 7.558279654617912e-07,
|
| 855 |
+
"loss": 0.0004396864678710699,
|
| 856 |
+
"step": 56,
|
| 857 |
+
"step_time": 74.89075421541929
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"clip_ratio/high_max": 0.0,
|
| 861 |
+
"clip_ratio/high_mean": 0.0,
|
| 862 |
+
"clip_ratio/low_mean": 0.0,
|
| 863 |
+
"clip_ratio/low_min": 0.0,
|
| 864 |
+
"clip_ratio/region_mean": 0.0,
|
| 865 |
+
"completions/clipped_ratio": 0.48046875,
|
| 866 |
+
"completions/max_length": 2048.0,
|
| 867 |
+
"completions/mean_length": 1569.734375,
|
| 868 |
+
"completions/min_length": 191.0,
|
| 869 |
+
"epoch": 0.35625,
|
| 870 |
+
"frac_reward_zero_std": 0.1875,
|
| 871 |
+
"grad_norm": 1.294716634314696,
|
| 872 |
+
"kl": 0.010782271448988467,
|
| 873 |
+
"learning_rate": 7.470688337985029e-07,
|
| 874 |
+
"loss": 0.00010784397454699501,
|
| 875 |
+
"reward": 0.39453125,
|
| 876 |
+
"reward_std": 0.3805779218673706,
|
| 877 |
+
"rewards/MajorityAccuracy/mean": 0.39453125,
|
| 878 |
+
"rewards/MajorityAccuracy/std": 0.48970717191696167,
|
| 879 |
+
"step": 57,
|
| 880 |
+
"step_time": 524.9491717852652
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"clip_ratio/high_max": 0.003033819841220975,
|
| 884 |
+
"clip_ratio/high_mean": 0.0011043790509575047,
|
| 885 |
+
"clip_ratio/low_mean": 0.0002875434001907706,
|
| 886 |
+
"clip_ratio/low_min": 0.0,
|
| 887 |
+
"clip_ratio/region_mean": 0.001391922458424233,
|
| 888 |
+
"epoch": 0.3625,
|
| 889 |
+
"grad_norm": 1.8748635488396295,
|
| 890 |
+
"kl": 0.010607728036120534,
|
| 891 |
+
"learning_rate": 7.382082083507225e-07,
|
| 892 |
+
"loss": 0.00017821788787841797,
|
| 893 |
+
"step": 58,
|
| 894 |
+
"step_time": 74.19262247905135
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"clip_ratio/high_max": 0.003974063089117408,
|
| 898 |
+
"clip_ratio/high_mean": 0.0012963942426722497,
|
| 899 |
+
"clip_ratio/low_mean": 0.0005912025117140729,
|
| 900 |
+
"clip_ratio/low_min": 0.0,
|
| 901 |
+
"clip_ratio/region_mean": 0.0018875967580243014,
|
| 902 |
+
"epoch": 0.36875,
|
| 903 |
+
"grad_norm": 2.885324257908364,
|
| 904 |
+
"kl": 0.01205208688043058,
|
| 905 |
+
"learning_rate": 7.292497289883432e-07,
|
| 906 |
+
"loss": 0.00014140456914901733,
|
| 907 |
+
"step": 59,
|
| 908 |
+
"step_time": 75.36175334081054
|
| 909 |
+
},
|
| 910 |
+
{
|
| 911 |
+
"clip_ratio/high_max": 0.0004996304051019251,
|
| 912 |
+
"clip_ratio/high_mean": 0.00012490760127548128,
|
| 913 |
+
"clip_ratio/low_mean": 0.0002288818359375,
|
| 914 |
+
"clip_ratio/low_min": 0.0,
|
| 915 |
+
"clip_ratio/region_mean": 0.0003537894372129813,
|
| 916 |
+
"epoch": 0.375,
|
| 917 |
+
"grad_norm": 1.1807708669868144,
|
| 918 |
+
"kl": 0.009168512071482837,
|
| 919 |
+
"learning_rate": 7.201970757788171e-07,
|
| 920 |
+
"loss": 0.00010514166206121445,
|
| 921 |
+
"step": 60,
|
| 922 |
+
"step_time": 74.87276351824403
|
| 923 |
+
},
|
| 924 |
+
{
|
| 925 |
+
"clip_ratio/high_max": 0.0023645634937565774,
|
| 926 |
+
"clip_ratio/high_mean": 0.0007008918983046897,
|
| 927 |
+
"clip_ratio/low_mean": 0.0008125346066663042,
|
| 928 |
+
"clip_ratio/low_min": 0.0,
|
| 929 |
+
"clip_ratio/region_mean": 0.001513426483143121,
|
| 930 |
+
"epoch": 0.38125,
|
| 931 |
+
"grad_norm": 1.4316256287108047,
|
| 932 |
+
"kl": 0.009021992445923388,
|
| 933 |
+
"learning_rate": 7.110539674754159e-07,
|
| 934 |
+
"loss": 6.683469109702855e-05,
|
| 935 |
+
"step": 61,
|
| 936 |
+
"step_time": 75.87214371934533
|
| 937 |
+
},
|
| 938 |
+
{
|
| 939 |
+
"clip_ratio/high_max": 0.0011609980938374065,
|
| 940 |
+
"clip_ratio/high_mean": 0.0003345206259837141,
|
| 941 |
+
"clip_ratio/low_mean": 0.0006591154742636718,
|
| 942 |
+
"clip_ratio/low_min": 0.0,
|
| 943 |
+
"clip_ratio/region_mean": 0.0009936361166182905,
|
| 944 |
+
"epoch": 0.3875,
|
| 945 |
+
"grad_norm": 1.4566554084426844,
|
| 946 |
+
"kl": 0.008537797490134835,
|
| 947 |
+
"learning_rate": 7.018241599895973e-07,
|
| 948 |
+
"loss": 0.00032491888850927353,
|
| 949 |
+
"step": 62,
|
| 950 |
+
"step_time": 74.8810028322041
|
| 951 |
+
},
|
| 952 |
+
{
|
| 953 |
+
"clip_ratio/high_max": 0.0032113665074575692,
|
| 954 |
+
"clip_ratio/high_mean": 0.0011742574934032746,
|
| 955 |
+
"clip_ratio/low_mean": 0.0016626243304926902,
|
| 956 |
+
"clip_ratio/low_min": 0.00018310546875,
|
| 957 |
+
"clip_ratio/region_mean": 0.0028368817875161767,
|
| 958 |
+
"epoch": 0.39375,
|
| 959 |
+
"grad_norm": 1.7239739656430717,
|
| 960 |
+
"kl": 0.01039774448145181,
|
| 961 |
+
"learning_rate": 6.925114448481088e-07,
|
| 962 |
+
"loss": -0.00023233331739902496,
|
| 963 |
+
"step": 63,
|
| 964 |
+
"step_time": 75.10972927883267
|
| 965 |
+
},
|
| 966 |
+
{
|
| 967 |
+
"clip_ratio/high_max": 0.003642104711616412,
|
| 968 |
+
"clip_ratio/high_mean": 0.0013460345508065075,
|
| 969 |
+
"clip_ratio/low_mean": 0.0006294465711107478,
|
| 970 |
+
"clip_ratio/low_min": 0.0,
|
| 971 |
+
"clip_ratio/region_mean": 0.0019754811219172552,
|
| 972 |
+
"epoch": 0.4,
|
| 973 |
+
"grad_norm": 1.1718441509061512,
|
| 974 |
+
"kl": 0.008446098188869655,
|
| 975 |
+
"learning_rate": 6.831196476354614e-07,
|
| 976 |
+
"loss": 0.0002567608607932925,
|
| 977 |
+
"step": 64,
|
| 978 |
+
"step_time": 74.69232140854001
|
| 979 |
+
},
|
| 980 |
+
{
|
| 981 |
+
"clip_ratio/high_max": 0.0,
|
| 982 |
+
"clip_ratio/high_mean": 0.0,
|
| 983 |
+
"clip_ratio/low_mean": 0.0,
|
| 984 |
+
"clip_ratio/low_min": 0.0,
|
| 985 |
+
"clip_ratio/region_mean": 0.0,
|
| 986 |
+
"completions/clipped_ratio": 0.41796875,
|
| 987 |
+
"completions/max_length": 2048.0,
|
| 988 |
+
"completions/mean_length": 1388.62109375,
|
| 989 |
+
"completions/min_length": 121.0,
|
| 990 |
+
"epoch": 0.40625,
|
| 991 |
+
"frac_reward_zero_std": 0.3125,
|
| 992 |
+
"grad_norm": 1.3615352694036993,
|
| 993 |
+
"kl": 0.011758966953493655,
|
| 994 |
+
"learning_rate": 6.7365262642241e-07,
|
| 995 |
+
"loss": 0.00011765056115109473,
|
| 996 |
+
"reward": 0.42578125,
|
| 997 |
+
"reward_std": 0.31971096992492676,
|
| 998 |
+
"rewards/MajorityAccuracy/mean": 0.42578125,
|
| 999 |
+
"rewards/MajorityAccuracy/std": 0.49542948603630066,
|
| 1000 |
+
"step": 65,
|
| 1001 |
+
"step_time": 518.0938603691757
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"clip_ratio/high_max": 0.002323765162145719,
|
| 1005 |
+
"clip_ratio/high_mean": 0.0005809412905364297,
|
| 1006 |
+
"clip_ratio/low_mean": 7.62939453125e-05,
|
| 1007 |
+
"clip_ratio/low_min": 0.0,
|
| 1008 |
+
"clip_ratio/region_mean": 0.0006572352358489297,
|
| 1009 |
+
"epoch": 0.4125,
|
| 1010 |
+
"grad_norm": 4.696711474444763,
|
| 1011 |
+
"kl": 0.011750384932383895,
|
| 1012 |
+
"learning_rate": 6.641142701810931e-07,
|
| 1013 |
+
"loss": 0.00012603143113665283,
|
| 1014 |
+
"step": 66,
|
| 1015 |
+
"step_time": 73.87216125801206
|
| 1016 |
+
},
|
| 1017 |
+
{
|
| 1018 |
+
"clip_ratio/high_max": 0.001618263660930097,
|
| 1019 |
+
"clip_ratio/high_mean": 0.0004904854140477255,
|
| 1020 |
+
"clip_ratio/low_mean": 0.00044589269964490086,
|
| 1021 |
+
"clip_ratio/low_min": 0.0,
|
| 1022 |
+
"clip_ratio/region_mean": 0.0009363780991407111,
|
| 1023 |
+
"epoch": 0.41875,
|
| 1024 |
+
"grad_norm": 1.6016987710419028,
|
| 1025 |
+
"kl": 0.009656625101342797,
|
| 1026 |
+
"learning_rate": 6.545084971874736e-07,
|
| 1027 |
+
"loss": 9.953970584319904e-05,
|
| 1028 |
+
"step": 67,
|
| 1029 |
+
"step_time": 74.07577167823911
|
| 1030 |
+
},
|
| 1031 |
+
{
|
| 1032 |
+
"clip_ratio/high_max": 0.0044829786056652665,
|
| 1033 |
+
"clip_ratio/high_mean": 0.0016512869697180577,
|
| 1034 |
+
"clip_ratio/low_mean": 0.0010233745088044088,
|
| 1035 |
+
"clip_ratio/low_min": 0.0,
|
| 1036 |
+
"clip_ratio/region_mean": 0.0026746614748844877,
|
| 1037 |
+
"epoch": 0.425,
|
| 1038 |
+
"grad_norm": 2.6111359103081258,
|
| 1039 |
+
"kl": 0.014279014547355473,
|
| 1040 |
+
"learning_rate": 6.448392534117461e-07,
|
| 1041 |
+
"loss": 0.0005701221525669098,
|
| 1042 |
+
"step": 68,
|
| 1043 |
+
"step_time": 74.03759878501296
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"clip_ratio/high_max": 0.0024384072530665435,
|
| 1047 |
+
"clip_ratio/high_mean": 0.0007156982701417292,
|
| 1048 |
+
"clip_ratio/low_mean": 0.0004992723115719855,
|
| 1049 |
+
"clip_ratio/low_min": 0.0,
|
| 1050 |
+
"clip_ratio/region_mean": 0.0012149705798947252,
|
| 1051 |
+
"epoch": 0.43125,
|
| 1052 |
+
"grad_norm": 2.9580277477662587,
|
| 1053 |
+
"kl": 0.015088776708580554,
|
| 1054 |
+
"learning_rate": 6.351105108973644e-07,
|
| 1055 |
+
"loss": 0.00012439738202374429,
|
| 1056 |
+
"step": 69,
|
| 1057 |
+
"step_time": 74.1509125046432
|
| 1058 |
+
},
|
| 1059 |
+
{
|
| 1060 |
+
"clip_ratio/high_max": 0.0028054171125404537,
|
| 1061 |
+
"clip_ratio/high_mean": 0.0007991190286702476,
|
| 1062 |
+
"clip_ratio/low_mean": 0.0003662109375,
|
| 1063 |
+
"clip_ratio/low_min": 0.0,
|
| 1064 |
+
"clip_ratio/region_mean": 0.00116532995889429,
|
| 1065 |
+
"epoch": 0.4375,
|
| 1066 |
+
"grad_norm": 1.671757539789347,
|
| 1067 |
+
"kl": 0.010453568072989583,
|
| 1068 |
+
"learning_rate": 6.253262661293602e-07,
|
| 1069 |
+
"loss": 0.0002570932556409389,
|
| 1070 |
+
"step": 70,
|
| 1071 |
+
"step_time": 75.53403532505035
|
| 1072 |
+
},
|
| 1073 |
+
{
|
| 1074 |
+
"clip_ratio/high_max": 0.0027060136198997498,
|
| 1075 |
+
"clip_ratio/high_mean": 0.0008483246274408884,
|
| 1076 |
+
"clip_ratio/low_mean": 0.00045102843432687223,
|
| 1077 |
+
"clip_ratio/low_min": 0.0,
|
| 1078 |
+
"clip_ratio/region_mean": 0.0012993530835956335,
|
| 1079 |
+
"epoch": 0.44375,
|
| 1080 |
+
"grad_norm": 1.0852356016538744,
|
| 1081 |
+
"kl": 0.011065920582041144,
|
| 1082 |
+
"learning_rate": 6.154905383926216e-07,
|
| 1083 |
+
"loss": 7.682408613618463e-05,
|
| 1084 |
+
"step": 71,
|
| 1085 |
+
"step_time": 75.00721688941121
|
| 1086 |
+
},
|
| 1087 |
+
{
|
| 1088 |
+
"clip_ratio/high_max": 0.00253958698885981,
|
| 1089 |
+
"clip_ratio/high_mean": 0.0007263874904310796,
|
| 1090 |
+
"clip_ratio/low_mean": 0.0013194926723372191,
|
| 1091 |
+
"clip_ratio/low_min": 0.0,
|
| 1092 |
+
"clip_ratio/region_mean": 0.002045880173682235,
|
| 1093 |
+
"epoch": 0.45,
|
| 1094 |
+
"grad_norm": 1.4111556472266575,
|
| 1095 |
+
"kl": 0.008942228683736175,
|
| 1096 |
+
"learning_rate": 6.056073681208037e-07,
|
| 1097 |
+
"loss": -8.830417937133461e-05,
|
| 1098 |
+
"step": 72,
|
| 1099 |
+
"step_time": 74.7511993125081
|
| 1100 |
+
},
|
| 1101 |
+
{
|
| 1102 |
+
"clip_ratio/high_max": 0.0,
|
| 1103 |
+
"clip_ratio/high_mean": 0.0,
|
| 1104 |
+
"clip_ratio/low_mean": 0.0,
|
| 1105 |
+
"clip_ratio/low_min": 0.0,
|
| 1106 |
+
"clip_ratio/region_mean": 0.0,
|
| 1107 |
+
"completions/clipped_ratio": 0.26953125,
|
| 1108 |
+
"completions/max_length": 2048.0,
|
| 1109 |
+
"completions/mean_length": 1186.81640625,
|
| 1110 |
+
"completions/min_length": 150.0,
|
| 1111 |
+
"epoch": 0.45625,
|
| 1112 |
+
"frac_reward_zero_std": 0.25,
|
| 1113 |
+
"grad_norm": 1.847113093689933,
|
| 1114 |
+
"kl": 0.009032989328261465,
|
| 1115 |
+
"learning_rate": 5.956808152365532e-07,
|
| 1116 |
+
"loss": 9.029309876495972e-05,
|
| 1117 |
+
"reward": 0.56640625,
|
| 1118 |
+
"reward_std": 0.3441670835018158,
|
| 1119 |
+
"rewards/MajorityAccuracy/mean": 0.56640625,
|
| 1120 |
+
"rewards/MajorityAccuracy/std": 0.4965413510799408,
|
| 1121 |
+
"step": 73,
|
| 1122 |
+
"step_time": 515.2636110149324
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"clip_ratio/high_max": 0.005440236607682891,
|
| 1126 |
+
"clip_ratio/high_mean": 0.0017128038562077563,
|
| 1127 |
+
"clip_ratio/low_mean": 0.00025836139684543014,
|
| 1128 |
+
"clip_ratio/low_min": 0.0,
|
| 1129 |
+
"clip_ratio/region_mean": 0.00197116524213925,
|
| 1130 |
+
"epoch": 0.4625,
|
| 1131 |
+
"grad_norm": 3.2072181911001185,
|
| 1132 |
+
"kl": 0.016334758489392698,
|
| 1133 |
+
"learning_rate": 5.857149574837268e-07,
|
| 1134 |
+
"loss": -0.00013885341468267143,
|
| 1135 |
+
"step": 74,
|
| 1136 |
+
"step_time": 73.84726579114795
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"clip_ratio/high_max": 0.005438593958388083,
|
| 1140 |
+
"clip_ratio/high_mean": 0.001487199519033311,
|
| 1141 |
+
"clip_ratio/low_mean": 0.0005589009469986195,
|
| 1142 |
+
"clip_ratio/low_min": 0.0,
|
| 1143 |
+
"clip_ratio/region_mean": 0.0020461004660319304,
|
| 1144 |
+
"epoch": 0.46875,
|
| 1145 |
+
"grad_norm": 3.197275917685857,
|
| 1146 |
+
"kl": 0.01910787494853139,
|
| 1147 |
+
"learning_rate": 5.757138887522883e-07,
|
| 1148 |
+
"loss": 0.0003203567466698587,
|
| 1149 |
+
"step": 75,
|
| 1150 |
+
"step_time": 72.564996432513
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"clip_ratio/high_max": 0.004228524572681636,
|
| 1154 |
+
"clip_ratio/high_mean": 0.0013021920167375356,
|
| 1155 |
+
"clip_ratio/low_mean": 0.00044065849942853674,
|
| 1156 |
+
"clip_ratio/low_min": 0.0,
|
| 1157 |
+
"clip_ratio/region_mean": 0.0017428505270800088,
|
| 1158 |
+
"epoch": 0.475,
|
| 1159 |
+
"grad_norm": 2.433542005660699,
|
| 1160 |
+
"kl": 0.012555921450257301,
|
| 1161 |
+
"learning_rate": 5.656817173965732e-07,
|
| 1162 |
+
"loss": -6.967849913053215e-05,
|
| 1163 |
+
"step": 76,
|
| 1164 |
+
"step_time": 73.60645537823439
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"clip_ratio/high_max": 0.0017181458824779838,
|
| 1168 |
+
"clip_ratio/high_mean": 0.00042953647061949596,
|
| 1169 |
+
"clip_ratio/low_mean": 0.00010915668826783076,
|
| 1170 |
+
"clip_ratio/low_min": 0.0,
|
| 1171 |
+
"clip_ratio/region_mean": 0.0005386931588873267,
|
| 1172 |
+
"epoch": 0.48125,
|
| 1173 |
+
"grad_norm": 0.6820289696793876,
|
| 1174 |
+
"kl": 0.01305523724295199,
|
| 1175 |
+
"learning_rate": 5.556225645476118e-07,
|
| 1176 |
+
"loss": 0.00021713657770305872,
|
| 1177 |
+
"step": 77,
|
| 1178 |
+
"step_time": 73.62196180596948
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"clip_ratio/high_max": 0.007542098464909941,
|
| 1182 |
+
"clip_ratio/high_mean": 0.0021191251507843845,
|
| 1183 |
+
"clip_ratio/low_mean": 0.000936978220124729,
|
| 1184 |
+
"clip_ratio/low_min": 0.0,
|
| 1185 |
+
"clip_ratio/region_mean": 0.003056103378185071,
|
| 1186 |
+
"epoch": 0.4875,
|
| 1187 |
+
"grad_norm": 2.315701832945261,
|
| 1188 |
+
"kl": 0.018608344835229218,
|
| 1189 |
+
"learning_rate": 5.455405624202032e-07,
|
| 1190 |
+
"loss": 0.00032984837889671326,
|
| 1191 |
+
"step": 78,
|
| 1192 |
+
"step_time": 75.43008405715227
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"clip_ratio/high_max": 0.0036515317697194405,
|
| 1196 |
+
"clip_ratio/high_mean": 0.0011704004009516211,
|
| 1197 |
+
"clip_ratio/low_mean": 0.0005198153958190233,
|
| 1198 |
+
"clip_ratio/low_min": 0.0,
|
| 1199 |
+
"clip_ratio/region_mean": 0.0016902158095035702,
|
| 1200 |
+
"epoch": 0.49375,
|
| 1201 |
+
"grad_norm": 1.4125328639233607,
|
| 1202 |
+
"kl": 0.01202479307539761,
|
| 1203 |
+
"learning_rate": 5.354398526154365e-07,
|
| 1204 |
+
"loss": 0.00029543257551267743,
|
| 1205 |
+
"step": 79,
|
| 1206 |
+
"step_time": 75.79501016438007
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"clip_ratio/high_max": 0.0046439422585535794,
|
| 1210 |
+
"clip_ratio/high_mean": 0.0018766334760584868,
|
| 1211 |
+
"clip_ratio/low_mean": 0.0017579132691025734,
|
| 1212 |
+
"clip_ratio/low_min": 0.0,
|
| 1213 |
+
"clip_ratio/region_mean": 0.003634546708781272,
|
| 1214 |
+
"epoch": 0.5,
|
| 1215 |
+
"grad_norm": 2.3458241525189822,
|
| 1216 |
+
"kl": 0.01274471974465996,
|
| 1217 |
+
"learning_rate": 5.253245844193564e-07,
|
| 1218 |
+
"loss": 0.00033188238739967346,
|
| 1219 |
+
"step": 80,
|
| 1220 |
+
"step_time": 75.15501376241446
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"clip_ratio/high_max": 0.0,
|
| 1224 |
+
"clip_ratio/high_mean": 0.0,
|
| 1225 |
+
"clip_ratio/low_mean": 0.0,
|
| 1226 |
+
"clip_ratio/low_min": 0.0,
|
| 1227 |
+
"clip_ratio/region_mean": 0.0,
|
| 1228 |
+
"completions/clipped_ratio": 0.29296875,
|
| 1229 |
+
"completions/max_length": 2048.0,
|
| 1230 |
+
"completions/mean_length": 1202.21875,
|
| 1231 |
+
"completions/min_length": 64.0,
|
| 1232 |
+
"epoch": 0.50625,
|
| 1233 |
+
"frac_reward_zero_std": 0.375,
|
| 1234 |
+
"grad_norm": 2.824814324162206,
|
| 1235 |
+
"kl": 0.01336689037270844,
|
| 1236 |
+
"learning_rate": 5.151989130984714e-07,
|
| 1237 |
+
"loss": 0.0001336345449090004,
|
| 1238 |
+
"reward": 0.578125,
|
| 1239 |
+
"reward_std": 0.28300705552101135,
|
| 1240 |
+
"rewards/MajorityAccuracy/mean": 0.578125,
|
| 1241 |
+
"rewards/MajorityAccuracy/std": 0.49482616782188416,
|
| 1242 |
+
"step": 81,
|
| 1243 |
+
"step_time": 519.394398715347
|
| 1244 |
+
},
|
| 1245 |
+
{
|
| 1246 |
+
"clip_ratio/high_max": 0.0011485914656077512,
|
| 1247 |
+
"clip_ratio/high_mean": 0.0003616117191995727,
|
| 1248 |
+
"clip_ratio/low_mean": 0.0005527293942577671,
|
| 1249 |
+
"clip_ratio/low_min": 0.0,
|
| 1250 |
+
"clip_ratio/region_mean": 0.000914341100724414,
|
| 1251 |
+
"epoch": 0.5125,
|
| 1252 |
+
"grad_norm": 1.2630021960059958,
|
| 1253 |
+
"kl": 0.014386181370355189,
|
| 1254 |
+
"learning_rate": 5.050669981928055e-07,
|
| 1255 |
+
"loss": 0.0002198715810663998,
|
| 1256 |
+
"step": 82,
|
| 1257 |
+
"step_time": 73.70060339197516
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"clip_ratio/high_max": 0.0,
|
| 1261 |
+
"clip_ratio/high_mean": 0.0,
|
| 1262 |
+
"clip_ratio/low_mean": 0.0,
|
| 1263 |
+
"clip_ratio/low_min": 0.0,
|
| 1264 |
+
"clip_ratio/region_mean": 0.0,
|
| 1265 |
+
"epoch": 0.51875,
|
| 1266 |
+
"grad_norm": 0.05290895644356498,
|
| 1267 |
+
"kl": 0.01761094736866653,
|
| 1268 |
+
"learning_rate": 4.949330018071946e-07,
|
| 1269 |
+
"loss": 0.00017613370437175035,
|
| 1270 |
+
"step": 83,
|
| 1271 |
+
"step_time": 74.04236837849021
|
| 1272 |
+
},
|
| 1273 |
+
{
|
| 1274 |
+
"clip_ratio/high_max": 0.0006514315682579763,
|
| 1275 |
+
"clip_ratio/high_mean": 0.00016285789206449408,
|
| 1276 |
+
"clip_ratio/low_mean": 0.0003204345703125,
|
| 1277 |
+
"clip_ratio/low_min": 0.0,
|
| 1278 |
+
"clip_ratio/region_mean": 0.00048329245692002587,
|
| 1279 |
+
"epoch": 0.525,
|
| 1280 |
+
"grad_norm": 1.2705999186574302,
|
| 1281 |
+
"kl": 0.026702288538217545,
|
| 1282 |
+
"learning_rate": 4.848010869015287e-07,
|
| 1283 |
+
"loss": 0.0003247161512263119,
|
| 1284 |
+
"step": 84,
|
| 1285 |
+
"step_time": 73.77489949390292
|
| 1286 |
+
},
|
| 1287 |
+
{
|
| 1288 |
+
"clip_ratio/high_max": 0.0021605894289677963,
|
| 1289 |
+
"clip_ratio/high_mean": 0.0006437489864765666,
|
| 1290 |
+
"clip_ratio/low_mean": 0.00039707346877548844,
|
| 1291 |
+
"clip_ratio/low_min": 0.0,
|
| 1292 |
+
"clip_ratio/region_mean": 0.0010408224479760975,
|
| 1293 |
+
"epoch": 0.53125,
|
| 1294 |
+
"grad_norm": 1.2397844171950152,
|
| 1295 |
+
"kl": 0.012526679434813559,
|
| 1296 |
+
"learning_rate": 4.746754155806437e-07,
|
| 1297 |
+
"loss": 0.00019235593208577484,
|
| 1298 |
+
"step": 85,
|
| 1299 |
+
"step_time": 74.66063784062862
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"clip_ratio/high_max": 0.0025378570426255465,
|
| 1303 |
+
"clip_ratio/high_mean": 0.0008219376250053756,
|
| 1304 |
+
"clip_ratio/low_mean": 0.0009623294463381171,
|
| 1305 |
+
"clip_ratio/low_min": 0.0,
|
| 1306 |
+
"clip_ratio/region_mean": 0.0017842670931713656,
|
| 1307 |
+
"epoch": 0.5375,
|
| 1308 |
+
"grad_norm": 2.7077538977693147,
|
| 1309 |
+
"kl": 0.017646656604483724,
|
| 1310 |
+
"learning_rate": 4.645601473845635e-07,
|
| 1311 |
+
"loss": 0.00025191486929543316,
|
| 1312 |
+
"step": 86,
|
| 1313 |
+
"step_time": 73.51122678816319
|
| 1314 |
+
},
|
| 1315 |
+
{
|
| 1316 |
+
"clip_ratio/high_max": 0.013768360018730164,
|
| 1317 |
+
"clip_ratio/high_mean": 0.004194453562377021,
|
| 1318 |
+
"clip_ratio/low_mean": 0.000741943276807433,
|
| 1319 |
+
"clip_ratio/low_min": 0.0,
|
| 1320 |
+
"clip_ratio/region_mean": 0.0049363968064426444,
|
| 1321 |
+
"epoch": 0.54375,
|
| 1322 |
+
"grad_norm": 2.3497879377163846,
|
| 1323 |
+
"kl": 0.01835615176241845,
|
| 1324 |
+
"learning_rate": 4.544594375797968e-07,
|
| 1325 |
+
"loss": 0.00011122896830784157,
|
| 1326 |
+
"step": 87,
|
| 1327 |
+
"step_time": 73.26023896411061
|
| 1328 |
+
},
|
| 1329 |
+
{
|
| 1330 |
+
"clip_ratio/high_max": 0.0020394847670104355,
|
| 1331 |
+
"clip_ratio/high_mean": 0.0006841149661340751,
|
| 1332 |
+
"clip_ratio/low_mean": 0.0006920517334947363,
|
| 1333 |
+
"clip_ratio/low_min": 0.0,
|
| 1334 |
+
"clip_ratio/region_mean": 0.0013761666923528537,
|
| 1335 |
+
"epoch": 0.55,
|
| 1336 |
+
"grad_norm": 1.4850848119134565,
|
| 1337 |
+
"kl": 0.014141770312562585,
|
| 1338 |
+
"learning_rate": 4.443774354523882e-07,
|
| 1339 |
+
"loss": -3.951950930058956e-06,
|
| 1340 |
+
"step": 88,
|
| 1341 |
+
"step_time": 73.44360157474875
|
| 1342 |
+
},
|
| 1343 |
+
{
|
| 1344 |
+
"clip_ratio/high_max": 0.0,
|
| 1345 |
+
"clip_ratio/high_mean": 0.0,
|
| 1346 |
+
"clip_ratio/low_mean": 0.0,
|
| 1347 |
+
"clip_ratio/low_min": 0.0,
|
| 1348 |
+
"clip_ratio/region_mean": 0.0,
|
| 1349 |
+
"completions/clipped_ratio": 0.2890625,
|
| 1350 |
+
"completions/max_length": 2048.0,
|
| 1351 |
+
"completions/mean_length": 1173.79296875,
|
| 1352 |
+
"completions/min_length": 80.0,
|
| 1353 |
+
"epoch": 0.55625,
|
| 1354 |
+
"frac_reward_zero_std": 0.28125,
|
| 1355 |
+
"grad_norm": 2.2401336730259995,
|
| 1356 |
+
"kl": 0.016683788155205548,
|
| 1357 |
+
"learning_rate": 4.3431828260342675e-07,
|
| 1358 |
+
"loss": 0.00016684446018189192,
|
| 1359 |
+
"reward": 0.53515625,
|
| 1360 |
+
"reward_std": 0.3394179940223694,
|
| 1361 |
+
"rewards/MajorityAccuracy/mean": 0.53515625,
|
| 1362 |
+
"rewards/MajorityAccuracy/std": 0.49973952770233154,
|
| 1363 |
+
"step": 89,
|
| 1364 |
+
"step_time": 512.5296324938536
|
| 1365 |
+
},
|
| 1366 |
+
{
|
| 1367 |
+
"clip_ratio/high_max": 0.0026401729555800557,
|
| 1368 |
+
"clip_ratio/high_mean": 0.0008625191330793314,
|
| 1369 |
+
"clip_ratio/low_mean": 0.00016182010585907847,
|
| 1370 |
+
"clip_ratio/low_min": 0.0,
|
| 1371 |
+
"clip_ratio/region_mean": 0.0010243392389384098,
|
| 1372 |
+
"epoch": 0.5625,
|
| 1373 |
+
"grad_norm": 2.5293187222903586,
|
| 1374 |
+
"kl": 0.016139961895532906,
|
| 1375 |
+
"learning_rate": 4.242861112477118e-07,
|
| 1376 |
+
"loss": 0.0006059574079699814,
|
| 1377 |
+
"step": 90,
|
| 1378 |
+
"step_time": 72.96825967729092
|
| 1379 |
+
},
|
| 1380 |
+
{
|
| 1381 |
+
"clip_ratio/high_max": 0.0028021819816785865,
|
| 1382 |
+
"clip_ratio/high_mean": 0.0008649781357235042,
|
| 1383 |
+
"clip_ratio/low_mean": 0.00048797928684507497,
|
| 1384 |
+
"clip_ratio/low_min": 0.0,
|
| 1385 |
+
"clip_ratio/region_mean": 0.0013529574207495898,
|
| 1386 |
+
"epoch": 0.56875,
|
| 1387 |
+
"grad_norm": 2.1042226897274716,
|
| 1388 |
+
"kl": 0.012991935480386019,
|
| 1389 |
+
"learning_rate": 4.1428504251627325e-07,
|
| 1390 |
+
"loss": 0.00028318638214841485,
|
| 1391 |
+
"step": 91,
|
| 1392 |
+
"step_time": 74.02629935741425
|
| 1393 |
+
},
|
| 1394 |
+
{
|
| 1395 |
+
"clip_ratio/high_max": 0.0012005973694613203,
|
| 1396 |
+
"clip_ratio/high_mean": 0.0003306074322608765,
|
| 1397 |
+
"clip_ratio/low_mean": 0.0005430040691862814,
|
| 1398 |
+
"clip_ratio/low_min": 0.0,
|
| 1399 |
+
"clip_ratio/region_mean": 0.0008736114923522109,
|
| 1400 |
+
"epoch": 0.575,
|
| 1401 |
+
"grad_norm": 1.7149402436291845,
|
| 1402 |
+
"kl": 0.023449595784768462,
|
| 1403 |
+
"learning_rate": 4.0431918476344685e-07,
|
| 1404 |
+
"loss": 0.00011021864338545129,
|
| 1405 |
+
"step": 92,
|
| 1406 |
+
"step_time": 74.13926462084055
|
| 1407 |
+
},
|
| 1408 |
+
{
|
| 1409 |
+
"clip_ratio/high_max": 0.0019526997639331967,
|
| 1410 |
+
"clip_ratio/high_mean": 0.0005093326981295832,
|
| 1411 |
+
"clip_ratio/low_mean": 7.62939453125e-05,
|
| 1412 |
+
"clip_ratio/low_min": 0.0,
|
| 1413 |
+
"clip_ratio/region_mean": 0.0005856266434420832,
|
| 1414 |
+
"epoch": 0.58125,
|
| 1415 |
+
"grad_norm": 0.9073865419044614,
|
| 1416 |
+
"kl": 0.012478606891818345,
|
| 1417 |
+
"learning_rate": 3.943926318791963e-07,
|
| 1418 |
+
"loss": 0.0002823451068252325,
|
| 1419 |
+
"step": 93,
|
| 1420 |
+
"step_time": 75.47891764342785
|
| 1421 |
+
},
|
| 1422 |
+
{
|
| 1423 |
+
"clip_ratio/high_max": 0.0020500093160080723,
|
| 1424 |
+
"clip_ratio/high_mean": 0.0006024501908541424,
|
| 1425 |
+
"clip_ratio/low_mean": 0.000643200910417363,
|
| 1426 |
+
"clip_ratio/low_min": 0.0,
|
| 1427 |
+
"clip_ratio/region_mean": 0.001245651081262622,
|
| 1428 |
+
"epoch": 0.5875,
|
| 1429 |
+
"grad_norm": 1.5254042604771518,
|
| 1430 |
+
"kl": 0.014944139285944402,
|
| 1431 |
+
"learning_rate": 3.845094616073783e-07,
|
| 1432 |
+
"loss": 0.0003860839642584324,
|
| 1433 |
+
"step": 94,
|
| 1434 |
+
"step_time": 75.34067194536328
|
| 1435 |
+
},
|
| 1436 |
+
{
|
| 1437 |
+
"clip_ratio/high_max": 0.0012221146607771516,
|
| 1438 |
+
"clip_ratio/high_mean": 0.0003207949048373848,
|
| 1439 |
+
"clip_ratio/low_mean": 0.00025373500648129266,
|
| 1440 |
+
"clip_ratio/low_min": 0.0,
|
| 1441 |
+
"clip_ratio/region_mean": 0.0005745299022237305,
|
| 1442 |
+
"epoch": 0.59375,
|
| 1443 |
+
"grad_norm": 2.595319221291584,
|
| 1444 |
+
"kl": 0.018806543666869402,
|
| 1445 |
+
"learning_rate": 3.7467373387063964e-07,
|
| 1446 |
+
"loss": 0.0005276034353300929,
|
| 1447 |
+
"step": 95,
|
| 1448 |
+
"step_time": 72.76824268326163
|
| 1449 |
+
},
|
| 1450 |
+
{
|
| 1451 |
+
"clip_ratio/high_max": 0.00413227899116464,
|
| 1452 |
+
"clip_ratio/high_mean": 0.0011529554758453742,
|
| 1453 |
+
"clip_ratio/low_mean": 0.0004239620757289231,
|
| 1454 |
+
"clip_ratio/low_min": 0.0,
|
| 1455 |
+
"clip_ratio/region_mean": 0.0015769175515742972,
|
| 1456 |
+
"epoch": 0.6,
|
| 1457 |
+
"grad_norm": 1.6757151275942275,
|
| 1458 |
+
"kl": 0.013757532113231719,
|
| 1459 |
+
"learning_rate": 3.648894891026358e-07,
|
| 1460 |
+
"loss": 0.00020213943207636476,
|
| 1461 |
+
"step": 96,
|
| 1462 |
+
"step_time": 75.19560718536377
|
| 1463 |
+
},
|
| 1464 |
+
{
|
| 1465 |
+
"clip_ratio/high_max": 0.0,
|
| 1466 |
+
"clip_ratio/high_mean": 0.0,
|
| 1467 |
+
"clip_ratio/low_mean": 0.0,
|
| 1468 |
+
"clip_ratio/low_min": 0.0,
|
| 1469 |
+
"clip_ratio/region_mean": 0.0,
|
| 1470 |
+
"completions/clipped_ratio": 0.27734375,
|
| 1471 |
+
"completions/max_length": 2048.0,
|
| 1472 |
+
"completions/mean_length": 1156.08984375,
|
| 1473 |
+
"completions/min_length": 76.0,
|
| 1474 |
+
"epoch": 0.60625,
|
| 1475 |
+
"frac_reward_zero_std": 0.28125,
|
| 1476 |
+
"grad_norm": 1.2807824372904737,
|
| 1477 |
+
"kl": 0.020676342770457268,
|
| 1478 |
+
"learning_rate": 3.5516074658825397e-07,
|
| 1479 |
+
"loss": 0.0002068376197712496,
|
| 1480 |
+
"reward": 0.5,
|
| 1481 |
+
"reward_std": 0.2990657091140747,
|
| 1482 |
+
"rewards/MajorityAccuracy/mean": 0.5,
|
| 1483 |
+
"rewards/MajorityAccuracy/std": 0.5009794235229492,
|
| 1484 |
+
"step": 97,
|
| 1485 |
+
"step_time": 516.070428751409
|
| 1486 |
+
},
|
| 1487 |
+
{
|
| 1488 |
+
"clip_ratio/high_max": 0.0014559215269400738,
|
| 1489 |
+
"clip_ratio/high_mean": 0.00043193167402932886,
|
| 1490 |
+
"clip_ratio/low_mean": 0.00019992608031316195,
|
| 1491 |
+
"clip_ratio/low_min": 0.0,
|
| 1492 |
+
"clip_ratio/region_mean": 0.0006318577470665332,
|
| 1493 |
+
"epoch": 0.6125,
|
| 1494 |
+
"grad_norm": 1.3472379702437391,
|
| 1495 |
+
"kl": 0.014394080266356468,
|
| 1496 |
+
"learning_rate": 3.454915028125263e-07,
|
| 1497 |
+
"loss": 0.00018879241542890668,
|
| 1498 |
+
"step": 98,
|
| 1499 |
+
"step_time": 75.43395470455289
|
| 1500 |
+
},
|
| 1501 |
+
{
|
| 1502 |
+
"clip_ratio/high_max": 0.0006106428118073381,
|
| 1503 |
+
"clip_ratio/high_mean": 0.0002520774669392267,
|
| 1504 |
+
"clip_ratio/low_mean": 0.000213623046875,
|
| 1505 |
+
"clip_ratio/low_min": 0.0,
|
| 1506 |
+
"clip_ratio/region_mean": 0.0004657005119952373,
|
| 1507 |
+
"epoch": 0.61875,
|
| 1508 |
+
"grad_norm": 1.561452506716982,
|
| 1509 |
+
"kl": 0.014319001697003841,
|
| 1510 |
+
"learning_rate": 3.3588572981890684e-07,
|
| 1511 |
+
"loss": 4.142448597121984e-05,
|
| 1512 |
+
"step": 99,
|
| 1513 |
+
"step_time": 73.45833129435778
|
| 1514 |
+
},
|
| 1515 |
+
{
|
| 1516 |
+
"clip_ratio/high_max": 0.001976332707272377,
|
| 1517 |
+
"clip_ratio/high_mean": 0.0005430643705039984,
|
| 1518 |
+
"clip_ratio/low_mean": 0.0005013573136238847,
|
| 1519 |
+
"clip_ratio/low_min": 0.0,
|
| 1520 |
+
"clip_ratio/region_mean": 0.0010444216823088937,
|
| 1521 |
+
"epoch": 0.625,
|
| 1522 |
+
"grad_norm": 2.3488160974695025,
|
| 1523 |
+
"kl": 0.016415129764936864,
|
| 1524 |
+
"learning_rate": 3.263473735775899e-07,
|
| 1525 |
+
"loss": 0.0004034433513879776,
|
| 1526 |
+
"step": 100,
|
| 1527 |
+
"step_time": 73.54472163692117
|
| 1528 |
+
},
|
| 1529 |
+
{
|
| 1530 |
+
"clip_ratio/high_max": 0.004820868402021006,
|
| 1531 |
+
"clip_ratio/high_mean": 0.0013645231010741554,
|
| 1532 |
+
"clip_ratio/low_mean": 0.0005557738768402487,
|
| 1533 |
+
"clip_ratio/low_min": 0.0,
|
| 1534 |
+
"clip_ratio/region_mean": 0.0019202969706384465,
|
| 1535 |
+
"epoch": 0.63125,
|
| 1536 |
+
"grad_norm": 5.177787756453676,
|
| 1537 |
+
"kl": 0.01794739905744791,
|
| 1538 |
+
"learning_rate": 3.1688035236453865e-07,
|
| 1539 |
+
"loss": 6.125308573246002e-05,
|
| 1540 |
+
"step": 101,
|
| 1541 |
+
"step_time": 72.92532254382968
|
| 1542 |
+
},
|
| 1543 |
+
{
|
| 1544 |
+
"clip_ratio/high_max": 0.0028177451458759606,
|
| 1545 |
+
"clip_ratio/high_mean": 0.0009176246094284579,
|
| 1546 |
+
"clip_ratio/low_mean": 0.00030242155480664223,
|
| 1547 |
+
"clip_ratio/low_min": 0.0,
|
| 1548 |
+
"clip_ratio/region_mean": 0.0012200461715110578,
|
| 1549 |
+
"epoch": 0.6375,
|
| 1550 |
+
"grad_norm": 2.710727686609032,
|
| 1551 |
+
"kl": 0.013163566123694181,
|
| 1552 |
+
"learning_rate": 3.0748855515189096e-07,
|
| 1553 |
+
"loss": 0.00031520426273345947,
|
| 1554 |
+
"step": 102,
|
| 1555 |
+
"step_time": 74.4597052000463
|
| 1556 |
+
},
|
| 1557 |
+
{
|
| 1558 |
+
"clip_ratio/high_max": 0.00172796147671761,
|
| 1559 |
+
"clip_ratio/high_mean": 0.0005168031912035076,
|
| 1560 |
+
"clip_ratio/low_mean": 6.103515625e-05,
|
| 1561 |
+
"clip_ratio/low_min": 0.0,
|
| 1562 |
+
"clip_ratio/region_mean": 0.000577838349272497,
|
| 1563 |
+
"epoch": 0.64375,
|
| 1564 |
+
"grad_norm": 1.0028679519984425,
|
| 1565 |
+
"kl": 0.01762246817816049,
|
| 1566 |
+
"learning_rate": 2.981758400104028e-07,
|
| 1567 |
+
"loss": 0.00014572852523997426,
|
| 1568 |
+
"step": 103,
|
| 1569 |
+
"step_time": 73.82820834964514
|
| 1570 |
+
},
|
| 1571 |
+
{
|
| 1572 |
+
"clip_ratio/high_max": 0.0010159287776332349,
|
| 1573 |
+
"clip_ratio/high_mean": 0.0002539821944083087,
|
| 1574 |
+
"clip_ratio/low_mean": 9.469696669839323e-05,
|
| 1575 |
+
"clip_ratio/low_min": 0.0,
|
| 1576 |
+
"clip_ratio/region_mean": 0.00034867916110670194,
|
| 1577 |
+
"epoch": 0.65,
|
| 1578 |
+
"grad_norm": 1.8909143427920203,
|
| 1579 |
+
"kl": 0.01686234213411808,
|
| 1580 |
+
"learning_rate": 2.8894603252458403e-07,
|
| 1581 |
+
"loss": 0.0002810780133586377,
|
| 1582 |
+
"step": 104,
|
| 1583 |
+
"step_time": 72.44602734223008
|
| 1584 |
+
},
|
| 1585 |
+
{
|
| 1586 |
+
"clip_ratio/high_max": 0.0,
|
| 1587 |
+
"clip_ratio/high_mean": 0.0,
|
| 1588 |
+
"clip_ratio/low_mean": 0.0,
|
| 1589 |
+
"clip_ratio/low_min": 0.0,
|
| 1590 |
+
"clip_ratio/region_mean": 0.0,
|
| 1591 |
+
"completions/clipped_ratio": 0.30078125,
|
| 1592 |
+
"completions/max_length": 2048.0,
|
| 1593 |
+
"completions/mean_length": 1233.30859375,
|
| 1594 |
+
"completions/min_length": 86.0,
|
| 1595 |
+
"epoch": 0.65625,
|
| 1596 |
+
"frac_reward_zero_std": 0.1875,
|
| 1597 |
+
"grad_norm": 1.6124512965159195,
|
| 1598 |
+
"kl": 0.013469588942825794,
|
| 1599 |
+
"learning_rate": 2.7980292422118277e-07,
|
| 1600 |
+
"loss": 0.00013475585728883743,
|
| 1601 |
+
"reward": 0.515625,
|
| 1602 |
+
"reward_std": 0.3548349142074585,
|
| 1603 |
+
"rewards/MajorityAccuracy/mean": 0.515625,
|
| 1604 |
+
"rewards/MajorityAccuracy/std": 0.5007347464561462,
|
| 1605 |
+
"step": 105,
|
| 1606 |
+
"step_time": 514.3105178587139
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"clip_ratio/high_max": 0.0027941544831264764,
|
| 1610 |
+
"clip_ratio/high_mean": 0.000735630608687643,
|
| 1611 |
+
"clip_ratio/low_mean": 0.00015286062807717826,
|
| 1612 |
+
"clip_ratio/low_min": 0.0,
|
| 1613 |
+
"clip_ratio/region_mean": 0.0008884912367648212,
|
| 1614 |
+
"epoch": 0.6625,
|
| 1615 |
+
"grad_norm": 1.7004907886086102,
|
| 1616 |
+
"kl": 0.014392888871952891,
|
| 1617 |
+
"learning_rate": 2.70750271011657e-07,
|
| 1618 |
+
"loss": 0.00018343181000091136,
|
| 1619 |
+
"step": 106,
|
| 1620 |
+
"step_time": 73.8561255633831
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"clip_ratio/high_max": 0.0027800172829302028,
|
| 1624 |
+
"clip_ratio/high_mean": 0.0007905700804258231,
|
| 1625 |
+
"clip_ratio/low_mean": 6.054864570614882e-05,
|
| 1626 |
+
"clip_ratio/low_min": 0.0,
|
| 1627 |
+
"clip_ratio/region_mean": 0.0008511187261319719,
|
| 1628 |
+
"epoch": 0.66875,
|
| 1629 |
+
"grad_norm": 1.9752735925883946,
|
| 1630 |
+
"kl": 0.02144697681069374,
|
| 1631 |
+
"learning_rate": 2.6179179164927754e-07,
|
| 1632 |
+
"loss": 0.0002503176801837981,
|
| 1633 |
+
"step": 107,
|
| 1634 |
+
"step_time": 72.71466437727213
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"clip_ratio/high_max": 0.0023569360564579256,
|
| 1638 |
+
"clip_ratio/high_mean": 0.000799030583948479,
|
| 1639 |
+
"clip_ratio/low_mean": 0.000285327885649167,
|
| 1640 |
+
"clip_ratio/low_min": 0.0,
|
| 1641 |
+
"clip_ratio/region_mean": 0.001084358478692593,
|
| 1642 |
+
"epoch": 0.675,
|
| 1643 |
+
"grad_norm": 1.51614727733094,
|
| 1644 |
+
"kl": 0.013899670215323567,
|
| 1645 |
+
"learning_rate": 2.529311662014972e-07,
|
| 1646 |
+
"loss": 0.00047684647142887115,
|
| 1647 |
+
"step": 108,
|
| 1648 |
+
"step_time": 73.87454456835985
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"clip_ratio/high_max": 0.002201883340603672,
|
| 1652 |
+
"clip_ratio/high_mean": 0.0007131404854590073,
|
| 1653 |
+
"clip_ratio/low_mean": 0.00019544691895134747,
|
| 1654 |
+
"clip_ratio/low_min": 0.0,
|
| 1655 |
+
"clip_ratio/region_mean": 0.0009085874044103548,
|
| 1656 |
+
"epoch": 0.68125,
|
| 1657 |
+
"grad_norm": 2.105189378466528,
|
| 1658 |
+
"kl": 0.015819149557501078,
|
| 1659 |
+
"learning_rate": 2.441720345382089e-07,
|
| 1660 |
+
"loss": 0.00034971858258359134,
|
| 1661 |
+
"step": 109,
|
| 1662 |
+
"step_time": 75.04459753260016
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"clip_ratio/high_max": 0.00032654128153808415,
|
| 1666 |
+
"clip_ratio/high_mean": 8.163532038452104e-05,
|
| 1667 |
+
"clip_ratio/low_mean": 0.0003225622931495309,
|
| 1668 |
+
"clip_ratio/low_min": 0.0,
|
| 1669 |
+
"clip_ratio/region_mean": 0.0004041976062580943,
|
| 1670 |
+
"epoch": 0.6875,
|
| 1671 |
+
"grad_norm": 1.0186783082871074,
|
| 1672 |
+
"kl": 0.010556956520304084,
|
| 1673 |
+
"learning_rate": 2.355179948365189e-07,
|
| 1674 |
+
"loss": -5.93688091612421e-05,
|
| 1675 |
+
"step": 110,
|
| 1676 |
+
"step_time": 75.33110052347183
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"clip_ratio/high_max": 0.0037064257776364684,
|
| 1680 |
+
"clip_ratio/high_mean": 0.0011136798711959273,
|
| 1681 |
+
"clip_ratio/low_mean": 0.000754187218262814,
|
| 1682 |
+
"clip_ratio/low_min": 0.0,
|
| 1683 |
+
"clip_ratio/region_mean": 0.001867867111286614,
|
| 1684 |
+
"epoch": 0.69375,
|
| 1685 |
+
"grad_norm": 4.72414117110905,
|
| 1686 |
+
"kl": 0.01859257509931922,
|
| 1687 |
+
"learning_rate": 2.26972602102645e-07,
|
| 1688 |
+
"loss": 0.00030503328889608383,
|
| 1689 |
+
"step": 111,
|
| 1690 |
+
"step_time": 72.76150969788432
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"clip_ratio/high_max": 0.0018552342808106914,
|
| 1694 |
+
"clip_ratio/high_mean": 0.0007037544637569226,
|
| 1695 |
+
"clip_ratio/low_mean": 0.0005709440338250715,
|
| 1696 |
+
"clip_ratio/low_min": 0.0,
|
| 1697 |
+
"clip_ratio/region_mean": 0.0012746985084959306,
|
| 1698 |
+
"epoch": 0.7,
|
| 1699 |
+
"grad_norm": 1.964354356848317,
|
| 1700 |
+
"kl": 0.01885864371433854,
|
| 1701 |
+
"learning_rate": 2.1853936671155127e-07,
|
| 1702 |
+
"loss": 0.000238846056163311,
|
| 1703 |
+
"step": 112,
|
| 1704 |
+
"step_time": 73.34909965097904
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"clip_ratio/high_max": 0.0,
|
| 1708 |
+
"clip_ratio/high_mean": 0.0,
|
| 1709 |
+
"clip_ratio/low_mean": 0.0,
|
| 1710 |
+
"clip_ratio/low_min": 0.0,
|
| 1711 |
+
"clip_ratio/region_mean": 0.0,
|
| 1712 |
+
"completions/clipped_ratio": 0.3359375,
|
| 1713 |
+
"completions/max_length": 2048.0,
|
| 1714 |
+
"completions/mean_length": 1254.86328125,
|
| 1715 |
+
"completions/min_length": 87.0,
|
| 1716 |
+
"epoch": 0.70625,
|
| 1717 |
+
"frac_reward_zero_std": 0.28125,
|
| 1718 |
+
"grad_norm": 4.453686822499959,
|
| 1719 |
+
"kl": 0.019711605506017804,
|
| 1720 |
+
"learning_rate": 2.1022175296491512e-07,
|
| 1721 |
+
"loss": 0.00019712270295713097,
|
| 1722 |
+
"reward": 0.44921875,
|
| 1723 |
+
"reward_std": 0.3262813091278076,
|
| 1724 |
+
"rewards/MajorityAccuracy/mean": 0.44921875,
|
| 1725 |
+
"rewards/MajorityAccuracy/std": 0.49838894605636597,
|
| 1726 |
+
"step": 113,
|
| 1727 |
+
"step_time": 521.804576471448
|
| 1728 |
+
},
|
| 1729 |
+
{
|
| 1730 |
+
"clip_ratio/high_max": 0.0006432364898500964,
|
| 1731 |
+
"clip_ratio/high_mean": 0.0001608091224625241,
|
| 1732 |
+
"clip_ratio/low_mean": 0.00012947287177667022,
|
| 1733 |
+
"clip_ratio/low_min": 0.0,
|
| 1734 |
+
"clip_ratio/region_mean": 0.0002902819942391943,
|
| 1735 |
+
"epoch": 0.7125,
|
| 1736 |
+
"grad_norm": 1.5280464046272169,
|
| 1737 |
+
"kl": 0.01700198941398412,
|
| 1738 |
+
"learning_rate": 2.0202317766802152e-07,
|
| 1739 |
+
"loss": 0.0002343460510019213,
|
| 1740 |
+
"step": 114,
|
| 1741 |
+
"step_time": 74.10264269262552
|
| 1742 |
+
},
|
| 1743 |
+
{
|
| 1744 |
+
"clip_ratio/high_max": 0.002756149275228381,
|
| 1745 |
+
"clip_ratio/high_mean": 0.0008415775046159979,
|
| 1746 |
+
"clip_ratio/low_mean": 0.00039004058635327965,
|
| 1747 |
+
"clip_ratio/low_min": 0.0,
|
| 1748 |
+
"clip_ratio/region_mean": 0.0012316180946072564,
|
| 1749 |
+
"epoch": 0.71875,
|
| 1750 |
+
"grad_norm": 2.644142561485859,
|
| 1751 |
+
"kl": 0.014950693701393902,
|
| 1752 |
+
"learning_rate": 1.9394700872616853e-07,
|
| 1753 |
+
"loss": 0.0005593271926045418,
|
| 1754 |
+
"step": 115,
|
| 1755 |
+
"step_time": 75.04527363553643
|
| 1756 |
+
},
|
| 1757 |
+
{
|
| 1758 |
+
"clip_ratio/high_max": 0.0031106895621633157,
|
| 1759 |
+
"clip_ratio/high_mean": 0.0009262974708690308,
|
| 1760 |
+
"clip_ratio/low_mean": 0.0010799261035572272,
|
| 1761 |
+
"clip_ratio/low_min": 0.0,
|
| 1762 |
+
"clip_ratio/region_mean": 0.002006223570788279,
|
| 1763 |
+
"epoch": 0.725,
|
| 1764 |
+
"grad_norm": 2.269318027610156,
|
| 1765 |
+
"kl": 0.01631354179698974,
|
| 1766 |
+
"learning_rate": 1.8599656376116024e-07,
|
| 1767 |
+
"loss": 0.00047273217933252454,
|
| 1768 |
+
"step": 116,
|
| 1769 |
+
"step_time": 75.04768951237202
|
| 1770 |
+
},
|
| 1771 |
+
{
|
| 1772 |
+
"clip_ratio/high_max": 0.00016066837997641414,
|
| 1773 |
+
"clip_ratio/high_mean": 4.0167094994103536e-05,
|
| 1774 |
+
"clip_ratio/low_mean": 0.000274658203125,
|
| 1775 |
+
"clip_ratio/low_min": 0.0,
|
| 1776 |
+
"clip_ratio/region_mean": 0.00031482529811910354,
|
| 1777 |
+
"epoch": 0.73125,
|
| 1778 |
+
"grad_norm": 1.1112874086848055,
|
| 1779 |
+
"kl": 0.01185278280172497,
|
| 1780 |
+
"learning_rate": 1.7817510874845582e-07,
|
| 1781 |
+
"loss": 0.00018813290807884187,
|
| 1782 |
+
"step": 117,
|
| 1783 |
+
"step_time": 74.60871486738324
|
| 1784 |
+
},
|
| 1785 |
+
{
|
| 1786 |
+
"clip_ratio/high_max": 0.0021475244575412944,
|
| 1787 |
+
"clip_ratio/high_mean": 0.0005540420170291327,
|
| 1788 |
+
"clip_ratio/low_mean": 0.0005245494830887765,
|
| 1789 |
+
"clip_ratio/low_min": 0.0,
|
| 1790 |
+
"clip_ratio/region_mean": 0.0010785914982989198,
|
| 1791 |
+
"epoch": 0.7375,
|
| 1792 |
+
"grad_norm": 2.1463325125611084,
|
| 1793 |
+
"kl": 0.012678552535362542,
|
| 1794 |
+
"learning_rate": 1.7048585667553412e-07,
|
| 1795 |
+
"loss": 0.00015098179574124515,
|
| 1796 |
+
"step": 118,
|
| 1797 |
+
"step_time": 74.65231367945671
|
| 1798 |
+
},
|
| 1799 |
+
{
|
| 1800 |
+
"clip_ratio/high_max": 0.0015881001891102642,
|
| 1801 |
+
"clip_ratio/high_mean": 0.00047015296877361834,
|
| 1802 |
+
"clip_ratio/low_mean": 0.00027825978395412676,
|
| 1803 |
+
"clip_ratio/low_min": 0.0,
|
| 1804 |
+
"clip_ratio/region_mean": 0.0007484127527277451,
|
| 1805 |
+
"epoch": 0.74375,
|
| 1806 |
+
"grad_norm": 1.7306809233295601,
|
| 1807 |
+
"kl": 0.013833149569109082,
|
| 1808 |
+
"learning_rate": 1.6293196622202632e-07,
|
| 1809 |
+
"loss": 5.002810212317854e-05,
|
| 1810 |
+
"step": 119,
|
| 1811 |
+
"step_time": 74.6779679954052
|
| 1812 |
+
},
|
| 1813 |
+
{
|
| 1814 |
+
"clip_ratio/high_max": 0.004361190221970901,
|
| 1815 |
+
"clip_ratio/high_mean": 0.001156505175458733,
|
| 1816 |
+
"clip_ratio/low_mean": 0.000335693359375,
|
| 1817 |
+
"clip_ratio/low_min": 0.0,
|
| 1818 |
+
"clip_ratio/region_mean": 0.001492198534833733,
|
| 1819 |
+
"epoch": 0.75,
|
| 1820 |
+
"grad_norm": 1.884931889393149,
|
| 1821 |
+
"kl": 0.018189109163358808,
|
| 1822 |
+
"learning_rate": 1.555165404621567e-07,
|
| 1823 |
+
"loss": 0.0001261005236301571,
|
| 1824 |
+
"step": 120,
|
| 1825 |
+
"step_time": 75.45197087153792
|
| 1826 |
+
},
|
| 1827 |
+
{
|
| 1828 |
+
"clip_ratio/high_max": 0.0,
|
| 1829 |
+
"clip_ratio/high_mean": 0.0,
|
| 1830 |
+
"clip_ratio/low_mean": 0.0,
|
| 1831 |
+
"clip_ratio/low_min": 0.0,
|
| 1832 |
+
"clip_ratio/region_mean": 0.0,
|
| 1833 |
+
"completions/clipped_ratio": 0.30859375,
|
| 1834 |
+
"completions/max_length": 2048.0,
|
| 1835 |
+
"completions/mean_length": 1216.51953125,
|
| 1836 |
+
"completions/min_length": 39.0,
|
| 1837 |
+
"epoch": 0.75625,
|
| 1838 |
+
"frac_reward_zero_std": 0.125,
|
| 1839 |
+
"grad_norm": 2.1141982341582315,
|
| 1840 |
+
"kl": 0.014289551181718707,
|
| 1841 |
+
"learning_rate": 1.4824262559002592e-07,
|
| 1842 |
+
"loss": 0.00014293193817138672,
|
| 1843 |
+
"reward": 0.55078125,
|
| 1844 |
+
"reward_std": 0.4083327651023865,
|
| 1845 |
+
"rewards/MajorityAccuracy/mean": 0.55078125,
|
| 1846 |
+
"rewards/MajorityAccuracy/std": 0.49838894605636597,
|
| 1847 |
+
"step": 121,
|
| 1848 |
+
"step_time": 516.1800138317049
|
| 1849 |
+
},
|
| 1850 |
+
{
|
| 1851 |
+
"clip_ratio/high_max": 0.0018894674140028656,
|
| 1852 |
+
"clip_ratio/high_mean": 0.0006553974017151631,
|
| 1853 |
+
"clip_ratio/low_mean": 0.0001983642578125,
|
| 1854 |
+
"clip_ratio/low_min": 0.0,
|
| 1855 |
+
"clip_ratio/region_mean": 0.0008537616577086737,
|
| 1856 |
+
"epoch": 0.7625,
|
| 1857 |
+
"grad_norm": 2.4619688443604097,
|
| 1858 |
+
"kl": 0.017472690204158425,
|
| 1859 |
+
"learning_rate": 1.4111320966826057e-07,
|
| 1860 |
+
"loss": -4.841946065425873e-05,
|
| 1861 |
+
"step": 122,
|
| 1862 |
+
"step_time": 73.8767229616642
|
| 1863 |
+
},
|
| 1864 |
+
{
|
| 1865 |
+
"clip_ratio/high_max": 0.0018210244743386284,
|
| 1866 |
+
"clip_ratio/high_mean": 0.0005070601000625174,
|
| 1867 |
+
"clip_ratio/low_mean": 0.00010993320029228926,
|
| 1868 |
+
"clip_ratio/low_min": 0.0,
|
| 1869 |
+
"clip_ratio/region_mean": 0.0006169933003548067,
|
| 1870 |
+
"epoch": 0.76875,
|
| 1871 |
+
"grad_norm": 1.3819693148079288,
|
| 1872 |
+
"kl": 0.016606250894255936,
|
| 1873 |
+
"learning_rate": 1.3413122140054217e-07,
|
| 1874 |
+
"loss": 0.00019602995598688722,
|
| 1875 |
+
"step": 123,
|
| 1876 |
+
"step_time": 74.98103069514036
|
| 1877 |
+
},
|
| 1878 |
+
{
|
| 1879 |
+
"clip_ratio/high_max": 0.004828446923056617,
|
| 1880 |
+
"clip_ratio/high_mean": 0.001853768299042713,
|
| 1881 |
+
"clip_ratio/low_mean": 0.00018310546875,
|
| 1882 |
+
"clip_ratio/low_min": 0.0,
|
| 1883 |
+
"clip_ratio/region_mean": 0.0020368737750686705,
|
| 1884 |
+
"epoch": 0.775,
|
| 1885 |
+
"grad_norm": 1.5490775573072801,
|
| 1886 |
+
"kl": 0.018943154136650264,
|
| 1887 |
+
"learning_rate": 1.272995289285202e-07,
|
| 1888 |
+
"loss": -3.704056143760681e-05,
|
| 1889 |
+
"step": 124,
|
| 1890 |
+
"step_time": 74.00208830460906
|
| 1891 |
+
},
|
| 1892 |
+
{
|
| 1893 |
+
"clip_ratio/high_max": 0.0018545659841038287,
|
| 1894 |
+
"clip_ratio/high_mean": 0.0005734845544793643,
|
| 1895 |
+
"clip_ratio/low_mean": 0.000335693359375,
|
| 1896 |
+
"clip_ratio/low_min": 0.0,
|
| 1897 |
+
"clip_ratio/region_mean": 0.0009091779138543643,
|
| 1898 |
+
"epoch": 0.78125,
|
| 1899 |
+
"grad_norm": 1.495056399527209,
|
| 1900 |
+
"kl": 0.011007337714545429,
|
| 1901 |
+
"learning_rate": 1.2062093865360457e-07,
|
| 1902 |
+
"loss": 9.245984256267548e-05,
|
| 1903 |
+
"step": 125,
|
| 1904 |
+
"step_time": 74.73828943818808
|
| 1905 |
+
},
|
| 1906 |
+
{
|
| 1907 |
+
"clip_ratio/high_max": 0.000885325025592465,
|
| 1908 |
+
"clip_ratio/high_mean": 0.0002933405739895534,
|
| 1909 |
+
"clip_ratio/low_mean": 0.00029055769482511096,
|
| 1910 |
+
"clip_ratio/low_min": 0.0,
|
| 1911 |
+
"clip_ratio/region_mean": 0.0005838982688146643,
|
| 1912 |
+
"epoch": 0.7875,
|
| 1913 |
+
"grad_norm": 1.3090214288172612,
|
| 1914 |
+
"kl": 0.01160969806369394,
|
| 1915 |
+
"learning_rate": 1.1409819408411897e-07,
|
| 1916 |
+
"loss": 0.00014726072549819946,
|
| 1917 |
+
"step": 126,
|
| 1918 |
+
"step_time": 75.2373479641974
|
| 1919 |
+
},
|
| 1920 |
+
{
|
| 1921 |
+
"clip_ratio/high_max": 0.004401224767207168,
|
| 1922 |
+
"clip_ratio/high_mean": 0.0014231072163966019,
|
| 1923 |
+
"clip_ratio/low_mean": 0.00020501745893852785,
|
| 1924 |
+
"clip_ratio/low_min": 0.0,
|
| 1925 |
+
"clip_ratio/region_mean": 0.0016281246244034264,
|
| 1926 |
+
"epoch": 0.79375,
|
| 1927 |
+
"grad_norm": 2.2604399567346496,
|
| 1928 |
+
"kl": 0.01724324410315603,
|
| 1929 |
+
"learning_rate": 1.0773397470829143e-07,
|
| 1930 |
+
"loss": 0.0003859587013721466,
|
| 1931 |
+
"step": 127,
|
| 1932 |
+
"step_time": 74.11185354739428
|
| 1933 |
+
},
|
| 1934 |
+
{
|
| 1935 |
+
"clip_ratio/high_max": 0.004197781992843375,
|
| 1936 |
+
"clip_ratio/high_mean": 0.001188401147373952,
|
| 1937 |
+
"clip_ratio/low_mean": 0.0004607444934663363,
|
| 1938 |
+
"clip_ratio/low_min": 0.0,
|
| 1939 |
+
"clip_ratio/region_mean": 0.0016491456408402883,
|
| 1940 |
+
"epoch": 0.8,
|
| 1941 |
+
"grad_norm": 1.9455995023515806,
|
| 1942 |
+
"kl": 0.03236217447556555,
|
| 1943 |
+
"learning_rate": 1.0153089489354256e-07,
|
| 1944 |
+
"loss": 0.00021363080304581672,
|
| 1945 |
+
"step": 128,
|
| 1946 |
+
"step_time": 75.64158493280411
|
| 1947 |
+
},
|
| 1948 |
+
{
|
| 1949 |
+
"clip_ratio/high_max": 0.0,
|
| 1950 |
+
"clip_ratio/high_mean": 0.0,
|
| 1951 |
+
"clip_ratio/low_mean": 0.0,
|
| 1952 |
+
"clip_ratio/low_min": 0.0,
|
| 1953 |
+
"clip_ratio/region_mean": 0.0,
|
| 1954 |
+
"completions/clipped_ratio": 0.265625,
|
| 1955 |
+
"completions/max_length": 2048.0,
|
| 1956 |
+
"completions/mean_length": 1237.78515625,
|
| 1957 |
+
"completions/min_length": 144.0,
|
| 1958 |
+
"epoch": 0.80625,
|
| 1959 |
+
"frac_reward_zero_std": 0.1875,
|
| 1960 |
+
"grad_norm": 1.636693303269506,
|
| 1961 |
+
"kl": 0.015208918019197881,
|
| 1962 |
+
"learning_rate": 9.549150281252632e-08,
|
| 1963 |
+
"loss": 0.00015214766608551145,
|
| 1964 |
+
"reward": 0.515625,
|
| 1965 |
+
"reward_std": 0.36060887575149536,
|
| 1966 |
+
"rewards/MajorityAccuracy/mean": 0.515625,
|
| 1967 |
+
"rewards/MajorityAccuracy/std": 0.5007347464561462,
|
| 1968 |
+
"step": 129,
|
| 1969 |
+
"step_time": 514.7536188103259
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"clip_ratio/high_max": 0.002930103786638938,
|
| 1973 |
+
"clip_ratio/high_mean": 0.0007505064095312264,
|
| 1974 |
+
"clip_ratio/low_mean": 0.00040740182157605886,
|
| 1975 |
+
"clip_ratio/low_min": 0.0,
|
| 1976 |
+
"clip_ratio/region_mean": 0.001157908198365476,
|
| 1977 |
+
"epoch": 0.8125,
|
| 1978 |
+
"grad_norm": 2.7707474089319013,
|
| 1979 |
+
"kl": 0.016175380675122142,
|
| 1980 |
+
"learning_rate": 8.961827939636196e-08,
|
| 1981 |
+
"loss": -4.0811486542224884e-05,
|
| 1982 |
+
"step": 130,
|
| 1983 |
+
"step_time": 74.78703932091594
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"clip_ratio/high_max": 0.0023499914022977464,
|
| 1987 |
+
"clip_ratio/high_mean": 0.0009669972878327826,
|
| 1988 |
+
"clip_ratio/low_mean": 0.00013374183436098974,
|
| 1989 |
+
"clip_ratio/low_min": 0.0,
|
| 1990 |
+
"clip_ratio/region_mean": 0.0011007391221937723,
|
| 1991 |
+
"epoch": 0.81875,
|
| 1992 |
+
"grad_norm": 2.0275609412619517,
|
| 1993 |
+
"kl": 0.01613416918553412,
|
| 1994 |
+
"learning_rate": 8.391363731548811e-08,
|
| 1995 |
+
"loss": 0.00014563556760549545,
|
| 1996 |
+
"step": 131,
|
| 1997 |
+
"step_time": 74.7958739772439
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"clip_ratio/high_max": 0.002675424242625013,
|
| 2001 |
+
"clip_ratio/high_mean": 0.000858895575220231,
|
| 2002 |
+
"clip_ratio/low_mean": 0.00038681228943460155,
|
| 2003 |
+
"clip_ratio/low_min": 0.0,
|
| 2004 |
+
"clip_ratio/region_mean": 0.0012457078701118007,
|
| 2005 |
+
"epoch": 0.825,
|
| 2006 |
+
"grad_norm": 1.9298130283190533,
|
| 2007 |
+
"kl": 0.015873914235271513,
|
| 2008 |
+
"learning_rate": 7.837991998855897e-08,
|
| 2009 |
+
"loss": 0.000243532529566437,
|
| 2010 |
+
"step": 132,
|
| 2011 |
+
"step_time": 74.20337053760886
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"clip_ratio/high_max": 0.0013162548348191194,
|
| 2015 |
+
"clip_ratio/high_mean": 0.00032906370870477986,
|
| 2016 |
+
"clip_ratio/low_mean": 9.29951056605205e-05,
|
| 2017 |
+
"clip_ratio/low_min": 0.0,
|
| 2018 |
+
"clip_ratio/region_mean": 0.00042205881436530035,
|
| 2019 |
+
"epoch": 0.83125,
|
| 2020 |
+
"grad_norm": 1.5314078654747647,
|
| 2021 |
+
"kl": 0.013434231863357127,
|
| 2022 |
+
"learning_rate": 7.301940061978722e-08,
|
| 2023 |
+
"loss": 0.00023119134129956365,
|
| 2024 |
+
"step": 133,
|
| 2025 |
+
"step_time": 73.6042508110404
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"clip_ratio/high_max": 0.0011792282602982596,
|
| 2029 |
+
"clip_ratio/high_mean": 0.00041436674291617237,
|
| 2030 |
+
"clip_ratio/low_mean": 0.0003480121013126336,
|
| 2031 |
+
"clip_ratio/low_min": 0.0,
|
| 2032 |
+
"clip_ratio/region_mean": 0.0007623788478667848,
|
| 2033 |
+
"epoch": 0.8375,
|
| 2034 |
+
"grad_norm": 2.3275152541003767,
|
| 2035 |
+
"kl": 0.015750460792332888,
|
| 2036 |
+
"learning_rate": 6.783428126513125e-08,
|
| 2037 |
+
"loss": -0.0001191955670947209,
|
| 2038 |
+
"step": 134,
|
| 2039 |
+
"step_time": 72.60192551836371
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"clip_ratio/high_max": 0.0007160825989558361,
|
| 2043 |
+
"clip_ratio/high_mean": 0.00022431050638260785,
|
| 2044 |
+
"clip_ratio/low_mean": 0.0001068115234375,
|
| 2045 |
+
"clip_ratio/low_min": 0.0,
|
| 2046 |
+
"clip_ratio/region_mean": 0.00033112202800111845,
|
| 2047 |
+
"epoch": 0.84375,
|
| 2048 |
+
"grad_norm": 1.1709326355216616,
|
| 2049 |
+
"kl": 0.010809578816406429,
|
| 2050 |
+
"learning_rate": 6.282669192770895e-08,
|
| 2051 |
+
"loss": 0.00020827597472816706,
|
| 2052 |
+
"step": 135,
|
| 2053 |
+
"step_time": 74.24207011982799
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"clip_ratio/high_max": 0.0026083550474140793,
|
| 2057 |
+
"clip_ratio/high_mean": 0.0009231938856828492,
|
| 2058 |
+
"clip_ratio/low_mean": 0.0006340222607832402,
|
| 2059 |
+
"clip_ratio/low_min": 0.0,
|
| 2060 |
+
"clip_ratio/region_mean": 0.0015572161501040682,
|
| 2061 |
+
"epoch": 0.85,
|
| 2062 |
+
"grad_norm": 1.930487724443307,
|
| 2063 |
+
"kl": 0.017073628958314657,
|
| 2064 |
+
"learning_rate": 5.799868968281074e-08,
|
| 2065 |
+
"loss": 0.00030478835105895996,
|
| 2066 |
+
"step": 136,
|
| 2067 |
+
"step_time": 74.89684385433793
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"clip_ratio/high_max": 0.0,
|
| 2071 |
+
"clip_ratio/high_mean": 0.0,
|
| 2072 |
+
"clip_ratio/low_mean": 0.0,
|
| 2073 |
+
"clip_ratio/low_min": 0.0,
|
| 2074 |
+
"clip_ratio/region_mean": 0.0,
|
| 2075 |
+
"completions/clipped_ratio": 0.26171875,
|
| 2076 |
+
"completions/max_length": 2048.0,
|
| 2077 |
+
"completions/mean_length": 1130.50390625,
|
| 2078 |
+
"completions/min_length": 56.0,
|
| 2079 |
+
"epoch": 0.85625,
|
| 2080 |
+
"frac_reward_zero_std": 0.0625,
|
| 2081 |
+
"grad_norm": 2.143006807008217,
|
| 2082 |
+
"kl": 0.013857169426046312,
|
| 2083 |
+
"learning_rate": 5.335225783287051e-08,
|
| 2084 |
+
"loss": 0.0001385073846904561,
|
| 2085 |
+
"reward": 0.5,
|
| 2086 |
+
"reward_std": 0.4116378724575043,
|
| 2087 |
+
"rewards/MajorityAccuracy/mean": 0.5,
|
| 2088 |
+
"rewards/MajorityAccuracy/std": 0.5009794235229492,
|
| 2089 |
+
"step": 137,
|
| 2090 |
+
"step_time": 511.61212066560984
|
| 2091 |
+
},
|
| 2092 |
+
{
|
| 2093 |
+
"clip_ratio/high_max": 0.004213077569147572,
|
| 2094 |
+
"clip_ratio/high_mean": 0.0014503759375656955,
|
| 2095 |
+
"clip_ratio/low_mean": 0.0002209111553383991,
|
| 2096 |
+
"clip_ratio/low_min": 0.0,
|
| 2097 |
+
"clip_ratio/region_mean": 0.0016712871001800522,
|
| 2098 |
+
"epoch": 0.8625,
|
| 2099 |
+
"grad_norm": 1.6546103566617825,
|
| 2100 |
+
"kl": 0.012210751883685589,
|
| 2101 |
+
"learning_rate": 4.888930509274125e-08,
|
| 2102 |
+
"loss": 9.114108979701996e-05,
|
| 2103 |
+
"step": 138,
|
| 2104 |
+
"step_time": 73.69021527096629
|
| 2105 |
+
},
|
| 2106 |
+
{
|
| 2107 |
+
"clip_ratio/high_max": 0.0027119406731799245,
|
| 2108 |
+
"clip_ratio/high_mean": 0.0008260437680291943,
|
| 2109 |
+
"clip_ratio/low_mean": 0.0001291257285629399,
|
| 2110 |
+
"clip_ratio/low_min": 0.0,
|
| 2111 |
+
"clip_ratio/region_mean": 0.0009551695038680919,
|
| 2112 |
+
"epoch": 0.86875,
|
| 2113 |
+
"grad_norm": 2.036597244902105,
|
| 2114 |
+
"kl": 0.016748178051784635,
|
| 2115 |
+
"learning_rate": 4.4611664805611794e-08,
|
| 2116 |
+
"loss": 0.0004185158759355545,
|
| 2117 |
+
"step": 139,
|
| 2118 |
+
"step_time": 73.58164126425982
|
| 2119 |
+
},
|
| 2120 |
+
{
|
| 2121 |
+
"clip_ratio/high_max": 0.001714406767860055,
|
| 2122 |
+
"clip_ratio/high_mean": 0.0005005978600820526,
|
| 2123 |
+
"clip_ratio/low_mean": 0.0007976595079526305,
|
| 2124 |
+
"clip_ratio/low_min": 0.0,
|
| 2125 |
+
"clip_ratio/region_mean": 0.0012982573680346832,
|
| 2126 |
+
"epoch": 0.875,
|
| 2127 |
+
"grad_norm": 1.718552428334268,
|
| 2128 |
+
"kl": 0.015153248445130885,
|
| 2129 |
+
"learning_rate": 4.0521094189884696e-08,
|
| 2130 |
+
"loss": 6.19981437921524e-05,
|
| 2131 |
+
"step": 140,
|
| 2132 |
+
"step_time": 74.48522027209401
|
| 2133 |
+
},
|
| 2134 |
+
{
|
| 2135 |
+
"clip_ratio/high_max": 0.0016459480830235407,
|
| 2136 |
+
"clip_ratio/high_mean": 0.000544462469406426,
|
| 2137 |
+
"clip_ratio/low_mean": 0.00018310546875,
|
| 2138 |
+
"clip_ratio/low_min": 0.0,
|
| 2139 |
+
"clip_ratio/region_mean": 0.000727567938156426,
|
| 2140 |
+
"epoch": 0.88125,
|
| 2141 |
+
"grad_norm": 1.619205406113154,
|
| 2142 |
+
"kl": 0.014125036424957216,
|
| 2143 |
+
"learning_rate": 3.6619273617325695e-08,
|
| 2144 |
+
"loss": 0.00016590580344200134,
|
| 2145 |
+
"step": 141,
|
| 2146 |
+
"step_time": 74.49084364622831
|
| 2147 |
+
},
|
| 2148 |
+
{
|
| 2149 |
+
"clip_ratio/high_max": 0.0012025919131701812,
|
| 2150 |
+
"clip_ratio/high_mean": 0.0004202575182716828,
|
| 2151 |
+
"clip_ratio/low_mean": 0.0006424478997359984,
|
| 2152 |
+
"clip_ratio/low_min": 0.0,
|
| 2153 |
+
"clip_ratio/region_mean": 0.00106270542164566,
|
| 2154 |
+
"epoch": 0.8875,
|
| 2155 |
+
"grad_norm": 2.9844551300109523,
|
| 2156 |
+
"kl": 0.015560108004137874,
|
| 2157 |
+
"learning_rate": 3.2907805922781476e-08,
|
| 2158 |
+
"loss": 0.00028870347887277603,
|
| 2159 |
+
"step": 142,
|
| 2160 |
+
"step_time": 72.70355377346277
|
| 2161 |
+
},
|
| 2162 |
+
{
|
| 2163 |
+
"clip_ratio/high_max": 0.002853073841833975,
|
| 2164 |
+
"clip_ratio/high_mean": 0.0008418692796112737,
|
| 2165 |
+
"clip_ratio/low_mean": 0.0003751957483473234,
|
| 2166 |
+
"clip_ratio/low_min": 0.0,
|
| 2167 |
+
"clip_ratio/region_mean": 0.0012170650334155653,
|
| 2168 |
+
"epoch": 0.89375,
|
| 2169 |
+
"grad_norm": 4.6521706454090666,
|
| 2170 |
+
"kl": 0.017092958209104836,
|
| 2171 |
+
"learning_rate": 2.9388215745748345e-08,
|
| 2172 |
+
"loss": 0.00014861988893244416,
|
| 2173 |
+
"step": 143,
|
| 2174 |
+
"step_time": 72.8293684348464
|
| 2175 |
+
},
|
| 2176 |
+
{
|
| 2177 |
+
"clip_ratio/high_max": 0.0020103333372389898,
|
| 2178 |
+
"clip_ratio/high_mean": 0.0005627317914331798,
|
| 2179 |
+
"clip_ratio/low_mean": 0.000352631279383786,
|
| 2180 |
+
"clip_ratio/low_min": 0.0,
|
| 2181 |
+
"clip_ratio/region_mean": 0.0009153630489890929,
|
| 2182 |
+
"epoch": 0.9,
|
| 2183 |
+
"grad_norm": 2.89700088593663,
|
| 2184 |
+
"kl": 0.012443124898709357,
|
| 2185 |
+
"learning_rate": 2.6061948904063658e-08,
|
| 2186 |
+
"loss": -0.0003589671105146408,
|
| 2187 |
+
"step": 144,
|
| 2188 |
+
"step_time": 72.64231837913394
|
| 2189 |
+
},
|
| 2190 |
+
{
|
| 2191 |
+
"clip_ratio/high_max": 0.0,
|
| 2192 |
+
"clip_ratio/high_mean": 0.0,
|
| 2193 |
+
"clip_ratio/low_mean": 0.0,
|
| 2194 |
+
"clip_ratio/low_min": 0.0,
|
| 2195 |
+
"clip_ratio/region_mean": 0.0,
|
| 2196 |
+
"completions/clipped_ratio": 0.3203125,
|
| 2197 |
+
"completions/max_length": 2048.0,
|
| 2198 |
+
"completions/mean_length": 1280.421875,
|
| 2199 |
+
"completions/min_length": 123.0,
|
| 2200 |
+
"epoch": 0.90625,
|
| 2201 |
+
"frac_reward_zero_std": 0.25,
|
| 2202 |
+
"grad_norm": 1.2434938725390732,
|
| 2203 |
+
"kl": 0.014555978705175221,
|
| 2204 |
+
"learning_rate": 2.293037179997559e-08,
|
| 2205 |
+
"loss": 0.00014554709196090698,
|
| 2206 |
+
"reward": 0.453125,
|
| 2207 |
+
"reward_std": 0.3533535301685333,
|
| 2208 |
+
"rewards/MajorityAccuracy/mean": 0.453125,
|
| 2209 |
+
"rewards/MajorityAccuracy/std": 0.4987730085849762,
|
| 2210 |
+
"step": 145,
|
| 2211 |
+
"step_time": 517.6411182805896
|
| 2212 |
+
},
|
| 2213 |
+
{
|
| 2214 |
+
"clip_ratio/high_max": 0.0007220203478937037,
|
| 2215 |
+
"clip_ratio/high_mean": 0.00018050508697342593,
|
| 2216 |
+
"clip_ratio/low_mean": 0.00019994521790067665,
|
| 2217 |
+
"clip_ratio/low_min": 0.0,
|
| 2218 |
+
"clip_ratio/region_mean": 0.0003804503012361238,
|
| 2219 |
+
"epoch": 0.9125,
|
| 2220 |
+
"grad_norm": 2.0313651764285616,
|
| 2221 |
+
"kl": 0.014359165681526065,
|
| 2222 |
+
"learning_rate": 1.9994770858837107e-08,
|
| 2223 |
+
"loss": 0.00043821538565680385,
|
| 2224 |
+
"step": 146,
|
| 2225 |
+
"step_time": 74.01912296935916
|
| 2226 |
+
},
|
| 2227 |
+
{
|
| 2228 |
+
"clip_ratio/high_max": 0.0009462006273679435,
|
| 2229 |
+
"clip_ratio/high_mean": 0.0003147515635646414,
|
| 2230 |
+
"clip_ratio/low_mean": 0.00011757271568058059,
|
| 2231 |
+
"clip_ratio/low_min": 0.0,
|
| 2232 |
+
"clip_ratio/region_mean": 0.0004323242774262326,
|
| 2233 |
+
"epoch": 0.91875,
|
| 2234 |
+
"grad_norm": 1.8831485299559545,
|
| 2235 |
+
"kl": 0.013082645717076957,
|
| 2236 |
+
"learning_rate": 1.725635200065323e-08,
|
| 2237 |
+
"loss": 0.00011767353862524033,
|
| 2238 |
+
"step": 147,
|
| 2239 |
+
"step_time": 74.52565836161375
|
| 2240 |
+
},
|
| 2241 |
+
{
|
| 2242 |
+
"clip_ratio/high_max": 0.00165591060067527,
|
| 2243 |
+
"clip_ratio/high_mean": 0.0004325567224441329,
|
| 2244 |
+
"clip_ratio/low_mean": 0.000491939477797132,
|
| 2245 |
+
"clip_ratio/low_min": 0.0,
|
| 2246 |
+
"clip_ratio/region_mean": 0.0009244961911463179,
|
| 2247 |
+
"epoch": 0.925,
|
| 2248 |
+
"grad_norm": 2.2469837605501666,
|
| 2249 |
+
"kl": 0.014070844277739525,
|
| 2250 |
+
"learning_rate": 1.4716240144699187e-08,
|
| 2251 |
+
"loss": 8.55992766446434e-05,
|
| 2252 |
+
"step": 148,
|
| 2253 |
+
"step_time": 75.28892084956169
|
| 2254 |
+
},
|
| 2255 |
+
{
|
| 2256 |
+
"clip_ratio/high_max": 0.0020678117289207876,
|
| 2257 |
+
"clip_ratio/high_mean": 0.0008662999316584319,
|
| 2258 |
+
"clip_ratio/low_mean": 0.0002419132233626442,
|
| 2259 |
+
"clip_ratio/low_min": 0.0,
|
| 2260 |
+
"clip_ratio/region_mean": 0.0011082131532020867,
|
| 2261 |
+
"epoch": 0.93125,
|
| 2262 |
+
"grad_norm": 2.619185615005676,
|
| 2263 |
+
"kl": 0.014924226910807192,
|
| 2264 |
+
"learning_rate": 1.2375478747413015e-08,
|
| 2265 |
+
"loss": 0.00019100838107988238,
|
| 2266 |
+
"step": 149,
|
| 2267 |
+
"step_time": 74.75358823314309
|
| 2268 |
+
},
|
| 2269 |
+
{
|
| 2270 |
+
"clip_ratio/high_max": 0.002543181646615267,
|
| 2271 |
+
"clip_ratio/high_mean": 0.0008062501801759936,
|
| 2272 |
+
"clip_ratio/low_mean": 0.0001900085699162446,
|
| 2273 |
+
"clip_ratio/low_min": 0.0,
|
| 2274 |
+
"clip_ratio/region_mean": 0.0009962587428162806,
|
| 2275 |
+
"epoch": 0.9375,
|
| 2276 |
+
"grad_norm": 1.6501184965785314,
|
| 2277 |
+
"kl": 0.013570723589509726,
|
| 2278 |
+
"learning_rate": 1.0235029373752757e-08,
|
| 2279 |
+
"loss": 0.0003025059122592211,
|
| 2280 |
+
"step": 150,
|
| 2281 |
+
"step_time": 73.02685219049454
|
| 2282 |
+
},
|
| 2283 |
+
{
|
| 2284 |
+
"clip_ratio/high_max": 0.002294555350090377,
|
| 2285 |
+
"clip_ratio/high_mean": 0.0007895195667515509,
|
| 2286 |
+
"clip_ratio/low_mean": 0.0003400865680305287,
|
| 2287 |
+
"clip_ratio/low_min": 0.0,
|
| 2288 |
+
"clip_ratio/region_mean": 0.001129606127506122,
|
| 2289 |
+
"epoch": 0.94375,
|
| 2290 |
+
"grad_norm": 3.082121435564892,
|
| 2291 |
+
"kl": 0.013305945554748178,
|
| 2292 |
+
"learning_rate": 8.295771302193721e-09,
|
| 2293 |
+
"loss": 0.0004080172802787274,
|
| 2294 |
+
"step": 151,
|
| 2295 |
+
"step_time": 73.3901127949357
|
| 2296 |
+
},
|
| 2297 |
+
{
|
| 2298 |
+
"clip_ratio/high_max": 0.00037347291072364897,
|
| 2299 |
+
"clip_ratio/high_mean": 9.336822768091224e-05,
|
| 2300 |
+
"clip_ratio/low_mean": 4.595623613568023e-05,
|
| 2301 |
+
"clip_ratio/low_min": 0.0,
|
| 2302 |
+
"clip_ratio/region_mean": 0.00013932446381659247,
|
| 2303 |
+
"epoch": 0.95,
|
| 2304 |
+
"grad_norm": 2.0224046723456803,
|
| 2305 |
+
"kl": 0.01788234047126025,
|
| 2306 |
+
"learning_rate": 6.558501163527963e-09,
|
| 2307 |
+
"loss": -8.157340198522434e-05,
|
| 2308 |
+
"step": 152,
|
| 2309 |
+
"step_time": 74.21607157588005
|
| 2310 |
+
},
|
| 2311 |
+
{
|
| 2312 |
+
"clip_ratio/high_max": 0.0,
|
| 2313 |
+
"clip_ratio/high_mean": 0.0,
|
| 2314 |
+
"clip_ratio/low_mean": 0.0,
|
| 2315 |
+
"clip_ratio/low_min": 0.0,
|
| 2316 |
+
"clip_ratio/region_mean": 0.0,
|
| 2317 |
+
"completions/clipped_ratio": 0.30859375,
|
| 2318 |
+
"completions/max_length": 2048.0,
|
| 2319 |
+
"completions/mean_length": 1254.30078125,
|
| 2320 |
+
"completions/min_length": 65.0,
|
| 2321 |
+
"epoch": 0.95625,
|
| 2322 |
+
"frac_reward_zero_std": 0.21875,
|
| 2323 |
+
"grad_norm": 1.8643086712594241,
|
| 2324 |
+
"kl": 0.01597043650690466,
|
| 2325 |
+
"learning_rate": 5.023932613615445e-09,
|
| 2326 |
+
"loss": 0.0001597777009010315,
|
| 2327 |
+
"reward": 0.50390625,
|
| 2328 |
+
"reward_std": 0.34219756722450256,
|
| 2329 |
+
"rewards/MajorityAccuracy/mean": 0.50390625,
|
| 2330 |
+
"rewards/MajorityAccuracy/std": 0.5009641051292419,
|
| 2331 |
+
"step": 153,
|
| 2332 |
+
"step_time": 515.7658118791878
|
| 2333 |
+
},
|
| 2334 |
+
{
|
| 2335 |
+
"clip_ratio/high_max": 0.0008875924104359001,
|
| 2336 |
+
"clip_ratio/high_mean": 0.0002707025887502823,
|
| 2337 |
+
"clip_ratio/low_mean": 0.00024533608302590437,
|
| 2338 |
+
"clip_ratio/low_min": 0.0,
|
| 2339 |
+
"clip_ratio/region_mean": 0.0005160386790521443,
|
| 2340 |
+
"epoch": 0.9625,
|
| 2341 |
+
"grad_norm": 1.8252270866706195,
|
| 2342 |
+
"kl": 0.015271487645804882,
|
| 2343 |
+
"learning_rate": 3.6926960402202666e-09,
|
| 2344 |
+
"loss": 0.00020401252550072968,
|
| 2345 |
+
"step": 154,
|
| 2346 |
+
"step_time": 73.92385125905275
|
| 2347 |
+
},
|
| 2348 |
+
{
|
| 2349 |
+
"clip_ratio/high_max": 0.0008958398102549836,
|
| 2350 |
+
"clip_ratio/high_mean": 0.0002239599525637459,
|
| 2351 |
+
"clip_ratio/low_mean": 0.0001765295019140467,
|
| 2352 |
+
"clip_ratio/low_min": 0.0,
|
| 2353 |
+
"clip_ratio/region_mean": 0.0004004894508398138,
|
| 2354 |
+
"epoch": 0.96875,
|
| 2355 |
+
"grad_norm": 1.252581910893551,
|
| 2356 |
+
"kl": 0.022681452333927155,
|
| 2357 |
+
"learning_rate": 2.5653383040524224e-09,
|
| 2358 |
+
"loss": 0.00021443299192469567,
|
| 2359 |
+
"step": 155,
|
| 2360 |
+
"step_time": 72.84839617088437
|
| 2361 |
+
},
|
| 2362 |
+
{
|
| 2363 |
+
"clip_ratio/high_max": 0.001389196448144503,
|
| 2364 |
+
"clip_ratio/high_mean": 0.0005048971688665915,
|
| 2365 |
+
"clip_ratio/low_mean": 0.00016494010014866944,
|
| 2366 |
+
"clip_ratio/low_min": 0.0,
|
| 2367 |
+
"clip_ratio/region_mean": 0.000669837278110208,
|
| 2368 |
+
"epoch": 0.975,
|
| 2369 |
+
"grad_norm": 1.7260137637842543,
|
| 2370 |
+
"kl": 0.023905686102807522,
|
| 2371 |
+
"learning_rate": 1.642322514122385e-09,
|
| 2372 |
+
"loss": 0.0002882710541598499,
|
| 2373 |
+
"step": 156,
|
| 2374 |
+
"step_time": 74.80902964994311
|
| 2375 |
+
},
|
| 2376 |
+
{
|
| 2377 |
+
"clip_ratio/high_max": 0.00235198660811875,
|
| 2378 |
+
"clip_ratio/high_mean": 0.0007292962582141627,
|
| 2379 |
+
"clip_ratio/low_mean": 1.52587890625e-05,
|
| 2380 |
+
"clip_ratio/low_min": 0.0,
|
| 2381 |
+
"clip_ratio/region_mean": 0.0007445550436386839,
|
| 2382 |
+
"epoch": 0.98125,
|
| 2383 |
+
"grad_norm": 1.0589700812977996,
|
| 2384 |
+
"kl": 0.016600233502686024,
|
| 2385 |
+
"learning_rate": 9.240278374995636e-10,
|
| 2386 |
+
"loss": 0.00013788865180686116,
|
| 2387 |
+
"step": 157,
|
| 2388 |
+
"step_time": 74.3064695596695
|
| 2389 |
+
},
|
| 2390 |
+
{
|
| 2391 |
+
"clip_ratio/high_max": 0.004126824263948947,
|
| 2392 |
+
"clip_ratio/high_mean": 0.0012889184145024046,
|
| 2393 |
+
"clip_ratio/low_mean": 0.000624177795543801,
|
| 2394 |
+
"clip_ratio/low_min": 0.0,
|
| 2395 |
+
"clip_ratio/region_mean": 0.001913096202770248,
|
| 2396 |
+
"epoch": 0.9875,
|
| 2397 |
+
"grad_norm": 1.963533190051781,
|
| 2398 |
+
"kl": 0.018561265198513865,
|
| 2399 |
+
"learning_rate": 4.1074934355384006e-10,
|
| 2400 |
+
"loss": 8.841510862112045e-05,
|
| 2401 |
+
"step": 158,
|
| 2402 |
+
"step_time": 74.9793388210237
|
| 2403 |
+
},
|
| 2404 |
+
{
|
| 2405 |
+
"clip_ratio/high_max": 0.0007600432654726319,
|
| 2406 |
+
"clip_ratio/high_mean": 0.00021356017350626644,
|
| 2407 |
+
"clip_ratio/low_mean": 0.0005439942760858685,
|
| 2408 |
+
"clip_ratio/low_min": 0.0,
|
| 2409 |
+
"clip_ratio/region_mean": 0.0007575544477731455,
|
| 2410 |
+
"epoch": 0.99375,
|
| 2411 |
+
"grad_norm": 2.056048571623334,
|
| 2412 |
+
"kl": 0.017109745647758245,
|
| 2413 |
+
"learning_rate": 1.0269788274358537e-10,
|
| 2414 |
+
"loss": 0.00013978696370031685,
|
| 2415 |
+
"step": 159,
|
| 2416 |
+
"step_time": 75.07948110252619
|
| 2417 |
+
},
|
| 2418 |
+
{
|
| 2419 |
+
"clip_ratio/high_max": 0.0014208552602212876,
|
| 2420 |
+
"clip_ratio/high_mean": 0.0005744461086578667,
|
| 2421 |
+
"clip_ratio/low_mean": 0.0008020315581234172,
|
| 2422 |
+
"clip_ratio/low_min": 0.0,
|
| 2423 |
+
"clip_ratio/region_mean": 0.0013764776813331991,
|
| 2424 |
+
"epoch": 1.0,
|
| 2425 |
+
"grad_norm": 1.9622077285827297,
|
| 2426 |
+
"kl": 0.01676316373050213,
|
| 2427 |
+
"learning_rate": 0.0,
|
| 2428 |
+
"loss": 0.0005262978374958038,
|
| 2429 |
+
"step": 160,
|
| 2430 |
+
"step_time": 75.42604714632034
|
| 2431 |
+
}
|
| 2432 |
+
],
|
| 2433 |
+
"logging_steps": 1,
|
| 2434 |
+
"max_steps": 160,
|
| 2435 |
+
"num_input_tokens_seen": 0,
|
| 2436 |
+
"num_train_epochs": 1,
|
| 2437 |
+
"save_steps": 80,
|
| 2438 |
+
"stateful_callbacks": {
|
| 2439 |
+
"TrainerControl": {
|
| 2440 |
+
"args": {
|
| 2441 |
+
"should_epoch_stop": false,
|
| 2442 |
+
"should_evaluate": false,
|
| 2443 |
+
"should_log": false,
|
| 2444 |
+
"should_save": true,
|
| 2445 |
+
"should_training_stop": true
|
| 2446 |
+
},
|
| 2447 |
+
"attributes": {}
|
| 2448 |
+
}
|
| 2449 |
+
},
|
| 2450 |
+
"total_flos": 0.0,
|
| 2451 |
+
"train_batch_size": 1,
|
| 2452 |
+
"trial_name": null,
|
| 2453 |
+
"trial_params": null
|
| 2454 |
+
}
|
c1_no_history/round_2/chat_template.jinja
ADDED
|
@@ -0,0 +1,154 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{%- set image_count = namespace(value=0) %}
|
| 2 |
+
{%- set video_count = namespace(value=0) %}
|
| 3 |
+
{%- macro render_content(content, do_vision_count, is_system_content=false) %}
|
| 4 |
+
{%- if content is string %}
|
| 5 |
+
{{- content }}
|
| 6 |
+
{%- elif content is iterable and content is not mapping %}
|
| 7 |
+
{%- for item in content %}
|
| 8 |
+
{%- if 'image' in item or 'image_url' in item or item.type == 'image' %}
|
| 9 |
+
{%- if is_system_content %}
|
| 10 |
+
{{- raise_exception('System message cannot contain images.') }}
|
| 11 |
+
{%- endif %}
|
| 12 |
+
{%- if do_vision_count %}
|
| 13 |
+
{%- set image_count.value = image_count.value + 1 %}
|
| 14 |
+
{%- endif %}
|
| 15 |
+
{%- if add_vision_id %}
|
| 16 |
+
{{- 'Picture ' ~ image_count.value ~ ': ' }}
|
| 17 |
+
{%- endif %}
|
| 18 |
+
{{- '<|vision_start|><|image_pad|><|vision_end|>' }}
|
| 19 |
+
{%- elif 'video' in item or item.type == 'video' %}
|
| 20 |
+
{%- if is_system_content %}
|
| 21 |
+
{{- raise_exception('System message cannot contain videos.') }}
|
| 22 |
+
{%- endif %}
|
| 23 |
+
{%- if do_vision_count %}
|
| 24 |
+
{%- set video_count.value = video_count.value + 1 %}
|
| 25 |
+
{%- endif %}
|
| 26 |
+
{%- if add_vision_id %}
|
| 27 |
+
{{- 'Video ' ~ video_count.value ~ ': ' }}
|
| 28 |
+
{%- endif %}
|
| 29 |
+
{{- '<|vision_start|><|video_pad|><|vision_end|>' }}
|
| 30 |
+
{%- elif 'text' in item %}
|
| 31 |
+
{{- item.text }}
|
| 32 |
+
{%- else %}
|
| 33 |
+
{{- raise_exception('Unexpected item type in content.') }}
|
| 34 |
+
{%- endif %}
|
| 35 |
+
{%- endfor %}
|
| 36 |
+
{%- elif content is none or content is undefined %}
|
| 37 |
+
{{- '' }}
|
| 38 |
+
{%- else %}
|
| 39 |
+
{{- raise_exception('Unexpected content type.') }}
|
| 40 |
+
{%- endif %}
|
| 41 |
+
{%- endmacro %}
|
| 42 |
+
{%- if not messages %}
|
| 43 |
+
{{- raise_exception('No messages provided.') }}
|
| 44 |
+
{%- endif %}
|
| 45 |
+
{%- if tools and tools is iterable and tools is not mapping %}
|
| 46 |
+
{{- '<|im_start|>system\n' }}
|
| 47 |
+
{{- "# Tools\n\nYou have access to the following functions:\n\n<tools>" }}
|
| 48 |
+
{%- for tool in tools %}
|
| 49 |
+
{{- "\n" }}
|
| 50 |
+
{{- tool | tojson }}
|
| 51 |
+
{%- endfor %}
|
| 52 |
+
{{- "\n</tools>" }}
|
| 53 |
+
{{- '\n\nIf you choose to call a function ONLY reply in the following format with NO suffix:\n\n<tool_call>\n<function=example_function_name>\n<parameter=example_parameter_1>\nvalue_1\n</parameter>\n<parameter=example_parameter_2>\nThis is the value for the second parameter\nthat can span\nmultiple lines\n</parameter>\n</function>\n</tool_call>\n\n<IMPORTANT>\nReminder:\n- Function calls MUST follow the specified format: an inner <function=...></function> block must be nested within <tool_call></tool_call> XML tags\n- Required parameters MUST be specified\n- You may provide optional reasoning for your function call in natural language BEFORE the function call, but NOT after\n- If there is no function call available, answer the question like normal with your current knowledge and do not tell the user about function calls\n</IMPORTANT>' }}
|
| 54 |
+
{%- if messages[0].role == 'system' %}
|
| 55 |
+
{%- set content = render_content(messages[0].content, false, true)|trim %}
|
| 56 |
+
{%- if content %}
|
| 57 |
+
{{- '\n\n' + content }}
|
| 58 |
+
{%- endif %}
|
| 59 |
+
{%- endif %}
|
| 60 |
+
{{- '<|im_end|>\n' }}
|
| 61 |
+
{%- else %}
|
| 62 |
+
{%- if messages[0].role == 'system' %}
|
| 63 |
+
{%- set content = render_content(messages[0].content, false, true)|trim %}
|
| 64 |
+
{{- '<|im_start|>system\n' + content + '<|im_end|>\n' }}
|
| 65 |
+
{%- endif %}
|
| 66 |
+
{%- endif %}
|
| 67 |
+
{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
|
| 68 |
+
{%- for message in messages[::-1] %}
|
| 69 |
+
{%- set index = (messages|length - 1) - loop.index0 %}
|
| 70 |
+
{%- if ns.multi_step_tool and message.role == "user" %}
|
| 71 |
+
{%- set content = render_content(message.content, false)|trim %}
|
| 72 |
+
{%- if not(content.startswith('<tool_response>') and content.endswith('</tool_response>')) %}
|
| 73 |
+
{%- set ns.multi_step_tool = false %}
|
| 74 |
+
{%- set ns.last_query_index = index %}
|
| 75 |
+
{%- endif %}
|
| 76 |
+
{%- endif %}
|
| 77 |
+
{%- endfor %}
|
| 78 |
+
{%- if ns.multi_step_tool %}
|
| 79 |
+
{{- raise_exception('No user query found in messages.') }}
|
| 80 |
+
{%- endif %}
|
| 81 |
+
{%- for message in messages %}
|
| 82 |
+
{%- set content = render_content(message.content, true)|trim %}
|
| 83 |
+
{%- if message.role == "system" %}
|
| 84 |
+
{%- if not loop.first %}
|
| 85 |
+
{{- raise_exception('System message must be at the beginning.') }}
|
| 86 |
+
{%- endif %}
|
| 87 |
+
{%- elif message.role == "user" %}
|
| 88 |
+
{{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
|
| 89 |
+
{%- elif message.role == "assistant" %}
|
| 90 |
+
{%- set reasoning_content = '' %}
|
| 91 |
+
{%- if message.reasoning_content is string %}
|
| 92 |
+
{%- set reasoning_content = message.reasoning_content %}
|
| 93 |
+
{%- else %}
|
| 94 |
+
{%- if '</think>' in content %}
|
| 95 |
+
{%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
|
| 96 |
+
{%- set content = content.split('</think>')[-1].lstrip('\n') %}
|
| 97 |
+
{%- endif %}
|
| 98 |
+
{%- endif %}
|
| 99 |
+
{%- set reasoning_content = reasoning_content|trim %}
|
| 100 |
+
{%- if loop.index0 > ns.last_query_index %}
|
| 101 |
+
{{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content + '\n</think>\n\n' + content }}
|
| 102 |
+
{%- else %}
|
| 103 |
+
{{- '<|im_start|>' + message.role + '\n' + content }}
|
| 104 |
+
{%- endif %}
|
| 105 |
+
{%- if message.tool_calls and message.tool_calls is iterable and message.tool_calls is not mapping %}
|
| 106 |
+
{%- for tool_call in message.tool_calls %}
|
| 107 |
+
{%- if tool_call.function is defined %}
|
| 108 |
+
{%- set tool_call = tool_call.function %}
|
| 109 |
+
{%- endif %}
|
| 110 |
+
{%- if loop.first %}
|
| 111 |
+
{%- if content|trim %}
|
| 112 |
+
{{- '\n\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
|
| 113 |
+
{%- else %}
|
| 114 |
+
{{- '<tool_call>\n<function=' + tool_call.name + '>\n' }}
|
| 115 |
+
{%- endif %}
|
| 116 |
+
{%- else %}
|
| 117 |
+
{{- '\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
|
| 118 |
+
{%- endif %}
|
| 119 |
+
{%- if tool_call.arguments is defined %}
|
| 120 |
+
{%- for args_name, args_value in tool_call.arguments|items %}
|
| 121 |
+
{{- '<parameter=' + args_name + '>\n' }}
|
| 122 |
+
{%- set args_value = args_value | tojson | safe if args_value is mapping or (args_value is sequence and args_value is not string) else args_value | string %}
|
| 123 |
+
{{- args_value }}
|
| 124 |
+
{{- '\n</parameter>\n' }}
|
| 125 |
+
{%- endfor %}
|
| 126 |
+
{%- endif %}
|
| 127 |
+
{{- '</function>\n</tool_call>' }}
|
| 128 |
+
{%- endfor %}
|
| 129 |
+
{%- endif %}
|
| 130 |
+
{{- '<|im_end|>\n' }}
|
| 131 |
+
{%- elif message.role == "tool" %}
|
| 132 |
+
{%- if loop.previtem and loop.previtem.role != "tool" %}
|
| 133 |
+
{{- '<|im_start|>user' }}
|
| 134 |
+
{%- endif %}
|
| 135 |
+
{{- '\n<tool_response>\n' }}
|
| 136 |
+
{{- content }}
|
| 137 |
+
{{- '\n</tool_response>' }}
|
| 138 |
+
{%- if not loop.last and loop.nextitem.role != "tool" %}
|
| 139 |
+
{{- '<|im_end|>\n' }}
|
| 140 |
+
{%- elif loop.last %}
|
| 141 |
+
{{- '<|im_end|>\n' }}
|
| 142 |
+
{%- endif %}
|
| 143 |
+
{%- else %}
|
| 144 |
+
{{- raise_exception('Unexpected message role.') }}
|
| 145 |
+
{%- endif %}
|
| 146 |
+
{%- endfor %}
|
| 147 |
+
{%- if add_generation_prompt %}
|
| 148 |
+
{{- '<|im_start|>assistant\n' }}
|
| 149 |
+
{%- if enable_thinking is defined and enable_thinking is false %}
|
| 150 |
+
{{- '<think>\n\n</think>\n\n' }}
|
| 151 |
+
{%- else %}
|
| 152 |
+
{{- '<think>\n' }}
|
| 153 |
+
{%- endif %}
|
| 154 |
+
{%- endif %}
|
c1_no_history/round_2/config.json
ADDED
|
@@ -0,0 +1,113 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"Qwen3_5ForConditionalGeneration"
|
| 4 |
+
],
|
| 5 |
+
"dtype": "bfloat16",
|
| 6 |
+
"eos_token_id": 248046,
|
| 7 |
+
"hidden_size": 4096,
|
| 8 |
+
"image_token_id": 248056,
|
| 9 |
+
"model_type": "qwen3_5",
|
| 10 |
+
"pad_token_id": 248044,
|
| 11 |
+
"text_config": {
|
| 12 |
+
"attention_bias": false,
|
| 13 |
+
"attention_dropout": 0.0,
|
| 14 |
+
"attn_output_gate": true,
|
| 15 |
+
"bos_token_id": null,
|
| 16 |
+
"dtype": "bfloat16",
|
| 17 |
+
"eos_token_id": 248044,
|
| 18 |
+
"full_attention_interval": 4,
|
| 19 |
+
"head_dim": 256,
|
| 20 |
+
"hidden_act": "silu",
|
| 21 |
+
"hidden_size": 4096,
|
| 22 |
+
"initializer_range": 0.02,
|
| 23 |
+
"intermediate_size": 12288,
|
| 24 |
+
"layer_types": [
|
| 25 |
+
"linear_attention",
|
| 26 |
+
"linear_attention",
|
| 27 |
+
"linear_attention",
|
| 28 |
+
"full_attention",
|
| 29 |
+
"linear_attention",
|
| 30 |
+
"linear_attention",
|
| 31 |
+
"linear_attention",
|
| 32 |
+
"full_attention",
|
| 33 |
+
"linear_attention",
|
| 34 |
+
"linear_attention",
|
| 35 |
+
"linear_attention",
|
| 36 |
+
"full_attention",
|
| 37 |
+
"linear_attention",
|
| 38 |
+
"linear_attention",
|
| 39 |
+
"linear_attention",
|
| 40 |
+
"full_attention",
|
| 41 |
+
"linear_attention",
|
| 42 |
+
"linear_attention",
|
| 43 |
+
"linear_attention",
|
| 44 |
+
"full_attention",
|
| 45 |
+
"linear_attention",
|
| 46 |
+
"linear_attention",
|
| 47 |
+
"linear_attention",
|
| 48 |
+
"full_attention",
|
| 49 |
+
"linear_attention",
|
| 50 |
+
"linear_attention",
|
| 51 |
+
"linear_attention",
|
| 52 |
+
"full_attention",
|
| 53 |
+
"linear_attention",
|
| 54 |
+
"linear_attention",
|
| 55 |
+
"linear_attention",
|
| 56 |
+
"full_attention"
|
| 57 |
+
],
|
| 58 |
+
"linear_conv_kernel_dim": 4,
|
| 59 |
+
"linear_key_head_dim": 128,
|
| 60 |
+
"linear_num_key_heads": 16,
|
| 61 |
+
"linear_num_value_heads": 32,
|
| 62 |
+
"linear_value_head_dim": 128,
|
| 63 |
+
"mamba_ssm_dtype": "float32",
|
| 64 |
+
"max_position_embeddings": 262144,
|
| 65 |
+
"mlp_only_layers": [],
|
| 66 |
+
"model_type": "qwen3_5_text",
|
| 67 |
+
"mtp_num_hidden_layers": 1,
|
| 68 |
+
"mtp_use_dedicated_embeddings": false,
|
| 69 |
+
"num_attention_heads": 16,
|
| 70 |
+
"num_hidden_layers": 32,
|
| 71 |
+
"num_key_value_heads": 4,
|
| 72 |
+
"pad_token_id": 248044,
|
| 73 |
+
"partial_rotary_factor": 0.25,
|
| 74 |
+
"rms_norm_eps": 1e-06,
|
| 75 |
+
"rope_parameters": {
|
| 76 |
+
"mrope_interleaved": true,
|
| 77 |
+
"mrope_section": [
|
| 78 |
+
11,
|
| 79 |
+
11,
|
| 80 |
+
10
|
| 81 |
+
],
|
| 82 |
+
"partial_rotary_factor": 0.25,
|
| 83 |
+
"rope_theta": 10000000,
|
| 84 |
+
"rope_type": "default"
|
| 85 |
+
},
|
| 86 |
+
"tie_word_embeddings": false,
|
| 87 |
+
"use_cache": false,
|
| 88 |
+
"vocab_size": 248320
|
| 89 |
+
},
|
| 90 |
+
"tie_word_embeddings": false,
|
| 91 |
+
"transformers_version": "5.10.4",
|
| 92 |
+
"use_cache": false,
|
| 93 |
+
"video_token_id": 248057,
|
| 94 |
+
"vision_config": {
|
| 95 |
+
"deepstack_visual_indexes": [],
|
| 96 |
+
"depth": 27,
|
| 97 |
+
"dtype": "bfloat16",
|
| 98 |
+
"hidden_act": "gelu_pytorch_tanh",
|
| 99 |
+
"hidden_size": 1152,
|
| 100 |
+
"in_channels": 3,
|
| 101 |
+
"initializer_range": 0.02,
|
| 102 |
+
"intermediate_size": 4304,
|
| 103 |
+
"model_type": "qwen3_5_vision",
|
| 104 |
+
"num_heads": 16,
|
| 105 |
+
"num_position_embeddings": 2304,
|
| 106 |
+
"out_hidden_size": 4096,
|
| 107 |
+
"patch_size": 16,
|
| 108 |
+
"spatial_merge_size": 2,
|
| 109 |
+
"temporal_patch_size": 2
|
| 110 |
+
},
|
| 111 |
+
"vision_end_token_id": 248054,
|
| 112 |
+
"vision_start_token_id": 248053
|
| 113 |
+
}
|
c1_no_history/round_2/generation_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"eos_token_id": [
|
| 4 |
+
248044,
|
| 5 |
+
248046
|
| 6 |
+
],
|
| 7 |
+
"transformers_version": "5.10.4",
|
| 8 |
+
"use_cache": true
|
| 9 |
+
}
|
c1_no_history/round_2/model.safetensors.index.json
ADDED
|
@@ -0,0 +1,768 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"metadata": {
|
| 3 |
+
"total_parameters": 1469680,
|
| 4 |
+
"total_size": 18819627488
|
| 5 |
+
},
|
| 6 |
+
"weight_map": {
|
| 7 |
+
"lm_head.weight": "model-00001-of-00004.safetensors",
|
| 8 |
+
"model.language_model.embed_tokens.weight": "model-00001-of-00004.safetensors",
|
| 9 |
+
"model.language_model.layers.0.input_layernorm.weight": "model-00001-of-00004.safetensors",
|
| 10 |
+
"model.language_model.layers.0.linear_attn.A_log": "model-00001-of-00004.safetensors",
|
| 11 |
+
"model.language_model.layers.0.linear_attn.conv1d.weight": "model-00001-of-00004.safetensors",
|
| 12 |
+
"model.language_model.layers.0.linear_attn.dt_bias": "model-00001-of-00004.safetensors",
|
| 13 |
+
"model.language_model.layers.0.linear_attn.in_proj_a.weight": "model-00001-of-00004.safetensors",
|
| 14 |
+
"model.language_model.layers.0.linear_attn.in_proj_b.weight": "model-00001-of-00004.safetensors",
|
| 15 |
+
"model.language_model.layers.0.linear_attn.in_proj_qkv.weight": "model-00001-of-00004.safetensors",
|
| 16 |
+
"model.language_model.layers.0.linear_attn.in_proj_z.weight": "model-00001-of-00004.safetensors",
|
| 17 |
+
"model.language_model.layers.0.linear_attn.norm.weight": "model-00001-of-00004.safetensors",
|
| 18 |
+
"model.language_model.layers.0.linear_attn.out_proj.weight": "model-00001-of-00004.safetensors",
|
| 19 |
+
"model.language_model.layers.0.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
|
| 20 |
+
"model.language_model.layers.0.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
|
| 21 |
+
"model.language_model.layers.0.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
|
| 22 |
+
"model.language_model.layers.0.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
|
| 23 |
+
"model.language_model.layers.1.input_layernorm.weight": "model-00001-of-00004.safetensors",
|
| 24 |
+
"model.language_model.layers.1.linear_attn.A_log": "model-00001-of-00004.safetensors",
|
| 25 |
+
"model.language_model.layers.1.linear_attn.conv1d.weight": "model-00001-of-00004.safetensors",
|
| 26 |
+
"model.language_model.layers.1.linear_attn.dt_bias": "model-00001-of-00004.safetensors",
|
| 27 |
+
"model.language_model.layers.1.linear_attn.in_proj_a.weight": "model-00001-of-00004.safetensors",
|
| 28 |
+
"model.language_model.layers.1.linear_attn.in_proj_b.weight": "model-00001-of-00004.safetensors",
|
| 29 |
+
"model.language_model.layers.1.linear_attn.in_proj_qkv.weight": "model-00001-of-00004.safetensors",
|
| 30 |
+
"model.language_model.layers.1.linear_attn.in_proj_z.weight": "model-00001-of-00004.safetensors",
|
| 31 |
+
"model.language_model.layers.1.linear_attn.norm.weight": "model-00001-of-00004.safetensors",
|
| 32 |
+
"model.language_model.layers.1.linear_attn.out_proj.weight": "model-00001-of-00004.safetensors",
|
| 33 |
+
"model.language_model.layers.1.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
|
| 34 |
+
"model.language_model.layers.1.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
|
| 35 |
+
"model.language_model.layers.1.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
|
| 36 |
+
"model.language_model.layers.1.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
|
| 37 |
+
"model.language_model.layers.10.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 38 |
+
"model.language_model.layers.10.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 39 |
+
"model.language_model.layers.10.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 40 |
+
"model.language_model.layers.10.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 41 |
+
"model.language_model.layers.10.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 42 |
+
"model.language_model.layers.10.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 43 |
+
"model.language_model.layers.10.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 44 |
+
"model.language_model.layers.10.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 45 |
+
"model.language_model.layers.10.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 46 |
+
"model.language_model.layers.10.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 47 |
+
"model.language_model.layers.10.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 48 |
+
"model.language_model.layers.10.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 49 |
+
"model.language_model.layers.10.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 50 |
+
"model.language_model.layers.10.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 51 |
+
"model.language_model.layers.11.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 52 |
+
"model.language_model.layers.11.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 53 |
+
"model.language_model.layers.11.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 54 |
+
"model.language_model.layers.11.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 55 |
+
"model.language_model.layers.11.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 56 |
+
"model.language_model.layers.11.self_attn.k_norm.weight": "model-00002-of-00004.safetensors",
|
| 57 |
+
"model.language_model.layers.11.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
|
| 58 |
+
"model.language_model.layers.11.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
|
| 59 |
+
"model.language_model.layers.11.self_attn.q_norm.weight": "model-00002-of-00004.safetensors",
|
| 60 |
+
"model.language_model.layers.11.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
|
| 61 |
+
"model.language_model.layers.11.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
|
| 62 |
+
"model.language_model.layers.12.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 63 |
+
"model.language_model.layers.12.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 64 |
+
"model.language_model.layers.12.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 65 |
+
"model.language_model.layers.12.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 66 |
+
"model.language_model.layers.12.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 67 |
+
"model.language_model.layers.12.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 68 |
+
"model.language_model.layers.12.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 69 |
+
"model.language_model.layers.12.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 70 |
+
"model.language_model.layers.12.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 71 |
+
"model.language_model.layers.12.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 72 |
+
"model.language_model.layers.12.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 73 |
+
"model.language_model.layers.12.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 74 |
+
"model.language_model.layers.12.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 75 |
+
"model.language_model.layers.12.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 76 |
+
"model.language_model.layers.13.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 77 |
+
"model.language_model.layers.13.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 78 |
+
"model.language_model.layers.13.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 79 |
+
"model.language_model.layers.13.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 80 |
+
"model.language_model.layers.13.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 81 |
+
"model.language_model.layers.13.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 82 |
+
"model.language_model.layers.13.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 83 |
+
"model.language_model.layers.13.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 84 |
+
"model.language_model.layers.13.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 85 |
+
"model.language_model.layers.13.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 86 |
+
"model.language_model.layers.13.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 87 |
+
"model.language_model.layers.13.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 88 |
+
"model.language_model.layers.13.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 89 |
+
"model.language_model.layers.13.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 90 |
+
"model.language_model.layers.14.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 91 |
+
"model.language_model.layers.14.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 92 |
+
"model.language_model.layers.14.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 93 |
+
"model.language_model.layers.14.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 94 |
+
"model.language_model.layers.14.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 95 |
+
"model.language_model.layers.14.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 96 |
+
"model.language_model.layers.14.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 97 |
+
"model.language_model.layers.14.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 98 |
+
"model.language_model.layers.14.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 99 |
+
"model.language_model.layers.14.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 100 |
+
"model.language_model.layers.14.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 101 |
+
"model.language_model.layers.14.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 102 |
+
"model.language_model.layers.14.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 103 |
+
"model.language_model.layers.14.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 104 |
+
"model.language_model.layers.15.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 105 |
+
"model.language_model.layers.15.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 106 |
+
"model.language_model.layers.15.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 107 |
+
"model.language_model.layers.15.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 108 |
+
"model.language_model.layers.15.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 109 |
+
"model.language_model.layers.15.self_attn.k_norm.weight": "model-00003-of-00004.safetensors",
|
| 110 |
+
"model.language_model.layers.15.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
|
| 111 |
+
"model.language_model.layers.15.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
|
| 112 |
+
"model.language_model.layers.15.self_attn.q_norm.weight": "model-00003-of-00004.safetensors",
|
| 113 |
+
"model.language_model.layers.15.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
|
| 114 |
+
"model.language_model.layers.15.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
|
| 115 |
+
"model.language_model.layers.16.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 116 |
+
"model.language_model.layers.16.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 117 |
+
"model.language_model.layers.16.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 118 |
+
"model.language_model.layers.16.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 119 |
+
"model.language_model.layers.16.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 120 |
+
"model.language_model.layers.16.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 121 |
+
"model.language_model.layers.16.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 122 |
+
"model.language_model.layers.16.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 123 |
+
"model.language_model.layers.16.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 124 |
+
"model.language_model.layers.16.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 125 |
+
"model.language_model.layers.16.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 126 |
+
"model.language_model.layers.16.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 127 |
+
"model.language_model.layers.16.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 128 |
+
"model.language_model.layers.16.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 129 |
+
"model.language_model.layers.17.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 130 |
+
"model.language_model.layers.17.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 131 |
+
"model.language_model.layers.17.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 132 |
+
"model.language_model.layers.17.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 133 |
+
"model.language_model.layers.17.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 134 |
+
"model.language_model.layers.17.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 135 |
+
"model.language_model.layers.17.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 136 |
+
"model.language_model.layers.17.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 137 |
+
"model.language_model.layers.17.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 138 |
+
"model.language_model.layers.17.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 139 |
+
"model.language_model.layers.17.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 140 |
+
"model.language_model.layers.17.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 141 |
+
"model.language_model.layers.17.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 142 |
+
"model.language_model.layers.17.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 143 |
+
"model.language_model.layers.18.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 144 |
+
"model.language_model.layers.18.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 145 |
+
"model.language_model.layers.18.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 146 |
+
"model.language_model.layers.18.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 147 |
+
"model.language_model.layers.18.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 148 |
+
"model.language_model.layers.18.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 149 |
+
"model.language_model.layers.18.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 150 |
+
"model.language_model.layers.18.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 151 |
+
"model.language_model.layers.18.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 152 |
+
"model.language_model.layers.18.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 153 |
+
"model.language_model.layers.18.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 154 |
+
"model.language_model.layers.18.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 155 |
+
"model.language_model.layers.18.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 156 |
+
"model.language_model.layers.18.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 157 |
+
"model.language_model.layers.19.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 158 |
+
"model.language_model.layers.19.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 159 |
+
"model.language_model.layers.19.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 160 |
+
"model.language_model.layers.19.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 161 |
+
"model.language_model.layers.19.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 162 |
+
"model.language_model.layers.19.self_attn.k_norm.weight": "model-00003-of-00004.safetensors",
|
| 163 |
+
"model.language_model.layers.19.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
|
| 164 |
+
"model.language_model.layers.19.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
|
| 165 |
+
"model.language_model.layers.19.self_attn.q_norm.weight": "model-00003-of-00004.safetensors",
|
| 166 |
+
"model.language_model.layers.19.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
|
| 167 |
+
"model.language_model.layers.19.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
|
| 168 |
+
"model.language_model.layers.2.input_layernorm.weight": "model-00001-of-00004.safetensors",
|
| 169 |
+
"model.language_model.layers.2.linear_attn.A_log": "model-00001-of-00004.safetensors",
|
| 170 |
+
"model.language_model.layers.2.linear_attn.conv1d.weight": "model-00001-of-00004.safetensors",
|
| 171 |
+
"model.language_model.layers.2.linear_attn.dt_bias": "model-00001-of-00004.safetensors",
|
| 172 |
+
"model.language_model.layers.2.linear_attn.in_proj_a.weight": "model-00001-of-00004.safetensors",
|
| 173 |
+
"model.language_model.layers.2.linear_attn.in_proj_b.weight": "model-00001-of-00004.safetensors",
|
| 174 |
+
"model.language_model.layers.2.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 175 |
+
"model.language_model.layers.2.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 176 |
+
"model.language_model.layers.2.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 177 |
+
"model.language_model.layers.2.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 178 |
+
"model.language_model.layers.2.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 179 |
+
"model.language_model.layers.2.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 180 |
+
"model.language_model.layers.2.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 181 |
+
"model.language_model.layers.2.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 182 |
+
"model.language_model.layers.20.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 183 |
+
"model.language_model.layers.20.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 184 |
+
"model.language_model.layers.20.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 185 |
+
"model.language_model.layers.20.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 186 |
+
"model.language_model.layers.20.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 187 |
+
"model.language_model.layers.20.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 188 |
+
"model.language_model.layers.20.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 189 |
+
"model.language_model.layers.20.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 190 |
+
"model.language_model.layers.20.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 191 |
+
"model.language_model.layers.20.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 192 |
+
"model.language_model.layers.20.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 193 |
+
"model.language_model.layers.20.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 194 |
+
"model.language_model.layers.20.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 195 |
+
"model.language_model.layers.20.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 196 |
+
"model.language_model.layers.21.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 197 |
+
"model.language_model.layers.21.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 198 |
+
"model.language_model.layers.21.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 199 |
+
"model.language_model.layers.21.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 200 |
+
"model.language_model.layers.21.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 201 |
+
"model.language_model.layers.21.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 202 |
+
"model.language_model.layers.21.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 203 |
+
"model.language_model.layers.21.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 204 |
+
"model.language_model.layers.21.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 205 |
+
"model.language_model.layers.21.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 206 |
+
"model.language_model.layers.21.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 207 |
+
"model.language_model.layers.21.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 208 |
+
"model.language_model.layers.21.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 209 |
+
"model.language_model.layers.21.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 210 |
+
"model.language_model.layers.22.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 211 |
+
"model.language_model.layers.22.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 212 |
+
"model.language_model.layers.22.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 213 |
+
"model.language_model.layers.22.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 214 |
+
"model.language_model.layers.22.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 215 |
+
"model.language_model.layers.22.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 216 |
+
"model.language_model.layers.22.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 217 |
+
"model.language_model.layers.22.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 218 |
+
"model.language_model.layers.22.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 219 |
+
"model.language_model.layers.22.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 220 |
+
"model.language_model.layers.22.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 221 |
+
"model.language_model.layers.22.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 222 |
+
"model.language_model.layers.22.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 223 |
+
"model.language_model.layers.22.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 224 |
+
"model.language_model.layers.23.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 225 |
+
"model.language_model.layers.23.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 226 |
+
"model.language_model.layers.23.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 227 |
+
"model.language_model.layers.23.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 228 |
+
"model.language_model.layers.23.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 229 |
+
"model.language_model.layers.23.self_attn.k_norm.weight": "model-00003-of-00004.safetensors",
|
| 230 |
+
"model.language_model.layers.23.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
|
| 231 |
+
"model.language_model.layers.23.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
|
| 232 |
+
"model.language_model.layers.23.self_attn.q_norm.weight": "model-00003-of-00004.safetensors",
|
| 233 |
+
"model.language_model.layers.23.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
|
| 234 |
+
"model.language_model.layers.23.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
|
| 235 |
+
"model.language_model.layers.24.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 236 |
+
"model.language_model.layers.24.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 237 |
+
"model.language_model.layers.24.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 238 |
+
"model.language_model.layers.24.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 239 |
+
"model.language_model.layers.24.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 240 |
+
"model.language_model.layers.24.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 241 |
+
"model.language_model.layers.24.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 242 |
+
"model.language_model.layers.24.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 243 |
+
"model.language_model.layers.24.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 244 |
+
"model.language_model.layers.24.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 245 |
+
"model.language_model.layers.24.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 246 |
+
"model.language_model.layers.24.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 247 |
+
"model.language_model.layers.24.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 248 |
+
"model.language_model.layers.24.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 249 |
+
"model.language_model.layers.25.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 250 |
+
"model.language_model.layers.25.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 251 |
+
"model.language_model.layers.25.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 252 |
+
"model.language_model.layers.25.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 253 |
+
"model.language_model.layers.25.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 254 |
+
"model.language_model.layers.25.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 255 |
+
"model.language_model.layers.25.linear_attn.in_proj_qkv.weight": "model-00004-of-00004.safetensors",
|
| 256 |
+
"model.language_model.layers.25.linear_attn.in_proj_z.weight": "model-00004-of-00004.safetensors",
|
| 257 |
+
"model.language_model.layers.25.linear_attn.norm.weight": "model-00004-of-00004.safetensors",
|
| 258 |
+
"model.language_model.layers.25.linear_attn.out_proj.weight": "model-00004-of-00004.safetensors",
|
| 259 |
+
"model.language_model.layers.25.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 260 |
+
"model.language_model.layers.25.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 261 |
+
"model.language_model.layers.25.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 262 |
+
"model.language_model.layers.25.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 263 |
+
"model.language_model.layers.26.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 264 |
+
"model.language_model.layers.26.linear_attn.A_log": "model-00004-of-00004.safetensors",
|
| 265 |
+
"model.language_model.layers.26.linear_attn.conv1d.weight": "model-00004-of-00004.safetensors",
|
| 266 |
+
"model.language_model.layers.26.linear_attn.dt_bias": "model-00004-of-00004.safetensors",
|
| 267 |
+
"model.language_model.layers.26.linear_attn.in_proj_a.weight": "model-00004-of-00004.safetensors",
|
| 268 |
+
"model.language_model.layers.26.linear_attn.in_proj_b.weight": "model-00004-of-00004.safetensors",
|
| 269 |
+
"model.language_model.layers.26.linear_attn.in_proj_qkv.weight": "model-00004-of-00004.safetensors",
|
| 270 |
+
"model.language_model.layers.26.linear_attn.in_proj_z.weight": "model-00004-of-00004.safetensors",
|
| 271 |
+
"model.language_model.layers.26.linear_attn.norm.weight": "model-00004-of-00004.safetensors",
|
| 272 |
+
"model.language_model.layers.26.linear_attn.out_proj.weight": "model-00004-of-00004.safetensors",
|
| 273 |
+
"model.language_model.layers.26.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 274 |
+
"model.language_model.layers.26.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 275 |
+
"model.language_model.layers.26.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 276 |
+
"model.language_model.layers.26.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 277 |
+
"model.language_model.layers.27.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 278 |
+
"model.language_model.layers.27.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 279 |
+
"model.language_model.layers.27.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 280 |
+
"model.language_model.layers.27.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 281 |
+
"model.language_model.layers.27.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 282 |
+
"model.language_model.layers.27.self_attn.k_norm.weight": "model-00004-of-00004.safetensors",
|
| 283 |
+
"model.language_model.layers.27.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
|
| 284 |
+
"model.language_model.layers.27.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
|
| 285 |
+
"model.language_model.layers.27.self_attn.q_norm.weight": "model-00004-of-00004.safetensors",
|
| 286 |
+
"model.language_model.layers.27.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
|
| 287 |
+
"model.language_model.layers.27.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
|
| 288 |
+
"model.language_model.layers.28.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 289 |
+
"model.language_model.layers.28.linear_attn.A_log": "model-00004-of-00004.safetensors",
|
| 290 |
+
"model.language_model.layers.28.linear_attn.conv1d.weight": "model-00004-of-00004.safetensors",
|
| 291 |
+
"model.language_model.layers.28.linear_attn.dt_bias": "model-00004-of-00004.safetensors",
|
| 292 |
+
"model.language_model.layers.28.linear_attn.in_proj_a.weight": "model-00004-of-00004.safetensors",
|
| 293 |
+
"model.language_model.layers.28.linear_attn.in_proj_b.weight": "model-00004-of-00004.safetensors",
|
| 294 |
+
"model.language_model.layers.28.linear_attn.in_proj_qkv.weight": "model-00004-of-00004.safetensors",
|
| 295 |
+
"model.language_model.layers.28.linear_attn.in_proj_z.weight": "model-00004-of-00004.safetensors",
|
| 296 |
+
"model.language_model.layers.28.linear_attn.norm.weight": "model-00004-of-00004.safetensors",
|
| 297 |
+
"model.language_model.layers.28.linear_attn.out_proj.weight": "model-00004-of-00004.safetensors",
|
| 298 |
+
"model.language_model.layers.28.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 299 |
+
"model.language_model.layers.28.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 300 |
+
"model.language_model.layers.28.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 301 |
+
"model.language_model.layers.28.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 302 |
+
"model.language_model.layers.29.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 303 |
+
"model.language_model.layers.29.linear_attn.A_log": "model-00004-of-00004.safetensors",
|
| 304 |
+
"model.language_model.layers.29.linear_attn.conv1d.weight": "model-00004-of-00004.safetensors",
|
| 305 |
+
"model.language_model.layers.29.linear_attn.dt_bias": "model-00004-of-00004.safetensors",
|
| 306 |
+
"model.language_model.layers.29.linear_attn.in_proj_a.weight": "model-00004-of-00004.safetensors",
|
| 307 |
+
"model.language_model.layers.29.linear_attn.in_proj_b.weight": "model-00004-of-00004.safetensors",
|
| 308 |
+
"model.language_model.layers.29.linear_attn.in_proj_qkv.weight": "model-00004-of-00004.safetensors",
|
| 309 |
+
"model.language_model.layers.29.linear_attn.in_proj_z.weight": "model-00004-of-00004.safetensors",
|
| 310 |
+
"model.language_model.layers.29.linear_attn.norm.weight": "model-00004-of-00004.safetensors",
|
| 311 |
+
"model.language_model.layers.29.linear_attn.out_proj.weight": "model-00004-of-00004.safetensors",
|
| 312 |
+
"model.language_model.layers.29.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 313 |
+
"model.language_model.layers.29.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 314 |
+
"model.language_model.layers.29.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 315 |
+
"model.language_model.layers.29.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 316 |
+
"model.language_model.layers.3.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 317 |
+
"model.language_model.layers.3.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 318 |
+
"model.language_model.layers.3.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 319 |
+
"model.language_model.layers.3.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 320 |
+
"model.language_model.layers.3.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 321 |
+
"model.language_model.layers.3.self_attn.k_norm.weight": "model-00002-of-00004.safetensors",
|
| 322 |
+
"model.language_model.layers.3.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
|
| 323 |
+
"model.language_model.layers.3.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
|
| 324 |
+
"model.language_model.layers.3.self_attn.q_norm.weight": "model-00002-of-00004.safetensors",
|
| 325 |
+
"model.language_model.layers.3.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
|
| 326 |
+
"model.language_model.layers.3.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
|
| 327 |
+
"model.language_model.layers.30.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 328 |
+
"model.language_model.layers.30.linear_attn.A_log": "model-00004-of-00004.safetensors",
|
| 329 |
+
"model.language_model.layers.30.linear_attn.conv1d.weight": "model-00004-of-00004.safetensors",
|
| 330 |
+
"model.language_model.layers.30.linear_attn.dt_bias": "model-00004-of-00004.safetensors",
|
| 331 |
+
"model.language_model.layers.30.linear_attn.in_proj_a.weight": "model-00004-of-00004.safetensors",
|
| 332 |
+
"model.language_model.layers.30.linear_attn.in_proj_b.weight": "model-00004-of-00004.safetensors",
|
| 333 |
+
"model.language_model.layers.30.linear_attn.in_proj_qkv.weight": "model-00004-of-00004.safetensors",
|
| 334 |
+
"model.language_model.layers.30.linear_attn.in_proj_z.weight": "model-00004-of-00004.safetensors",
|
| 335 |
+
"model.language_model.layers.30.linear_attn.norm.weight": "model-00004-of-00004.safetensors",
|
| 336 |
+
"model.language_model.layers.30.linear_attn.out_proj.weight": "model-00004-of-00004.safetensors",
|
| 337 |
+
"model.language_model.layers.30.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 338 |
+
"model.language_model.layers.30.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 339 |
+
"model.language_model.layers.30.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 340 |
+
"model.language_model.layers.30.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 341 |
+
"model.language_model.layers.31.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 342 |
+
"model.language_model.layers.31.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 343 |
+
"model.language_model.layers.31.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 344 |
+
"model.language_model.layers.31.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 345 |
+
"model.language_model.layers.31.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 346 |
+
"model.language_model.layers.31.self_attn.k_norm.weight": "model-00004-of-00004.safetensors",
|
| 347 |
+
"model.language_model.layers.31.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
|
| 348 |
+
"model.language_model.layers.31.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
|
| 349 |
+
"model.language_model.layers.31.self_attn.q_norm.weight": "model-00004-of-00004.safetensors",
|
| 350 |
+
"model.language_model.layers.31.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
|
| 351 |
+
"model.language_model.layers.31.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
|
| 352 |
+
"model.language_model.layers.4.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 353 |
+
"model.language_model.layers.4.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 354 |
+
"model.language_model.layers.4.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 355 |
+
"model.language_model.layers.4.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 356 |
+
"model.language_model.layers.4.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 357 |
+
"model.language_model.layers.4.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 358 |
+
"model.language_model.layers.4.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 359 |
+
"model.language_model.layers.4.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 360 |
+
"model.language_model.layers.4.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 361 |
+
"model.language_model.layers.4.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 362 |
+
"model.language_model.layers.4.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 363 |
+
"model.language_model.layers.4.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 364 |
+
"model.language_model.layers.4.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 365 |
+
"model.language_model.layers.4.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 366 |
+
"model.language_model.layers.5.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 367 |
+
"model.language_model.layers.5.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 368 |
+
"model.language_model.layers.5.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 369 |
+
"model.language_model.layers.5.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 370 |
+
"model.language_model.layers.5.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 371 |
+
"model.language_model.layers.5.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 372 |
+
"model.language_model.layers.5.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 373 |
+
"model.language_model.layers.5.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 374 |
+
"model.language_model.layers.5.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 375 |
+
"model.language_model.layers.5.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 376 |
+
"model.language_model.layers.5.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 377 |
+
"model.language_model.layers.5.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 378 |
+
"model.language_model.layers.5.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 379 |
+
"model.language_model.layers.5.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 380 |
+
"model.language_model.layers.6.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 381 |
+
"model.language_model.layers.6.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 382 |
+
"model.language_model.layers.6.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 383 |
+
"model.language_model.layers.6.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 384 |
+
"model.language_model.layers.6.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 385 |
+
"model.language_model.layers.6.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 386 |
+
"model.language_model.layers.6.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 387 |
+
"model.language_model.layers.6.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 388 |
+
"model.language_model.layers.6.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 389 |
+
"model.language_model.layers.6.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 390 |
+
"model.language_model.layers.6.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 391 |
+
"model.language_model.layers.6.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 392 |
+
"model.language_model.layers.6.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 393 |
+
"model.language_model.layers.6.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 394 |
+
"model.language_model.layers.7.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 395 |
+
"model.language_model.layers.7.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 396 |
+
"model.language_model.layers.7.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 397 |
+
"model.language_model.layers.7.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 398 |
+
"model.language_model.layers.7.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 399 |
+
"model.language_model.layers.7.self_attn.k_norm.weight": "model-00002-of-00004.safetensors",
|
| 400 |
+
"model.language_model.layers.7.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
|
| 401 |
+
"model.language_model.layers.7.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
|
| 402 |
+
"model.language_model.layers.7.self_attn.q_norm.weight": "model-00002-of-00004.safetensors",
|
| 403 |
+
"model.language_model.layers.7.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
|
| 404 |
+
"model.language_model.layers.7.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
|
| 405 |
+
"model.language_model.layers.8.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 406 |
+
"model.language_model.layers.8.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 407 |
+
"model.language_model.layers.8.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 408 |
+
"model.language_model.layers.8.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 409 |
+
"model.language_model.layers.8.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 410 |
+
"model.language_model.layers.8.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 411 |
+
"model.language_model.layers.8.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 412 |
+
"model.language_model.layers.8.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 413 |
+
"model.language_model.layers.8.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 414 |
+
"model.language_model.layers.8.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 415 |
+
"model.language_model.layers.8.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 416 |
+
"model.language_model.layers.8.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 417 |
+
"model.language_model.layers.8.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 418 |
+
"model.language_model.layers.8.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 419 |
+
"model.language_model.layers.9.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 420 |
+
"model.language_model.layers.9.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 421 |
+
"model.language_model.layers.9.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 422 |
+
"model.language_model.layers.9.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 423 |
+
"model.language_model.layers.9.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 424 |
+
"model.language_model.layers.9.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 425 |
+
"model.language_model.layers.9.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 426 |
+
"model.language_model.layers.9.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 427 |
+
"model.language_model.layers.9.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 428 |
+
"model.language_model.layers.9.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 429 |
+
"model.language_model.layers.9.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 430 |
+
"model.language_model.layers.9.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 431 |
+
"model.language_model.layers.9.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 432 |
+
"model.language_model.layers.9.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 433 |
+
"model.language_model.norm.weight": "model-00004-of-00004.safetensors",
|
| 434 |
+
"model.visual.blocks.0.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 435 |
+
"model.visual.blocks.0.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 436 |
+
"model.visual.blocks.0.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 437 |
+
"model.visual.blocks.0.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 438 |
+
"model.visual.blocks.0.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 439 |
+
"model.visual.blocks.0.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 440 |
+
"model.visual.blocks.0.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 441 |
+
"model.visual.blocks.0.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 442 |
+
"model.visual.blocks.0.norm1.bias": "model-00004-of-00004.safetensors",
|
| 443 |
+
"model.visual.blocks.0.norm1.weight": "model-00004-of-00004.safetensors",
|
| 444 |
+
"model.visual.blocks.0.norm2.bias": "model-00004-of-00004.safetensors",
|
| 445 |
+
"model.visual.blocks.0.norm2.weight": "model-00004-of-00004.safetensors",
|
| 446 |
+
"model.visual.blocks.1.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 447 |
+
"model.visual.blocks.1.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 448 |
+
"model.visual.blocks.1.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 449 |
+
"model.visual.blocks.1.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 450 |
+
"model.visual.blocks.1.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 451 |
+
"model.visual.blocks.1.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 452 |
+
"model.visual.blocks.1.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 453 |
+
"model.visual.blocks.1.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 454 |
+
"model.visual.blocks.1.norm1.bias": "model-00004-of-00004.safetensors",
|
| 455 |
+
"model.visual.blocks.1.norm1.weight": "model-00004-of-00004.safetensors",
|
| 456 |
+
"model.visual.blocks.1.norm2.bias": "model-00004-of-00004.safetensors",
|
| 457 |
+
"model.visual.blocks.1.norm2.weight": "model-00004-of-00004.safetensors",
|
| 458 |
+
"model.visual.blocks.10.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 459 |
+
"model.visual.blocks.10.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 460 |
+
"model.visual.blocks.10.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 461 |
+
"model.visual.blocks.10.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 462 |
+
"model.visual.blocks.10.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 463 |
+
"model.visual.blocks.10.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 464 |
+
"model.visual.blocks.10.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 465 |
+
"model.visual.blocks.10.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 466 |
+
"model.visual.blocks.10.norm1.bias": "model-00004-of-00004.safetensors",
|
| 467 |
+
"model.visual.blocks.10.norm1.weight": "model-00004-of-00004.safetensors",
|
| 468 |
+
"model.visual.blocks.10.norm2.bias": "model-00004-of-00004.safetensors",
|
| 469 |
+
"model.visual.blocks.10.norm2.weight": "model-00004-of-00004.safetensors",
|
| 470 |
+
"model.visual.blocks.11.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 471 |
+
"model.visual.blocks.11.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 472 |
+
"model.visual.blocks.11.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 473 |
+
"model.visual.blocks.11.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 474 |
+
"model.visual.blocks.11.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 475 |
+
"model.visual.blocks.11.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 476 |
+
"model.visual.blocks.11.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 477 |
+
"model.visual.blocks.11.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 478 |
+
"model.visual.blocks.11.norm1.bias": "model-00004-of-00004.safetensors",
|
| 479 |
+
"model.visual.blocks.11.norm1.weight": "model-00004-of-00004.safetensors",
|
| 480 |
+
"model.visual.blocks.11.norm2.bias": "model-00004-of-00004.safetensors",
|
| 481 |
+
"model.visual.blocks.11.norm2.weight": "model-00004-of-00004.safetensors",
|
| 482 |
+
"model.visual.blocks.12.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 483 |
+
"model.visual.blocks.12.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 484 |
+
"model.visual.blocks.12.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 485 |
+
"model.visual.blocks.12.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 486 |
+
"model.visual.blocks.12.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 487 |
+
"model.visual.blocks.12.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 488 |
+
"model.visual.blocks.12.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 489 |
+
"model.visual.blocks.12.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 490 |
+
"model.visual.blocks.12.norm1.bias": "model-00004-of-00004.safetensors",
|
| 491 |
+
"model.visual.blocks.12.norm1.weight": "model-00004-of-00004.safetensors",
|
| 492 |
+
"model.visual.blocks.12.norm2.bias": "model-00004-of-00004.safetensors",
|
| 493 |
+
"model.visual.blocks.12.norm2.weight": "model-00004-of-00004.safetensors",
|
| 494 |
+
"model.visual.blocks.13.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 495 |
+
"model.visual.blocks.13.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 496 |
+
"model.visual.blocks.13.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 497 |
+
"model.visual.blocks.13.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 498 |
+
"model.visual.blocks.13.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 499 |
+
"model.visual.blocks.13.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 500 |
+
"model.visual.blocks.13.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 501 |
+
"model.visual.blocks.13.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 502 |
+
"model.visual.blocks.13.norm1.bias": "model-00004-of-00004.safetensors",
|
| 503 |
+
"model.visual.blocks.13.norm1.weight": "model-00004-of-00004.safetensors",
|
| 504 |
+
"model.visual.blocks.13.norm2.bias": "model-00004-of-00004.safetensors",
|
| 505 |
+
"model.visual.blocks.13.norm2.weight": "model-00004-of-00004.safetensors",
|
| 506 |
+
"model.visual.blocks.14.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 507 |
+
"model.visual.blocks.14.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 508 |
+
"model.visual.blocks.14.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 509 |
+
"model.visual.blocks.14.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 510 |
+
"model.visual.blocks.14.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 511 |
+
"model.visual.blocks.14.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 512 |
+
"model.visual.blocks.14.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 513 |
+
"model.visual.blocks.14.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 514 |
+
"model.visual.blocks.14.norm1.bias": "model-00004-of-00004.safetensors",
|
| 515 |
+
"model.visual.blocks.14.norm1.weight": "model-00004-of-00004.safetensors",
|
| 516 |
+
"model.visual.blocks.14.norm2.bias": "model-00004-of-00004.safetensors",
|
| 517 |
+
"model.visual.blocks.14.norm2.weight": "model-00004-of-00004.safetensors",
|
| 518 |
+
"model.visual.blocks.15.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 519 |
+
"model.visual.blocks.15.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 520 |
+
"model.visual.blocks.15.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 521 |
+
"model.visual.blocks.15.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 522 |
+
"model.visual.blocks.15.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 523 |
+
"model.visual.blocks.15.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 524 |
+
"model.visual.blocks.15.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 525 |
+
"model.visual.blocks.15.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 526 |
+
"model.visual.blocks.15.norm1.bias": "model-00004-of-00004.safetensors",
|
| 527 |
+
"model.visual.blocks.15.norm1.weight": "model-00004-of-00004.safetensors",
|
| 528 |
+
"model.visual.blocks.15.norm2.bias": "model-00004-of-00004.safetensors",
|
| 529 |
+
"model.visual.blocks.15.norm2.weight": "model-00004-of-00004.safetensors",
|
| 530 |
+
"model.visual.blocks.16.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 531 |
+
"model.visual.blocks.16.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 532 |
+
"model.visual.blocks.16.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 533 |
+
"model.visual.blocks.16.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 534 |
+
"model.visual.blocks.16.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 535 |
+
"model.visual.blocks.16.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 536 |
+
"model.visual.blocks.16.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 537 |
+
"model.visual.blocks.16.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 538 |
+
"model.visual.blocks.16.norm1.bias": "model-00004-of-00004.safetensors",
|
| 539 |
+
"model.visual.blocks.16.norm1.weight": "model-00004-of-00004.safetensors",
|
| 540 |
+
"model.visual.blocks.16.norm2.bias": "model-00004-of-00004.safetensors",
|
| 541 |
+
"model.visual.blocks.16.norm2.weight": "model-00004-of-00004.safetensors",
|
| 542 |
+
"model.visual.blocks.17.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 543 |
+
"model.visual.blocks.17.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 544 |
+
"model.visual.blocks.17.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 545 |
+
"model.visual.blocks.17.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 546 |
+
"model.visual.blocks.17.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 547 |
+
"model.visual.blocks.17.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 548 |
+
"model.visual.blocks.17.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 549 |
+
"model.visual.blocks.17.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 550 |
+
"model.visual.blocks.17.norm1.bias": "model-00004-of-00004.safetensors",
|
| 551 |
+
"model.visual.blocks.17.norm1.weight": "model-00004-of-00004.safetensors",
|
| 552 |
+
"model.visual.blocks.17.norm2.bias": "model-00004-of-00004.safetensors",
|
| 553 |
+
"model.visual.blocks.17.norm2.weight": "model-00004-of-00004.safetensors",
|
| 554 |
+
"model.visual.blocks.18.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 555 |
+
"model.visual.blocks.18.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 556 |
+
"model.visual.blocks.18.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 557 |
+
"model.visual.blocks.18.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 558 |
+
"model.visual.blocks.18.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 559 |
+
"model.visual.blocks.18.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 560 |
+
"model.visual.blocks.18.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 561 |
+
"model.visual.blocks.18.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 562 |
+
"model.visual.blocks.18.norm1.bias": "model-00004-of-00004.safetensors",
|
| 563 |
+
"model.visual.blocks.18.norm1.weight": "model-00004-of-00004.safetensors",
|
| 564 |
+
"model.visual.blocks.18.norm2.bias": "model-00004-of-00004.safetensors",
|
| 565 |
+
"model.visual.blocks.18.norm2.weight": "model-00004-of-00004.safetensors",
|
| 566 |
+
"model.visual.blocks.19.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 567 |
+
"model.visual.blocks.19.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 568 |
+
"model.visual.blocks.19.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 569 |
+
"model.visual.blocks.19.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 570 |
+
"model.visual.blocks.19.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 571 |
+
"model.visual.blocks.19.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 572 |
+
"model.visual.blocks.19.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 573 |
+
"model.visual.blocks.19.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 574 |
+
"model.visual.blocks.19.norm1.bias": "model-00004-of-00004.safetensors",
|
| 575 |
+
"model.visual.blocks.19.norm1.weight": "model-00004-of-00004.safetensors",
|
| 576 |
+
"model.visual.blocks.19.norm2.bias": "model-00004-of-00004.safetensors",
|
| 577 |
+
"model.visual.blocks.19.norm2.weight": "model-00004-of-00004.safetensors",
|
| 578 |
+
"model.visual.blocks.2.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 579 |
+
"model.visual.blocks.2.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 580 |
+
"model.visual.blocks.2.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 581 |
+
"model.visual.blocks.2.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 582 |
+
"model.visual.blocks.2.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 583 |
+
"model.visual.blocks.2.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 584 |
+
"model.visual.blocks.2.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 585 |
+
"model.visual.blocks.2.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 586 |
+
"model.visual.blocks.2.norm1.bias": "model-00004-of-00004.safetensors",
|
| 587 |
+
"model.visual.blocks.2.norm1.weight": "model-00004-of-00004.safetensors",
|
| 588 |
+
"model.visual.blocks.2.norm2.bias": "model-00004-of-00004.safetensors",
|
| 589 |
+
"model.visual.blocks.2.norm2.weight": "model-00004-of-00004.safetensors",
|
| 590 |
+
"model.visual.blocks.20.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 591 |
+
"model.visual.blocks.20.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 592 |
+
"model.visual.blocks.20.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 593 |
+
"model.visual.blocks.20.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 594 |
+
"model.visual.blocks.20.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 595 |
+
"model.visual.blocks.20.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 596 |
+
"model.visual.blocks.20.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 597 |
+
"model.visual.blocks.20.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 598 |
+
"model.visual.blocks.20.norm1.bias": "model-00004-of-00004.safetensors",
|
| 599 |
+
"model.visual.blocks.20.norm1.weight": "model-00004-of-00004.safetensors",
|
| 600 |
+
"model.visual.blocks.20.norm2.bias": "model-00004-of-00004.safetensors",
|
| 601 |
+
"model.visual.blocks.20.norm2.weight": "model-00004-of-00004.safetensors",
|
| 602 |
+
"model.visual.blocks.21.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 603 |
+
"model.visual.blocks.21.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 604 |
+
"model.visual.blocks.21.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 605 |
+
"model.visual.blocks.21.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 606 |
+
"model.visual.blocks.21.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 607 |
+
"model.visual.blocks.21.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 608 |
+
"model.visual.blocks.21.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 609 |
+
"model.visual.blocks.21.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 610 |
+
"model.visual.blocks.21.norm1.bias": "model-00004-of-00004.safetensors",
|
| 611 |
+
"model.visual.blocks.21.norm1.weight": "model-00004-of-00004.safetensors",
|
| 612 |
+
"model.visual.blocks.21.norm2.bias": "model-00004-of-00004.safetensors",
|
| 613 |
+
"model.visual.blocks.21.norm2.weight": "model-00004-of-00004.safetensors",
|
| 614 |
+
"model.visual.blocks.22.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 615 |
+
"model.visual.blocks.22.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 616 |
+
"model.visual.blocks.22.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 617 |
+
"model.visual.blocks.22.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 618 |
+
"model.visual.blocks.22.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 619 |
+
"model.visual.blocks.22.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 620 |
+
"model.visual.blocks.22.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 621 |
+
"model.visual.blocks.22.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 622 |
+
"model.visual.blocks.22.norm1.bias": "model-00004-of-00004.safetensors",
|
| 623 |
+
"model.visual.blocks.22.norm1.weight": "model-00004-of-00004.safetensors",
|
| 624 |
+
"model.visual.blocks.22.norm2.bias": "model-00004-of-00004.safetensors",
|
| 625 |
+
"model.visual.blocks.22.norm2.weight": "model-00004-of-00004.safetensors",
|
| 626 |
+
"model.visual.blocks.23.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 627 |
+
"model.visual.blocks.23.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 628 |
+
"model.visual.blocks.23.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 629 |
+
"model.visual.blocks.23.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 630 |
+
"model.visual.blocks.23.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 631 |
+
"model.visual.blocks.23.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 632 |
+
"model.visual.blocks.23.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 633 |
+
"model.visual.blocks.23.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 634 |
+
"model.visual.blocks.23.norm1.bias": "model-00004-of-00004.safetensors",
|
| 635 |
+
"model.visual.blocks.23.norm1.weight": "model-00004-of-00004.safetensors",
|
| 636 |
+
"model.visual.blocks.23.norm2.bias": "model-00004-of-00004.safetensors",
|
| 637 |
+
"model.visual.blocks.23.norm2.weight": "model-00004-of-00004.safetensors",
|
| 638 |
+
"model.visual.blocks.24.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 639 |
+
"model.visual.blocks.24.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 640 |
+
"model.visual.blocks.24.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 641 |
+
"model.visual.blocks.24.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 642 |
+
"model.visual.blocks.24.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 643 |
+
"model.visual.blocks.24.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 644 |
+
"model.visual.blocks.24.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 645 |
+
"model.visual.blocks.24.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 646 |
+
"model.visual.blocks.24.norm1.bias": "model-00004-of-00004.safetensors",
|
| 647 |
+
"model.visual.blocks.24.norm1.weight": "model-00004-of-00004.safetensors",
|
| 648 |
+
"model.visual.blocks.24.norm2.bias": "model-00004-of-00004.safetensors",
|
| 649 |
+
"model.visual.blocks.24.norm2.weight": "model-00004-of-00004.safetensors",
|
| 650 |
+
"model.visual.blocks.25.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 651 |
+
"model.visual.blocks.25.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 652 |
+
"model.visual.blocks.25.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 653 |
+
"model.visual.blocks.25.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 654 |
+
"model.visual.blocks.25.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 655 |
+
"model.visual.blocks.25.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 656 |
+
"model.visual.blocks.25.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 657 |
+
"model.visual.blocks.25.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 658 |
+
"model.visual.blocks.25.norm1.bias": "model-00004-of-00004.safetensors",
|
| 659 |
+
"model.visual.blocks.25.norm1.weight": "model-00004-of-00004.safetensors",
|
| 660 |
+
"model.visual.blocks.25.norm2.bias": "model-00004-of-00004.safetensors",
|
| 661 |
+
"model.visual.blocks.25.norm2.weight": "model-00004-of-00004.safetensors",
|
| 662 |
+
"model.visual.blocks.26.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 663 |
+
"model.visual.blocks.26.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 664 |
+
"model.visual.blocks.26.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 665 |
+
"model.visual.blocks.26.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 666 |
+
"model.visual.blocks.26.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 667 |
+
"model.visual.blocks.26.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 668 |
+
"model.visual.blocks.26.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 669 |
+
"model.visual.blocks.26.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 670 |
+
"model.visual.blocks.26.norm1.bias": "model-00004-of-00004.safetensors",
|
| 671 |
+
"model.visual.blocks.26.norm1.weight": "model-00004-of-00004.safetensors",
|
| 672 |
+
"model.visual.blocks.26.norm2.bias": "model-00004-of-00004.safetensors",
|
| 673 |
+
"model.visual.blocks.26.norm2.weight": "model-00004-of-00004.safetensors",
|
| 674 |
+
"model.visual.blocks.3.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 675 |
+
"model.visual.blocks.3.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 676 |
+
"model.visual.blocks.3.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 677 |
+
"model.visual.blocks.3.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 678 |
+
"model.visual.blocks.3.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 679 |
+
"model.visual.blocks.3.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 680 |
+
"model.visual.blocks.3.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 681 |
+
"model.visual.blocks.3.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 682 |
+
"model.visual.blocks.3.norm1.bias": "model-00004-of-00004.safetensors",
|
| 683 |
+
"model.visual.blocks.3.norm1.weight": "model-00004-of-00004.safetensors",
|
| 684 |
+
"model.visual.blocks.3.norm2.bias": "model-00004-of-00004.safetensors",
|
| 685 |
+
"model.visual.blocks.3.norm2.weight": "model-00004-of-00004.safetensors",
|
| 686 |
+
"model.visual.blocks.4.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 687 |
+
"model.visual.blocks.4.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 688 |
+
"model.visual.blocks.4.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 689 |
+
"model.visual.blocks.4.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 690 |
+
"model.visual.blocks.4.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 691 |
+
"model.visual.blocks.4.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 692 |
+
"model.visual.blocks.4.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 693 |
+
"model.visual.blocks.4.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 694 |
+
"model.visual.blocks.4.norm1.bias": "model-00004-of-00004.safetensors",
|
| 695 |
+
"model.visual.blocks.4.norm1.weight": "model-00004-of-00004.safetensors",
|
| 696 |
+
"model.visual.blocks.4.norm2.bias": "model-00004-of-00004.safetensors",
|
| 697 |
+
"model.visual.blocks.4.norm2.weight": "model-00004-of-00004.safetensors",
|
| 698 |
+
"model.visual.blocks.5.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 699 |
+
"model.visual.blocks.5.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 700 |
+
"model.visual.blocks.5.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 701 |
+
"model.visual.blocks.5.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 702 |
+
"model.visual.blocks.5.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 703 |
+
"model.visual.blocks.5.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 704 |
+
"model.visual.blocks.5.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 705 |
+
"model.visual.blocks.5.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 706 |
+
"model.visual.blocks.5.norm1.bias": "model-00004-of-00004.safetensors",
|
| 707 |
+
"model.visual.blocks.5.norm1.weight": "model-00004-of-00004.safetensors",
|
| 708 |
+
"model.visual.blocks.5.norm2.bias": "model-00004-of-00004.safetensors",
|
| 709 |
+
"model.visual.blocks.5.norm2.weight": "model-00004-of-00004.safetensors",
|
| 710 |
+
"model.visual.blocks.6.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 711 |
+
"model.visual.blocks.6.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 712 |
+
"model.visual.blocks.6.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 713 |
+
"model.visual.blocks.6.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 714 |
+
"model.visual.blocks.6.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 715 |
+
"model.visual.blocks.6.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 716 |
+
"model.visual.blocks.6.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 717 |
+
"model.visual.blocks.6.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 718 |
+
"model.visual.blocks.6.norm1.bias": "model-00004-of-00004.safetensors",
|
| 719 |
+
"model.visual.blocks.6.norm1.weight": "model-00004-of-00004.safetensors",
|
| 720 |
+
"model.visual.blocks.6.norm2.bias": "model-00004-of-00004.safetensors",
|
| 721 |
+
"model.visual.blocks.6.norm2.weight": "model-00004-of-00004.safetensors",
|
| 722 |
+
"model.visual.blocks.7.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 723 |
+
"model.visual.blocks.7.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 724 |
+
"model.visual.blocks.7.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 725 |
+
"model.visual.blocks.7.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 726 |
+
"model.visual.blocks.7.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 727 |
+
"model.visual.blocks.7.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 728 |
+
"model.visual.blocks.7.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 729 |
+
"model.visual.blocks.7.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 730 |
+
"model.visual.blocks.7.norm1.bias": "model-00004-of-00004.safetensors",
|
| 731 |
+
"model.visual.blocks.7.norm1.weight": "model-00004-of-00004.safetensors",
|
| 732 |
+
"model.visual.blocks.7.norm2.bias": "model-00004-of-00004.safetensors",
|
| 733 |
+
"model.visual.blocks.7.norm2.weight": "model-00004-of-00004.safetensors",
|
| 734 |
+
"model.visual.blocks.8.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 735 |
+
"model.visual.blocks.8.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 736 |
+
"model.visual.blocks.8.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 737 |
+
"model.visual.blocks.8.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 738 |
+
"model.visual.blocks.8.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 739 |
+
"model.visual.blocks.8.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 740 |
+
"model.visual.blocks.8.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 741 |
+
"model.visual.blocks.8.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 742 |
+
"model.visual.blocks.8.norm1.bias": "model-00004-of-00004.safetensors",
|
| 743 |
+
"model.visual.blocks.8.norm1.weight": "model-00004-of-00004.safetensors",
|
| 744 |
+
"model.visual.blocks.8.norm2.bias": "model-00004-of-00004.safetensors",
|
| 745 |
+
"model.visual.blocks.8.norm2.weight": "model-00004-of-00004.safetensors",
|
| 746 |
+
"model.visual.blocks.9.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 747 |
+
"model.visual.blocks.9.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 748 |
+
"model.visual.blocks.9.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 749 |
+
"model.visual.blocks.9.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 750 |
+
"model.visual.blocks.9.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 751 |
+
"model.visual.blocks.9.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 752 |
+
"model.visual.blocks.9.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 753 |
+
"model.visual.blocks.9.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 754 |
+
"model.visual.blocks.9.norm1.bias": "model-00004-of-00004.safetensors",
|
| 755 |
+
"model.visual.blocks.9.norm1.weight": "model-00004-of-00004.safetensors",
|
| 756 |
+
"model.visual.blocks.9.norm2.bias": "model-00004-of-00004.safetensors",
|
| 757 |
+
"model.visual.blocks.9.norm2.weight": "model-00004-of-00004.safetensors",
|
| 758 |
+
"model.visual.merger.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 759 |
+
"model.visual.merger.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 760 |
+
"model.visual.merger.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 761 |
+
"model.visual.merger.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 762 |
+
"model.visual.merger.norm.bias": "model-00004-of-00004.safetensors",
|
| 763 |
+
"model.visual.merger.norm.weight": "model-00004-of-00004.safetensors",
|
| 764 |
+
"model.visual.patch_embed.proj.bias": "model-00004-of-00004.safetensors",
|
| 765 |
+
"model.visual.patch_embed.proj.weight": "model-00004-of-00004.safetensors",
|
| 766 |
+
"model.visual.pos_embed.weight": "model-00004-of-00004.safetensors"
|
| 767 |
+
}
|
| 768 |
+
}
|
c1_no_history/round_2/preprocessor_config.json
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"size": {
|
| 3 |
+
"longest_edge": 16777216,
|
| 4 |
+
"shortest_edge": 65536
|
| 5 |
+
},
|
| 6 |
+
"patch_size": 16,
|
| 7 |
+
"temporal_patch_size": 2,
|
| 8 |
+
"merge_size": 2,
|
| 9 |
+
"image_mean": [
|
| 10 |
+
0.5,
|
| 11 |
+
0.5,
|
| 12 |
+
0.5
|
| 13 |
+
],
|
| 14 |
+
"image_std": [
|
| 15 |
+
0.5,
|
| 16 |
+
0.5,
|
| 17 |
+
0.5
|
| 18 |
+
],
|
| 19 |
+
"processor_class": "Qwen3VLProcessor",
|
| 20 |
+
"image_processor_type": "Qwen2VLImageProcessorFast"
|
| 21 |
+
}
|
c1_no_history/round_2/processor_config.json
ADDED
|
@@ -0,0 +1,60 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"image_processor": {
|
| 3 |
+
"do_convert_rgb": true,
|
| 4 |
+
"do_normalize": true,
|
| 5 |
+
"do_rescale": true,
|
| 6 |
+
"do_resize": true,
|
| 7 |
+
"image_mean": [
|
| 8 |
+
0.5,
|
| 9 |
+
0.5,
|
| 10 |
+
0.5
|
| 11 |
+
],
|
| 12 |
+
"image_processor_type": "Qwen2VLImageProcessor",
|
| 13 |
+
"image_std": [
|
| 14 |
+
0.5,
|
| 15 |
+
0.5,
|
| 16 |
+
0.5
|
| 17 |
+
],
|
| 18 |
+
"merge_size": 2,
|
| 19 |
+
"patch_size": 16,
|
| 20 |
+
"resample": 3,
|
| 21 |
+
"rescale_factor": 0.00392156862745098,
|
| 22 |
+
"size": {
|
| 23 |
+
"longest_edge": 16777216,
|
| 24 |
+
"shortest_edge": 65536
|
| 25 |
+
},
|
| 26 |
+
"temporal_patch_size": 2
|
| 27 |
+
},
|
| 28 |
+
"processor_class": "Qwen3VLProcessor",
|
| 29 |
+
"video_processor": {
|
| 30 |
+
"do_convert_rgb": true,
|
| 31 |
+
"do_normalize": true,
|
| 32 |
+
"do_rescale": true,
|
| 33 |
+
"do_resize": true,
|
| 34 |
+
"do_sample_frames": true,
|
| 35 |
+
"fps": 2,
|
| 36 |
+
"image_mean": [
|
| 37 |
+
0.5,
|
| 38 |
+
0.5,
|
| 39 |
+
0.5
|
| 40 |
+
],
|
| 41 |
+
"image_std": [
|
| 42 |
+
0.5,
|
| 43 |
+
0.5,
|
| 44 |
+
0.5
|
| 45 |
+
],
|
| 46 |
+
"max_frames": 768,
|
| 47 |
+
"merge_size": 2,
|
| 48 |
+
"min_frames": 4,
|
| 49 |
+
"patch_size": 16,
|
| 50 |
+
"resample": 3,
|
| 51 |
+
"rescale_factor": 0.00392156862745098,
|
| 52 |
+
"return_metadata": false,
|
| 53 |
+
"size": {
|
| 54 |
+
"longest_edge": 25165824,
|
| 55 |
+
"shortest_edge": 4096
|
| 56 |
+
},
|
| 57 |
+
"temporal_patch_size": 2,
|
| 58 |
+
"video_processor_type": "Qwen3VLVideoProcessor"
|
| 59 |
+
}
|
| 60 |
+
}
|
c1_no_history/round_2/tokenizer_config.json
ADDED
|
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"audio_bos_token": "<|audio_start|>",
|
| 4 |
+
"audio_eos_token": "<|audio_end|>",
|
| 5 |
+
"audio_token": "<|audio_pad|>",
|
| 6 |
+
"backend": "tokenizers",
|
| 7 |
+
"bos_token": null,
|
| 8 |
+
"clean_up_tokenization_spaces": false,
|
| 9 |
+
"eos_token": "<|im_end|>",
|
| 10 |
+
"errors": "replace",
|
| 11 |
+
"image_token": "<|image_pad|>",
|
| 12 |
+
"is_local": true,
|
| 13 |
+
"local_files_only": true,
|
| 14 |
+
"model_max_length": 262144,
|
| 15 |
+
"model_specific_special_tokens": {
|
| 16 |
+
"audio_bos_token": "<|audio_start|>",
|
| 17 |
+
"audio_eos_token": "<|audio_end|>",
|
| 18 |
+
"audio_token": "<|audio_pad|>",
|
| 19 |
+
"image_token": "<|image_pad|>",
|
| 20 |
+
"video_token": "<|video_pad|>",
|
| 21 |
+
"vision_bos_token": "<|vision_start|>",
|
| 22 |
+
"vision_eos_token": "<|vision_end|>"
|
| 23 |
+
},
|
| 24 |
+
"pad_token": "<|endoftext|>",
|
| 25 |
+
"pretokenize_regex": "(?i:'s|'t|'re|'ve|'m|'ll|'d)|[^\\r\\n\\p{L}\\p{N}]?[\\p{L}\\p{M}]+|\\p{N}| ?[^\\s\\p{L}\\p{M}\\p{N}]+[\\r\\n]*|\\s*[\\r\\n]+|\\s+(?!\\S)|\\s+",
|
| 26 |
+
"processor_class": "Qwen3VLProcessor",
|
| 27 |
+
"split_special_tokens": false,
|
| 28 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 29 |
+
"unk_token": null,
|
| 30 |
+
"video_token": "<|video_pad|>",
|
| 31 |
+
"vision_bos_token": "<|vision_start|>",
|
| 32 |
+
"vision_eos_token": "<|vision_end|>"
|
| 33 |
+
}
|
c1_no_history/round_2/trainer_state.json
ADDED
|
@@ -0,0 +1,2454 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 1.0,
|
| 6 |
+
"eval_steps": 80.0,
|
| 7 |
+
"global_step": 160,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"clip_ratio/high_max": 0.0,
|
| 14 |
+
"clip_ratio/high_mean": 0.0,
|
| 15 |
+
"clip_ratio/low_mean": 0.0,
|
| 16 |
+
"clip_ratio/low_min": 0.0,
|
| 17 |
+
"clip_ratio/region_mean": 0.0,
|
| 18 |
+
"completions/clipped_ratio": 0.5390625,
|
| 19 |
+
"completions/max_length": 2048.0,
|
| 20 |
+
"completions/mean_length": 1469.59765625,
|
| 21 |
+
"completions/min_length": 96.0,
|
| 22 |
+
"epoch": 0.00625,
|
| 23 |
+
"frac_reward_zero_std": 0.40625,
|
| 24 |
+
"grad_norm": 1.1241474627466976,
|
| 25 |
+
"kl": 0.0,
|
| 26 |
+
"learning_rate": 2e-07,
|
| 27 |
+
"loss": 1.862645149230957e-09,
|
| 28 |
+
"reward": 0.2578125,
|
| 29 |
+
"reward_std": 0.25434356927871704,
|
| 30 |
+
"rewards/MajorityAccuracy/mean": 0.2578125,
|
| 31 |
+
"rewards/MajorityAccuracy/std": 0.4382871091365814,
|
| 32 |
+
"step": 1,
|
| 33 |
+
"step_time": 587.5223547592759
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"clip_ratio/high_max": 0.0,
|
| 37 |
+
"clip_ratio/high_mean": 0.0,
|
| 38 |
+
"clip_ratio/low_mean": 0.0,
|
| 39 |
+
"clip_ratio/low_min": 0.0,
|
| 40 |
+
"clip_ratio/region_mean": 0.0,
|
| 41 |
+
"epoch": 0.0125,
|
| 42 |
+
"grad_norm": 0.9459657809558545,
|
| 43 |
+
"kl": 0.0,
|
| 44 |
+
"learning_rate": 4e-07,
|
| 45 |
+
"loss": 1.30385160446167e-08,
|
| 46 |
+
"step": 2,
|
| 47 |
+
"step_time": 73.55425463244319
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"clip_ratio/high_max": 0.00013787433999823406,
|
| 51 |
+
"clip_ratio/high_mean": 3.4468584999558516e-05,
|
| 52 |
+
"clip_ratio/low_mean": 0.00023351544223260134,
|
| 53 |
+
"clip_ratio/low_min": 0.0,
|
| 54 |
+
"clip_ratio/region_mean": 0.00026798402905114926,
|
| 55 |
+
"epoch": 0.01875,
|
| 56 |
+
"grad_norm": 1.30467860106837,
|
| 57 |
+
"kl": 0.00020495090939220972,
|
| 58 |
+
"learning_rate": 6e-07,
|
| 59 |
+
"loss": 0.00012717577919829637,
|
| 60 |
+
"step": 3,
|
| 61 |
+
"step_time": 72.84012562781572
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"clip_ratio/high_max": 0.00013858093007002026,
|
| 65 |
+
"clip_ratio/high_mean": 3.4645232517505065e-05,
|
| 66 |
+
"clip_ratio/low_mean": 7.62939453125e-05,
|
| 67 |
+
"clip_ratio/low_min": 0.0,
|
| 68 |
+
"clip_ratio/region_mean": 0.00011093917783000506,
|
| 69 |
+
"epoch": 0.025,
|
| 70 |
+
"grad_norm": 1.1138960555775923,
|
| 71 |
+
"kl": 0.00015524327136517968,
|
| 72 |
+
"learning_rate": 8e-07,
|
| 73 |
+
"loss": 6.386158929672092e-05,
|
| 74 |
+
"step": 4,
|
| 75 |
+
"step_time": 74.99929020181298
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"clip_ratio/high_max": 0.00252091878792271,
|
| 79 |
+
"clip_ratio/high_mean": 0.0006463462486863136,
|
| 80 |
+
"clip_ratio/low_mean": 0.0011124031516374089,
|
| 81 |
+
"clip_ratio/low_min": 0.0,
|
| 82 |
+
"clip_ratio/region_mean": 0.0017587494221515954,
|
| 83 |
+
"epoch": 0.03125,
|
| 84 |
+
"grad_norm": 4.79883269535261,
|
| 85 |
+
"kl": 0.0003803490926657105,
|
| 86 |
+
"learning_rate": 1e-06,
|
| 87 |
+
"loss": 0.00038909108843654394,
|
| 88 |
+
"step": 5,
|
| 89 |
+
"step_time": 73.31161722913384
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"clip_ratio/high_max": 0.0008073716089711525,
|
| 93 |
+
"clip_ratio/high_mean": 0.00021739790281571914,
|
| 94 |
+
"clip_ratio/low_mean": 0.0002899169921875,
|
| 95 |
+
"clip_ratio/low_min": 0.0,
|
| 96 |
+
"clip_ratio/region_mean": 0.0005073148931842297,
|
| 97 |
+
"epoch": 0.0375,
|
| 98 |
+
"grad_norm": 1.3774844283510703,
|
| 99 |
+
"kl": 0.00020736013902933337,
|
| 100 |
+
"learning_rate": 9.998973021172564e-07,
|
| 101 |
+
"loss": 0.00015363124839495867,
|
| 102 |
+
"step": 6,
|
| 103 |
+
"step_time": 74.78102323040366
|
| 104 |
+
},
|
| 105 |
+
{
|
| 106 |
+
"clip_ratio/high_max": 0.0014103966677794233,
|
| 107 |
+
"clip_ratio/high_mean": 0.0003525991669448558,
|
| 108 |
+
"clip_ratio/low_mean": 0.0007186873990576714,
|
| 109 |
+
"clip_ratio/low_min": 0.0,
|
| 110 |
+
"clip_ratio/region_mean": 0.001071286569640506,
|
| 111 |
+
"epoch": 0.04375,
|
| 112 |
+
"grad_norm": 2.594642601681158,
|
| 113 |
+
"kl": 0.0003412158403079957,
|
| 114 |
+
"learning_rate": 9.99589250656446e-07,
|
| 115 |
+
"loss": 0.00042259017936885357,
|
| 116 |
+
"step": 7,
|
| 117 |
+
"step_time": 73.97499135881662
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"clip_ratio/high_max": 0.0006687593195238151,
|
| 121 |
+
"clip_ratio/high_mean": 0.00016718982988095377,
|
| 122 |
+
"clip_ratio/low_mean": 0.0004885946400463581,
|
| 123 |
+
"clip_ratio/low_min": 0.0,
|
| 124 |
+
"clip_ratio/region_mean": 0.0006557844681083225,
|
| 125 |
+
"epoch": 0.05,
|
| 126 |
+
"grad_norm": 2.997302661420687,
|
| 127 |
+
"kl": 0.0002366066910326481,
|
| 128 |
+
"learning_rate": 9.990759721625005e-07,
|
| 129 |
+
"loss": 0.00012835719098802656,
|
| 130 |
+
"step": 8,
|
| 131 |
+
"step_time": 73.20104898884892
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"clip_ratio/high_max": 0.0,
|
| 135 |
+
"clip_ratio/high_mean": 0.0,
|
| 136 |
+
"clip_ratio/low_mean": 0.0,
|
| 137 |
+
"clip_ratio/low_min": 0.0,
|
| 138 |
+
"clip_ratio/region_mean": 0.0,
|
| 139 |
+
"completions/clipped_ratio": 0.3984375,
|
| 140 |
+
"completions/max_length": 2048.0,
|
| 141 |
+
"completions/mean_length": 1348.92578125,
|
| 142 |
+
"completions/min_length": 149.0,
|
| 143 |
+
"epoch": 0.05625,
|
| 144 |
+
"frac_reward_zero_std": 0.5,
|
| 145 |
+
"grad_norm": 0.9946026922972496,
|
| 146 |
+
"kl": 0.0002653920601005666,
|
| 147 |
+
"learning_rate": 9.983576774858775e-07,
|
| 148 |
+
"loss": 2.668411070771981e-06,
|
| 149 |
+
"reward": 0.3671875,
|
| 150 |
+
"reward_std": 0.2210792601108551,
|
| 151 |
+
"rewards/MajorityAccuracy/mean": 0.3671875,
|
| 152 |
+
"rewards/MajorityAccuracy/std": 0.48298245668411255,
|
| 153 |
+
"step": 9,
|
| 154 |
+
"step_time": 521.1360748670995
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"clip_ratio/high_max": 0.0007365329074673355,
|
| 158 |
+
"clip_ratio/high_mean": 0.00018413322686683387,
|
| 159 |
+
"clip_ratio/low_mean": 9.1552734375e-05,
|
| 160 |
+
"clip_ratio/low_min": 0.0,
|
| 161 |
+
"clip_ratio/region_mean": 0.00027568596124183387,
|
| 162 |
+
"epoch": 0.0625,
|
| 163 |
+
"grad_norm": 1.1125473405674269,
|
| 164 |
+
"kl": 0.00046436276170425117,
|
| 165 |
+
"learning_rate": 9.974346616959475e-07,
|
| 166 |
+
"loss": -8.042917033890262e-05,
|
| 167 |
+
"step": 10,
|
| 168 |
+
"step_time": 75.32976933568716
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"clip_ratio/high_max": 0.00105183370760642,
|
| 172 |
+
"clip_ratio/high_mean": 0.00031171037699095905,
|
| 173 |
+
"clip_ratio/low_mean": 0.00031523796496912837,
|
| 174 |
+
"clip_ratio/low_min": 0.0,
|
| 175 |
+
"clip_ratio/region_mean": 0.0006269483419600874,
|
| 176 |
+
"epoch": 0.06875,
|
| 177 |
+
"grad_norm": 1.7267943958433216,
|
| 178 |
+
"kl": 0.0005767217298853211,
|
| 179 |
+
"learning_rate": 9.963073039597796e-07,
|
| 180 |
+
"loss": 0.00012182450882392004,
|
| 181 |
+
"step": 11,
|
| 182 |
+
"step_time": 72.00949835032225
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"clip_ratio/high_max": 0.0015034668758744374,
|
| 186 |
+
"clip_ratio/high_mean": 0.0004660788290493656,
|
| 187 |
+
"clip_ratio/low_mean": 0.0004783690092153847,
|
| 188 |
+
"clip_ratio/low_min": 0.0,
|
| 189 |
+
"clip_ratio/region_mean": 0.0009444478419027291,
|
| 190 |
+
"epoch": 0.075,
|
| 191 |
+
"grad_norm": 1.3226552963959308,
|
| 192 |
+
"kl": 0.0005502448984771036,
|
| 193 |
+
"learning_rate": 9.949760673863846e-07,
|
| 194 |
+
"loss": -8.652787073515356e-05,
|
| 195 |
+
"step": 12,
|
| 196 |
+
"step_time": 74.34446802735329
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"clip_ratio/high_max": 0.002365104330237955,
|
| 200 |
+
"clip_ratio/high_mean": 0.0005912760825594887,
|
| 201 |
+
"clip_ratio/low_mean": 0.0005813279349240474,
|
| 202 |
+
"clip_ratio/low_min": 0.0,
|
| 203 |
+
"clip_ratio/region_mean": 0.0011726040174835362,
|
| 204 |
+
"epoch": 0.08125,
|
| 205 |
+
"grad_norm": 1.6355447194533486,
|
| 206 |
+
"kl": 0.0005874192356714047,
|
| 207 |
+
"learning_rate": 9.934414988364722e-07,
|
| 208 |
+
"loss": 0.0001525637781014666,
|
| 209 |
+
"step": 13,
|
| 210 |
+
"step_time": 75.45086678490043
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"clip_ratio/high_max": 0.002128555381204933,
|
| 214 |
+
"clip_ratio/high_mean": 0.0006181085773278028,
|
| 215 |
+
"clip_ratio/low_mean": 0.0004372063522168901,
|
| 216 |
+
"clip_ratio/low_min": 0.0,
|
| 217 |
+
"clip_ratio/region_mean": 0.001055314925906714,
|
| 218 |
+
"epoch": 0.0875,
|
| 219 |
+
"grad_norm": 1.9329924894413262,
|
| 220 |
+
"kl": 0.0006940279636182822,
|
| 221 |
+
"learning_rate": 9.917042286978063e-07,
|
| 222 |
+
"loss": 2.9903028917033225e-05,
|
| 223 |
+
"step": 14,
|
| 224 |
+
"step_time": 75.48207560926676
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"clip_ratio/high_max": 0.005292479589115828,
|
| 228 |
+
"clip_ratio/high_mean": 0.001323119897278957,
|
| 229 |
+
"clip_ratio/low_mean": 0.0012423099906300195,
|
| 230 |
+
"clip_ratio/low_min": 0.0,
|
| 231 |
+
"clip_ratio/region_mean": 0.002565429953392595,
|
| 232 |
+
"epoch": 0.09375,
|
| 233 |
+
"grad_norm": 1.746210277313782,
|
| 234 |
+
"kl": 0.0011365032696630806,
|
| 235 |
+
"learning_rate": 9.897649706262473e-07,
|
| 236 |
+
"loss": 0.00017141042917501181,
|
| 237 |
+
"step": 15,
|
| 238 |
+
"step_time": 74.02667892351747
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"clip_ratio/high_max": 0.005609305982943624,
|
| 242 |
+
"clip_ratio/high_mean": 0.002503583731595427,
|
| 243 |
+
"clip_ratio/low_mean": 0.0008624937036074698,
|
| 244 |
+
"clip_ratio/low_min": 0.0,
|
| 245 |
+
"clip_ratio/region_mean": 0.0033660774352028966,
|
| 246 |
+
"epoch": 0.1,
|
| 247 |
+
"grad_norm": 1.740458356023467,
|
| 248 |
+
"kl": 0.001083006769476924,
|
| 249 |
+
"learning_rate": 9.87624521252587e-07,
|
| 250 |
+
"loss": 0.0003503815387375653,
|
| 251 |
+
"step": 16,
|
| 252 |
+
"step_time": 73.8888233602047
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"clip_ratio/high_max": 0.0,
|
| 256 |
+
"clip_ratio/high_mean": 0.0,
|
| 257 |
+
"clip_ratio/low_mean": 0.0,
|
| 258 |
+
"clip_ratio/low_min": 0.0,
|
| 259 |
+
"clip_ratio/region_mean": 0.0,
|
| 260 |
+
"completions/clipped_ratio": 0.53125,
|
| 261 |
+
"completions/max_length": 2048.0,
|
| 262 |
+
"completions/mean_length": 1558.2578125,
|
| 263 |
+
"completions/min_length": 87.0,
|
| 264 |
+
"epoch": 0.10625,
|
| 265 |
+
"frac_reward_zero_std": 0.40625,
|
| 266 |
+
"grad_norm": 1.6007394614371044,
|
| 267 |
+
"kl": 0.0010908053518505767,
|
| 268 |
+
"learning_rate": 9.852837598553008e-07,
|
| 269 |
+
"loss": 1.0928706615231931e-05,
|
| 270 |
+
"reward": 0.3125,
|
| 271 |
+
"reward_std": 0.2661821246147156,
|
| 272 |
+
"rewards/MajorityAccuracy/mean": 0.3125,
|
| 273 |
+
"rewards/MajorityAccuracy/std": 0.4644203782081604,
|
| 274 |
+
"step": 17,
|
| 275 |
+
"step_time": 517.3998123966157
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"clip_ratio/high_max": 0.01139346236595884,
|
| 279 |
+
"clip_ratio/high_mean": 0.00284836559148971,
|
| 280 |
+
"clip_ratio/low_mean": 0.0009136572480201721,
|
| 281 |
+
"clip_ratio/low_min": 0.0,
|
| 282 |
+
"clip_ratio/region_mean": 0.003762022839509882,
|
| 283 |
+
"epoch": 0.1125,
|
| 284 |
+
"grad_norm": 4.877469711467282,
|
| 285 |
+
"kl": 0.0018789936730172485,
|
| 286 |
+
"learning_rate": 9.827436479993468e-07,
|
| 287 |
+
"loss": 0.00013730433420278132,
|
| 288 |
+
"step": 18,
|
| 289 |
+
"step_time": 73.47609467804432
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"clip_ratio/high_max": 0.002615714882267639,
|
| 293 |
+
"clip_ratio/high_mean": 0.0006539287205669098,
|
| 294 |
+
"clip_ratio/low_mean": 0.0006118500023148954,
|
| 295 |
+
"clip_ratio/low_min": 6.103515625e-05,
|
| 296 |
+
"clip_ratio/region_mean": 0.0012657787156058475,
|
| 297 |
+
"epoch": 0.11875,
|
| 298 |
+
"grad_norm": 1.9164024823671022,
|
| 299 |
+
"kl": 0.0011639377480605617,
|
| 300 |
+
"learning_rate": 9.800052291411629e-07,
|
| 301 |
+
"loss": 0.0003384137526154518,
|
| 302 |
+
"step": 19,
|
| 303 |
+
"step_time": 74.72441589459777
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"clip_ratio/high_max": 0.0023127737804315984,
|
| 307 |
+
"clip_ratio/high_mean": 0.0006888782227179036,
|
| 308 |
+
"clip_ratio/low_mean": 0.0013053212314844131,
|
| 309 |
+
"clip_ratio/low_min": 0.0,
|
| 310 |
+
"clip_ratio/region_mean": 0.0019941994396504015,
|
| 311 |
+
"epoch": 0.125,
|
| 312 |
+
"grad_norm": 1.9129612635200044,
|
| 313 |
+
"kl": 0.001775556811480783,
|
| 314 |
+
"learning_rate": 9.770696282000244e-07,
|
| 315 |
+
"loss": 0.0001273558591492474,
|
| 316 |
+
"step": 20,
|
| 317 |
+
"step_time": 73.22417965903878
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"clip_ratio/high_max": 0.0021240458663669415,
|
| 321 |
+
"clip_ratio/high_mean": 0.0005310114665917354,
|
| 322 |
+
"clip_ratio/low_mean": 0.00021817372180521488,
|
| 323 |
+
"clip_ratio/low_min": 0.0,
|
| 324 |
+
"clip_ratio/region_mean": 0.0007491851883969503,
|
| 325 |
+
"epoch": 0.13125,
|
| 326 |
+
"grad_norm": 1.4123371705205006,
|
| 327 |
+
"kl": 0.0030850676412228495,
|
| 328 |
+
"learning_rate": 9.739380510959364e-07,
|
| 329 |
+
"loss": 0.0001571828906890005,
|
| 330 |
+
"step": 21,
|
| 331 |
+
"step_time": 73.4378209412098
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"clip_ratio/high_max": 0.0010753278329502791,
|
| 335 |
+
"clip_ratio/high_mean": 0.0002688319582375698,
|
| 336 |
+
"clip_ratio/low_mean": 0.0004730224609375,
|
| 337 |
+
"clip_ratio/low_min": 0.0,
|
| 338 |
+
"clip_ratio/region_mean": 0.0007418544264510274,
|
| 339 |
+
"epoch": 0.1375,
|
| 340 |
+
"grad_norm": 1.0660216961650202,
|
| 341 |
+
"kl": 0.0014997907564975321,
|
| 342 |
+
"learning_rate": 9.706117842542516e-07,
|
| 343 |
+
"loss": 8.50217038532719e-05,
|
| 344 |
+
"step": 22,
|
| 345 |
+
"step_time": 74.86454071104527
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"clip_ratio/high_max": 0.0023321895132539794,
|
| 349 |
+
"clip_ratio/high_mean": 0.0006900679254613351,
|
| 350 |
+
"clip_ratio/low_mean": 0.000913863696041517,
|
| 351 |
+
"clip_ratio/low_min": 0.0,
|
| 352 |
+
"clip_ratio/region_mean": 0.0016039316469687037,
|
| 353 |
+
"epoch": 0.14375,
|
| 354 |
+
"grad_norm": 0.9697439573022547,
|
| 355 |
+
"kl": 0.0016111107834149152,
|
| 356 |
+
"learning_rate": 9.670921940772186e-07,
|
| 357 |
+
"loss": 0.00023680657614022493,
|
| 358 |
+
"step": 23,
|
| 359 |
+
"step_time": 74.42688927054405
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"clip_ratio/high_max": 0.005745383474277332,
|
| 363 |
+
"clip_ratio/high_mean": 0.0019300065687275492,
|
| 364 |
+
"clip_ratio/low_mean": 0.001005318212264683,
|
| 365 |
+
"clip_ratio/low_min": 0.0,
|
| 366 |
+
"clip_ratio/region_mean": 0.002935324766440317,
|
| 367 |
+
"epoch": 0.15,
|
| 368 |
+
"grad_norm": 1.6811455960802044,
|
| 369 |
+
"kl": 0.0020049167214892805,
|
| 370 |
+
"learning_rate": 9.633807263826744e-07,
|
| 371 |
+
"loss": 0.00027306011179462075,
|
| 372 |
+
"step": 24,
|
| 373 |
+
"step_time": 73.37336959317327
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"clip_ratio/high_max": 0.0,
|
| 377 |
+
"clip_ratio/high_mean": 0.0,
|
| 378 |
+
"clip_ratio/low_mean": 0.0,
|
| 379 |
+
"clip_ratio/low_min": 0.0,
|
| 380 |
+
"clip_ratio/region_mean": 0.0,
|
| 381 |
+
"completions/clipped_ratio": 0.51171875,
|
| 382 |
+
"completions/max_length": 2048.0,
|
| 383 |
+
"completions/mean_length": 1502.671875,
|
| 384 |
+
"completions/min_length": 117.0,
|
| 385 |
+
"epoch": 0.15625,
|
| 386 |
+
"frac_reward_zero_std": 0.375,
|
| 387 |
+
"grad_norm": 1.9056297708929744,
|
| 388 |
+
"kl": 0.0020018220675410703,
|
| 389 |
+
"learning_rate": 9.594789058101153e-07,
|
| 390 |
+
"loss": 2.001370012294501e-05,
|
| 391 |
+
"reward": 0.25,
|
| 392 |
+
"reward_std": 0.27552318572998047,
|
| 393 |
+
"rewards/MajorityAccuracy/mean": 0.25,
|
| 394 |
+
"rewards/MajorityAccuracy/std": 0.4338609278202057,
|
| 395 |
+
"step": 25,
|
| 396 |
+
"step_time": 519.4811908528209
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"clip_ratio/high_max": 0.001000058138743043,
|
| 400 |
+
"clip_ratio/high_mean": 0.00025001453468576074,
|
| 401 |
+
"clip_ratio/low_mean": 0.00013991654668643605,
|
| 402 |
+
"clip_ratio/low_min": 0.0,
|
| 403 |
+
"clip_ratio/region_mean": 0.0003899310831911862,
|
| 404 |
+
"epoch": 0.1625,
|
| 405 |
+
"grad_norm": 1.5259251617349252,
|
| 406 |
+
"kl": 0.002029442781349644,
|
| 407 |
+
"learning_rate": 9.55388335194388e-07,
|
| 408 |
+
"loss": -0.00011907667794730514,
|
| 409 |
+
"step": 26,
|
| 410 |
+
"step_time": 75.2788129709661
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"clip_ratio/high_max": 0.001166992587968707,
|
| 414 |
+
"clip_ratio/high_mean": 0.0003837501135421917,
|
| 415 |
+
"clip_ratio/low_mean": 0.0008414265030296519,
|
| 416 |
+
"clip_ratio/low_min": 0.0,
|
| 417 |
+
"clip_ratio/region_mean": 0.0012251766165718436,
|
| 418 |
+
"epoch": 0.16875,
|
| 419 |
+
"grad_norm": 1.7676959308999227,
|
| 420 |
+
"kl": 0.0026849806890822947,
|
| 421 |
+
"learning_rate": 9.511106949072587e-07,
|
| 422 |
+
"loss": 0.00018171517876908183,
|
| 423 |
+
"step": 27,
|
| 424 |
+
"step_time": 73.11744230613112
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"clip_ratio/high_max": 0.0008336667960975319,
|
| 428 |
+
"clip_ratio/high_mean": 0.0002732171124080196,
|
| 429 |
+
"clip_ratio/low_mean": 0.0006234441461856477,
|
| 430 |
+
"clip_ratio/low_min": 0.0,
|
| 431 |
+
"clip_ratio/region_mean": 0.0008966612585936673,
|
| 432 |
+
"epoch": 0.175,
|
| 433 |
+
"grad_norm": 1.0908037914908766,
|
| 434 |
+
"kl": 0.003132696554530412,
|
| 435 |
+
"learning_rate": 9.466477421671295e-07,
|
| 436 |
+
"loss": 8.278353197965771e-05,
|
| 437 |
+
"step": 28,
|
| 438 |
+
"step_time": 73.71278984844685
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"clip_ratio/high_max": 0.002535978885134682,
|
| 442 |
+
"clip_ratio/high_mean": 0.0007381830582744442,
|
| 443 |
+
"clip_ratio/low_mean": 0.002084880616166629,
|
| 444 |
+
"clip_ratio/low_min": 0.0,
|
| 445 |
+
"clip_ratio/region_mean": 0.0028230636671651155,
|
| 446 |
+
"epoch": 0.18125,
|
| 447 |
+
"grad_norm": 1.7592582646554116,
|
| 448 |
+
"kl": 0.003540548466844484,
|
| 449 |
+
"learning_rate": 9.420013103171891e-07,
|
| 450 |
+
"loss": 1.9709586922544986e-05,
|
| 451 |
+
"step": 29,
|
| 452 |
+
"step_time": 73.13300959765911
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"clip_ratio/high_max": 0.0033833983325166628,
|
| 456 |
+
"clip_ratio/high_mean": 0.0009477530729782302,
|
| 457 |
+
"clip_ratio/low_mean": 0.0023876800551079214,
|
| 458 |
+
"clip_ratio/low_min": 0.0,
|
| 459 |
+
"clip_ratio/region_mean": 0.0033354331681039184,
|
| 460 |
+
"epoch": 0.1875,
|
| 461 |
+
"grad_norm": 1.8225399963570847,
|
| 462 |
+
"kl": 0.003134117287117988,
|
| 463 |
+
"learning_rate": 9.37173308072291e-07,
|
| 464 |
+
"loss": 0.0005150716751813889,
|
| 465 |
+
"step": 30,
|
| 466 |
+
"step_time": 74.00688955560327
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"clip_ratio/high_max": 0.0,
|
| 470 |
+
"clip_ratio/high_mean": 0.0,
|
| 471 |
+
"clip_ratio/low_mean": 0.0,
|
| 472 |
+
"clip_ratio/low_min": 0.0,
|
| 473 |
+
"clip_ratio/region_mean": 0.0,
|
| 474 |
+
"epoch": 0.19375,
|
| 475 |
+
"grad_norm": 0.008685960915726126,
|
| 476 |
+
"kl": 0.0026164232986047864,
|
| 477 |
+
"learning_rate": 9.321657187348688e-07,
|
| 478 |
+
"loss": 2.617505015223287e-05,
|
| 479 |
+
"step": 31,
|
| 480 |
+
"step_time": 75.22697420790792
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"clip_ratio/high_max": 0.0019941278151236475,
|
| 484 |
+
"clip_ratio/high_mean": 0.0007269826455740258,
|
| 485 |
+
"clip_ratio/low_mean": 0.002218474997789599,
|
| 486 |
+
"clip_ratio/low_min": 0.0,
|
| 487 |
+
"clip_ratio/region_mean": 0.002945457643363625,
|
| 488 |
+
"epoch": 0.2,
|
| 489 |
+
"grad_norm": 1.8198979357393932,
|
| 490 |
+
"kl": 0.003297060029581189,
|
| 491 |
+
"learning_rate": 9.269805993802128e-07,
|
| 492 |
+
"loss": 0.00048082199646160007,
|
| 493 |
+
"step": 32,
|
| 494 |
+
"step_time": 74.8325846195221
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"clip_ratio/high_max": 0.0,
|
| 498 |
+
"clip_ratio/high_mean": 0.0,
|
| 499 |
+
"clip_ratio/low_mean": 0.0,
|
| 500 |
+
"clip_ratio/low_min": 0.0,
|
| 501 |
+
"clip_ratio/region_mean": 0.0,
|
| 502 |
+
"completions/clipped_ratio": 0.453125,
|
| 503 |
+
"completions/max_length": 2048.0,
|
| 504 |
+
"completions/mean_length": 1418.2578125,
|
| 505 |
+
"completions/min_length": 137.0,
|
| 506 |
+
"epoch": 0.20625,
|
| 507 |
+
"frac_reward_zero_std": 0.34375,
|
| 508 |
+
"grad_norm": 3.5280315339765296,
|
| 509 |
+
"kl": 0.0043433274258859456,
|
| 510 |
+
"learning_rate": 9.216200800114411e-07,
|
| 511 |
+
"loss": 4.3431296944618225e-05,
|
| 512 |
+
"reward": 0.33203125,
|
| 513 |
+
"reward_std": 0.287099689245224,
|
| 514 |
+
"rewards/MajorityAccuracy/mean": 0.33203125,
|
| 515 |
+
"rewards/MajorityAccuracy/std": 0.47186464071273804,
|
| 516 |
+
"step": 33,
|
| 517 |
+
"step_time": 517.0760045945644
|
| 518 |
+
},
|
| 519 |
+
{
|
| 520 |
+
"clip_ratio/high_max": 0.0008250083410530351,
|
| 521 |
+
"clip_ratio/high_mean": 0.0002259557440993376,
|
| 522 |
+
"clip_ratio/low_mean": 0.00020950000180164352,
|
| 523 |
+
"clip_ratio/low_min": 0.0,
|
| 524 |
+
"clip_ratio/region_mean": 0.00043545574953895994,
|
| 525 |
+
"epoch": 0.2125,
|
| 526 |
+
"grad_norm": 0.9961738957822602,
|
| 527 |
+
"kl": 0.003357837616931647,
|
| 528 |
+
"learning_rate": 9.160863626845119e-07,
|
| 529 |
+
"loss": -8.374909157282673e-07,
|
| 530 |
+
"step": 34,
|
| 531 |
+
"step_time": 74.67387718707323
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"clip_ratio/high_max": 0.0006233565218281001,
|
| 535 |
+
"clip_ratio/high_mean": 0.00015583913045702502,
|
| 536 |
+
"clip_ratio/low_mean": 0.00026032312598545104,
|
| 537 |
+
"clip_ratio/low_min": 0.0,
|
| 538 |
+
"clip_ratio/region_mean": 0.00041616225644247606,
|
| 539 |
+
"epoch": 0.21875,
|
| 540 |
+
"grad_norm": 2.7621642373469952,
|
| 541 |
+
"kl": 0.003123411675915122,
|
| 542 |
+
"learning_rate": 9.103817206036382e-07,
|
| 543 |
+
"loss": 7.218452810775489e-05,
|
| 544 |
+
"step": 35,
|
| 545 |
+
"step_time": 74.69708428531885
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"clip_ratio/high_max": 0.002714538626605645,
|
| 549 |
+
"clip_ratio/high_mean": 0.0007643270073458552,
|
| 550 |
+
"clip_ratio/low_mean": 0.00019464710931060836,
|
| 551 |
+
"clip_ratio/low_min": 0.0,
|
| 552 |
+
"clip_ratio/region_mean": 0.0009589741166564636,
|
| 553 |
+
"epoch": 0.225,
|
| 554 |
+
"grad_norm": 1.7433732825457273,
|
| 555 |
+
"kl": 0.003681796893943101,
|
| 556 |
+
"learning_rate": 9.045084971874737e-07,
|
| 557 |
+
"loss": 0.000480028975289315,
|
| 558 |
+
"step": 36,
|
| 559 |
+
"step_time": 73.06108717247844
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"clip_ratio/high_max": 0.004595063393935561,
|
| 563 |
+
"clip_ratio/high_mean": 0.001308612379943952,
|
| 564 |
+
"clip_ratio/low_mean": 0.0006579212822543923,
|
| 565 |
+
"clip_ratio/low_min": 0.0,
|
| 566 |
+
"clip_ratio/region_mean": 0.0019665336585603654,
|
| 567 |
+
"epoch": 0.23125,
|
| 568 |
+
"grad_norm": 1.9397026663329515,
|
| 569 |
+
"kl": 0.005285477731376886,
|
| 570 |
+
"learning_rate": 8.984691051064574e-07,
|
| 571 |
+
"loss": 3.5799737815978006e-05,
|
| 572 |
+
"step": 37,
|
| 573 |
+
"step_time": 72.80702435225248
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"clip_ratio/high_max": 0.0021414044458651915,
|
| 577 |
+
"clip_ratio/high_mean": 0.0006601417662750464,
|
| 578 |
+
"clip_ratio/low_mean": 0.0015340583340730518,
|
| 579 |
+
"clip_ratio/low_min": 0.0,
|
| 580 |
+
"clip_ratio/region_mean": 0.002194200103986077,
|
| 581 |
+
"epoch": 0.2375,
|
| 582 |
+
"grad_norm": 2.142389203365126,
|
| 583 |
+
"kl": 0.004392585105961189,
|
| 584 |
+
"learning_rate": 8.922660252917087e-07,
|
| 585 |
+
"loss": 0.00045010127360001206,
|
| 586 |
+
"step": 38,
|
| 587 |
+
"step_time": 74.77272975817323
|
| 588 |
+
},
|
| 589 |
+
{
|
| 590 |
+
"clip_ratio/high_max": 0.002902920765336603,
|
| 591 |
+
"clip_ratio/high_mean": 0.0007918677583802491,
|
| 592 |
+
"clip_ratio/low_mean": 0.0013848157686879858,
|
| 593 |
+
"clip_ratio/low_min": 0.0,
|
| 594 |
+
"clip_ratio/region_mean": 0.0021766835125163198,
|
| 595 |
+
"epoch": 0.24375,
|
| 596 |
+
"grad_norm": 1.6250895797671419,
|
| 597 |
+
"kl": 0.003375674656126648,
|
| 598 |
+
"learning_rate": 8.859018059158809e-07,
|
| 599 |
+
"loss": 0.00021965500491205603,
|
| 600 |
+
"step": 39,
|
| 601 |
+
"step_time": 73.48269414529204
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"clip_ratio/high_max": 0.006917939201230183,
|
| 605 |
+
"clip_ratio/high_mean": 0.002284060617967043,
|
| 606 |
+
"clip_ratio/low_mean": 0.0005010104941902682,
|
| 607 |
+
"clip_ratio/low_min": 0.0,
|
| 608 |
+
"clip_ratio/region_mean": 0.0027850710757775232,
|
| 609 |
+
"epoch": 0.25,
|
| 610 |
+
"grad_norm": 2.188761097359433,
|
| 611 |
+
"kl": 0.005054004432167858,
|
| 612 |
+
"learning_rate": 8.793790613463954e-07,
|
| 613 |
+
"loss": 0.0003297346120234579,
|
| 614 |
+
"step": 40,
|
| 615 |
+
"step_time": 73.4439012221992
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"clip_ratio/high_max": 0.0,
|
| 619 |
+
"clip_ratio/high_mean": 0.0,
|
| 620 |
+
"clip_ratio/low_mean": 0.0,
|
| 621 |
+
"clip_ratio/low_min": 0.0,
|
| 622 |
+
"clip_ratio/region_mean": 0.0,
|
| 623 |
+
"completions/clipped_ratio": 0.546875,
|
| 624 |
+
"completions/max_length": 2048.0,
|
| 625 |
+
"completions/mean_length": 1569.2578125,
|
| 626 |
+
"completions/min_length": 102.0,
|
| 627 |
+
"epoch": 0.25625,
|
| 628 |
+
"frac_reward_zero_std": 0.375,
|
| 629 |
+
"grad_norm": 3.560953419181603,
|
| 630 |
+
"kl": 0.00709361769258976,
|
| 631 |
+
"learning_rate": 8.727004710714798e-07,
|
| 632 |
+
"loss": 7.0963054895401e-05,
|
| 633 |
+
"reward": 0.2421875,
|
| 634 |
+
"reward_std": 0.27552318572998047,
|
| 635 |
+
"rewards/MajorityAccuracy/mean": 0.2421875,
|
| 636 |
+
"rewards/MajorityAccuracy/std": 0.4292463958263397,
|
| 637 |
+
"step": 41,
|
| 638 |
+
"step_time": 516.8003982715309
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"clip_ratio/high_max": 0.000788942925282754,
|
| 642 |
+
"clip_ratio/high_mean": 0.00022951878781896085,
|
| 643 |
+
"clip_ratio/low_mean": 0.00022395513951778412,
|
| 644 |
+
"clip_ratio/low_min": 0.0,
|
| 645 |
+
"clip_ratio/region_mean": 0.0004534739346127026,
|
| 646 |
+
"epoch": 0.2625,
|
| 647 |
+
"grad_norm": 1.6310971735167341,
|
| 648 |
+
"kl": 0.004359257174655795,
|
| 649 |
+
"learning_rate": 8.658687785994578e-07,
|
| 650 |
+
"loss": 0.00018209085101261735,
|
| 651 |
+
"step": 42,
|
| 652 |
+
"step_time": 73.48767914995551
|
| 653 |
+
},
|
| 654 |
+
{
|
| 655 |
+
"clip_ratio/high_max": 0.00014810427092015743,
|
| 656 |
+
"clip_ratio/high_mean": 3.702606773003936e-05,
|
| 657 |
+
"clip_ratio/low_mean": 7.62939453125e-05,
|
| 658 |
+
"clip_ratio/low_min": 0.0,
|
| 659 |
+
"clip_ratio/region_mean": 0.00011332001304253936,
|
| 660 |
+
"epoch": 0.26875,
|
| 661 |
+
"grad_norm": 0.7076925675402148,
|
| 662 |
+
"kl": 0.004095586598850787,
|
| 663 |
+
"learning_rate": 8.588867903317394e-07,
|
| 664 |
+
"loss": 2.1422167264972813e-05,
|
| 665 |
+
"step": 43,
|
| 666 |
+
"step_time": 75.41750577837229
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"clip_ratio/high_max": 0.0017757921305019408,
|
| 670 |
+
"clip_ratio/high_mean": 0.0005022502664360218,
|
| 671 |
+
"clip_ratio/low_mean": 0.00041227911424357444,
|
| 672 |
+
"clip_ratio/low_min": 0.0,
|
| 673 |
+
"clip_ratio/region_mean": 0.0009145293806795962,
|
| 674 |
+
"epoch": 0.275,
|
| 675 |
+
"grad_norm": 1.240797088502556,
|
| 676 |
+
"kl": 0.005239230813458562,
|
| 677 |
+
"learning_rate": 8.51757374409974e-07,
|
| 678 |
+
"loss": 0.0001312796666752547,
|
| 679 |
+
"step": 44,
|
| 680 |
+
"step_time": 75.07205566763878
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"clip_ratio/high_max": 0.00396852963604033,
|
| 684 |
+
"clip_ratio/high_mean": 0.001217511082359124,
|
| 685 |
+
"clip_ratio/low_mean": 0.001209133377415128,
|
| 686 |
+
"clip_ratio/low_min": 0.0,
|
| 687 |
+
"clip_ratio/region_mean": 0.0024266444524982944,
|
| 688 |
+
"epoch": 0.28125,
|
| 689 |
+
"grad_norm": 2.6285133323696406,
|
| 690 |
+
"kl": 0.004023105022497475,
|
| 691 |
+
"learning_rate": 8.444834595378433e-07,
|
| 692 |
+
"loss": -3.817494143731892e-05,
|
| 693 |
+
"step": 45,
|
| 694 |
+
"step_time": 73.23190971836448
|
| 695 |
+
},
|
| 696 |
+
{
|
| 697 |
+
"clip_ratio/high_max": 0.0033378129883203655,
|
| 698 |
+
"clip_ratio/high_mean": 0.0011150640566484071,
|
| 699 |
+
"clip_ratio/low_mean": 0.0012092852193745784,
|
| 700 |
+
"clip_ratio/low_min": 0.0,
|
| 701 |
+
"clip_ratio/region_mean": 0.0023243492760229856,
|
| 702 |
+
"epoch": 0.2875,
|
| 703 |
+
"grad_norm": 2.2505738628640892,
|
| 704 |
+
"kl": 0.008873901912011206,
|
| 705 |
+
"learning_rate": 8.370680337779736e-07,
|
| 706 |
+
"loss": 0.000795513391494751,
|
| 707 |
+
"step": 46,
|
| 708 |
+
"step_time": 73.41908664628863
|
| 709 |
+
},
|
| 710 |
+
{
|
| 711 |
+
"clip_ratio/high_max": 0.0025950818962883204,
|
| 712 |
+
"clip_ratio/high_mean": 0.0007081434960127808,
|
| 713 |
+
"clip_ratio/low_mean": 0.00102339743898483,
|
| 714 |
+
"clip_ratio/low_min": 0.0,
|
| 715 |
+
"clip_ratio/region_mean": 0.0017315409422735684,
|
| 716 |
+
"epoch": 0.29375,
|
| 717 |
+
"grad_norm": 1.330621272177699,
|
| 718 |
+
"kl": 0.006177376781124622,
|
| 719 |
+
"learning_rate": 8.295141433244659e-07,
|
| 720 |
+
"loss": 0.000267393741523847,
|
| 721 |
+
"step": 47,
|
| 722 |
+
"step_time": 73.48475310578942
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"clip_ratio/high_max": 0.003605342179071158,
|
| 726 |
+
"clip_ratio/high_mean": 0.0009732159523991868,
|
| 727 |
+
"clip_ratio/low_mean": 0.0014871827879687771,
|
| 728 |
+
"clip_ratio/low_min": 0.0,
|
| 729 |
+
"clip_ratio/region_mean": 0.0024603987112641335,
|
| 730 |
+
"epoch": 0.3,
|
| 731 |
+
"grad_norm": 1.3669504148723979,
|
| 732 |
+
"kl": 0.005506305024027824,
|
| 733 |
+
"learning_rate": 8.218248912515442e-07,
|
| 734 |
+
"loss": -5.289263208396733e-05,
|
| 735 |
+
"step": 48,
|
| 736 |
+
"step_time": 75.12205986306071
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"clip_ratio/high_max": 0.0,
|
| 740 |
+
"clip_ratio/high_mean": 0.0,
|
| 741 |
+
"clip_ratio/low_mean": 0.0,
|
| 742 |
+
"clip_ratio/low_min": 0.0,
|
| 743 |
+
"clip_ratio/region_mean": 0.0,
|
| 744 |
+
"completions/clipped_ratio": 0.484375,
|
| 745 |
+
"completions/max_length": 2048.0,
|
| 746 |
+
"completions/mean_length": 1479.77734375,
|
| 747 |
+
"completions/min_length": 223.0,
|
| 748 |
+
"epoch": 0.30625,
|
| 749 |
+
"frac_reward_zero_std": 0.3125,
|
| 750 |
+
"grad_norm": 2.0622994318044032,
|
| 751 |
+
"kl": 0.005408838973380625,
|
| 752 |
+
"learning_rate": 8.140034362388397e-07,
|
| 753 |
+
"loss": 5.410988524090499e-05,
|
| 754 |
+
"reward": 0.296875,
|
| 755 |
+
"reward_std": 0.3062812089920044,
|
| 756 |
+
"rewards/MajorityAccuracy/mean": 0.296875,
|
| 757 |
+
"rewards/MajorityAccuracy/std": 0.45777595043182373,
|
| 758 |
+
"step": 49,
|
| 759 |
+
"step_time": 519.5137377269566
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"clip_ratio/high_max": 0.0026001971709774807,
|
| 763 |
+
"clip_ratio/high_mean": 0.0008267014709417708,
|
| 764 |
+
"clip_ratio/low_mean": 0.00032194448067457415,
|
| 765 |
+
"clip_ratio/low_min": 0.0,
|
| 766 |
+
"clip_ratio/region_mean": 0.0011486459443403874,
|
| 767 |
+
"epoch": 0.3125,
|
| 768 |
+
"grad_norm": 1.6635727317722462,
|
| 769 |
+
"kl": 0.005176276550628245,
|
| 770 |
+
"learning_rate": 8.060529912738314e-07,
|
| 771 |
+
"loss": 0.000208328478038311,
|
| 772 |
+
"step": 50,
|
| 773 |
+
"step_time": 73.46314207464457
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"clip_ratio/high_max": 0.0028284856016398408,
|
| 777 |
+
"clip_ratio/high_mean": 0.0007803857970429817,
|
| 778 |
+
"clip_ratio/low_mean": 0.0005973704610369168,
|
| 779 |
+
"clip_ratio/low_min": 0.0,
|
| 780 |
+
"clip_ratio/region_mean": 0.0013777562562609091,
|
| 781 |
+
"epoch": 0.31875,
|
| 782 |
+
"grad_norm": 1.3855811305630168,
|
| 783 |
+
"kl": 0.006209813465829939,
|
| 784 |
+
"learning_rate": 7.979768223319785e-07,
|
| 785 |
+
"loss": 0.000343125342624262,
|
| 786 |
+
"step": 51,
|
| 787 |
+
"step_time": 75.06738603487611
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"clip_ratio/high_max": 0.0009015696487040259,
|
| 791 |
+
"clip_ratio/high_mean": 0.0003309696894575609,
|
| 792 |
+
"clip_ratio/low_mean": 0.0003817197502939962,
|
| 793 |
+
"clip_ratio/low_min": 0.0,
|
| 794 |
+
"clip_ratio/region_mean": 0.0007126894452085253,
|
| 795 |
+
"epoch": 0.325,
|
| 796 |
+
"grad_norm": 1.1650759536719177,
|
| 797 |
+
"kl": 0.006465354294050485,
|
| 798 |
+
"learning_rate": 7.897782470350849e-07,
|
| 799 |
+
"loss": 0.00014516718511004,
|
| 800 |
+
"step": 52,
|
| 801 |
+
"step_time": 75.17007875442505
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"clip_ratio/high_max": 0.008496318099787459,
|
| 805 |
+
"clip_ratio/high_mean": 0.002401231315161567,
|
| 806 |
+
"clip_ratio/low_mean": 0.0027177518350072205,
|
| 807 |
+
"clip_ratio/low_min": 0.0,
|
| 808 |
+
"clip_ratio/region_mean": 0.005118983186548576,
|
| 809 |
+
"epoch": 0.33125,
|
| 810 |
+
"grad_norm": 4.30678366598094,
|
| 811 |
+
"kl": 0.00909659347962588,
|
| 812 |
+
"learning_rate": 7.814606332884488e-07,
|
| 813 |
+
"loss": 0.00031716161174699664,
|
| 814 |
+
"step": 53,
|
| 815 |
+
"step_time": 73.48575103282928
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"clip_ratio/high_max": 0.004987899446859956,
|
| 819 |
+
"clip_ratio/high_mean": 0.001246974861714989,
|
| 820 |
+
"clip_ratio/low_mean": 0.0009175230443361215,
|
| 821 |
+
"clip_ratio/low_min": 0.0,
|
| 822 |
+
"clip_ratio/region_mean": 0.002164497898775153,
|
| 823 |
+
"epoch": 0.3375,
|
| 824 |
+
"grad_norm": 3.233499274862055,
|
| 825 |
+
"kl": 0.008112531621009111,
|
| 826 |
+
"learning_rate": 7.730273978973552e-07,
|
| 827 |
+
"loss": 0.00010565328557277098,
|
| 828 |
+
"step": 54,
|
| 829 |
+
"step_time": 73.17641399428248
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"clip_ratio/high_max": 0.003960966147133149,
|
| 833 |
+
"clip_ratio/high_mean": 0.0011415708540880587,
|
| 834 |
+
"clip_ratio/low_mean": 0.0005705947114620358,
|
| 835 |
+
"clip_ratio/low_min": 0.0,
|
| 836 |
+
"clip_ratio/region_mean": 0.0017121655691880733,
|
| 837 |
+
"epoch": 0.34375,
|
| 838 |
+
"grad_norm": 1.452087291727142,
|
| 839 |
+
"kl": 0.007018269330728799,
|
| 840 |
+
"learning_rate": 7.644820051634812e-07,
|
| 841 |
+
"loss": -2.5527318939566612e-05,
|
| 842 |
+
"step": 55,
|
| 843 |
+
"step_time": 75.14776101335883
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"clip_ratio/high_max": 0.0033280476927757263,
|
| 847 |
+
"clip_ratio/high_mean": 0.0010957905542454682,
|
| 848 |
+
"clip_ratio/low_mean": 0.0006241830415092409,
|
| 849 |
+
"clip_ratio/low_min": 0.0,
|
| 850 |
+
"clip_ratio/region_mean": 0.0017199736030306667,
|
| 851 |
+
"epoch": 0.35,
|
| 852 |
+
"grad_norm": 1.4159811531962458,
|
| 853 |
+
"kl": 0.005303874262608588,
|
| 854 |
+
"learning_rate": 7.558279654617912e-07,
|
| 855 |
+
"loss": 0.00010682064748834819,
|
| 856 |
+
"step": 56,
|
| 857 |
+
"step_time": 75.27423724532127
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"clip_ratio/high_max": 0.0,
|
| 861 |
+
"clip_ratio/high_mean": 0.0,
|
| 862 |
+
"clip_ratio/low_mean": 0.0,
|
| 863 |
+
"clip_ratio/low_min": 0.0,
|
| 864 |
+
"clip_ratio/region_mean": 0.0,
|
| 865 |
+
"completions/clipped_ratio": 0.46875,
|
| 866 |
+
"completions/max_length": 2048.0,
|
| 867 |
+
"completions/mean_length": 1457.265625,
|
| 868 |
+
"completions/min_length": 81.0,
|
| 869 |
+
"epoch": 0.35625,
|
| 870 |
+
"frac_reward_zero_std": 0.28125,
|
| 871 |
+
"grad_norm": 1.8872441581523447,
|
| 872 |
+
"kl": 0.005777926824521273,
|
| 873 |
+
"learning_rate": 7.470688337985029e-07,
|
| 874 |
+
"loss": 5.7812780141830444e-05,
|
| 875 |
+
"reward": 0.31640625,
|
| 876 |
+
"reward_std": 0.32799121737480164,
|
| 877 |
+
"rewards/MajorityAccuracy/mean": 0.31640625,
|
| 878 |
+
"rewards/MajorityAccuracy/std": 0.4659844934940338,
|
| 879 |
+
"step": 57,
|
| 880 |
+
"step_time": 517.2475496456027
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"clip_ratio/high_max": 9.682416566647589e-05,
|
| 884 |
+
"clip_ratio/high_mean": 2.4206041416618973e-05,
|
| 885 |
+
"clip_ratio/low_mean": 0.0001759087863320019,
|
| 886 |
+
"clip_ratio/low_min": 0.0,
|
| 887 |
+
"clip_ratio/region_mean": 0.00020011482774862088,
|
| 888 |
+
"epoch": 0.3625,
|
| 889 |
+
"grad_norm": 0.7958223925677695,
|
| 890 |
+
"kl": 0.006586018018424511,
|
| 891 |
+
"learning_rate": 7.382082083507225e-07,
|
| 892 |
+
"loss": 0.00015770539175719023,
|
| 893 |
+
"step": 58,
|
| 894 |
+
"step_time": 73.49105106666684
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"clip_ratio/high_max": 0.002435207410599105,
|
| 898 |
+
"clip_ratio/high_mean": 0.0008617622188467067,
|
| 899 |
+
"clip_ratio/low_mean": 0.00036808526783715934,
|
| 900 |
+
"clip_ratio/low_min": 0.0,
|
| 901 |
+
"clip_ratio/region_mean": 0.0012298474975978024,
|
| 902 |
+
"epoch": 0.36875,
|
| 903 |
+
"grad_norm": 1.7599979493226297,
|
| 904 |
+
"kl": 0.00523577252170071,
|
| 905 |
+
"learning_rate": 7.292497289883432e-07,
|
| 906 |
+
"loss": 0.0002776154433377087,
|
| 907 |
+
"step": 59,
|
| 908 |
+
"step_time": 73.40984213724732
|
| 909 |
+
},
|
| 910 |
+
{
|
| 911 |
+
"clip_ratio/high_max": 0.002041251558694057,
|
| 912 |
+
"clip_ratio/high_mean": 0.0005513043324754108,
|
| 913 |
+
"clip_ratio/low_mean": 0.001034867449561716,
|
| 914 |
+
"clip_ratio/low_min": 6.103515625e-05,
|
| 915 |
+
"clip_ratio/region_mean": 0.0015861718129599467,
|
| 916 |
+
"epoch": 0.375,
|
| 917 |
+
"grad_norm": 2.3346919207601857,
|
| 918 |
+
"kl": 0.006138660421129316,
|
| 919 |
+
"learning_rate": 7.201970757788171e-07,
|
| 920 |
+
"loss": 0.0002382323145866394,
|
| 921 |
+
"step": 60,
|
| 922 |
+
"step_time": 75.00094489753246
|
| 923 |
+
},
|
| 924 |
+
{
|
| 925 |
+
"clip_ratio/high_max": 0.001766975037753582,
|
| 926 |
+
"clip_ratio/high_mean": 0.00047582227853126824,
|
| 927 |
+
"clip_ratio/low_mean": 0.0010020178597187623,
|
| 928 |
+
"clip_ratio/low_min": 0.0,
|
| 929 |
+
"clip_ratio/region_mean": 0.0014778401164221577,
|
| 930 |
+
"epoch": 0.38125,
|
| 931 |
+
"grad_norm": 1.6479319548897207,
|
| 932 |
+
"kl": 0.00678858399624005,
|
| 933 |
+
"learning_rate": 7.110539674754159e-07,
|
| 934 |
+
"loss": 0.00019870793039444834,
|
| 935 |
+
"step": 61,
|
| 936 |
+
"step_time": 74.26738295331597
|
| 937 |
+
},
|
| 938 |
+
{
|
| 939 |
+
"clip_ratio/high_max": 0.0020797776960534975,
|
| 940 |
+
"clip_ratio/high_mean": 0.0005199444240133744,
|
| 941 |
+
"clip_ratio/low_mean": 0.0005633395267068408,
|
| 942 |
+
"clip_ratio/low_min": 0.0,
|
| 943 |
+
"clip_ratio/region_mean": 0.001083283954358194,
|
| 944 |
+
"epoch": 0.3875,
|
| 945 |
+
"grad_norm": 1.1551951490830197,
|
| 946 |
+
"kl": 0.005675450142007321,
|
| 947 |
+
"learning_rate": 7.018241599895973e-07,
|
| 948 |
+
"loss": 0.00021209912665653974,
|
| 949 |
+
"step": 62,
|
| 950 |
+
"step_time": 74.59427855163813
|
| 951 |
+
},
|
| 952 |
+
{
|
| 953 |
+
"clip_ratio/high_max": 0.005505284294486046,
|
| 954 |
+
"clip_ratio/high_mean": 0.0024661104544065893,
|
| 955 |
+
"clip_ratio/low_mean": 0.003516978722473141,
|
| 956 |
+
"clip_ratio/low_min": 0.0006174813897814602,
|
| 957 |
+
"clip_ratio/region_mean": 0.005983089184155688,
|
| 958 |
+
"epoch": 0.39375,
|
| 959 |
+
"grad_norm": 3.6573495676011114,
|
| 960 |
+
"kl": 0.014393504476174712,
|
| 961 |
+
"learning_rate": 6.925114448481088e-07,
|
| 962 |
+
"loss": 0.0010254168882966042,
|
| 963 |
+
"step": 63,
|
| 964 |
+
"step_time": 72.63141867890954
|
| 965 |
+
},
|
| 966 |
+
{
|
| 967 |
+
"clip_ratio/high_max": 0.0017844388930825517,
|
| 968 |
+
"clip_ratio/high_mean": 0.0004461097232706379,
|
| 969 |
+
"clip_ratio/low_mean": 0.001217032433487475,
|
| 970 |
+
"clip_ratio/low_min": 0.0,
|
| 971 |
+
"clip_ratio/region_mean": 0.0016631421458441764,
|
| 972 |
+
"epoch": 0.4,
|
| 973 |
+
"grad_norm": 1.452181195216457,
|
| 974 |
+
"kl": 0.01043098943773657,
|
| 975 |
+
"learning_rate": 6.831196476354614e-07,
|
| 976 |
+
"loss": 0.0003088266821578145,
|
| 977 |
+
"step": 64,
|
| 978 |
+
"step_time": 75.3654635772109
|
| 979 |
+
},
|
| 980 |
+
{
|
| 981 |
+
"clip_ratio/high_max": 0.0,
|
| 982 |
+
"clip_ratio/high_mean": 0.0,
|
| 983 |
+
"clip_ratio/low_mean": 0.0,
|
| 984 |
+
"clip_ratio/low_min": 0.0,
|
| 985 |
+
"clip_ratio/region_mean": 0.0,
|
| 986 |
+
"completions/clipped_ratio": 0.36328125,
|
| 987 |
+
"completions/max_length": 2048.0,
|
| 988 |
+
"completions/mean_length": 1284.5625,
|
| 989 |
+
"completions/min_length": 139.0,
|
| 990 |
+
"epoch": 0.40625,
|
| 991 |
+
"frac_reward_zero_std": 0.1875,
|
| 992 |
+
"grad_norm": 1.3364059167984172,
|
| 993 |
+
"kl": 0.006581229681614786,
|
| 994 |
+
"learning_rate": 6.7365262642241e-07,
|
| 995 |
+
"loss": 6.580284389201552e-05,
|
| 996 |
+
"reward": 0.44140625,
|
| 997 |
+
"reward_std": 0.36679214239120483,
|
| 998 |
+
"rewards/MajorityAccuracy/mean": 0.44140625,
|
| 999 |
+
"rewards/MajorityAccuracy/std": 0.4975275695323944,
|
| 1000 |
+
"step": 65,
|
| 1001 |
+
"step_time": 512.667545478791
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"clip_ratio/high_max": 0.0036375748604768887,
|
| 1005 |
+
"clip_ratio/high_mean": 0.0011029609195247758,
|
| 1006 |
+
"clip_ratio/low_mean": 0.0005945381562924013,
|
| 1007 |
+
"clip_ratio/low_min": 0.0,
|
| 1008 |
+
"clip_ratio/region_mean": 0.0016974990867311135,
|
| 1009 |
+
"epoch": 0.4125,
|
| 1010 |
+
"grad_norm": 1.5803846940023683,
|
| 1011 |
+
"kl": 0.007414525782223791,
|
| 1012 |
+
"learning_rate": 6.641142701810931e-07,
|
| 1013 |
+
"loss": 5.692944978363812e-05,
|
| 1014 |
+
"step": 66,
|
| 1015 |
+
"step_time": 74.33141418546438
|
| 1016 |
+
},
|
| 1017 |
+
{
|
| 1018 |
+
"clip_ratio/high_max": 0.0020199251594021916,
|
| 1019 |
+
"clip_ratio/high_mean": 0.0005049812898505479,
|
| 1020 |
+
"clip_ratio/low_mean": 0.0007356425194302574,
|
| 1021 |
+
"clip_ratio/low_min": 0.0,
|
| 1022 |
+
"clip_ratio/region_mean": 0.001240623812918784,
|
| 1023 |
+
"epoch": 0.41875,
|
| 1024 |
+
"grad_norm": 2.188878387275931,
|
| 1025 |
+
"kl": 0.007878637290559709,
|
| 1026 |
+
"learning_rate": 6.545084971874736e-07,
|
| 1027 |
+
"loss": 0.000116347597213462,
|
| 1028 |
+
"step": 67,
|
| 1029 |
+
"step_time": 74.82191213220358
|
| 1030 |
+
},
|
| 1031 |
+
{
|
| 1032 |
+
"clip_ratio/high_max": 0.0036529545031953603,
|
| 1033 |
+
"clip_ratio/high_mean": 0.0009720898597151972,
|
| 1034 |
+
"clip_ratio/low_mean": 0.00032160232512978837,
|
| 1035 |
+
"clip_ratio/low_min": 0.0,
|
| 1036 |
+
"clip_ratio/region_mean": 0.0012936921848449856,
|
| 1037 |
+
"epoch": 0.425,
|
| 1038 |
+
"grad_norm": 2.1829734221900927,
|
| 1039 |
+
"kl": 0.006164729944430292,
|
| 1040 |
+
"learning_rate": 6.448392534117461e-07,
|
| 1041 |
+
"loss": 2.5652116164565086e-06,
|
| 1042 |
+
"step": 68,
|
| 1043 |
+
"step_time": 75.37111585959792
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"clip_ratio/high_max": 0.0015615543234162033,
|
| 1047 |
+
"clip_ratio/high_mean": 0.0004322786699049175,
|
| 1048 |
+
"clip_ratio/low_mean": 0.0015808784664841369,
|
| 1049 |
+
"clip_ratio/low_min": 0.0,
|
| 1050 |
+
"clip_ratio/region_mean": 0.0020131571363890544,
|
| 1051 |
+
"epoch": 0.43125,
|
| 1052 |
+
"grad_norm": 2.7131838591882547,
|
| 1053 |
+
"kl": 0.008115546079352498,
|
| 1054 |
+
"learning_rate": 6.351105108973644e-07,
|
| 1055 |
+
"loss": 0.00015989529492799193,
|
| 1056 |
+
"step": 69,
|
| 1057 |
+
"step_time": 73.92649275809526
|
| 1058 |
+
},
|
| 1059 |
+
{
|
| 1060 |
+
"clip_ratio/high_max": 0.004396992415422574,
|
| 1061 |
+
"clip_ratio/high_mean": 0.0016422588159912266,
|
| 1062 |
+
"clip_ratio/low_mean": 0.0003944867057725787,
|
| 1063 |
+
"clip_ratio/low_min": 0.0,
|
| 1064 |
+
"clip_ratio/region_mean": 0.0020367455217638053,
|
| 1065 |
+
"epoch": 0.4375,
|
| 1066 |
+
"grad_norm": 2.4229514607658054,
|
| 1067 |
+
"kl": 0.007655080873519182,
|
| 1068 |
+
"learning_rate": 6.253262661293602e-07,
|
| 1069 |
+
"loss": 0.0005161119624972343,
|
| 1070 |
+
"step": 70,
|
| 1071 |
+
"step_time": 74.57587697729468
|
| 1072 |
+
},
|
| 1073 |
+
{
|
| 1074 |
+
"clip_ratio/high_max": 0.007809678791090846,
|
| 1075 |
+
"clip_ratio/high_mean": 0.003787895548157394,
|
| 1076 |
+
"clip_ratio/low_mean": 0.00047851866111159325,
|
| 1077 |
+
"clip_ratio/low_min": 0.0,
|
| 1078 |
+
"clip_ratio/region_mean": 0.004266414209268987,
|
| 1079 |
+
"epoch": 0.44375,
|
| 1080 |
+
"grad_norm": 2.3165247500208332,
|
| 1081 |
+
"kl": 0.008834616397507489,
|
| 1082 |
+
"learning_rate": 6.154905383926216e-07,
|
| 1083 |
+
"loss": 0.00038796960143372416,
|
| 1084 |
+
"step": 71,
|
| 1085 |
+
"step_time": 73.1556283198297
|
| 1086 |
+
},
|
| 1087 |
+
{
|
| 1088 |
+
"clip_ratio/high_max": 0.010272878222167492,
|
| 1089 |
+
"clip_ratio/high_mean": 0.004277804546291009,
|
| 1090 |
+
"clip_ratio/low_mean": 0.0011350003878760617,
|
| 1091 |
+
"clip_ratio/low_min": 0.0,
|
| 1092 |
+
"clip_ratio/region_mean": 0.0054128048941493034,
|
| 1093 |
+
"epoch": 0.45,
|
| 1094 |
+
"grad_norm": 3.2223180646637095,
|
| 1095 |
+
"kl": 0.008534728782251477,
|
| 1096 |
+
"learning_rate": 6.056073681208037e-07,
|
| 1097 |
+
"loss": 0.0003769584000110626,
|
| 1098 |
+
"step": 72,
|
| 1099 |
+
"step_time": 73.03810304030776
|
| 1100 |
+
},
|
| 1101 |
+
{
|
| 1102 |
+
"clip_ratio/high_max": 0.0,
|
| 1103 |
+
"clip_ratio/high_mean": 0.0,
|
| 1104 |
+
"clip_ratio/low_mean": 0.0,
|
| 1105 |
+
"clip_ratio/low_min": 0.0,
|
| 1106 |
+
"clip_ratio/region_mean": 0.0,
|
| 1107 |
+
"completions/clipped_ratio": 0.33984375,
|
| 1108 |
+
"completions/max_length": 2048.0,
|
| 1109 |
+
"completions/mean_length": 1223.4375,
|
| 1110 |
+
"completions/min_length": 86.0,
|
| 1111 |
+
"epoch": 0.45625,
|
| 1112 |
+
"frac_reward_zero_std": 0.375,
|
| 1113 |
+
"grad_norm": 2.691355645393165,
|
| 1114 |
+
"kl": 0.014940242748707533,
|
| 1115 |
+
"learning_rate": 5.956808152365532e-07,
|
| 1116 |
+
"loss": 0.00014938956883270293,
|
| 1117 |
+
"reward": 0.3125,
|
| 1118 |
+
"reward_std": 0.2847120761871338,
|
| 1119 |
+
"rewards/MajorityAccuracy/mean": 0.3125,
|
| 1120 |
+
"rewards/MajorityAccuracy/std": 0.4644203782081604,
|
| 1121 |
+
"step": 73,
|
| 1122 |
+
"step_time": 516.3145757801831
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"clip_ratio/high_max": 0.002089770627208054,
|
| 1126 |
+
"clip_ratio/high_mean": 0.0005224426568020135,
|
| 1127 |
+
"clip_ratio/low_mean": 0.00014671361714135855,
|
| 1128 |
+
"clip_ratio/low_min": 0.0,
|
| 1129 |
+
"clip_ratio/region_mean": 0.0006691562739433721,
|
| 1130 |
+
"epoch": 0.4625,
|
| 1131 |
+
"grad_norm": 1.9042760017102665,
|
| 1132 |
+
"kl": 0.00857027096208185,
|
| 1133 |
+
"learning_rate": 5.857149574837268e-07,
|
| 1134 |
+
"loss": 0.0003301405522506684,
|
| 1135 |
+
"step": 74,
|
| 1136 |
+
"step_time": 75.19721806049347
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"clip_ratio/high_max": 0.001051479637681041,
|
| 1140 |
+
"clip_ratio/high_mean": 0.00030916620926291216,
|
| 1141 |
+
"clip_ratio/low_mean": 0.000525127074070042,
|
| 1142 |
+
"clip_ratio/low_min": 0.0,
|
| 1143 |
+
"clip_ratio/region_mean": 0.0008342932760569965,
|
| 1144 |
+
"epoch": 0.46875,
|
| 1145 |
+
"grad_norm": 1.255909951375418,
|
| 1146 |
+
"kl": 0.008368810696993023,
|
| 1147 |
+
"learning_rate": 5.757138887522883e-07,
|
| 1148 |
+
"loss": -5.229597809375264e-05,
|
| 1149 |
+
"step": 75,
|
| 1150 |
+
"step_time": 74.1100598089397
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"clip_ratio/high_max": 0.002612207259517163,
|
| 1154 |
+
"clip_ratio/high_mean": 0.0008699412646819837,
|
| 1155 |
+
"clip_ratio/low_mean": 0.0012997117810300551,
|
| 1156 |
+
"clip_ratio/low_min": 0.0,
|
| 1157 |
+
"clip_ratio/region_mean": 0.0021696530311601236,
|
| 1158 |
+
"epoch": 0.475,
|
| 1159 |
+
"grad_norm": 2.4053703072813764,
|
| 1160 |
+
"kl": 0.011279026162810624,
|
| 1161 |
+
"learning_rate": 5.656817173965732e-07,
|
| 1162 |
+
"loss": -3.375578671693802e-05,
|
| 1163 |
+
"step": 76,
|
| 1164 |
+
"step_time": 72.89710001647472
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"clip_ratio/high_max": 0.0009904781327350065,
|
| 1168 |
+
"clip_ratio/high_mean": 0.00034436876376275904,
|
| 1169 |
+
"clip_ratio/low_mean": 0.00018123549671145156,
|
| 1170 |
+
"clip_ratio/low_min": 0.0,
|
| 1171 |
+
"clip_ratio/region_mean": 0.0005256042604742106,
|
| 1172 |
+
"epoch": 0.48125,
|
| 1173 |
+
"grad_norm": 1.4915335281577575,
|
| 1174 |
+
"kl": 0.01310815301258117,
|
| 1175 |
+
"learning_rate": 5.556225645476118e-07,
|
| 1176 |
+
"loss": 0.0001548215514048934,
|
| 1177 |
+
"step": 77,
|
| 1178 |
+
"step_time": 73.68724466487765
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"clip_ratio/high_max": 0.00275398297526408,
|
| 1182 |
+
"clip_ratio/high_mean": 0.00068849574381602,
|
| 1183 |
+
"clip_ratio/low_mean": 0.0012751572648994625,
|
| 1184 |
+
"clip_ratio/low_min": 0.0,
|
| 1185 |
+
"clip_ratio/region_mean": 0.0019636530087154824,
|
| 1186 |
+
"epoch": 0.4875,
|
| 1187 |
+
"grad_norm": 2.283836602622986,
|
| 1188 |
+
"kl": 0.01163852191530168,
|
| 1189 |
+
"learning_rate": 5.455405624202032e-07,
|
| 1190 |
+
"loss": 0.0005704909563064575,
|
| 1191 |
+
"step": 78,
|
| 1192 |
+
"step_time": 75.18435535952449
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"clip_ratio/high_max": 0.007776132028084248,
|
| 1196 |
+
"clip_ratio/high_mean": 0.0028873406699858606,
|
| 1197 |
+
"clip_ratio/low_mean": 0.0008151664878823794,
|
| 1198 |
+
"clip_ratio/low_min": 0.0,
|
| 1199 |
+
"clip_ratio/region_mean": 0.003702507179696113,
|
| 1200 |
+
"epoch": 0.49375,
|
| 1201 |
+
"grad_norm": 2.2305996364328697,
|
| 1202 |
+
"kl": 0.013354019611142576,
|
| 1203 |
+
"learning_rate": 5.354398526154365e-07,
|
| 1204 |
+
"loss": 0.00026892032474279404,
|
| 1205 |
+
"step": 79,
|
| 1206 |
+
"step_time": 74.04837746545672
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"clip_ratio/high_max": 0.001811735681258142,
|
| 1210 |
+
"clip_ratio/high_mean": 0.0004529339203145355,
|
| 1211 |
+
"clip_ratio/low_mean": 0.0009240979088644963,
|
| 1212 |
+
"clip_ratio/low_min": 0.0,
|
| 1213 |
+
"clip_ratio/region_mean": 0.0013770318291790318,
|
| 1214 |
+
"epoch": 0.5,
|
| 1215 |
+
"grad_norm": 2.1600010560922387,
|
| 1216 |
+
"kl": 0.009653882938437164,
|
| 1217 |
+
"learning_rate": 5.253245844193564e-07,
|
| 1218 |
+
"loss": 8.321736822836101e-05,
|
| 1219 |
+
"step": 80,
|
| 1220 |
+
"step_time": 73.54326060041785
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"clip_ratio/high_max": 0.0,
|
| 1224 |
+
"clip_ratio/high_mean": 0.0,
|
| 1225 |
+
"clip_ratio/low_mean": 0.0,
|
| 1226 |
+
"clip_ratio/low_min": 0.0,
|
| 1227 |
+
"clip_ratio/region_mean": 0.0,
|
| 1228 |
+
"completions/clipped_ratio": 0.41015625,
|
| 1229 |
+
"completions/max_length": 2048.0,
|
| 1230 |
+
"completions/mean_length": 1437.50390625,
|
| 1231 |
+
"completions/min_length": 83.0,
|
| 1232 |
+
"epoch": 0.50625,
|
| 1233 |
+
"frac_reward_zero_std": 0.25,
|
| 1234 |
+
"grad_norm": 1.5395705637510826,
|
| 1235 |
+
"kl": 0.010538910748437047,
|
| 1236 |
+
"learning_rate": 5.151989130984714e-07,
|
| 1237 |
+
"loss": 0.00010537529306020588,
|
| 1238 |
+
"reward": 0.3984375,
|
| 1239 |
+
"reward_std": 0.33338692784309387,
|
| 1240 |
+
"rewards/MajorityAccuracy/mean": 0.3984375,
|
| 1241 |
+
"rewards/MajorityAccuracy/std": 0.4905354082584381,
|
| 1242 |
+
"step": 81,
|
| 1243 |
+
"step_time": 516.5280490703881
|
| 1244 |
+
},
|
| 1245 |
+
{
|
| 1246 |
+
"clip_ratio/high_max": 0.0038096612406661734,
|
| 1247 |
+
"clip_ratio/high_mean": 0.0012034254759782925,
|
| 1248 |
+
"clip_ratio/low_mean": 0.0006539701425936073,
|
| 1249 |
+
"clip_ratio/low_min": 0.0,
|
| 1250 |
+
"clip_ratio/region_mean": 0.0018573955821921118,
|
| 1251 |
+
"epoch": 0.5125,
|
| 1252 |
+
"grad_norm": 3.0705510338446027,
|
| 1253 |
+
"kl": 0.018357954220846295,
|
| 1254 |
+
"learning_rate": 5.050669981928055e-07,
|
| 1255 |
+
"loss": 0.0005182726308703423,
|
| 1256 |
+
"step": 82,
|
| 1257 |
+
"step_time": 72.81564075127244
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"clip_ratio/high_max": 0.0009407761099282652,
|
| 1261 |
+
"clip_ratio/high_mean": 0.00032062282116385177,
|
| 1262 |
+
"clip_ratio/low_mean": 0.00022988911223364994,
|
| 1263 |
+
"clip_ratio/low_min": 0.0,
|
| 1264 |
+
"clip_ratio/region_mean": 0.0005505119333975017,
|
| 1265 |
+
"epoch": 0.51875,
|
| 1266 |
+
"grad_norm": 1.4355676671464088,
|
| 1267 |
+
"kl": 0.007934427063446492,
|
| 1268 |
+
"learning_rate": 4.949330018071946e-07,
|
| 1269 |
+
"loss": 0.00017895565542858094,
|
| 1270 |
+
"step": 83,
|
| 1271 |
+
"step_time": 73.94411617144942
|
| 1272 |
+
},
|
| 1273 |
+
{
|
| 1274 |
+
"clip_ratio/high_max": 0.00118467936408706,
|
| 1275 |
+
"clip_ratio/high_mean": 0.000296169841021765,
|
| 1276 |
+
"clip_ratio/low_mean": 0.00018310546875,
|
| 1277 |
+
"clip_ratio/low_min": 0.0,
|
| 1278 |
+
"clip_ratio/region_mean": 0.000479275309771765,
|
| 1279 |
+
"epoch": 0.525,
|
| 1280 |
+
"grad_norm": 1.1375574363920582,
|
| 1281 |
+
"kl": 0.01179729017894715,
|
| 1282 |
+
"learning_rate": 4.848010869015287e-07,
|
| 1283 |
+
"loss": 4.941665247315541e-05,
|
| 1284 |
+
"step": 84,
|
| 1285 |
+
"step_time": 74.93465014919639
|
| 1286 |
+
},
|
| 1287 |
+
{
|
| 1288 |
+
"clip_ratio/high_max": 0.002245798532385379,
|
| 1289 |
+
"clip_ratio/high_mean": 0.0006518479494843632,
|
| 1290 |
+
"clip_ratio/low_mean": 0.001090767553250771,
|
| 1291 |
+
"clip_ratio/low_min": 0.0,
|
| 1292 |
+
"clip_ratio/region_mean": 0.0017426155027351342,
|
| 1293 |
+
"epoch": 0.53125,
|
| 1294 |
+
"grad_norm": 3.0955020140117657,
|
| 1295 |
+
"kl": 0.015114621259272099,
|
| 1296 |
+
"learning_rate": 4.746754155806437e-07,
|
| 1297 |
+
"loss": -8.948786125984043e-05,
|
| 1298 |
+
"step": 85,
|
| 1299 |
+
"step_time": 74.94300713762641
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"clip_ratio/high_max": 0.002869865405955352,
|
| 1303 |
+
"clip_ratio/high_mean": 0.0013302585248311516,
|
| 1304 |
+
"clip_ratio/low_mean": 0.0006994259892962873,
|
| 1305 |
+
"clip_ratio/low_min": 6.753106572432443e-05,
|
| 1306 |
+
"clip_ratio/region_mean": 0.0020296845177654177,
|
| 1307 |
+
"epoch": 0.5375,
|
| 1308 |
+
"grad_norm": 1.5942721352456317,
|
| 1309 |
+
"kl": 0.010655982187017798,
|
| 1310 |
+
"learning_rate": 4.645601473845635e-07,
|
| 1311 |
+
"loss": 0.0003238450735807419,
|
| 1312 |
+
"step": 86,
|
| 1313 |
+
"step_time": 74.21203067526221
|
| 1314 |
+
},
|
| 1315 |
+
{
|
| 1316 |
+
"clip_ratio/high_max": 0.0005551733338506892,
|
| 1317 |
+
"clip_ratio/high_mean": 0.0001387933334626723,
|
| 1318 |
+
"clip_ratio/low_mean": 0.000910097238374874,
|
| 1319 |
+
"clip_ratio/low_min": 0.0,
|
| 1320 |
+
"clip_ratio/region_mean": 0.0010488905681995675,
|
| 1321 |
+
"epoch": 0.54375,
|
| 1322 |
+
"grad_norm": 1.014957276287081,
|
| 1323 |
+
"kl": 0.010844826814718544,
|
| 1324 |
+
"learning_rate": 4.544594375797968e-07,
|
| 1325 |
+
"loss": 0.00027926662005484104,
|
| 1326 |
+
"step": 87,
|
| 1327 |
+
"step_time": 74.60808426886797
|
| 1328 |
+
},
|
| 1329 |
+
{
|
| 1330 |
+
"clip_ratio/high_max": 0.0031506004597758874,
|
| 1331 |
+
"clip_ratio/high_mean": 0.0009538097619952168,
|
| 1332 |
+
"clip_ratio/low_mean": 0.0008956710298662074,
|
| 1333 |
+
"clip_ratio/low_min": 0.0,
|
| 1334 |
+
"clip_ratio/region_mean": 0.001849480831879191,
|
| 1335 |
+
"epoch": 0.55,
|
| 1336 |
+
"grad_norm": 2.237298283730702,
|
| 1337 |
+
"kl": 0.012738962890580297,
|
| 1338 |
+
"learning_rate": 4.443774354523882e-07,
|
| 1339 |
+
"loss": 0.0005584406317211688,
|
| 1340 |
+
"step": 88,
|
| 1341 |
+
"step_time": 73.30752529576421
|
| 1342 |
+
},
|
| 1343 |
+
{
|
| 1344 |
+
"clip_ratio/high_max": 0.0,
|
| 1345 |
+
"clip_ratio/high_mean": 0.0,
|
| 1346 |
+
"clip_ratio/low_mean": 0.0,
|
| 1347 |
+
"clip_ratio/low_min": 0.0,
|
| 1348 |
+
"clip_ratio/region_mean": 0.0,
|
| 1349 |
+
"completions/clipped_ratio": 0.41015625,
|
| 1350 |
+
"completions/max_length": 2048.0,
|
| 1351 |
+
"completions/mean_length": 1334.34765625,
|
| 1352 |
+
"completions/min_length": 87.0,
|
| 1353 |
+
"epoch": 0.55625,
|
| 1354 |
+
"frac_reward_zero_std": 0.375,
|
| 1355 |
+
"grad_norm": 2.1514553815284643,
|
| 1356 |
+
"kl": 0.014167973422445357,
|
| 1357 |
+
"learning_rate": 4.3431828260342675e-07,
|
| 1358 |
+
"loss": 0.00014173518866300583,
|
| 1359 |
+
"reward": 0.41796875,
|
| 1360 |
+
"reward_std": 0.26645296812057495,
|
| 1361 |
+
"rewards/MajorityAccuracy/mean": 0.41796875,
|
| 1362 |
+
"rewards/MajorityAccuracy/std": 0.49419113993644714,
|
| 1363 |
+
"step": 89,
|
| 1364 |
+
"step_time": 512.9642349258065
|
| 1365 |
+
},
|
| 1366 |
+
{
|
| 1367 |
+
"clip_ratio/high_max": 0.000117481205961667,
|
| 1368 |
+
"clip_ratio/high_mean": 2.937030149041675e-05,
|
| 1369 |
+
"clip_ratio/low_mean": 0.00036854849895462394,
|
| 1370 |
+
"clip_ratio/low_min": 0.0,
|
| 1371 |
+
"clip_ratio/region_mean": 0.0003979188040830195,
|
| 1372 |
+
"epoch": 0.5625,
|
| 1373 |
+
"grad_norm": 1.7054495336646436,
|
| 1374 |
+
"kl": 0.012596273212693632,
|
| 1375 |
+
"learning_rate": 4.242861112477118e-07,
|
| 1376 |
+
"loss": 0.00047689516213722527,
|
| 1377 |
+
"step": 90,
|
| 1378 |
+
"step_time": 74.41515237465501
|
| 1379 |
+
},
|
| 1380 |
+
{
|
| 1381 |
+
"clip_ratio/high_max": 0.0048589422658551484,
|
| 1382 |
+
"clip_ratio/high_mean": 0.0014133229851722717,
|
| 1383 |
+
"clip_ratio/low_mean": 0.00013657853196491487,
|
| 1384 |
+
"clip_ratio/low_min": 0.0,
|
| 1385 |
+
"clip_ratio/region_mean": 0.0015499015171371866,
|
| 1386 |
+
"epoch": 0.56875,
|
| 1387 |
+
"grad_norm": 1.9824187367051838,
|
| 1388 |
+
"kl": 0.010325115057639778,
|
| 1389 |
+
"learning_rate": 4.1428504251627325e-07,
|
| 1390 |
+
"loss": 0.00014215378905646503,
|
| 1391 |
+
"step": 91,
|
| 1392 |
+
"step_time": 75.99112362414598
|
| 1393 |
+
},
|
| 1394 |
+
{
|
| 1395 |
+
"clip_ratio/high_max": 0.0008261156617663801,
|
| 1396 |
+
"clip_ratio/high_mean": 0.00020652891544159502,
|
| 1397 |
+
"clip_ratio/low_mean": 0.00027394998323870823,
|
| 1398 |
+
"clip_ratio/low_min": 0.0,
|
| 1399 |
+
"clip_ratio/region_mean": 0.00048047889140434563,
|
| 1400 |
+
"epoch": 0.575,
|
| 1401 |
+
"grad_norm": 1.2531171222109287,
|
| 1402 |
+
"kl": 0.009100776223931462,
|
| 1403 |
+
"learning_rate": 4.0431918476344685e-07,
|
| 1404 |
+
"loss": -4.722527228295803e-05,
|
| 1405 |
+
"step": 92,
|
| 1406 |
+
"step_time": 73.713692933321
|
| 1407 |
+
},
|
| 1408 |
+
{
|
| 1409 |
+
"clip_ratio/high_max": 0.005234318057773635,
|
| 1410 |
+
"clip_ratio/high_mean": 0.0020389309065649286,
|
| 1411 |
+
"clip_ratio/low_mean": 0.00036352900133351795,
|
| 1412 |
+
"clip_ratio/low_min": 0.0,
|
| 1413 |
+
"clip_ratio/region_mean": 0.0024024599260883406,
|
| 1414 |
+
"epoch": 0.58125,
|
| 1415 |
+
"grad_norm": 2.0412351401805213,
|
| 1416 |
+
"kl": 0.011358004878275096,
|
| 1417 |
+
"learning_rate": 3.943926318791963e-07,
|
| 1418 |
+
"loss": 0.00033540837466716766,
|
| 1419 |
+
"step": 93,
|
| 1420 |
+
"step_time": 73.29895859211683
|
| 1421 |
+
},
|
| 1422 |
+
{
|
| 1423 |
+
"clip_ratio/high_max": 0.0009145967342192307,
|
| 1424 |
+
"clip_ratio/high_mean": 0.00022864918355480768,
|
| 1425 |
+
"clip_ratio/low_mean": 0.0007005560182733461,
|
| 1426 |
+
"clip_ratio/low_min": 0.0,
|
| 1427 |
+
"clip_ratio/region_mean": 0.0009292052054661326,
|
| 1428 |
+
"epoch": 0.5875,
|
| 1429 |
+
"grad_norm": 1.3916304861651063,
|
| 1430 |
+
"kl": 0.01160808268468827,
|
| 1431 |
+
"learning_rate": 3.845094616073783e-07,
|
| 1432 |
+
"loss": 0.00023558946850243956,
|
| 1433 |
+
"step": 94,
|
| 1434 |
+
"step_time": 74.17752105370164
|
| 1435 |
+
},
|
| 1436 |
+
{
|
| 1437 |
+
"clip_ratio/high_max": 0.00364518872811459,
|
| 1438 |
+
"clip_ratio/high_mean": 0.0012264611141290516,
|
| 1439 |
+
"clip_ratio/low_mean": 0.0005248154993751086,
|
| 1440 |
+
"clip_ratio/low_min": 0.0,
|
| 1441 |
+
"clip_ratio/region_mean": 0.0017512766062282026,
|
| 1442 |
+
"epoch": 0.59375,
|
| 1443 |
+
"grad_norm": 1.7613781473724905,
|
| 1444 |
+
"kl": 0.01445352123118937,
|
| 1445 |
+
"learning_rate": 3.7467373387063964e-07,
|
| 1446 |
+
"loss": 0.00011201974120922387,
|
| 1447 |
+
"step": 95,
|
| 1448 |
+
"step_time": 75.35093042999506
|
| 1449 |
+
},
|
| 1450 |
+
{
|
| 1451 |
+
"clip_ratio/high_max": 0.007222329586511478,
|
| 1452 |
+
"clip_ratio/high_mean": 0.002221020360593684,
|
| 1453 |
+
"clip_ratio/low_mean": 0.0005088787875138223,
|
| 1454 |
+
"clip_ratio/low_min": 0.0,
|
| 1455 |
+
"clip_ratio/region_mean": 0.002729899133555591,
|
| 1456 |
+
"epoch": 0.6,
|
| 1457 |
+
"grad_norm": 1.8805397595416136,
|
| 1458 |
+
"kl": 0.013379543088376522,
|
| 1459 |
+
"learning_rate": 3.648894891026358e-07,
|
| 1460 |
+
"loss": 0.00010419571481179446,
|
| 1461 |
+
"step": 96,
|
| 1462 |
+
"step_time": 75.34189929440618
|
| 1463 |
+
},
|
| 1464 |
+
{
|
| 1465 |
+
"clip_ratio/high_max": 0.0,
|
| 1466 |
+
"clip_ratio/high_mean": 0.0,
|
| 1467 |
+
"clip_ratio/low_mean": 0.0,
|
| 1468 |
+
"clip_ratio/low_min": 0.0,
|
| 1469 |
+
"clip_ratio/region_mean": 0.0,
|
| 1470 |
+
"completions/clipped_ratio": 0.38671875,
|
| 1471 |
+
"completions/max_length": 2048.0,
|
| 1472 |
+
"completions/mean_length": 1394.359375,
|
| 1473 |
+
"completions/min_length": 87.0,
|
| 1474 |
+
"epoch": 0.60625,
|
| 1475 |
+
"frac_reward_zero_std": 0.25,
|
| 1476 |
+
"grad_norm": 1.144513999930649,
|
| 1477 |
+
"kl": 0.00871997233480215,
|
| 1478 |
+
"learning_rate": 3.5516074658825397e-07,
|
| 1479 |
+
"loss": 8.716429874766618e-05,
|
| 1480 |
+
"reward": 0.36328125,
|
| 1481 |
+
"reward_std": 0.33114659786224365,
|
| 1482 |
+
"rewards/MajorityAccuracy/mean": 0.36328125,
|
| 1483 |
+
"rewards/MajorityAccuracy/std": 0.48188698291778564,
|
| 1484 |
+
"step": 97,
|
| 1485 |
+
"step_time": 517.1117947772145
|
| 1486 |
+
},
|
| 1487 |
+
{
|
| 1488 |
+
"clip_ratio/high_max": 0.0016795597184682265,
|
| 1489 |
+
"clip_ratio/high_mean": 0.0005400027985160705,
|
| 1490 |
+
"clip_ratio/low_mean": 0.00019572357450670097,
|
| 1491 |
+
"clip_ratio/low_min": 0.0,
|
| 1492 |
+
"clip_ratio/region_mean": 0.0007357263675658032,
|
| 1493 |
+
"epoch": 0.6125,
|
| 1494 |
+
"grad_norm": 1.3419738181514245,
|
| 1495 |
+
"kl": 0.010603868635371327,
|
| 1496 |
+
"learning_rate": 3.454915028125263e-07,
|
| 1497 |
+
"loss": 0.0001621129922568798,
|
| 1498 |
+
"step": 98,
|
| 1499 |
+
"step_time": 74.19324538111687
|
| 1500 |
+
},
|
| 1501 |
+
{
|
| 1502 |
+
"clip_ratio/high_max": 0.0019879528699675575,
|
| 1503 |
+
"clip_ratio/high_mean": 0.000587672293477226,
|
| 1504 |
+
"clip_ratio/low_mean": 0.0002593994140625,
|
| 1505 |
+
"clip_ratio/low_min": 0.0,
|
| 1506 |
+
"clip_ratio/region_mean": 0.0008470717002637684,
|
| 1507 |
+
"epoch": 0.61875,
|
| 1508 |
+
"grad_norm": 1.3150972882327816,
|
| 1509 |
+
"kl": 0.008666269364766777,
|
| 1510 |
+
"learning_rate": 3.3588572981890684e-07,
|
| 1511 |
+
"loss": -7.207151793409139e-05,
|
| 1512 |
+
"step": 99,
|
| 1513 |
+
"step_time": 73.58873805776238
|
| 1514 |
+
},
|
| 1515 |
+
{
|
| 1516 |
+
"clip_ratio/high_max": 0.004101540456758812,
|
| 1517 |
+
"clip_ratio/high_mean": 0.0017278023588005453,
|
| 1518 |
+
"clip_ratio/low_mean": 0.0006920996092958376,
|
| 1519 |
+
"clip_ratio/low_min": 0.0,
|
| 1520 |
+
"clip_ratio/region_mean": 0.0024199019535444677,
|
| 1521 |
+
"epoch": 0.625,
|
| 1522 |
+
"grad_norm": 2.4274128145707095,
|
| 1523 |
+
"kl": 0.014385863905772567,
|
| 1524 |
+
"learning_rate": 3.263473735775899e-07,
|
| 1525 |
+
"loss": -4.488509148359299e-05,
|
| 1526 |
+
"step": 100,
|
| 1527 |
+
"step_time": 72.66440482065082
|
| 1528 |
+
},
|
| 1529 |
+
{
|
| 1530 |
+
"clip_ratio/high_max": 0.0025767695769900456,
|
| 1531 |
+
"clip_ratio/high_mean": 0.0006441923942475114,
|
| 1532 |
+
"clip_ratio/low_mean": 0.00010471813584445044,
|
| 1533 |
+
"clip_ratio/low_min": 0.0,
|
| 1534 |
+
"clip_ratio/region_mean": 0.0007489105300919618,
|
| 1535 |
+
"epoch": 0.63125,
|
| 1536 |
+
"grad_norm": 2.447058871112701,
|
| 1537 |
+
"kl": 0.011274750111624599,
|
| 1538 |
+
"learning_rate": 3.1688035236453865e-07,
|
| 1539 |
+
"loss": 0.00022696080850437284,
|
| 1540 |
+
"step": 101,
|
| 1541 |
+
"step_time": 72.97443969920278
|
| 1542 |
+
},
|
| 1543 |
+
{
|
| 1544 |
+
"clip_ratio/high_max": 0.003905233956174925,
|
| 1545 |
+
"clip_ratio/high_mean": 0.0011746129675884731,
|
| 1546 |
+
"clip_ratio/low_mean": 0.0005347017868189141,
|
| 1547 |
+
"clip_ratio/low_min": 0.0,
|
| 1548 |
+
"clip_ratio/region_mean": 0.0017093147107516415,
|
| 1549 |
+
"epoch": 0.6375,
|
| 1550 |
+
"grad_norm": 2.199235509612732,
|
| 1551 |
+
"kl": 0.013855923665687442,
|
| 1552 |
+
"learning_rate": 3.0748855515189096e-07,
|
| 1553 |
+
"loss": 0.00022735632956027985,
|
| 1554 |
+
"step": 102,
|
| 1555 |
+
"step_time": 74.68020544201136
|
| 1556 |
+
},
|
| 1557 |
+
{
|
| 1558 |
+
"clip_ratio/high_max": 0.0024828914465615526,
|
| 1559 |
+
"clip_ratio/high_mean": 0.0006835623244114686,
|
| 1560 |
+
"clip_ratio/low_mean": 0.000274658203125,
|
| 1561 |
+
"clip_ratio/low_min": 0.0,
|
| 1562 |
+
"clip_ratio/region_mean": 0.0009582205311744474,
|
| 1563 |
+
"epoch": 0.64375,
|
| 1564 |
+
"grad_norm": 1.4244120126702584,
|
| 1565 |
+
"kl": 0.008463046222459525,
|
| 1566 |
+
"learning_rate": 2.981758400104028e-07,
|
| 1567 |
+
"loss": 1.590247848071158e-06,
|
| 1568 |
+
"step": 103,
|
| 1569 |
+
"step_time": 74.17703925073147
|
| 1570 |
+
},
|
| 1571 |
+
{
|
| 1572 |
+
"clip_ratio/high_max": 0.006886150280479342,
|
| 1573 |
+
"clip_ratio/high_mean": 0.001758280661306344,
|
| 1574 |
+
"clip_ratio/low_mean": 0.0016090536446426995,
|
| 1575 |
+
"clip_ratio/low_min": 0.0,
|
| 1576 |
+
"clip_ratio/region_mean": 0.003367334298673086,
|
| 1577 |
+
"epoch": 0.65,
|
| 1578 |
+
"grad_norm": 3.8368428205117104,
|
| 1579 |
+
"kl": 0.016008354257792234,
|
| 1580 |
+
"learning_rate": 2.8894603252458403e-07,
|
| 1581 |
+
"loss": 0.00046537647722288966,
|
| 1582 |
+
"step": 104,
|
| 1583 |
+
"step_time": 74.05075707286596
|
| 1584 |
+
},
|
| 1585 |
+
{
|
| 1586 |
+
"clip_ratio/high_max": 0.0,
|
| 1587 |
+
"clip_ratio/high_mean": 0.0,
|
| 1588 |
+
"clip_ratio/low_mean": 0.0,
|
| 1589 |
+
"clip_ratio/low_min": 0.0,
|
| 1590 |
+
"clip_ratio/region_mean": 0.0,
|
| 1591 |
+
"completions/clipped_ratio": 0.41015625,
|
| 1592 |
+
"completions/max_length": 2048.0,
|
| 1593 |
+
"completions/mean_length": 1335.65625,
|
| 1594 |
+
"completions/min_length": 99.0,
|
| 1595 |
+
"epoch": 0.65625,
|
| 1596 |
+
"frac_reward_zero_std": 0.375,
|
| 1597 |
+
"grad_norm": 1.220803662592034,
|
| 1598 |
+
"kl": 0.011086618062108755,
|
| 1599 |
+
"learning_rate": 2.7980292422118277e-07,
|
| 1600 |
+
"loss": 0.00011092099157394841,
|
| 1601 |
+
"reward": 0.3359375,
|
| 1602 |
+
"reward_std": 0.28694993257522583,
|
| 1603 |
+
"rewards/MajorityAccuracy/mean": 0.3359375,
|
| 1604 |
+
"rewards/MajorityAccuracy/std": 0.4732423424720764,
|
| 1605 |
+
"step": 105,
|
| 1606 |
+
"step_time": 514.5763935446739
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"clip_ratio/high_max": 0.0015038371057016775,
|
| 1610 |
+
"clip_ratio/high_mean": 0.00041272397720604204,
|
| 1611 |
+
"clip_ratio/low_mean": 0.00037203674582997337,
|
| 1612 |
+
"clip_ratio/low_min": 0.0,
|
| 1613 |
+
"clip_ratio/region_mean": 0.0007847607230360154,
|
| 1614 |
+
"epoch": 0.6625,
|
| 1615 |
+
"grad_norm": 2.352785723301752,
|
| 1616 |
+
"kl": 0.015654265182092786,
|
| 1617 |
+
"learning_rate": 2.70750271011657e-07,
|
| 1618 |
+
"loss": 0.00047953726607374847,
|
| 1619 |
+
"step": 106,
|
| 1620 |
+
"step_time": 73.12891332432628
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"clip_ratio/high_max": 0.0005753696386818774,
|
| 1624 |
+
"clip_ratio/high_mean": 0.00014384240967046935,
|
| 1625 |
+
"clip_ratio/low_mean": 0.0002481974079273641,
|
| 1626 |
+
"clip_ratio/low_min": 0.0,
|
| 1627 |
+
"clip_ratio/region_mean": 0.00039203981941682287,
|
| 1628 |
+
"epoch": 0.66875,
|
| 1629 |
+
"grad_norm": 1.7562269389405571,
|
| 1630 |
+
"kl": 0.011776745086535811,
|
| 1631 |
+
"learning_rate": 2.6179179164927754e-07,
|
| 1632 |
+
"loss": 0.0001632406929275021,
|
| 1633 |
+
"step": 107,
|
| 1634 |
+
"step_time": 73.30582751706243
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"clip_ratio/high_max": 0.0007596563955303282,
|
| 1638 |
+
"clip_ratio/high_mean": 0.00018991409888258204,
|
| 1639 |
+
"clip_ratio/low_mean": 0.0004135703929932788,
|
| 1640 |
+
"clip_ratio/low_min": 0.0,
|
| 1641 |
+
"clip_ratio/region_mean": 0.0006034844918758608,
|
| 1642 |
+
"epoch": 0.675,
|
| 1643 |
+
"grad_norm": 1.5112397358897378,
|
| 1644 |
+
"kl": 0.010366878472268581,
|
| 1645 |
+
"learning_rate": 2.529311662014972e-07,
|
| 1646 |
+
"loss": 0.00013845987268723547,
|
| 1647 |
+
"step": 108,
|
| 1648 |
+
"step_time": 75.03023574873805
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"clip_ratio/high_max": 0.0018026804536930285,
|
| 1652 |
+
"clip_ratio/high_mean": 0.0005415131436166121,
|
| 1653 |
+
"clip_ratio/low_mean": 0.0003935994755011052,
|
| 1654 |
+
"clip_ratio/low_min": 0.0,
|
| 1655 |
+
"clip_ratio/region_mean": 0.0009351125900138868,
|
| 1656 |
+
"epoch": 0.68125,
|
| 1657 |
+
"grad_norm": 1.8636733911233074,
|
| 1658 |
+
"kl": 0.012884767376817763,
|
| 1659 |
+
"learning_rate": 2.441720345382089e-07,
|
| 1660 |
+
"loss": 0.00026943825650960207,
|
| 1661 |
+
"step": 109,
|
| 1662 |
+
"step_time": 73.79918487742543
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"clip_ratio/high_max": 0.0015904313841019757,
|
| 1666 |
+
"clip_ratio/high_mean": 0.00042812542415049393,
|
| 1667 |
+
"clip_ratio/low_mean": 0.0004523911193246022,
|
| 1668 |
+
"clip_ratio/low_min": 0.0,
|
| 1669 |
+
"clip_ratio/region_mean": 0.0008805165434750961,
|
| 1670 |
+
"epoch": 0.6875,
|
| 1671 |
+
"grad_norm": 1.9066419158874415,
|
| 1672 |
+
"kl": 0.016152666648849845,
|
| 1673 |
+
"learning_rate": 2.355179948365189e-07,
|
| 1674 |
+
"loss": 0.0003760265826713294,
|
| 1675 |
+
"step": 110,
|
| 1676 |
+
"step_time": 74.65123998746276
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"clip_ratio/high_max": 0.0021794729109387845,
|
| 1680 |
+
"clip_ratio/high_mean": 0.0006537532390211709,
|
| 1681 |
+
"clip_ratio/low_mean": 0.00027097413840238005,
|
| 1682 |
+
"clip_ratio/low_min": 0.0,
|
| 1683 |
+
"clip_ratio/region_mean": 0.0009247273774235509,
|
| 1684 |
+
"epoch": 0.69375,
|
| 1685 |
+
"grad_norm": 1.5942121998170051,
|
| 1686 |
+
"kl": 0.016347182681784034,
|
| 1687 |
+
"learning_rate": 2.26972602102645e-07,
|
| 1688 |
+
"loss": 6.41259248368442e-05,
|
| 1689 |
+
"step": 111,
|
| 1690 |
+
"step_time": 74.39852786064148
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"clip_ratio/high_max": 0.0017805842217057943,
|
| 1694 |
+
"clip_ratio/high_mean": 0.0006332865996228065,
|
| 1695 |
+
"clip_ratio/low_mean": 0.0002116532741638366,
|
| 1696 |
+
"clip_ratio/low_min": 0.0,
|
| 1697 |
+
"clip_ratio/region_mean": 0.0008449398628727067,
|
| 1698 |
+
"epoch": 0.7,
|
| 1699 |
+
"grad_norm": 1.8452608620889128,
|
| 1700 |
+
"kl": 0.01217535138130188,
|
| 1701 |
+
"learning_rate": 2.1853936671155127e-07,
|
| 1702 |
+
"loss": 0.00010027207463281229,
|
| 1703 |
+
"step": 112,
|
| 1704 |
+
"step_time": 73.00480085611343
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"clip_ratio/high_max": 0.0,
|
| 1708 |
+
"clip_ratio/high_mean": 0.0,
|
| 1709 |
+
"clip_ratio/low_mean": 0.0,
|
| 1710 |
+
"clip_ratio/low_min": 0.0,
|
| 1711 |
+
"clip_ratio/region_mean": 0.0,
|
| 1712 |
+
"completions/clipped_ratio": 0.4453125,
|
| 1713 |
+
"completions/max_length": 2048.0,
|
| 1714 |
+
"completions/mean_length": 1456.93359375,
|
| 1715 |
+
"completions/min_length": 98.0,
|
| 1716 |
+
"epoch": 0.70625,
|
| 1717 |
+
"frac_reward_zero_std": 0.125,
|
| 1718 |
+
"grad_norm": 3.3450078068390594,
|
| 1719 |
+
"kl": 0.01337020902428776,
|
| 1720 |
+
"learning_rate": 2.1022175296491512e-07,
|
| 1721 |
+
"loss": 0.0001337010762654245,
|
| 1722 |
+
"reward": 0.4375,
|
| 1723 |
+
"reward_std": 0.39914143085479736,
|
| 1724 |
+
"rewards/MajorityAccuracy/mean": 0.4375,
|
| 1725 |
+
"rewards/MajorityAccuracy/std": 0.49705013632774353,
|
| 1726 |
+
"step": 113,
|
| 1727 |
+
"step_time": 516.7641914002597
|
| 1728 |
+
},
|
| 1729 |
+
{
|
| 1730 |
+
"clip_ratio/high_max": 0.0013083629964967258,
|
| 1731 |
+
"clip_ratio/high_mean": 0.00032709074912418146,
|
| 1732 |
+
"clip_ratio/low_mean": 0.000268772870185785,
|
| 1733 |
+
"clip_ratio/low_min": 0.0,
|
| 1734 |
+
"clip_ratio/region_mean": 0.000595863617490977,
|
| 1735 |
+
"epoch": 0.7125,
|
| 1736 |
+
"grad_norm": 1.7456095616822471,
|
| 1737 |
+
"kl": 0.011847492074593902,
|
| 1738 |
+
"learning_rate": 2.0202317766802152e-07,
|
| 1739 |
+
"loss": -1.6437421436421573e-05,
|
| 1740 |
+
"step": 114,
|
| 1741 |
+
"step_time": 73.7632961422205
|
| 1742 |
+
},
|
| 1743 |
+
{
|
| 1744 |
+
"clip_ratio/high_max": 0.003050226587220095,
|
| 1745 |
+
"clip_ratio/high_mean": 0.0009138742389040999,
|
| 1746 |
+
"clip_ratio/low_mean": 0.0003643500676844269,
|
| 1747 |
+
"clip_ratio/low_min": 0.0,
|
| 1748 |
+
"clip_ratio/region_mean": 0.001278224284760654,
|
| 1749 |
+
"epoch": 0.71875,
|
| 1750 |
+
"grad_norm": 2.8210950148631913,
|
| 1751 |
+
"kl": 0.011898136581294239,
|
| 1752 |
+
"learning_rate": 1.9394700872616853e-07,
|
| 1753 |
+
"loss": 0.00019844160124193877,
|
| 1754 |
+
"step": 115,
|
| 1755 |
+
"step_time": 75.0574233084917
|
| 1756 |
+
},
|
| 1757 |
+
{
|
| 1758 |
+
"clip_ratio/high_max": 0.0042421633552294225,
|
| 1759 |
+
"clip_ratio/high_mean": 0.0013892828865209594,
|
| 1760 |
+
"clip_ratio/low_mean": 0.00020345939265098423,
|
| 1761 |
+
"clip_ratio/low_min": 0.0,
|
| 1762 |
+
"clip_ratio/region_mean": 0.001592742293723859,
|
| 1763 |
+
"epoch": 0.725,
|
| 1764 |
+
"grad_norm": 1.780746688052519,
|
| 1765 |
+
"kl": 0.011460053967311978,
|
| 1766 |
+
"learning_rate": 1.8599656376116024e-07,
|
| 1767 |
+
"loss": 0.000269891694188118,
|
| 1768 |
+
"step": 116,
|
| 1769 |
+
"step_time": 73.74287265911698
|
| 1770 |
+
},
|
| 1771 |
+
{
|
| 1772 |
+
"clip_ratio/high_max": 0.002519528701668605,
|
| 1773 |
+
"clip_ratio/high_mean": 0.0007462826251867227,
|
| 1774 |
+
"clip_ratio/low_mean": 0.0005035943759139627,
|
| 1775 |
+
"clip_ratio/low_min": 0.0,
|
| 1776 |
+
"clip_ratio/region_mean": 0.0012498770156526007,
|
| 1777 |
+
"epoch": 0.73125,
|
| 1778 |
+
"grad_norm": 2.0171807677667535,
|
| 1779 |
+
"kl": 0.010149350855499506,
|
| 1780 |
+
"learning_rate": 1.7817510874845582e-07,
|
| 1781 |
+
"loss": 0.00022140517830848694,
|
| 1782 |
+
"step": 117,
|
| 1783 |
+
"step_time": 74.93591320887208
|
| 1784 |
+
},
|
| 1785 |
+
{
|
| 1786 |
+
"clip_ratio/high_max": 0.0027540519149624743,
|
| 1787 |
+
"clip_ratio/high_mean": 0.001106336234443006,
|
| 1788 |
+
"clip_ratio/low_mean": 0.00043273763731122017,
|
| 1789 |
+
"clip_ratio/low_min": 0.0,
|
| 1790 |
+
"clip_ratio/region_mean": 0.0015390738553833216,
|
| 1791 |
+
"epoch": 0.7375,
|
| 1792 |
+
"grad_norm": 1.629515132954119,
|
| 1793 |
+
"kl": 0.010879330220632255,
|
| 1794 |
+
"learning_rate": 1.7048585667553412e-07,
|
| 1795 |
+
"loss": 0.0002068839967250824,
|
| 1796 |
+
"step": 118,
|
| 1797 |
+
"step_time": 74.2498884536326
|
| 1798 |
+
},
|
| 1799 |
+
{
|
| 1800 |
+
"clip_ratio/high_max": 0.0011986348035861738,
|
| 1801 |
+
"clip_ratio/high_mean": 0.00037898078335274477,
|
| 1802 |
+
"clip_ratio/low_mean": 0.0009493520665273536,
|
| 1803 |
+
"clip_ratio/low_min": 0.00018310546875,
|
| 1804 |
+
"clip_ratio/region_mean": 0.0013283328589750454,
|
| 1805 |
+
"epoch": 0.74375,
|
| 1806 |
+
"grad_norm": 1.7673747876103736,
|
| 1807 |
+
"kl": 0.011658430797979236,
|
| 1808 |
+
"learning_rate": 1.6293196622202632e-07,
|
| 1809 |
+
"loss": 8.380413055419922e-05,
|
| 1810 |
+
"step": 119,
|
| 1811 |
+
"step_time": 73.73457884415984
|
| 1812 |
+
},
|
| 1813 |
+
{
|
| 1814 |
+
"clip_ratio/high_max": 0.0017452376487199217,
|
| 1815 |
+
"clip_ratio/high_mean": 0.0005964017254882492,
|
| 1816 |
+
"clip_ratio/low_mean": 0.0006427646949305199,
|
| 1817 |
+
"clip_ratio/low_min": 6.103515625e-05,
|
| 1818 |
+
"clip_ratio/region_mean": 0.0012391664276947267,
|
| 1819 |
+
"epoch": 0.75,
|
| 1820 |
+
"grad_norm": 2.0417681382819484,
|
| 1821 |
+
"kl": 0.011751449899747968,
|
| 1822 |
+
"learning_rate": 1.555165404621567e-07,
|
| 1823 |
+
"loss": 0.00014110654592514038,
|
| 1824 |
+
"step": 120,
|
| 1825 |
+
"step_time": 74.40912755206227
|
| 1826 |
+
},
|
| 1827 |
+
{
|
| 1828 |
+
"clip_ratio/high_max": 0.0,
|
| 1829 |
+
"clip_ratio/high_mean": 0.0,
|
| 1830 |
+
"clip_ratio/low_mean": 0.0,
|
| 1831 |
+
"clip_ratio/low_min": 0.0,
|
| 1832 |
+
"clip_ratio/region_mean": 0.0,
|
| 1833 |
+
"completions/clipped_ratio": 0.34765625,
|
| 1834 |
+
"completions/max_length": 2048.0,
|
| 1835 |
+
"completions/mean_length": 1317.078125,
|
| 1836 |
+
"completions/min_length": 145.0,
|
| 1837 |
+
"epoch": 0.75625,
|
| 1838 |
+
"frac_reward_zero_std": 0.25,
|
| 1839 |
+
"grad_norm": 2.4889636148189513,
|
| 1840 |
+
"kl": 0.011757506988942623,
|
| 1841 |
+
"learning_rate": 1.4824262559002592e-07,
|
| 1842 |
+
"loss": 0.00011754453589674085,
|
| 1843 |
+
"reward": 0.37890625,
|
| 1844 |
+
"reward_std": 0.35653358697891235,
|
| 1845 |
+
"rewards/MajorityAccuracy/mean": 0.37890625,
|
| 1846 |
+
"rewards/MajorityAccuracy/std": 0.48606500029563904,
|
| 1847 |
+
"step": 121,
|
| 1848 |
+
"step_time": 515.2314801923931
|
| 1849 |
+
},
|
| 1850 |
+
{
|
| 1851 |
+
"clip_ratio/high_max": 0.0021932368545094505,
|
| 1852 |
+
"clip_ratio/high_mean": 0.0007729964781901799,
|
| 1853 |
+
"clip_ratio/low_mean": 0.0003293624431535136,
|
| 1854 |
+
"clip_ratio/low_min": 0.0,
|
| 1855 |
+
"clip_ratio/region_mean": 0.00110235893225763,
|
| 1856 |
+
"epoch": 0.7625,
|
| 1857 |
+
"grad_norm": 1.7840325388840008,
|
| 1858 |
+
"kl": 0.011552143027074635,
|
| 1859 |
+
"learning_rate": 1.4111320966826057e-07,
|
| 1860 |
+
"loss": 6.229865539353341e-05,
|
| 1861 |
+
"step": 122,
|
| 1862 |
+
"step_time": 73.1909502595663
|
| 1863 |
+
},
|
| 1864 |
+
{
|
| 1865 |
+
"clip_ratio/high_max": 0.0017688935913611203,
|
| 1866 |
+
"clip_ratio/high_mean": 0.0004997740397811867,
|
| 1867 |
+
"clip_ratio/low_mean": 0.0004974332696292549,
|
| 1868 |
+
"clip_ratio/low_min": 0.0,
|
| 1869 |
+
"clip_ratio/region_mean": 0.0009972073094104417,
|
| 1870 |
+
"epoch": 0.76875,
|
| 1871 |
+
"grad_norm": 2.2831085673060403,
|
| 1872 |
+
"kl": 0.015406810212880373,
|
| 1873 |
+
"learning_rate": 1.3413122140054217e-07,
|
| 1874 |
+
"loss": 0.00029231482767499983,
|
| 1875 |
+
"step": 123,
|
| 1876 |
+
"step_time": 73.22497332841158
|
| 1877 |
+
},
|
| 1878 |
+
{
|
| 1879 |
+
"clip_ratio/high_max": 0.003100613976130262,
|
| 1880 |
+
"clip_ratio/high_mean": 0.0012567395897349343,
|
| 1881 |
+
"clip_ratio/low_mean": 0.0005439641281554941,
|
| 1882 |
+
"clip_ratio/low_min": 0.0,
|
| 1883 |
+
"clip_ratio/region_mean": 0.0018007037215284072,
|
| 1884 |
+
"epoch": 0.775,
|
| 1885 |
+
"grad_norm": 2.657219750295393,
|
| 1886 |
+
"kl": 0.016323628718964756,
|
| 1887 |
+
"learning_rate": 1.272995289285202e-07,
|
| 1888 |
+
"loss": -2.002343535423279e-06,
|
| 1889 |
+
"step": 124,
|
| 1890 |
+
"step_time": 72.05297258123755
|
| 1891 |
+
},
|
| 1892 |
+
{
|
| 1893 |
+
"clip_ratio/high_max": 0.0005551312933675945,
|
| 1894 |
+
"clip_ratio/high_mean": 0.00017152247892227024,
|
| 1895 |
+
"clip_ratio/low_mean": 0.0004119868463021703,
|
| 1896 |
+
"clip_ratio/low_min": 0.0,
|
| 1897 |
+
"clip_ratio/region_mean": 0.0005835093179484829,
|
| 1898 |
+
"epoch": 0.78125,
|
| 1899 |
+
"grad_norm": 1.609740344633677,
|
| 1900 |
+
"kl": 0.009846860193647444,
|
| 1901 |
+
"learning_rate": 1.2062093865360457e-07,
|
| 1902 |
+
"loss": 0.00022564176470041275,
|
| 1903 |
+
"step": 125,
|
| 1904 |
+
"step_time": 75.04421076923609
|
| 1905 |
+
},
|
| 1906 |
+
{
|
| 1907 |
+
"clip_ratio/high_max": 0.0012583268544403836,
|
| 1908 |
+
"clip_ratio/high_mean": 0.0003145817136100959,
|
| 1909 |
+
"clip_ratio/low_mean": 0.0005487042435561307,
|
| 1910 |
+
"clip_ratio/low_min": 0.0,
|
| 1911 |
+
"clip_ratio/region_mean": 0.0008632859571662266,
|
| 1912 |
+
"epoch": 0.7875,
|
| 1913 |
+
"grad_norm": 3.1310803781013896,
|
| 1914 |
+
"kl": 0.015399142750538886,
|
| 1915 |
+
"learning_rate": 1.1409819408411897e-07,
|
| 1916 |
+
"loss": 0.0001448571856599301,
|
| 1917 |
+
"step": 126,
|
| 1918 |
+
"step_time": 75.39871926233172
|
| 1919 |
+
},
|
| 1920 |
+
{
|
| 1921 |
+
"clip_ratio/high_max": 0.004204335724352859,
|
| 1922 |
+
"clip_ratio/high_mean": 0.0010510839310882147,
|
| 1923 |
+
"clip_ratio/low_mean": 0.00045586173655465245,
|
| 1924 |
+
"clip_ratio/low_min": 0.0,
|
| 1925 |
+
"clip_ratio/region_mean": 0.0015069456676428672,
|
| 1926 |
+
"epoch": 0.79375,
|
| 1927 |
+
"grad_norm": 2.869585953065354,
|
| 1928 |
+
"kl": 0.013336818083189428,
|
| 1929 |
+
"learning_rate": 1.0773397470829143e-07,
|
| 1930 |
+
"loss": 4.010299744550139e-05,
|
| 1931 |
+
"step": 127,
|
| 1932 |
+
"step_time": 75.2492504902184
|
| 1933 |
+
},
|
| 1934 |
+
{
|
| 1935 |
+
"clip_ratio/high_max": 0.0017749118123902008,
|
| 1936 |
+
"clip_ratio/high_mean": 0.0006324241767288186,
|
| 1937 |
+
"clip_ratio/low_mean": 9.1552734375e-05,
|
| 1938 |
+
"clip_ratio/low_min": 0.0,
|
| 1939 |
+
"clip_ratio/region_mean": 0.0007239769111038186,
|
| 1940 |
+
"epoch": 0.8,
|
| 1941 |
+
"grad_norm": 1.8548093987288556,
|
| 1942 |
+
"kl": 0.011208364507183433,
|
| 1943 |
+
"learning_rate": 1.0153089489354256e-07,
|
| 1944 |
+
"loss": 7.110345177352428e-05,
|
| 1945 |
+
"step": 128,
|
| 1946 |
+
"step_time": 75.15141194686294
|
| 1947 |
+
},
|
| 1948 |
+
{
|
| 1949 |
+
"clip_ratio/high_max": 0.0,
|
| 1950 |
+
"clip_ratio/high_mean": 0.0,
|
| 1951 |
+
"clip_ratio/low_mean": 0.0,
|
| 1952 |
+
"clip_ratio/low_min": 0.0,
|
| 1953 |
+
"clip_ratio/region_mean": 0.0,
|
| 1954 |
+
"completions/clipped_ratio": 0.42578125,
|
| 1955 |
+
"completions/max_length": 2048.0,
|
| 1956 |
+
"completions/mean_length": 1429.54296875,
|
| 1957 |
+
"completions/min_length": 110.0,
|
| 1958 |
+
"epoch": 0.80625,
|
| 1959 |
+
"frac_reward_zero_std": 0.28125,
|
| 1960 |
+
"grad_norm": 2.138888046810375,
|
| 1961 |
+
"kl": 0.011759641347452998,
|
| 1962 |
+
"learning_rate": 9.549150281252632e-08,
|
| 1963 |
+
"loss": 0.00011761952191591263,
|
| 1964 |
+
"reward": 0.37109375,
|
| 1965 |
+
"reward_std": 0.3399508595466614,
|
| 1966 |
+
"rewards/MajorityAccuracy/mean": 0.37109375,
|
| 1967 |
+
"rewards/MajorityAccuracy/std": 0.48404380679130554,
|
| 1968 |
+
"step": 129,
|
| 1969 |
+
"step_time": 517.2581111751497
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"clip_ratio/high_max": 0.0018661861977307126,
|
| 1973 |
+
"clip_ratio/high_mean": 0.0006171361383167095,
|
| 1974 |
+
"clip_ratio/low_mean": 0.0003204345703125,
|
| 1975 |
+
"clip_ratio/low_min": 0.0,
|
| 1976 |
+
"clip_ratio/region_mean": 0.0009375707086292095,
|
| 1977 |
+
"epoch": 0.8125,
|
| 1978 |
+
"grad_norm": 1.211301661457545,
|
| 1979 |
+
"kl": 0.011604880914092064,
|
| 1980 |
+
"learning_rate": 8.961827939636196e-08,
|
| 1981 |
+
"loss": 0.00018641927454154938,
|
| 1982 |
+
"step": 130,
|
| 1983 |
+
"step_time": 74.08386950567365
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"clip_ratio/high_max": 0.0018379736575298011,
|
| 1987 |
+
"clip_ratio/high_mean": 0.0005890145657758694,
|
| 1988 |
+
"clip_ratio/low_mean": 0.00011597550837905146,
|
| 1989 |
+
"clip_ratio/low_min": 0.0,
|
| 1990 |
+
"clip_ratio/region_mean": 0.0007049900705169421,
|
| 1991 |
+
"epoch": 0.81875,
|
| 1992 |
+
"grad_norm": 1.6112925838854508,
|
| 1993 |
+
"kl": 0.012096593505702913,
|
| 1994 |
+
"learning_rate": 8.391363731548811e-08,
|
| 1995 |
+
"loss": 5.52123601664789e-05,
|
| 1996 |
+
"step": 131,
|
| 1997 |
+
"step_time": 73.43572547286749
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"clip_ratio/high_max": 0.0037292905180947855,
|
| 2001 |
+
"clip_ratio/high_mean": 0.0009941120551957283,
|
| 2002 |
+
"clip_ratio/low_mean": 0.0004347985850472469,
|
| 2003 |
+
"clip_ratio/low_min": 0.0,
|
| 2004 |
+
"clip_ratio/region_mean": 0.0014289106111391447,
|
| 2005 |
+
"epoch": 0.825,
|
| 2006 |
+
"grad_norm": 14.256946432834056,
|
| 2007 |
+
"kl": 0.012659716536290944,
|
| 2008 |
+
"learning_rate": 7.837991998855897e-08,
|
| 2009 |
+
"loss": 0.0004249491903465241,
|
| 2010 |
+
"step": 132,
|
| 2011 |
+
"step_time": 73.67545045167208
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"clip_ratio/high_max": 0.0023109152680262923,
|
| 2015 |
+
"clip_ratio/high_mean": 0.0006982129052630626,
|
| 2016 |
+
"clip_ratio/low_mean": 0.0006271040092542535,
|
| 2017 |
+
"clip_ratio/low_min": 0.0,
|
| 2018 |
+
"clip_ratio/region_mean": 0.0013253169217932737,
|
| 2019 |
+
"epoch": 0.83125,
|
| 2020 |
+
"grad_norm": 2.2295583492424518,
|
| 2021 |
+
"kl": 0.012115559657104313,
|
| 2022 |
+
"learning_rate": 7.301940061978722e-08,
|
| 2023 |
+
"loss": -6.144275539554656e-05,
|
| 2024 |
+
"step": 133,
|
| 2025 |
+
"step_time": 74.51327027752995
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"clip_ratio/high_max": 0.003820090933004394,
|
| 2029 |
+
"clip_ratio/high_mean": 0.0013424298085737973,
|
| 2030 |
+
"clip_ratio/low_mean": 0.0010082749504363164,
|
| 2031 |
+
"clip_ratio/low_min": 0.0,
|
| 2032 |
+
"clip_ratio/region_mean": 0.002350704773562029,
|
| 2033 |
+
"epoch": 0.8375,
|
| 2034 |
+
"grad_norm": 3.0783760978105357,
|
| 2035 |
+
"kl": 0.014531118562445045,
|
| 2036 |
+
"learning_rate": 6.783428126513125e-08,
|
| 2037 |
+
"loss": 0.0007252329960465431,
|
| 2038 |
+
"step": 134,
|
| 2039 |
+
"step_time": 74.86898485943675
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"clip_ratio/high_max": 0.001929215926793404,
|
| 2043 |
+
"clip_ratio/high_mean": 0.000482303981698351,
|
| 2044 |
+
"clip_ratio/low_mean": 0.0002767360128927976,
|
| 2045 |
+
"clip_ratio/low_min": 0.0,
|
| 2046 |
+
"clip_ratio/region_mean": 0.0007590399945911486,
|
| 2047 |
+
"epoch": 0.84375,
|
| 2048 |
+
"grad_norm": 1.635305131974605,
|
| 2049 |
+
"kl": 0.011638955096714199,
|
| 2050 |
+
"learning_rate": 6.282669192770895e-08,
|
| 2051 |
+
"loss": 0.0003322207776363939,
|
| 2052 |
+
"step": 135,
|
| 2053 |
+
"step_time": 75.14104739204049
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"clip_ratio/high_max": 0.0011947421153308824,
|
| 2057 |
+
"clip_ratio/high_mean": 0.0003148104042338673,
|
| 2058 |
+
"clip_ratio/low_mean": 0.00012849126869696192,
|
| 2059 |
+
"clip_ratio/low_min": 0.0,
|
| 2060 |
+
"clip_ratio/region_mean": 0.0004433016692928504,
|
| 2061 |
+
"epoch": 0.85,
|
| 2062 |
+
"grad_norm": 1.7119664428389896,
|
| 2063 |
+
"kl": 0.010013362276367843,
|
| 2064 |
+
"learning_rate": 5.799868968281074e-08,
|
| 2065 |
+
"loss": 6.719694647472352e-05,
|
| 2066 |
+
"step": 136,
|
| 2067 |
+
"step_time": 74.13166991993785
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"clip_ratio/high_max": 0.0,
|
| 2071 |
+
"clip_ratio/high_mean": 0.0,
|
| 2072 |
+
"clip_ratio/low_mean": 0.0,
|
| 2073 |
+
"clip_ratio/low_min": 0.0,
|
| 2074 |
+
"clip_ratio/region_mean": 0.0,
|
| 2075 |
+
"completions/clipped_ratio": 0.42578125,
|
| 2076 |
+
"completions/max_length": 2048.0,
|
| 2077 |
+
"completions/mean_length": 1388.60546875,
|
| 2078 |
+
"completions/min_length": 144.0,
|
| 2079 |
+
"epoch": 0.85625,
|
| 2080 |
+
"frac_reward_zero_std": 0.3125,
|
| 2081 |
+
"grad_norm": 2.230867782036166,
|
| 2082 |
+
"kl": 0.012941621709614992,
|
| 2083 |
+
"learning_rate": 5.335225783287051e-08,
|
| 2084 |
+
"loss": 0.00012943055480718613,
|
| 2085 |
+
"reward": 0.39453125,
|
| 2086 |
+
"reward_std": 0.2970898449420929,
|
| 2087 |
+
"rewards/MajorityAccuracy/mean": 0.39453125,
|
| 2088 |
+
"rewards/MajorityAccuracy/std": 0.48970717191696167,
|
| 2089 |
+
"step": 137,
|
| 2090 |
+
"step_time": 512.8900125995278
|
| 2091 |
+
},
|
| 2092 |
+
{
|
| 2093 |
+
"clip_ratio/high_max": 0.0025262467097491026,
|
| 2094 |
+
"clip_ratio/high_mean": 0.0007618230411026161,
|
| 2095 |
+
"clip_ratio/low_mean": 0.000152587890625,
|
| 2096 |
+
"clip_ratio/low_min": 0.0,
|
| 2097 |
+
"clip_ratio/region_mean": 0.0009144109426415525,
|
| 2098 |
+
"epoch": 0.8625,
|
| 2099 |
+
"grad_norm": 1.2584154344786542,
|
| 2100 |
+
"kl": 0.010506410500966012,
|
| 2101 |
+
"learning_rate": 4.888930509274125e-08,
|
| 2102 |
+
"loss": 4.6425251639448106e-05,
|
| 2103 |
+
"step": 138,
|
| 2104 |
+
"step_time": 73.9938875362277
|
| 2105 |
+
},
|
| 2106 |
+
{
|
| 2107 |
+
"clip_ratio/high_max": 0.0006807730969740078,
|
| 2108 |
+
"clip_ratio/high_mean": 0.0002528270852053538,
|
| 2109 |
+
"clip_ratio/low_mean": 9.067579230759293e-05,
|
| 2110 |
+
"clip_ratio/low_min": 0.0,
|
| 2111 |
+
"clip_ratio/region_mean": 0.0003435028775129467,
|
| 2112 |
+
"epoch": 0.86875,
|
| 2113 |
+
"grad_norm": 1.3783052013898802,
|
| 2114 |
+
"kl": 0.010039298213087022,
|
| 2115 |
+
"learning_rate": 4.4611664805611794e-08,
|
| 2116 |
+
"loss": -2.8838501748396084e-05,
|
| 2117 |
+
"step": 139,
|
| 2118 |
+
"step_time": 73.570172265172
|
| 2119 |
+
},
|
| 2120 |
+
{
|
| 2121 |
+
"clip_ratio/high_max": 0.0020387282929732464,
|
| 2122 |
+
"clip_ratio/high_mean": 0.0006771115549781825,
|
| 2123 |
+
"clip_ratio/low_mean": 0.00037247453292366117,
|
| 2124 |
+
"clip_ratio/low_min": 0.0,
|
| 2125 |
+
"clip_ratio/region_mean": 0.0010495860915398225,
|
| 2126 |
+
"epoch": 0.875,
|
| 2127 |
+
"grad_norm": 1.5434170087200714,
|
| 2128 |
+
"kl": 0.012984891305677593,
|
| 2129 |
+
"learning_rate": 4.0521094189884696e-08,
|
| 2130 |
+
"loss": 0.00040381215512752533,
|
| 2131 |
+
"step": 140,
|
| 2132 |
+
"step_time": 75.07217965647578
|
| 2133 |
+
},
|
| 2134 |
+
{
|
| 2135 |
+
"clip_ratio/high_max": 0.0034280881518498063,
|
| 2136 |
+
"clip_ratio/high_mean": 0.0010269850608892739,
|
| 2137 |
+
"clip_ratio/low_mean": 0.0007718576525803655,
|
| 2138 |
+
"clip_ratio/low_min": 0.0,
|
| 2139 |
+
"clip_ratio/region_mean": 0.0017988426989177242,
|
| 2140 |
+
"epoch": 0.88125,
|
| 2141 |
+
"grad_norm": 3.887345609517078,
|
| 2142 |
+
"kl": 0.016007773810997605,
|
| 2143 |
+
"learning_rate": 3.6619273617325695e-08,
|
| 2144 |
+
"loss": 0.00037208799039945006,
|
| 2145 |
+
"step": 141,
|
| 2146 |
+
"step_time": 72.63409619033337
|
| 2147 |
+
},
|
| 2148 |
+
{
|
| 2149 |
+
"clip_ratio/high_max": 0.0016112564771901816,
|
| 2150 |
+
"clip_ratio/high_mean": 0.0004028141192975454,
|
| 2151 |
+
"clip_ratio/low_mean": 0.0003183513217663858,
|
| 2152 |
+
"clip_ratio/low_min": 0.0,
|
| 2153 |
+
"clip_ratio/region_mean": 0.0007211654374259524,
|
| 2154 |
+
"epoch": 0.8875,
|
| 2155 |
+
"grad_norm": 1.0901807385222138,
|
| 2156 |
+
"kl": 0.01953425374813378,
|
| 2157 |
+
"learning_rate": 3.2907805922781476e-08,
|
| 2158 |
+
"loss": 0.00033988914219662547,
|
| 2159 |
+
"step": 142,
|
| 2160 |
+
"step_time": 72.96996520087123
|
| 2161 |
+
},
|
| 2162 |
+
{
|
| 2163 |
+
"clip_ratio/high_max": 0.00046113126882119104,
|
| 2164 |
+
"clip_ratio/high_mean": 0.0001348140649497509,
|
| 2165 |
+
"clip_ratio/low_mean": 0.0003204345703125,
|
| 2166 |
+
"clip_ratio/low_min": 0.0,
|
| 2167 |
+
"clip_ratio/region_mean": 0.0004552486352622509,
|
| 2168 |
+
"epoch": 0.89375,
|
| 2169 |
+
"grad_norm": 1.3324528742075952,
|
| 2170 |
+
"kl": 0.009650545485783368,
|
| 2171 |
+
"learning_rate": 2.9388215745748345e-08,
|
| 2172 |
+
"loss": 0.00014820303476881236,
|
| 2173 |
+
"step": 143,
|
| 2174 |
+
"step_time": 74.72494164854288
|
| 2175 |
+
},
|
| 2176 |
+
{
|
| 2177 |
+
"clip_ratio/high_max": 0.002269154880195856,
|
| 2178 |
+
"clip_ratio/high_mean": 0.0005868199659744278,
|
| 2179 |
+
"clip_ratio/low_mean": 0.00038469276478281245,
|
| 2180 |
+
"clip_ratio/low_min": 0.0,
|
| 2181 |
+
"clip_ratio/region_mean": 0.0009715127453091554,
|
| 2182 |
+
"epoch": 0.9,
|
| 2183 |
+
"grad_norm": 1.8813002628580513,
|
| 2184 |
+
"kl": 0.012995302327908576,
|
| 2185 |
+
"learning_rate": 2.6061948904063658e-08,
|
| 2186 |
+
"loss": 0.00038667654735036194,
|
| 2187 |
+
"step": 144,
|
| 2188 |
+
"step_time": 74.2004100382328
|
| 2189 |
+
},
|
| 2190 |
+
{
|
| 2191 |
+
"clip_ratio/high_max": 0.0,
|
| 2192 |
+
"clip_ratio/high_mean": 0.0,
|
| 2193 |
+
"clip_ratio/low_mean": 0.0,
|
| 2194 |
+
"clip_ratio/low_min": 0.0,
|
| 2195 |
+
"clip_ratio/region_mean": 0.0,
|
| 2196 |
+
"completions/clipped_ratio": 0.4375,
|
| 2197 |
+
"completions/max_length": 2048.0,
|
| 2198 |
+
"completions/mean_length": 1440.1796875,
|
| 2199 |
+
"completions/min_length": 187.0,
|
| 2200 |
+
"epoch": 0.90625,
|
| 2201 |
+
"frac_reward_zero_std": 0.34375,
|
| 2202 |
+
"grad_norm": 1.065368555111494,
|
| 2203 |
+
"kl": 0.010083240456879139,
|
| 2204 |
+
"learning_rate": 2.293037179997559e-08,
|
| 2205 |
+
"loss": 0.000100809644209221,
|
| 2206 |
+
"reward": 0.32421875,
|
| 2207 |
+
"reward_std": 0.30194634199142456,
|
| 2208 |
+
"rewards/MajorityAccuracy/mean": 0.32421875,
|
| 2209 |
+
"rewards/MajorityAccuracy/std": 0.46899911761283875,
|
| 2210 |
+
"step": 145,
|
| 2211 |
+
"step_time": 518.8105758987367
|
| 2212 |
+
},
|
| 2213 |
+
{
|
| 2214 |
+
"clip_ratio/high_max": 0.003193992539308965,
|
| 2215 |
+
"clip_ratio/high_mean": 0.0008736635063542053,
|
| 2216 |
+
"clip_ratio/low_mean": 0.0006895240658195689,
|
| 2217 |
+
"clip_ratio/low_min": 0.0,
|
| 2218 |
+
"clip_ratio/region_mean": 0.0015631875721737742,
|
| 2219 |
+
"epoch": 0.9125,
|
| 2220 |
+
"grad_norm": 4.020428481774001,
|
| 2221 |
+
"kl": 0.013928839238360524,
|
| 2222 |
+
"learning_rate": 1.9994770858837107e-08,
|
| 2223 |
+
"loss": 0.00020919437520205975,
|
| 2224 |
+
"step": 146,
|
| 2225 |
+
"step_time": 74.39246311411262
|
| 2226 |
+
},
|
| 2227 |
+
{
|
| 2228 |
+
"clip_ratio/high_max": 0.0029897911299485713,
|
| 2229 |
+
"clip_ratio/high_mean": 0.0009192799116135575,
|
| 2230 |
+
"clip_ratio/low_mean": 0.0007092542800819501,
|
| 2231 |
+
"clip_ratio/low_min": 6.103515625e-05,
|
| 2232 |
+
"clip_ratio/region_mean": 0.0016285341625916772,
|
| 2233 |
+
"epoch": 0.91875,
|
| 2234 |
+
"grad_norm": 2.3804664695422617,
|
| 2235 |
+
"kl": 0.014371934812515974,
|
| 2236 |
+
"learning_rate": 1.725635200065323e-08,
|
| 2237 |
+
"loss": 0.00025225523859262466,
|
| 2238 |
+
"step": 147,
|
| 2239 |
+
"step_time": 75.1100714057684
|
| 2240 |
+
},
|
| 2241 |
+
{
|
| 2242 |
+
"clip_ratio/high_max": 0.002689582121092826,
|
| 2243 |
+
"clip_ratio/high_mean": 0.0007682544237468392,
|
| 2244 |
+
"clip_ratio/low_mean": 0.00036163448385195807,
|
| 2245 |
+
"clip_ratio/low_min": 0.0,
|
| 2246 |
+
"clip_ratio/region_mean": 0.0011298889003228396,
|
| 2247 |
+
"epoch": 0.925,
|
| 2248 |
+
"grad_norm": 2.07021466483763,
|
| 2249 |
+
"kl": 0.011867468012496829,
|
| 2250 |
+
"learning_rate": 1.4716240144699187e-08,
|
| 2251 |
+
"loss": 0.00012768342276103795,
|
| 2252 |
+
"step": 148,
|
| 2253 |
+
"step_time": 74.7243696115911
|
| 2254 |
+
},
|
| 2255 |
+
{
|
| 2256 |
+
"clip_ratio/high_max": 0.002056548066320829,
|
| 2257 |
+
"clip_ratio/high_mean": 0.0006517047331726644,
|
| 2258 |
+
"clip_ratio/low_mean": 0.0002158319312002277,
|
| 2259 |
+
"clip_ratio/low_min": 0.0,
|
| 2260 |
+
"clip_ratio/region_mean": 0.0008675366661918815,
|
| 2261 |
+
"epoch": 0.93125,
|
| 2262 |
+
"grad_norm": 1.1457868632052457,
|
| 2263 |
+
"kl": 0.010700110928155482,
|
| 2264 |
+
"learning_rate": 1.2375478747413015e-08,
|
| 2265 |
+
"loss": 0.00020325295918155462,
|
| 2266 |
+
"step": 149,
|
| 2267 |
+
"step_time": 75.06848003342748
|
| 2268 |
+
},
|
| 2269 |
+
{
|
| 2270 |
+
"clip_ratio/high_max": 0.001326192490523681,
|
| 2271 |
+
"clip_ratio/high_mean": 0.0005196733545744792,
|
| 2272 |
+
"clip_ratio/low_mean": 4.083162093593273e-05,
|
| 2273 |
+
"clip_ratio/low_min": 0.0,
|
| 2274 |
+
"clip_ratio/region_mean": 0.0005605049755104119,
|
| 2275 |
+
"epoch": 0.9375,
|
| 2276 |
+
"grad_norm": 1.4894562862471945,
|
| 2277 |
+
"kl": 0.01614364783745259,
|
| 2278 |
+
"learning_rate": 1.0235029373752757e-08,
|
| 2279 |
+
"loss": 0.00019657165103126317,
|
| 2280 |
+
"step": 150,
|
| 2281 |
+
"step_time": 72.77896462008357
|
| 2282 |
+
},
|
| 2283 |
+
{
|
| 2284 |
+
"clip_ratio/high_max": 0.0005099616682855412,
|
| 2285 |
+
"clip_ratio/high_mean": 0.00015947608699207194,
|
| 2286 |
+
"clip_ratio/low_mean": 0.0001983642578125,
|
| 2287 |
+
"clip_ratio/low_min": 0.0,
|
| 2288 |
+
"clip_ratio/region_mean": 0.00035784034480457194,
|
| 2289 |
+
"epoch": 0.94375,
|
| 2290 |
+
"grad_norm": 1.6958606339576152,
|
| 2291 |
+
"kl": 0.010462523438036442,
|
| 2292 |
+
"learning_rate": 8.295771302193721e-09,
|
| 2293 |
+
"loss": 7.445565279340371e-05,
|
| 2294 |
+
"step": 151,
|
| 2295 |
+
"step_time": 73.22191849723458
|
| 2296 |
+
},
|
| 2297 |
+
{
|
| 2298 |
+
"clip_ratio/high_max": 0.001179445935122203,
|
| 2299 |
+
"clip_ratio/high_mean": 0.00044367102782416623,
|
| 2300 |
+
"clip_ratio/low_mean": 0.0007282171209226362,
|
| 2301 |
+
"clip_ratio/low_min": 0.0,
|
| 2302 |
+
"clip_ratio/region_mean": 0.0011718881360138766,
|
| 2303 |
+
"epoch": 0.95,
|
| 2304 |
+
"grad_norm": 1.5017095834950092,
|
| 2305 |
+
"kl": 0.011120545561425388,
|
| 2306 |
+
"learning_rate": 6.558501163527963e-09,
|
| 2307 |
+
"loss": 4.9191319703822955e-05,
|
| 2308 |
+
"step": 152,
|
| 2309 |
+
"step_time": 73.74227200448513
|
| 2310 |
+
},
|
| 2311 |
+
{
|
| 2312 |
+
"clip_ratio/high_max": 0.0,
|
| 2313 |
+
"clip_ratio/high_mean": 0.0,
|
| 2314 |
+
"clip_ratio/low_mean": 0.0,
|
| 2315 |
+
"clip_ratio/low_min": 0.0,
|
| 2316 |
+
"clip_ratio/region_mean": 0.0,
|
| 2317 |
+
"completions/clipped_ratio": 0.47265625,
|
| 2318 |
+
"completions/max_length": 2048.0,
|
| 2319 |
+
"completions/mean_length": 1503.6953125,
|
| 2320 |
+
"completions/min_length": 147.0,
|
| 2321 |
+
"epoch": 0.95625,
|
| 2322 |
+
"frac_reward_zero_std": 0.28125,
|
| 2323 |
+
"grad_norm": 2.7483107827600923,
|
| 2324 |
+
"kl": 0.01531501195859164,
|
| 2325 |
+
"learning_rate": 5.023932613615445e-09,
|
| 2326 |
+
"loss": 0.0001531355082988739,
|
| 2327 |
+
"reward": 0.29296875,
|
| 2328 |
+
"reward_std": 0.32970115542411804,
|
| 2329 |
+
"rewards/MajorityAccuracy/mean": 0.29296875,
|
| 2330 |
+
"rewards/MajorityAccuracy/std": 0.45601576566696167,
|
| 2331 |
+
"step": 153,
|
| 2332 |
+
"step_time": 519.8760307505727
|
| 2333 |
+
},
|
| 2334 |
+
{
|
| 2335 |
+
"clip_ratio/high_max": 0.002972931251861155,
|
| 2336 |
+
"clip_ratio/high_mean": 0.00114014891005354,
|
| 2337 |
+
"clip_ratio/low_mean": 0.00018906751211034134,
|
| 2338 |
+
"clip_ratio/low_min": 0.0,
|
| 2339 |
+
"clip_ratio/region_mean": 0.0013292164039739873,
|
| 2340 |
+
"epoch": 0.9625,
|
| 2341 |
+
"grad_norm": 2.6522856587457295,
|
| 2342 |
+
"kl": 0.011618947144597769,
|
| 2343 |
+
"learning_rate": 3.6926960402202666e-09,
|
| 2344 |
+
"loss": 0.00017663277685642242,
|
| 2345 |
+
"step": 154,
|
| 2346 |
+
"step_time": 74.2137517221272
|
| 2347 |
+
},
|
| 2348 |
+
{
|
| 2349 |
+
"clip_ratio/high_max": 0.0016927705437410623,
|
| 2350 |
+
"clip_ratio/high_mean": 0.0005715724910260178,
|
| 2351 |
+
"clip_ratio/low_mean": 0.000244140625,
|
| 2352 |
+
"clip_ratio/low_min": 0.0,
|
| 2353 |
+
"clip_ratio/region_mean": 0.0008157131233019754,
|
| 2354 |
+
"epoch": 0.96875,
|
| 2355 |
+
"grad_norm": 1.345928995786229,
|
| 2356 |
+
"kl": 0.010836946894414723,
|
| 2357 |
+
"learning_rate": 2.5653383040524224e-09,
|
| 2358 |
+
"loss": 3.0436804081546143e-05,
|
| 2359 |
+
"step": 155,
|
| 2360 |
+
"step_time": 74.77166337519884
|
| 2361 |
+
},
|
| 2362 |
+
{
|
| 2363 |
+
"clip_ratio/high_max": 0.00012450199574232101,
|
| 2364 |
+
"clip_ratio/high_mean": 3.1125498935580254e-05,
|
| 2365 |
+
"clip_ratio/low_mean": 0.0007331070846703369,
|
| 2366 |
+
"clip_ratio/low_min": 0.0,
|
| 2367 |
+
"clip_ratio/region_mean": 0.0007642325836059172,
|
| 2368 |
+
"epoch": 0.975,
|
| 2369 |
+
"grad_norm": 1.8037494426366838,
|
| 2370 |
+
"kl": 0.009005147381685674,
|
| 2371 |
+
"learning_rate": 1.642322514122385e-09,
|
| 2372 |
+
"loss": 0.00013722004950977862,
|
| 2373 |
+
"step": 156,
|
| 2374 |
+
"step_time": 75.32932732254267
|
| 2375 |
+
},
|
| 2376 |
+
{
|
| 2377 |
+
"clip_ratio/high_max": 0.0007718660199316218,
|
| 2378 |
+
"clip_ratio/high_mean": 0.00019296650498290546,
|
| 2379 |
+
"clip_ratio/low_mean": 0.0004927884292555973,
|
| 2380 |
+
"clip_ratio/low_min": 0.0,
|
| 2381 |
+
"clip_ratio/region_mean": 0.000685754930600524,
|
| 2382 |
+
"epoch": 0.98125,
|
| 2383 |
+
"grad_norm": 1.575244850484651,
|
| 2384 |
+
"kl": 0.012953114230185747,
|
| 2385 |
+
"learning_rate": 9.240278374995636e-10,
|
| 2386 |
+
"loss": 0.000149703148053959,
|
| 2387 |
+
"step": 157,
|
| 2388 |
+
"step_time": 74.8022028543055
|
| 2389 |
+
},
|
| 2390 |
+
{
|
| 2391 |
+
"clip_ratio/high_max": 0.0018839943295461126,
|
| 2392 |
+
"clip_ratio/high_mean": 0.00047099858238652814,
|
| 2393 |
+
"clip_ratio/low_mean": 0.0005254395218798891,
|
| 2394 |
+
"clip_ratio/low_min": 0.0,
|
| 2395 |
+
"clip_ratio/region_mean": 0.0009964381188183324,
|
| 2396 |
+
"epoch": 0.9875,
|
| 2397 |
+
"grad_norm": 2.2833119714455874,
|
| 2398 |
+
"kl": 0.008678545011207461,
|
| 2399 |
+
"learning_rate": 4.1074934355384006e-10,
|
| 2400 |
+
"loss": 0.000215997570194304,
|
| 2401 |
+
"step": 158,
|
| 2402 |
+
"step_time": 75.02634674310684
|
| 2403 |
+
},
|
| 2404 |
+
{
|
| 2405 |
+
"clip_ratio/high_max": 0.0008463287522317842,
|
| 2406 |
+
"clip_ratio/high_mean": 0.00028776706676580943,
|
| 2407 |
+
"clip_ratio/low_mean": 0.0003690479243232403,
|
| 2408 |
+
"clip_ratio/low_min": 0.0,
|
| 2409 |
+
"clip_ratio/region_mean": 0.0006568149874510709,
|
| 2410 |
+
"epoch": 0.99375,
|
| 2411 |
+
"grad_norm": 2.322769149442764,
|
| 2412 |
+
"kl": 0.011222293484024704,
|
| 2413 |
+
"learning_rate": 1.0269788274358537e-10,
|
| 2414 |
+
"loss": 0.0001906205725390464,
|
| 2415 |
+
"step": 159,
|
| 2416 |
+
"step_time": 75.16437526419759
|
| 2417 |
+
},
|
| 2418 |
+
{
|
| 2419 |
+
"clip_ratio/high_max": 0.0014598329144064337,
|
| 2420 |
+
"clip_ratio/high_mean": 0.0003649582286016084,
|
| 2421 |
+
"clip_ratio/low_mean": 0.0005142891022842377,
|
| 2422 |
+
"clip_ratio/low_min": 0.0,
|
| 2423 |
+
"clip_ratio/region_mean": 0.0008792473308858462,
|
| 2424 |
+
"epoch": 1.0,
|
| 2425 |
+
"grad_norm": 1.9238437054433697,
|
| 2426 |
+
"kl": 0.014265156467445195,
|
| 2427 |
+
"learning_rate": 0.0,
|
| 2428 |
+
"loss": 0.00033854079083539546,
|
| 2429 |
+
"step": 160,
|
| 2430 |
+
"step_time": 74.7471311353147
|
| 2431 |
+
}
|
| 2432 |
+
],
|
| 2433 |
+
"logging_steps": 1,
|
| 2434 |
+
"max_steps": 160,
|
| 2435 |
+
"num_input_tokens_seen": 0,
|
| 2436 |
+
"num_train_epochs": 1,
|
| 2437 |
+
"save_steps": 80,
|
| 2438 |
+
"stateful_callbacks": {
|
| 2439 |
+
"TrainerControl": {
|
| 2440 |
+
"args": {
|
| 2441 |
+
"should_epoch_stop": false,
|
| 2442 |
+
"should_evaluate": false,
|
| 2443 |
+
"should_log": false,
|
| 2444 |
+
"should_save": true,
|
| 2445 |
+
"should_training_stop": true
|
| 2446 |
+
},
|
| 2447 |
+
"attributes": {}
|
| 2448 |
+
}
|
| 2449 |
+
},
|
| 2450 |
+
"total_flos": 0.0,
|
| 2451 |
+
"train_batch_size": 1,
|
| 2452 |
+
"trial_name": null,
|
| 2453 |
+
"trial_params": null
|
| 2454 |
+
}
|
c1_no_history/round_3/chat_template.jinja
ADDED
|
@@ -0,0 +1,154 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{%- set image_count = namespace(value=0) %}
|
| 2 |
+
{%- set video_count = namespace(value=0) %}
|
| 3 |
+
{%- macro render_content(content, do_vision_count, is_system_content=false) %}
|
| 4 |
+
{%- if content is string %}
|
| 5 |
+
{{- content }}
|
| 6 |
+
{%- elif content is iterable and content is not mapping %}
|
| 7 |
+
{%- for item in content %}
|
| 8 |
+
{%- if 'image' in item or 'image_url' in item or item.type == 'image' %}
|
| 9 |
+
{%- if is_system_content %}
|
| 10 |
+
{{- raise_exception('System message cannot contain images.') }}
|
| 11 |
+
{%- endif %}
|
| 12 |
+
{%- if do_vision_count %}
|
| 13 |
+
{%- set image_count.value = image_count.value + 1 %}
|
| 14 |
+
{%- endif %}
|
| 15 |
+
{%- if add_vision_id %}
|
| 16 |
+
{{- 'Picture ' ~ image_count.value ~ ': ' }}
|
| 17 |
+
{%- endif %}
|
| 18 |
+
{{- '<|vision_start|><|image_pad|><|vision_end|>' }}
|
| 19 |
+
{%- elif 'video' in item or item.type == 'video' %}
|
| 20 |
+
{%- if is_system_content %}
|
| 21 |
+
{{- raise_exception('System message cannot contain videos.') }}
|
| 22 |
+
{%- endif %}
|
| 23 |
+
{%- if do_vision_count %}
|
| 24 |
+
{%- set video_count.value = video_count.value + 1 %}
|
| 25 |
+
{%- endif %}
|
| 26 |
+
{%- if add_vision_id %}
|
| 27 |
+
{{- 'Video ' ~ video_count.value ~ ': ' }}
|
| 28 |
+
{%- endif %}
|
| 29 |
+
{{- '<|vision_start|><|video_pad|><|vision_end|>' }}
|
| 30 |
+
{%- elif 'text' in item %}
|
| 31 |
+
{{- item.text }}
|
| 32 |
+
{%- else %}
|
| 33 |
+
{{- raise_exception('Unexpected item type in content.') }}
|
| 34 |
+
{%- endif %}
|
| 35 |
+
{%- endfor %}
|
| 36 |
+
{%- elif content is none or content is undefined %}
|
| 37 |
+
{{- '' }}
|
| 38 |
+
{%- else %}
|
| 39 |
+
{{- raise_exception('Unexpected content type.') }}
|
| 40 |
+
{%- endif %}
|
| 41 |
+
{%- endmacro %}
|
| 42 |
+
{%- if not messages %}
|
| 43 |
+
{{- raise_exception('No messages provided.') }}
|
| 44 |
+
{%- endif %}
|
| 45 |
+
{%- if tools and tools is iterable and tools is not mapping %}
|
| 46 |
+
{{- '<|im_start|>system\n' }}
|
| 47 |
+
{{- "# Tools\n\nYou have access to the following functions:\n\n<tools>" }}
|
| 48 |
+
{%- for tool in tools %}
|
| 49 |
+
{{- "\n" }}
|
| 50 |
+
{{- tool | tojson }}
|
| 51 |
+
{%- endfor %}
|
| 52 |
+
{{- "\n</tools>" }}
|
| 53 |
+
{{- '\n\nIf you choose to call a function ONLY reply in the following format with NO suffix:\n\n<tool_call>\n<function=example_function_name>\n<parameter=example_parameter_1>\nvalue_1\n</parameter>\n<parameter=example_parameter_2>\nThis is the value for the second parameter\nthat can span\nmultiple lines\n</parameter>\n</function>\n</tool_call>\n\n<IMPORTANT>\nReminder:\n- Function calls MUST follow the specified format: an inner <function=...></function> block must be nested within <tool_call></tool_call> XML tags\n- Required parameters MUST be specified\n- You may provide optional reasoning for your function call in natural language BEFORE the function call, but NOT after\n- If there is no function call available, answer the question like normal with your current knowledge and do not tell the user about function calls\n</IMPORTANT>' }}
|
| 54 |
+
{%- if messages[0].role == 'system' %}
|
| 55 |
+
{%- set content = render_content(messages[0].content, false, true)|trim %}
|
| 56 |
+
{%- if content %}
|
| 57 |
+
{{- '\n\n' + content }}
|
| 58 |
+
{%- endif %}
|
| 59 |
+
{%- endif %}
|
| 60 |
+
{{- '<|im_end|>\n' }}
|
| 61 |
+
{%- else %}
|
| 62 |
+
{%- if messages[0].role == 'system' %}
|
| 63 |
+
{%- set content = render_content(messages[0].content, false, true)|trim %}
|
| 64 |
+
{{- '<|im_start|>system\n' + content + '<|im_end|>\n' }}
|
| 65 |
+
{%- endif %}
|
| 66 |
+
{%- endif %}
|
| 67 |
+
{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
|
| 68 |
+
{%- for message in messages[::-1] %}
|
| 69 |
+
{%- set index = (messages|length - 1) - loop.index0 %}
|
| 70 |
+
{%- if ns.multi_step_tool and message.role == "user" %}
|
| 71 |
+
{%- set content = render_content(message.content, false)|trim %}
|
| 72 |
+
{%- if not(content.startswith('<tool_response>') and content.endswith('</tool_response>')) %}
|
| 73 |
+
{%- set ns.multi_step_tool = false %}
|
| 74 |
+
{%- set ns.last_query_index = index %}
|
| 75 |
+
{%- endif %}
|
| 76 |
+
{%- endif %}
|
| 77 |
+
{%- endfor %}
|
| 78 |
+
{%- if ns.multi_step_tool %}
|
| 79 |
+
{{- raise_exception('No user query found in messages.') }}
|
| 80 |
+
{%- endif %}
|
| 81 |
+
{%- for message in messages %}
|
| 82 |
+
{%- set content = render_content(message.content, true)|trim %}
|
| 83 |
+
{%- if message.role == "system" %}
|
| 84 |
+
{%- if not loop.first %}
|
| 85 |
+
{{- raise_exception('System message must be at the beginning.') }}
|
| 86 |
+
{%- endif %}
|
| 87 |
+
{%- elif message.role == "user" %}
|
| 88 |
+
{{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
|
| 89 |
+
{%- elif message.role == "assistant" %}
|
| 90 |
+
{%- set reasoning_content = '' %}
|
| 91 |
+
{%- if message.reasoning_content is string %}
|
| 92 |
+
{%- set reasoning_content = message.reasoning_content %}
|
| 93 |
+
{%- else %}
|
| 94 |
+
{%- if '</think>' in content %}
|
| 95 |
+
{%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
|
| 96 |
+
{%- set content = content.split('</think>')[-1].lstrip('\n') %}
|
| 97 |
+
{%- endif %}
|
| 98 |
+
{%- endif %}
|
| 99 |
+
{%- set reasoning_content = reasoning_content|trim %}
|
| 100 |
+
{%- if loop.index0 > ns.last_query_index %}
|
| 101 |
+
{{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content + '\n</think>\n\n' + content }}
|
| 102 |
+
{%- else %}
|
| 103 |
+
{{- '<|im_start|>' + message.role + '\n' + content }}
|
| 104 |
+
{%- endif %}
|
| 105 |
+
{%- if message.tool_calls and message.tool_calls is iterable and message.tool_calls is not mapping %}
|
| 106 |
+
{%- for tool_call in message.tool_calls %}
|
| 107 |
+
{%- if tool_call.function is defined %}
|
| 108 |
+
{%- set tool_call = tool_call.function %}
|
| 109 |
+
{%- endif %}
|
| 110 |
+
{%- if loop.first %}
|
| 111 |
+
{%- if content|trim %}
|
| 112 |
+
{{- '\n\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
|
| 113 |
+
{%- else %}
|
| 114 |
+
{{- '<tool_call>\n<function=' + tool_call.name + '>\n' }}
|
| 115 |
+
{%- endif %}
|
| 116 |
+
{%- else %}
|
| 117 |
+
{{- '\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
|
| 118 |
+
{%- endif %}
|
| 119 |
+
{%- if tool_call.arguments is defined %}
|
| 120 |
+
{%- for args_name, args_value in tool_call.arguments|items %}
|
| 121 |
+
{{- '<parameter=' + args_name + '>\n' }}
|
| 122 |
+
{%- set args_value = args_value | tojson | safe if args_value is mapping or (args_value is sequence and args_value is not string) else args_value | string %}
|
| 123 |
+
{{- args_value }}
|
| 124 |
+
{{- '\n</parameter>\n' }}
|
| 125 |
+
{%- endfor %}
|
| 126 |
+
{%- endif %}
|
| 127 |
+
{{- '</function>\n</tool_call>' }}
|
| 128 |
+
{%- endfor %}
|
| 129 |
+
{%- endif %}
|
| 130 |
+
{{- '<|im_end|>\n' }}
|
| 131 |
+
{%- elif message.role == "tool" %}
|
| 132 |
+
{%- if loop.previtem and loop.previtem.role != "tool" %}
|
| 133 |
+
{{- '<|im_start|>user' }}
|
| 134 |
+
{%- endif %}
|
| 135 |
+
{{- '\n<tool_response>\n' }}
|
| 136 |
+
{{- content }}
|
| 137 |
+
{{- '\n</tool_response>' }}
|
| 138 |
+
{%- if not loop.last and loop.nextitem.role != "tool" %}
|
| 139 |
+
{{- '<|im_end|>\n' }}
|
| 140 |
+
{%- elif loop.last %}
|
| 141 |
+
{{- '<|im_end|>\n' }}
|
| 142 |
+
{%- endif %}
|
| 143 |
+
{%- else %}
|
| 144 |
+
{{- raise_exception('Unexpected message role.') }}
|
| 145 |
+
{%- endif %}
|
| 146 |
+
{%- endfor %}
|
| 147 |
+
{%- if add_generation_prompt %}
|
| 148 |
+
{{- '<|im_start|>assistant\n' }}
|
| 149 |
+
{%- if enable_thinking is defined and enable_thinking is false %}
|
| 150 |
+
{{- '<think>\n\n</think>\n\n' }}
|
| 151 |
+
{%- else %}
|
| 152 |
+
{{- '<think>\n' }}
|
| 153 |
+
{%- endif %}
|
| 154 |
+
{%- endif %}
|
c1_no_history/round_3/config.json
ADDED
|
@@ -0,0 +1,113 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"Qwen3_5ForConditionalGeneration"
|
| 4 |
+
],
|
| 5 |
+
"dtype": "bfloat16",
|
| 6 |
+
"eos_token_id": 248046,
|
| 7 |
+
"hidden_size": 4096,
|
| 8 |
+
"image_token_id": 248056,
|
| 9 |
+
"model_type": "qwen3_5",
|
| 10 |
+
"pad_token_id": 248044,
|
| 11 |
+
"text_config": {
|
| 12 |
+
"attention_bias": false,
|
| 13 |
+
"attention_dropout": 0.0,
|
| 14 |
+
"attn_output_gate": true,
|
| 15 |
+
"bos_token_id": null,
|
| 16 |
+
"dtype": "bfloat16",
|
| 17 |
+
"eos_token_id": 248044,
|
| 18 |
+
"full_attention_interval": 4,
|
| 19 |
+
"head_dim": 256,
|
| 20 |
+
"hidden_act": "silu",
|
| 21 |
+
"hidden_size": 4096,
|
| 22 |
+
"initializer_range": 0.02,
|
| 23 |
+
"intermediate_size": 12288,
|
| 24 |
+
"layer_types": [
|
| 25 |
+
"linear_attention",
|
| 26 |
+
"linear_attention",
|
| 27 |
+
"linear_attention",
|
| 28 |
+
"full_attention",
|
| 29 |
+
"linear_attention",
|
| 30 |
+
"linear_attention",
|
| 31 |
+
"linear_attention",
|
| 32 |
+
"full_attention",
|
| 33 |
+
"linear_attention",
|
| 34 |
+
"linear_attention",
|
| 35 |
+
"linear_attention",
|
| 36 |
+
"full_attention",
|
| 37 |
+
"linear_attention",
|
| 38 |
+
"linear_attention",
|
| 39 |
+
"linear_attention",
|
| 40 |
+
"full_attention",
|
| 41 |
+
"linear_attention",
|
| 42 |
+
"linear_attention",
|
| 43 |
+
"linear_attention",
|
| 44 |
+
"full_attention",
|
| 45 |
+
"linear_attention",
|
| 46 |
+
"linear_attention",
|
| 47 |
+
"linear_attention",
|
| 48 |
+
"full_attention",
|
| 49 |
+
"linear_attention",
|
| 50 |
+
"linear_attention",
|
| 51 |
+
"linear_attention",
|
| 52 |
+
"full_attention",
|
| 53 |
+
"linear_attention",
|
| 54 |
+
"linear_attention",
|
| 55 |
+
"linear_attention",
|
| 56 |
+
"full_attention"
|
| 57 |
+
],
|
| 58 |
+
"linear_conv_kernel_dim": 4,
|
| 59 |
+
"linear_key_head_dim": 128,
|
| 60 |
+
"linear_num_key_heads": 16,
|
| 61 |
+
"linear_num_value_heads": 32,
|
| 62 |
+
"linear_value_head_dim": 128,
|
| 63 |
+
"mamba_ssm_dtype": "float32",
|
| 64 |
+
"max_position_embeddings": 262144,
|
| 65 |
+
"mlp_only_layers": [],
|
| 66 |
+
"model_type": "qwen3_5_text",
|
| 67 |
+
"mtp_num_hidden_layers": 1,
|
| 68 |
+
"mtp_use_dedicated_embeddings": false,
|
| 69 |
+
"num_attention_heads": 16,
|
| 70 |
+
"num_hidden_layers": 32,
|
| 71 |
+
"num_key_value_heads": 4,
|
| 72 |
+
"pad_token_id": 248044,
|
| 73 |
+
"partial_rotary_factor": 0.25,
|
| 74 |
+
"rms_norm_eps": 1e-06,
|
| 75 |
+
"rope_parameters": {
|
| 76 |
+
"mrope_interleaved": true,
|
| 77 |
+
"mrope_section": [
|
| 78 |
+
11,
|
| 79 |
+
11,
|
| 80 |
+
10
|
| 81 |
+
],
|
| 82 |
+
"partial_rotary_factor": 0.25,
|
| 83 |
+
"rope_theta": 10000000,
|
| 84 |
+
"rope_type": "default"
|
| 85 |
+
},
|
| 86 |
+
"tie_word_embeddings": false,
|
| 87 |
+
"use_cache": false,
|
| 88 |
+
"vocab_size": 248320
|
| 89 |
+
},
|
| 90 |
+
"tie_word_embeddings": false,
|
| 91 |
+
"transformers_version": "5.10.4",
|
| 92 |
+
"use_cache": false,
|
| 93 |
+
"video_token_id": 248057,
|
| 94 |
+
"vision_config": {
|
| 95 |
+
"deepstack_visual_indexes": [],
|
| 96 |
+
"depth": 27,
|
| 97 |
+
"dtype": "bfloat16",
|
| 98 |
+
"hidden_act": "gelu_pytorch_tanh",
|
| 99 |
+
"hidden_size": 1152,
|
| 100 |
+
"in_channels": 3,
|
| 101 |
+
"initializer_range": 0.02,
|
| 102 |
+
"intermediate_size": 4304,
|
| 103 |
+
"model_type": "qwen3_5_vision",
|
| 104 |
+
"num_heads": 16,
|
| 105 |
+
"num_position_embeddings": 2304,
|
| 106 |
+
"out_hidden_size": 4096,
|
| 107 |
+
"patch_size": 16,
|
| 108 |
+
"spatial_merge_size": 2,
|
| 109 |
+
"temporal_patch_size": 2
|
| 110 |
+
},
|
| 111 |
+
"vision_end_token_id": 248054,
|
| 112 |
+
"vision_start_token_id": 248053
|
| 113 |
+
}
|
c1_no_history/round_3/generation_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"eos_token_id": [
|
| 4 |
+
248044,
|
| 5 |
+
248046
|
| 6 |
+
],
|
| 7 |
+
"transformers_version": "5.10.4",
|
| 8 |
+
"use_cache": true
|
| 9 |
+
}
|
c1_no_history/round_3/model.safetensors.index.json
ADDED
|
@@ -0,0 +1,768 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"metadata": {
|
| 3 |
+
"total_parameters": 1469680,
|
| 4 |
+
"total_size": 18819627488
|
| 5 |
+
},
|
| 6 |
+
"weight_map": {
|
| 7 |
+
"lm_head.weight": "model-00001-of-00004.safetensors",
|
| 8 |
+
"model.language_model.embed_tokens.weight": "model-00001-of-00004.safetensors",
|
| 9 |
+
"model.language_model.layers.0.input_layernorm.weight": "model-00001-of-00004.safetensors",
|
| 10 |
+
"model.language_model.layers.0.linear_attn.A_log": "model-00001-of-00004.safetensors",
|
| 11 |
+
"model.language_model.layers.0.linear_attn.conv1d.weight": "model-00001-of-00004.safetensors",
|
| 12 |
+
"model.language_model.layers.0.linear_attn.dt_bias": "model-00001-of-00004.safetensors",
|
| 13 |
+
"model.language_model.layers.0.linear_attn.in_proj_a.weight": "model-00001-of-00004.safetensors",
|
| 14 |
+
"model.language_model.layers.0.linear_attn.in_proj_b.weight": "model-00001-of-00004.safetensors",
|
| 15 |
+
"model.language_model.layers.0.linear_attn.in_proj_qkv.weight": "model-00001-of-00004.safetensors",
|
| 16 |
+
"model.language_model.layers.0.linear_attn.in_proj_z.weight": "model-00001-of-00004.safetensors",
|
| 17 |
+
"model.language_model.layers.0.linear_attn.norm.weight": "model-00001-of-00004.safetensors",
|
| 18 |
+
"model.language_model.layers.0.linear_attn.out_proj.weight": "model-00001-of-00004.safetensors",
|
| 19 |
+
"model.language_model.layers.0.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
|
| 20 |
+
"model.language_model.layers.0.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
|
| 21 |
+
"model.language_model.layers.0.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
|
| 22 |
+
"model.language_model.layers.0.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
|
| 23 |
+
"model.language_model.layers.1.input_layernorm.weight": "model-00001-of-00004.safetensors",
|
| 24 |
+
"model.language_model.layers.1.linear_attn.A_log": "model-00001-of-00004.safetensors",
|
| 25 |
+
"model.language_model.layers.1.linear_attn.conv1d.weight": "model-00001-of-00004.safetensors",
|
| 26 |
+
"model.language_model.layers.1.linear_attn.dt_bias": "model-00001-of-00004.safetensors",
|
| 27 |
+
"model.language_model.layers.1.linear_attn.in_proj_a.weight": "model-00001-of-00004.safetensors",
|
| 28 |
+
"model.language_model.layers.1.linear_attn.in_proj_b.weight": "model-00001-of-00004.safetensors",
|
| 29 |
+
"model.language_model.layers.1.linear_attn.in_proj_qkv.weight": "model-00001-of-00004.safetensors",
|
| 30 |
+
"model.language_model.layers.1.linear_attn.in_proj_z.weight": "model-00001-of-00004.safetensors",
|
| 31 |
+
"model.language_model.layers.1.linear_attn.norm.weight": "model-00001-of-00004.safetensors",
|
| 32 |
+
"model.language_model.layers.1.linear_attn.out_proj.weight": "model-00001-of-00004.safetensors",
|
| 33 |
+
"model.language_model.layers.1.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
|
| 34 |
+
"model.language_model.layers.1.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
|
| 35 |
+
"model.language_model.layers.1.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
|
| 36 |
+
"model.language_model.layers.1.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
|
| 37 |
+
"model.language_model.layers.10.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 38 |
+
"model.language_model.layers.10.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 39 |
+
"model.language_model.layers.10.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 40 |
+
"model.language_model.layers.10.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 41 |
+
"model.language_model.layers.10.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 42 |
+
"model.language_model.layers.10.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 43 |
+
"model.language_model.layers.10.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 44 |
+
"model.language_model.layers.10.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 45 |
+
"model.language_model.layers.10.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 46 |
+
"model.language_model.layers.10.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 47 |
+
"model.language_model.layers.10.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 48 |
+
"model.language_model.layers.10.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 49 |
+
"model.language_model.layers.10.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 50 |
+
"model.language_model.layers.10.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 51 |
+
"model.language_model.layers.11.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 52 |
+
"model.language_model.layers.11.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 53 |
+
"model.language_model.layers.11.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 54 |
+
"model.language_model.layers.11.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 55 |
+
"model.language_model.layers.11.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 56 |
+
"model.language_model.layers.11.self_attn.k_norm.weight": "model-00002-of-00004.safetensors",
|
| 57 |
+
"model.language_model.layers.11.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
|
| 58 |
+
"model.language_model.layers.11.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
|
| 59 |
+
"model.language_model.layers.11.self_attn.q_norm.weight": "model-00002-of-00004.safetensors",
|
| 60 |
+
"model.language_model.layers.11.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
|
| 61 |
+
"model.language_model.layers.11.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
|
| 62 |
+
"model.language_model.layers.12.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 63 |
+
"model.language_model.layers.12.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 64 |
+
"model.language_model.layers.12.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 65 |
+
"model.language_model.layers.12.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 66 |
+
"model.language_model.layers.12.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 67 |
+
"model.language_model.layers.12.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 68 |
+
"model.language_model.layers.12.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 69 |
+
"model.language_model.layers.12.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 70 |
+
"model.language_model.layers.12.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 71 |
+
"model.language_model.layers.12.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 72 |
+
"model.language_model.layers.12.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 73 |
+
"model.language_model.layers.12.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 74 |
+
"model.language_model.layers.12.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 75 |
+
"model.language_model.layers.12.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 76 |
+
"model.language_model.layers.13.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 77 |
+
"model.language_model.layers.13.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 78 |
+
"model.language_model.layers.13.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 79 |
+
"model.language_model.layers.13.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 80 |
+
"model.language_model.layers.13.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 81 |
+
"model.language_model.layers.13.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 82 |
+
"model.language_model.layers.13.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 83 |
+
"model.language_model.layers.13.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 84 |
+
"model.language_model.layers.13.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 85 |
+
"model.language_model.layers.13.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 86 |
+
"model.language_model.layers.13.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 87 |
+
"model.language_model.layers.13.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 88 |
+
"model.language_model.layers.13.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 89 |
+
"model.language_model.layers.13.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 90 |
+
"model.language_model.layers.14.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 91 |
+
"model.language_model.layers.14.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 92 |
+
"model.language_model.layers.14.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 93 |
+
"model.language_model.layers.14.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 94 |
+
"model.language_model.layers.14.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 95 |
+
"model.language_model.layers.14.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 96 |
+
"model.language_model.layers.14.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 97 |
+
"model.language_model.layers.14.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 98 |
+
"model.language_model.layers.14.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 99 |
+
"model.language_model.layers.14.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 100 |
+
"model.language_model.layers.14.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 101 |
+
"model.language_model.layers.14.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 102 |
+
"model.language_model.layers.14.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 103 |
+
"model.language_model.layers.14.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 104 |
+
"model.language_model.layers.15.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 105 |
+
"model.language_model.layers.15.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 106 |
+
"model.language_model.layers.15.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 107 |
+
"model.language_model.layers.15.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 108 |
+
"model.language_model.layers.15.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 109 |
+
"model.language_model.layers.15.self_attn.k_norm.weight": "model-00003-of-00004.safetensors",
|
| 110 |
+
"model.language_model.layers.15.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
|
| 111 |
+
"model.language_model.layers.15.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
|
| 112 |
+
"model.language_model.layers.15.self_attn.q_norm.weight": "model-00003-of-00004.safetensors",
|
| 113 |
+
"model.language_model.layers.15.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
|
| 114 |
+
"model.language_model.layers.15.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
|
| 115 |
+
"model.language_model.layers.16.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 116 |
+
"model.language_model.layers.16.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 117 |
+
"model.language_model.layers.16.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 118 |
+
"model.language_model.layers.16.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 119 |
+
"model.language_model.layers.16.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 120 |
+
"model.language_model.layers.16.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 121 |
+
"model.language_model.layers.16.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 122 |
+
"model.language_model.layers.16.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 123 |
+
"model.language_model.layers.16.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 124 |
+
"model.language_model.layers.16.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 125 |
+
"model.language_model.layers.16.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 126 |
+
"model.language_model.layers.16.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 127 |
+
"model.language_model.layers.16.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 128 |
+
"model.language_model.layers.16.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 129 |
+
"model.language_model.layers.17.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 130 |
+
"model.language_model.layers.17.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 131 |
+
"model.language_model.layers.17.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 132 |
+
"model.language_model.layers.17.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 133 |
+
"model.language_model.layers.17.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 134 |
+
"model.language_model.layers.17.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 135 |
+
"model.language_model.layers.17.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 136 |
+
"model.language_model.layers.17.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 137 |
+
"model.language_model.layers.17.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 138 |
+
"model.language_model.layers.17.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 139 |
+
"model.language_model.layers.17.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 140 |
+
"model.language_model.layers.17.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 141 |
+
"model.language_model.layers.17.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 142 |
+
"model.language_model.layers.17.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 143 |
+
"model.language_model.layers.18.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 144 |
+
"model.language_model.layers.18.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 145 |
+
"model.language_model.layers.18.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 146 |
+
"model.language_model.layers.18.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 147 |
+
"model.language_model.layers.18.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 148 |
+
"model.language_model.layers.18.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 149 |
+
"model.language_model.layers.18.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 150 |
+
"model.language_model.layers.18.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 151 |
+
"model.language_model.layers.18.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 152 |
+
"model.language_model.layers.18.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 153 |
+
"model.language_model.layers.18.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 154 |
+
"model.language_model.layers.18.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 155 |
+
"model.language_model.layers.18.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 156 |
+
"model.language_model.layers.18.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 157 |
+
"model.language_model.layers.19.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 158 |
+
"model.language_model.layers.19.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 159 |
+
"model.language_model.layers.19.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 160 |
+
"model.language_model.layers.19.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 161 |
+
"model.language_model.layers.19.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 162 |
+
"model.language_model.layers.19.self_attn.k_norm.weight": "model-00003-of-00004.safetensors",
|
| 163 |
+
"model.language_model.layers.19.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
|
| 164 |
+
"model.language_model.layers.19.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
|
| 165 |
+
"model.language_model.layers.19.self_attn.q_norm.weight": "model-00003-of-00004.safetensors",
|
| 166 |
+
"model.language_model.layers.19.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
|
| 167 |
+
"model.language_model.layers.19.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
|
| 168 |
+
"model.language_model.layers.2.input_layernorm.weight": "model-00001-of-00004.safetensors",
|
| 169 |
+
"model.language_model.layers.2.linear_attn.A_log": "model-00001-of-00004.safetensors",
|
| 170 |
+
"model.language_model.layers.2.linear_attn.conv1d.weight": "model-00001-of-00004.safetensors",
|
| 171 |
+
"model.language_model.layers.2.linear_attn.dt_bias": "model-00001-of-00004.safetensors",
|
| 172 |
+
"model.language_model.layers.2.linear_attn.in_proj_a.weight": "model-00001-of-00004.safetensors",
|
| 173 |
+
"model.language_model.layers.2.linear_attn.in_proj_b.weight": "model-00001-of-00004.safetensors",
|
| 174 |
+
"model.language_model.layers.2.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 175 |
+
"model.language_model.layers.2.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 176 |
+
"model.language_model.layers.2.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 177 |
+
"model.language_model.layers.2.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 178 |
+
"model.language_model.layers.2.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 179 |
+
"model.language_model.layers.2.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 180 |
+
"model.language_model.layers.2.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 181 |
+
"model.language_model.layers.2.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 182 |
+
"model.language_model.layers.20.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 183 |
+
"model.language_model.layers.20.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 184 |
+
"model.language_model.layers.20.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 185 |
+
"model.language_model.layers.20.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 186 |
+
"model.language_model.layers.20.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 187 |
+
"model.language_model.layers.20.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 188 |
+
"model.language_model.layers.20.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 189 |
+
"model.language_model.layers.20.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 190 |
+
"model.language_model.layers.20.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 191 |
+
"model.language_model.layers.20.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 192 |
+
"model.language_model.layers.20.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 193 |
+
"model.language_model.layers.20.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 194 |
+
"model.language_model.layers.20.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 195 |
+
"model.language_model.layers.20.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 196 |
+
"model.language_model.layers.21.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 197 |
+
"model.language_model.layers.21.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 198 |
+
"model.language_model.layers.21.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 199 |
+
"model.language_model.layers.21.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 200 |
+
"model.language_model.layers.21.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 201 |
+
"model.language_model.layers.21.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 202 |
+
"model.language_model.layers.21.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 203 |
+
"model.language_model.layers.21.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 204 |
+
"model.language_model.layers.21.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 205 |
+
"model.language_model.layers.21.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 206 |
+
"model.language_model.layers.21.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 207 |
+
"model.language_model.layers.21.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 208 |
+
"model.language_model.layers.21.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 209 |
+
"model.language_model.layers.21.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 210 |
+
"model.language_model.layers.22.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 211 |
+
"model.language_model.layers.22.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 212 |
+
"model.language_model.layers.22.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 213 |
+
"model.language_model.layers.22.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 214 |
+
"model.language_model.layers.22.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 215 |
+
"model.language_model.layers.22.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 216 |
+
"model.language_model.layers.22.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 217 |
+
"model.language_model.layers.22.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 218 |
+
"model.language_model.layers.22.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 219 |
+
"model.language_model.layers.22.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 220 |
+
"model.language_model.layers.22.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 221 |
+
"model.language_model.layers.22.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 222 |
+
"model.language_model.layers.22.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 223 |
+
"model.language_model.layers.22.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 224 |
+
"model.language_model.layers.23.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 225 |
+
"model.language_model.layers.23.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 226 |
+
"model.language_model.layers.23.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 227 |
+
"model.language_model.layers.23.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 228 |
+
"model.language_model.layers.23.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 229 |
+
"model.language_model.layers.23.self_attn.k_norm.weight": "model-00003-of-00004.safetensors",
|
| 230 |
+
"model.language_model.layers.23.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
|
| 231 |
+
"model.language_model.layers.23.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
|
| 232 |
+
"model.language_model.layers.23.self_attn.q_norm.weight": "model-00003-of-00004.safetensors",
|
| 233 |
+
"model.language_model.layers.23.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
|
| 234 |
+
"model.language_model.layers.23.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
|
| 235 |
+
"model.language_model.layers.24.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 236 |
+
"model.language_model.layers.24.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 237 |
+
"model.language_model.layers.24.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 238 |
+
"model.language_model.layers.24.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 239 |
+
"model.language_model.layers.24.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 240 |
+
"model.language_model.layers.24.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 241 |
+
"model.language_model.layers.24.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 242 |
+
"model.language_model.layers.24.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 243 |
+
"model.language_model.layers.24.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 244 |
+
"model.language_model.layers.24.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 245 |
+
"model.language_model.layers.24.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 246 |
+
"model.language_model.layers.24.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 247 |
+
"model.language_model.layers.24.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 248 |
+
"model.language_model.layers.24.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 249 |
+
"model.language_model.layers.25.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 250 |
+
"model.language_model.layers.25.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 251 |
+
"model.language_model.layers.25.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 252 |
+
"model.language_model.layers.25.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 253 |
+
"model.language_model.layers.25.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 254 |
+
"model.language_model.layers.25.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 255 |
+
"model.language_model.layers.25.linear_attn.in_proj_qkv.weight": "model-00004-of-00004.safetensors",
|
| 256 |
+
"model.language_model.layers.25.linear_attn.in_proj_z.weight": "model-00004-of-00004.safetensors",
|
| 257 |
+
"model.language_model.layers.25.linear_attn.norm.weight": "model-00004-of-00004.safetensors",
|
| 258 |
+
"model.language_model.layers.25.linear_attn.out_proj.weight": "model-00004-of-00004.safetensors",
|
| 259 |
+
"model.language_model.layers.25.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 260 |
+
"model.language_model.layers.25.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 261 |
+
"model.language_model.layers.25.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 262 |
+
"model.language_model.layers.25.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 263 |
+
"model.language_model.layers.26.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 264 |
+
"model.language_model.layers.26.linear_attn.A_log": "model-00004-of-00004.safetensors",
|
| 265 |
+
"model.language_model.layers.26.linear_attn.conv1d.weight": "model-00004-of-00004.safetensors",
|
| 266 |
+
"model.language_model.layers.26.linear_attn.dt_bias": "model-00004-of-00004.safetensors",
|
| 267 |
+
"model.language_model.layers.26.linear_attn.in_proj_a.weight": "model-00004-of-00004.safetensors",
|
| 268 |
+
"model.language_model.layers.26.linear_attn.in_proj_b.weight": "model-00004-of-00004.safetensors",
|
| 269 |
+
"model.language_model.layers.26.linear_attn.in_proj_qkv.weight": "model-00004-of-00004.safetensors",
|
| 270 |
+
"model.language_model.layers.26.linear_attn.in_proj_z.weight": "model-00004-of-00004.safetensors",
|
| 271 |
+
"model.language_model.layers.26.linear_attn.norm.weight": "model-00004-of-00004.safetensors",
|
| 272 |
+
"model.language_model.layers.26.linear_attn.out_proj.weight": "model-00004-of-00004.safetensors",
|
| 273 |
+
"model.language_model.layers.26.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 274 |
+
"model.language_model.layers.26.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 275 |
+
"model.language_model.layers.26.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 276 |
+
"model.language_model.layers.26.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 277 |
+
"model.language_model.layers.27.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 278 |
+
"model.language_model.layers.27.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 279 |
+
"model.language_model.layers.27.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 280 |
+
"model.language_model.layers.27.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 281 |
+
"model.language_model.layers.27.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 282 |
+
"model.language_model.layers.27.self_attn.k_norm.weight": "model-00004-of-00004.safetensors",
|
| 283 |
+
"model.language_model.layers.27.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
|
| 284 |
+
"model.language_model.layers.27.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
|
| 285 |
+
"model.language_model.layers.27.self_attn.q_norm.weight": "model-00004-of-00004.safetensors",
|
| 286 |
+
"model.language_model.layers.27.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
|
| 287 |
+
"model.language_model.layers.27.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
|
| 288 |
+
"model.language_model.layers.28.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 289 |
+
"model.language_model.layers.28.linear_attn.A_log": "model-00004-of-00004.safetensors",
|
| 290 |
+
"model.language_model.layers.28.linear_attn.conv1d.weight": "model-00004-of-00004.safetensors",
|
| 291 |
+
"model.language_model.layers.28.linear_attn.dt_bias": "model-00004-of-00004.safetensors",
|
| 292 |
+
"model.language_model.layers.28.linear_attn.in_proj_a.weight": "model-00004-of-00004.safetensors",
|
| 293 |
+
"model.language_model.layers.28.linear_attn.in_proj_b.weight": "model-00004-of-00004.safetensors",
|
| 294 |
+
"model.language_model.layers.28.linear_attn.in_proj_qkv.weight": "model-00004-of-00004.safetensors",
|
| 295 |
+
"model.language_model.layers.28.linear_attn.in_proj_z.weight": "model-00004-of-00004.safetensors",
|
| 296 |
+
"model.language_model.layers.28.linear_attn.norm.weight": "model-00004-of-00004.safetensors",
|
| 297 |
+
"model.language_model.layers.28.linear_attn.out_proj.weight": "model-00004-of-00004.safetensors",
|
| 298 |
+
"model.language_model.layers.28.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 299 |
+
"model.language_model.layers.28.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 300 |
+
"model.language_model.layers.28.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 301 |
+
"model.language_model.layers.28.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 302 |
+
"model.language_model.layers.29.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 303 |
+
"model.language_model.layers.29.linear_attn.A_log": "model-00004-of-00004.safetensors",
|
| 304 |
+
"model.language_model.layers.29.linear_attn.conv1d.weight": "model-00004-of-00004.safetensors",
|
| 305 |
+
"model.language_model.layers.29.linear_attn.dt_bias": "model-00004-of-00004.safetensors",
|
| 306 |
+
"model.language_model.layers.29.linear_attn.in_proj_a.weight": "model-00004-of-00004.safetensors",
|
| 307 |
+
"model.language_model.layers.29.linear_attn.in_proj_b.weight": "model-00004-of-00004.safetensors",
|
| 308 |
+
"model.language_model.layers.29.linear_attn.in_proj_qkv.weight": "model-00004-of-00004.safetensors",
|
| 309 |
+
"model.language_model.layers.29.linear_attn.in_proj_z.weight": "model-00004-of-00004.safetensors",
|
| 310 |
+
"model.language_model.layers.29.linear_attn.norm.weight": "model-00004-of-00004.safetensors",
|
| 311 |
+
"model.language_model.layers.29.linear_attn.out_proj.weight": "model-00004-of-00004.safetensors",
|
| 312 |
+
"model.language_model.layers.29.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 313 |
+
"model.language_model.layers.29.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 314 |
+
"model.language_model.layers.29.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 315 |
+
"model.language_model.layers.29.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 316 |
+
"model.language_model.layers.3.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 317 |
+
"model.language_model.layers.3.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 318 |
+
"model.language_model.layers.3.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 319 |
+
"model.language_model.layers.3.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 320 |
+
"model.language_model.layers.3.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 321 |
+
"model.language_model.layers.3.self_attn.k_norm.weight": "model-00002-of-00004.safetensors",
|
| 322 |
+
"model.language_model.layers.3.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
|
| 323 |
+
"model.language_model.layers.3.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
|
| 324 |
+
"model.language_model.layers.3.self_attn.q_norm.weight": "model-00002-of-00004.safetensors",
|
| 325 |
+
"model.language_model.layers.3.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
|
| 326 |
+
"model.language_model.layers.3.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
|
| 327 |
+
"model.language_model.layers.30.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 328 |
+
"model.language_model.layers.30.linear_attn.A_log": "model-00004-of-00004.safetensors",
|
| 329 |
+
"model.language_model.layers.30.linear_attn.conv1d.weight": "model-00004-of-00004.safetensors",
|
| 330 |
+
"model.language_model.layers.30.linear_attn.dt_bias": "model-00004-of-00004.safetensors",
|
| 331 |
+
"model.language_model.layers.30.linear_attn.in_proj_a.weight": "model-00004-of-00004.safetensors",
|
| 332 |
+
"model.language_model.layers.30.linear_attn.in_proj_b.weight": "model-00004-of-00004.safetensors",
|
| 333 |
+
"model.language_model.layers.30.linear_attn.in_proj_qkv.weight": "model-00004-of-00004.safetensors",
|
| 334 |
+
"model.language_model.layers.30.linear_attn.in_proj_z.weight": "model-00004-of-00004.safetensors",
|
| 335 |
+
"model.language_model.layers.30.linear_attn.norm.weight": "model-00004-of-00004.safetensors",
|
| 336 |
+
"model.language_model.layers.30.linear_attn.out_proj.weight": "model-00004-of-00004.safetensors",
|
| 337 |
+
"model.language_model.layers.30.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 338 |
+
"model.language_model.layers.30.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 339 |
+
"model.language_model.layers.30.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 340 |
+
"model.language_model.layers.30.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 341 |
+
"model.language_model.layers.31.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 342 |
+
"model.language_model.layers.31.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 343 |
+
"model.language_model.layers.31.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 344 |
+
"model.language_model.layers.31.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 345 |
+
"model.language_model.layers.31.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 346 |
+
"model.language_model.layers.31.self_attn.k_norm.weight": "model-00004-of-00004.safetensors",
|
| 347 |
+
"model.language_model.layers.31.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
|
| 348 |
+
"model.language_model.layers.31.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
|
| 349 |
+
"model.language_model.layers.31.self_attn.q_norm.weight": "model-00004-of-00004.safetensors",
|
| 350 |
+
"model.language_model.layers.31.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
|
| 351 |
+
"model.language_model.layers.31.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
|
| 352 |
+
"model.language_model.layers.4.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 353 |
+
"model.language_model.layers.4.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 354 |
+
"model.language_model.layers.4.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 355 |
+
"model.language_model.layers.4.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 356 |
+
"model.language_model.layers.4.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 357 |
+
"model.language_model.layers.4.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 358 |
+
"model.language_model.layers.4.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 359 |
+
"model.language_model.layers.4.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 360 |
+
"model.language_model.layers.4.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 361 |
+
"model.language_model.layers.4.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 362 |
+
"model.language_model.layers.4.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 363 |
+
"model.language_model.layers.4.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 364 |
+
"model.language_model.layers.4.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 365 |
+
"model.language_model.layers.4.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 366 |
+
"model.language_model.layers.5.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 367 |
+
"model.language_model.layers.5.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 368 |
+
"model.language_model.layers.5.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 369 |
+
"model.language_model.layers.5.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 370 |
+
"model.language_model.layers.5.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 371 |
+
"model.language_model.layers.5.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 372 |
+
"model.language_model.layers.5.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 373 |
+
"model.language_model.layers.5.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 374 |
+
"model.language_model.layers.5.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 375 |
+
"model.language_model.layers.5.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 376 |
+
"model.language_model.layers.5.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 377 |
+
"model.language_model.layers.5.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 378 |
+
"model.language_model.layers.5.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 379 |
+
"model.language_model.layers.5.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 380 |
+
"model.language_model.layers.6.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 381 |
+
"model.language_model.layers.6.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 382 |
+
"model.language_model.layers.6.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 383 |
+
"model.language_model.layers.6.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 384 |
+
"model.language_model.layers.6.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 385 |
+
"model.language_model.layers.6.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 386 |
+
"model.language_model.layers.6.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 387 |
+
"model.language_model.layers.6.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 388 |
+
"model.language_model.layers.6.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 389 |
+
"model.language_model.layers.6.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 390 |
+
"model.language_model.layers.6.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 391 |
+
"model.language_model.layers.6.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 392 |
+
"model.language_model.layers.6.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 393 |
+
"model.language_model.layers.6.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 394 |
+
"model.language_model.layers.7.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 395 |
+
"model.language_model.layers.7.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 396 |
+
"model.language_model.layers.7.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 397 |
+
"model.language_model.layers.7.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 398 |
+
"model.language_model.layers.7.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 399 |
+
"model.language_model.layers.7.self_attn.k_norm.weight": "model-00002-of-00004.safetensors",
|
| 400 |
+
"model.language_model.layers.7.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
|
| 401 |
+
"model.language_model.layers.7.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
|
| 402 |
+
"model.language_model.layers.7.self_attn.q_norm.weight": "model-00002-of-00004.safetensors",
|
| 403 |
+
"model.language_model.layers.7.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
|
| 404 |
+
"model.language_model.layers.7.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
|
| 405 |
+
"model.language_model.layers.8.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 406 |
+
"model.language_model.layers.8.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 407 |
+
"model.language_model.layers.8.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 408 |
+
"model.language_model.layers.8.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 409 |
+
"model.language_model.layers.8.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 410 |
+
"model.language_model.layers.8.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 411 |
+
"model.language_model.layers.8.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 412 |
+
"model.language_model.layers.8.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 413 |
+
"model.language_model.layers.8.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 414 |
+
"model.language_model.layers.8.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 415 |
+
"model.language_model.layers.8.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 416 |
+
"model.language_model.layers.8.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 417 |
+
"model.language_model.layers.8.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 418 |
+
"model.language_model.layers.8.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 419 |
+
"model.language_model.layers.9.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 420 |
+
"model.language_model.layers.9.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 421 |
+
"model.language_model.layers.9.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 422 |
+
"model.language_model.layers.9.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 423 |
+
"model.language_model.layers.9.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 424 |
+
"model.language_model.layers.9.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 425 |
+
"model.language_model.layers.9.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 426 |
+
"model.language_model.layers.9.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 427 |
+
"model.language_model.layers.9.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 428 |
+
"model.language_model.layers.9.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 429 |
+
"model.language_model.layers.9.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 430 |
+
"model.language_model.layers.9.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 431 |
+
"model.language_model.layers.9.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 432 |
+
"model.language_model.layers.9.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 433 |
+
"model.language_model.norm.weight": "model-00004-of-00004.safetensors",
|
| 434 |
+
"model.visual.blocks.0.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 435 |
+
"model.visual.blocks.0.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 436 |
+
"model.visual.blocks.0.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 437 |
+
"model.visual.blocks.0.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 438 |
+
"model.visual.blocks.0.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 439 |
+
"model.visual.blocks.0.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 440 |
+
"model.visual.blocks.0.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 441 |
+
"model.visual.blocks.0.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 442 |
+
"model.visual.blocks.0.norm1.bias": "model-00004-of-00004.safetensors",
|
| 443 |
+
"model.visual.blocks.0.norm1.weight": "model-00004-of-00004.safetensors",
|
| 444 |
+
"model.visual.blocks.0.norm2.bias": "model-00004-of-00004.safetensors",
|
| 445 |
+
"model.visual.blocks.0.norm2.weight": "model-00004-of-00004.safetensors",
|
| 446 |
+
"model.visual.blocks.1.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 447 |
+
"model.visual.blocks.1.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 448 |
+
"model.visual.blocks.1.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 449 |
+
"model.visual.blocks.1.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 450 |
+
"model.visual.blocks.1.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 451 |
+
"model.visual.blocks.1.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 452 |
+
"model.visual.blocks.1.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 453 |
+
"model.visual.blocks.1.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 454 |
+
"model.visual.blocks.1.norm1.bias": "model-00004-of-00004.safetensors",
|
| 455 |
+
"model.visual.blocks.1.norm1.weight": "model-00004-of-00004.safetensors",
|
| 456 |
+
"model.visual.blocks.1.norm2.bias": "model-00004-of-00004.safetensors",
|
| 457 |
+
"model.visual.blocks.1.norm2.weight": "model-00004-of-00004.safetensors",
|
| 458 |
+
"model.visual.blocks.10.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 459 |
+
"model.visual.blocks.10.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 460 |
+
"model.visual.blocks.10.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 461 |
+
"model.visual.blocks.10.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 462 |
+
"model.visual.blocks.10.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 463 |
+
"model.visual.blocks.10.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 464 |
+
"model.visual.blocks.10.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 465 |
+
"model.visual.blocks.10.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 466 |
+
"model.visual.blocks.10.norm1.bias": "model-00004-of-00004.safetensors",
|
| 467 |
+
"model.visual.blocks.10.norm1.weight": "model-00004-of-00004.safetensors",
|
| 468 |
+
"model.visual.blocks.10.norm2.bias": "model-00004-of-00004.safetensors",
|
| 469 |
+
"model.visual.blocks.10.norm2.weight": "model-00004-of-00004.safetensors",
|
| 470 |
+
"model.visual.blocks.11.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 471 |
+
"model.visual.blocks.11.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 472 |
+
"model.visual.blocks.11.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 473 |
+
"model.visual.blocks.11.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 474 |
+
"model.visual.blocks.11.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 475 |
+
"model.visual.blocks.11.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 476 |
+
"model.visual.blocks.11.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 477 |
+
"model.visual.blocks.11.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 478 |
+
"model.visual.blocks.11.norm1.bias": "model-00004-of-00004.safetensors",
|
| 479 |
+
"model.visual.blocks.11.norm1.weight": "model-00004-of-00004.safetensors",
|
| 480 |
+
"model.visual.blocks.11.norm2.bias": "model-00004-of-00004.safetensors",
|
| 481 |
+
"model.visual.blocks.11.norm2.weight": "model-00004-of-00004.safetensors",
|
| 482 |
+
"model.visual.blocks.12.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 483 |
+
"model.visual.blocks.12.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 484 |
+
"model.visual.blocks.12.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 485 |
+
"model.visual.blocks.12.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 486 |
+
"model.visual.blocks.12.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 487 |
+
"model.visual.blocks.12.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 488 |
+
"model.visual.blocks.12.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 489 |
+
"model.visual.blocks.12.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 490 |
+
"model.visual.blocks.12.norm1.bias": "model-00004-of-00004.safetensors",
|
| 491 |
+
"model.visual.blocks.12.norm1.weight": "model-00004-of-00004.safetensors",
|
| 492 |
+
"model.visual.blocks.12.norm2.bias": "model-00004-of-00004.safetensors",
|
| 493 |
+
"model.visual.blocks.12.norm2.weight": "model-00004-of-00004.safetensors",
|
| 494 |
+
"model.visual.blocks.13.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 495 |
+
"model.visual.blocks.13.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 496 |
+
"model.visual.blocks.13.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 497 |
+
"model.visual.blocks.13.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 498 |
+
"model.visual.blocks.13.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 499 |
+
"model.visual.blocks.13.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 500 |
+
"model.visual.blocks.13.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 501 |
+
"model.visual.blocks.13.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 502 |
+
"model.visual.blocks.13.norm1.bias": "model-00004-of-00004.safetensors",
|
| 503 |
+
"model.visual.blocks.13.norm1.weight": "model-00004-of-00004.safetensors",
|
| 504 |
+
"model.visual.blocks.13.norm2.bias": "model-00004-of-00004.safetensors",
|
| 505 |
+
"model.visual.blocks.13.norm2.weight": "model-00004-of-00004.safetensors",
|
| 506 |
+
"model.visual.blocks.14.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 507 |
+
"model.visual.blocks.14.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 508 |
+
"model.visual.blocks.14.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 509 |
+
"model.visual.blocks.14.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 510 |
+
"model.visual.blocks.14.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 511 |
+
"model.visual.blocks.14.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 512 |
+
"model.visual.blocks.14.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 513 |
+
"model.visual.blocks.14.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 514 |
+
"model.visual.blocks.14.norm1.bias": "model-00004-of-00004.safetensors",
|
| 515 |
+
"model.visual.blocks.14.norm1.weight": "model-00004-of-00004.safetensors",
|
| 516 |
+
"model.visual.blocks.14.norm2.bias": "model-00004-of-00004.safetensors",
|
| 517 |
+
"model.visual.blocks.14.norm2.weight": "model-00004-of-00004.safetensors",
|
| 518 |
+
"model.visual.blocks.15.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 519 |
+
"model.visual.blocks.15.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 520 |
+
"model.visual.blocks.15.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 521 |
+
"model.visual.blocks.15.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 522 |
+
"model.visual.blocks.15.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 523 |
+
"model.visual.blocks.15.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 524 |
+
"model.visual.blocks.15.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 525 |
+
"model.visual.blocks.15.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 526 |
+
"model.visual.blocks.15.norm1.bias": "model-00004-of-00004.safetensors",
|
| 527 |
+
"model.visual.blocks.15.norm1.weight": "model-00004-of-00004.safetensors",
|
| 528 |
+
"model.visual.blocks.15.norm2.bias": "model-00004-of-00004.safetensors",
|
| 529 |
+
"model.visual.blocks.15.norm2.weight": "model-00004-of-00004.safetensors",
|
| 530 |
+
"model.visual.blocks.16.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 531 |
+
"model.visual.blocks.16.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 532 |
+
"model.visual.blocks.16.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 533 |
+
"model.visual.blocks.16.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 534 |
+
"model.visual.blocks.16.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 535 |
+
"model.visual.blocks.16.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 536 |
+
"model.visual.blocks.16.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 537 |
+
"model.visual.blocks.16.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 538 |
+
"model.visual.blocks.16.norm1.bias": "model-00004-of-00004.safetensors",
|
| 539 |
+
"model.visual.blocks.16.norm1.weight": "model-00004-of-00004.safetensors",
|
| 540 |
+
"model.visual.blocks.16.norm2.bias": "model-00004-of-00004.safetensors",
|
| 541 |
+
"model.visual.blocks.16.norm2.weight": "model-00004-of-00004.safetensors",
|
| 542 |
+
"model.visual.blocks.17.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 543 |
+
"model.visual.blocks.17.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 544 |
+
"model.visual.blocks.17.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 545 |
+
"model.visual.blocks.17.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 546 |
+
"model.visual.blocks.17.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 547 |
+
"model.visual.blocks.17.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 548 |
+
"model.visual.blocks.17.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 549 |
+
"model.visual.blocks.17.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 550 |
+
"model.visual.blocks.17.norm1.bias": "model-00004-of-00004.safetensors",
|
| 551 |
+
"model.visual.blocks.17.norm1.weight": "model-00004-of-00004.safetensors",
|
| 552 |
+
"model.visual.blocks.17.norm2.bias": "model-00004-of-00004.safetensors",
|
| 553 |
+
"model.visual.blocks.17.norm2.weight": "model-00004-of-00004.safetensors",
|
| 554 |
+
"model.visual.blocks.18.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 555 |
+
"model.visual.blocks.18.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 556 |
+
"model.visual.blocks.18.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 557 |
+
"model.visual.blocks.18.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 558 |
+
"model.visual.blocks.18.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 559 |
+
"model.visual.blocks.18.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 560 |
+
"model.visual.blocks.18.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 561 |
+
"model.visual.blocks.18.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 562 |
+
"model.visual.blocks.18.norm1.bias": "model-00004-of-00004.safetensors",
|
| 563 |
+
"model.visual.blocks.18.norm1.weight": "model-00004-of-00004.safetensors",
|
| 564 |
+
"model.visual.blocks.18.norm2.bias": "model-00004-of-00004.safetensors",
|
| 565 |
+
"model.visual.blocks.18.norm2.weight": "model-00004-of-00004.safetensors",
|
| 566 |
+
"model.visual.blocks.19.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 567 |
+
"model.visual.blocks.19.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 568 |
+
"model.visual.blocks.19.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 569 |
+
"model.visual.blocks.19.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 570 |
+
"model.visual.blocks.19.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 571 |
+
"model.visual.blocks.19.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 572 |
+
"model.visual.blocks.19.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 573 |
+
"model.visual.blocks.19.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 574 |
+
"model.visual.blocks.19.norm1.bias": "model-00004-of-00004.safetensors",
|
| 575 |
+
"model.visual.blocks.19.norm1.weight": "model-00004-of-00004.safetensors",
|
| 576 |
+
"model.visual.blocks.19.norm2.bias": "model-00004-of-00004.safetensors",
|
| 577 |
+
"model.visual.blocks.19.norm2.weight": "model-00004-of-00004.safetensors",
|
| 578 |
+
"model.visual.blocks.2.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 579 |
+
"model.visual.blocks.2.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 580 |
+
"model.visual.blocks.2.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 581 |
+
"model.visual.blocks.2.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 582 |
+
"model.visual.blocks.2.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 583 |
+
"model.visual.blocks.2.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 584 |
+
"model.visual.blocks.2.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 585 |
+
"model.visual.blocks.2.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 586 |
+
"model.visual.blocks.2.norm1.bias": "model-00004-of-00004.safetensors",
|
| 587 |
+
"model.visual.blocks.2.norm1.weight": "model-00004-of-00004.safetensors",
|
| 588 |
+
"model.visual.blocks.2.norm2.bias": "model-00004-of-00004.safetensors",
|
| 589 |
+
"model.visual.blocks.2.norm2.weight": "model-00004-of-00004.safetensors",
|
| 590 |
+
"model.visual.blocks.20.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 591 |
+
"model.visual.blocks.20.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 592 |
+
"model.visual.blocks.20.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 593 |
+
"model.visual.blocks.20.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 594 |
+
"model.visual.blocks.20.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 595 |
+
"model.visual.blocks.20.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 596 |
+
"model.visual.blocks.20.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 597 |
+
"model.visual.blocks.20.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 598 |
+
"model.visual.blocks.20.norm1.bias": "model-00004-of-00004.safetensors",
|
| 599 |
+
"model.visual.blocks.20.norm1.weight": "model-00004-of-00004.safetensors",
|
| 600 |
+
"model.visual.blocks.20.norm2.bias": "model-00004-of-00004.safetensors",
|
| 601 |
+
"model.visual.blocks.20.norm2.weight": "model-00004-of-00004.safetensors",
|
| 602 |
+
"model.visual.blocks.21.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 603 |
+
"model.visual.blocks.21.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 604 |
+
"model.visual.blocks.21.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 605 |
+
"model.visual.blocks.21.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 606 |
+
"model.visual.blocks.21.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 607 |
+
"model.visual.blocks.21.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 608 |
+
"model.visual.blocks.21.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 609 |
+
"model.visual.blocks.21.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 610 |
+
"model.visual.blocks.21.norm1.bias": "model-00004-of-00004.safetensors",
|
| 611 |
+
"model.visual.blocks.21.norm1.weight": "model-00004-of-00004.safetensors",
|
| 612 |
+
"model.visual.blocks.21.norm2.bias": "model-00004-of-00004.safetensors",
|
| 613 |
+
"model.visual.blocks.21.norm2.weight": "model-00004-of-00004.safetensors",
|
| 614 |
+
"model.visual.blocks.22.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 615 |
+
"model.visual.blocks.22.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 616 |
+
"model.visual.blocks.22.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 617 |
+
"model.visual.blocks.22.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 618 |
+
"model.visual.blocks.22.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 619 |
+
"model.visual.blocks.22.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 620 |
+
"model.visual.blocks.22.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 621 |
+
"model.visual.blocks.22.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 622 |
+
"model.visual.blocks.22.norm1.bias": "model-00004-of-00004.safetensors",
|
| 623 |
+
"model.visual.blocks.22.norm1.weight": "model-00004-of-00004.safetensors",
|
| 624 |
+
"model.visual.blocks.22.norm2.bias": "model-00004-of-00004.safetensors",
|
| 625 |
+
"model.visual.blocks.22.norm2.weight": "model-00004-of-00004.safetensors",
|
| 626 |
+
"model.visual.blocks.23.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 627 |
+
"model.visual.blocks.23.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 628 |
+
"model.visual.blocks.23.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 629 |
+
"model.visual.blocks.23.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 630 |
+
"model.visual.blocks.23.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 631 |
+
"model.visual.blocks.23.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 632 |
+
"model.visual.blocks.23.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 633 |
+
"model.visual.blocks.23.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 634 |
+
"model.visual.blocks.23.norm1.bias": "model-00004-of-00004.safetensors",
|
| 635 |
+
"model.visual.blocks.23.norm1.weight": "model-00004-of-00004.safetensors",
|
| 636 |
+
"model.visual.blocks.23.norm2.bias": "model-00004-of-00004.safetensors",
|
| 637 |
+
"model.visual.blocks.23.norm2.weight": "model-00004-of-00004.safetensors",
|
| 638 |
+
"model.visual.blocks.24.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 639 |
+
"model.visual.blocks.24.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 640 |
+
"model.visual.blocks.24.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 641 |
+
"model.visual.blocks.24.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 642 |
+
"model.visual.blocks.24.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 643 |
+
"model.visual.blocks.24.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 644 |
+
"model.visual.blocks.24.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 645 |
+
"model.visual.blocks.24.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 646 |
+
"model.visual.blocks.24.norm1.bias": "model-00004-of-00004.safetensors",
|
| 647 |
+
"model.visual.blocks.24.norm1.weight": "model-00004-of-00004.safetensors",
|
| 648 |
+
"model.visual.blocks.24.norm2.bias": "model-00004-of-00004.safetensors",
|
| 649 |
+
"model.visual.blocks.24.norm2.weight": "model-00004-of-00004.safetensors",
|
| 650 |
+
"model.visual.blocks.25.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 651 |
+
"model.visual.blocks.25.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 652 |
+
"model.visual.blocks.25.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 653 |
+
"model.visual.blocks.25.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 654 |
+
"model.visual.blocks.25.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 655 |
+
"model.visual.blocks.25.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 656 |
+
"model.visual.blocks.25.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 657 |
+
"model.visual.blocks.25.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 658 |
+
"model.visual.blocks.25.norm1.bias": "model-00004-of-00004.safetensors",
|
| 659 |
+
"model.visual.blocks.25.norm1.weight": "model-00004-of-00004.safetensors",
|
| 660 |
+
"model.visual.blocks.25.norm2.bias": "model-00004-of-00004.safetensors",
|
| 661 |
+
"model.visual.blocks.25.norm2.weight": "model-00004-of-00004.safetensors",
|
| 662 |
+
"model.visual.blocks.26.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 663 |
+
"model.visual.blocks.26.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 664 |
+
"model.visual.blocks.26.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 665 |
+
"model.visual.blocks.26.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 666 |
+
"model.visual.blocks.26.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 667 |
+
"model.visual.blocks.26.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 668 |
+
"model.visual.blocks.26.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 669 |
+
"model.visual.blocks.26.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 670 |
+
"model.visual.blocks.26.norm1.bias": "model-00004-of-00004.safetensors",
|
| 671 |
+
"model.visual.blocks.26.norm1.weight": "model-00004-of-00004.safetensors",
|
| 672 |
+
"model.visual.blocks.26.norm2.bias": "model-00004-of-00004.safetensors",
|
| 673 |
+
"model.visual.blocks.26.norm2.weight": "model-00004-of-00004.safetensors",
|
| 674 |
+
"model.visual.blocks.3.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 675 |
+
"model.visual.blocks.3.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 676 |
+
"model.visual.blocks.3.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 677 |
+
"model.visual.blocks.3.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 678 |
+
"model.visual.blocks.3.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 679 |
+
"model.visual.blocks.3.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 680 |
+
"model.visual.blocks.3.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 681 |
+
"model.visual.blocks.3.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 682 |
+
"model.visual.blocks.3.norm1.bias": "model-00004-of-00004.safetensors",
|
| 683 |
+
"model.visual.blocks.3.norm1.weight": "model-00004-of-00004.safetensors",
|
| 684 |
+
"model.visual.blocks.3.norm2.bias": "model-00004-of-00004.safetensors",
|
| 685 |
+
"model.visual.blocks.3.norm2.weight": "model-00004-of-00004.safetensors",
|
| 686 |
+
"model.visual.blocks.4.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 687 |
+
"model.visual.blocks.4.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 688 |
+
"model.visual.blocks.4.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 689 |
+
"model.visual.blocks.4.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 690 |
+
"model.visual.blocks.4.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 691 |
+
"model.visual.blocks.4.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 692 |
+
"model.visual.blocks.4.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 693 |
+
"model.visual.blocks.4.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 694 |
+
"model.visual.blocks.4.norm1.bias": "model-00004-of-00004.safetensors",
|
| 695 |
+
"model.visual.blocks.4.norm1.weight": "model-00004-of-00004.safetensors",
|
| 696 |
+
"model.visual.blocks.4.norm2.bias": "model-00004-of-00004.safetensors",
|
| 697 |
+
"model.visual.blocks.4.norm2.weight": "model-00004-of-00004.safetensors",
|
| 698 |
+
"model.visual.blocks.5.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 699 |
+
"model.visual.blocks.5.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 700 |
+
"model.visual.blocks.5.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 701 |
+
"model.visual.blocks.5.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 702 |
+
"model.visual.blocks.5.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 703 |
+
"model.visual.blocks.5.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 704 |
+
"model.visual.blocks.5.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 705 |
+
"model.visual.blocks.5.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 706 |
+
"model.visual.blocks.5.norm1.bias": "model-00004-of-00004.safetensors",
|
| 707 |
+
"model.visual.blocks.5.norm1.weight": "model-00004-of-00004.safetensors",
|
| 708 |
+
"model.visual.blocks.5.norm2.bias": "model-00004-of-00004.safetensors",
|
| 709 |
+
"model.visual.blocks.5.norm2.weight": "model-00004-of-00004.safetensors",
|
| 710 |
+
"model.visual.blocks.6.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 711 |
+
"model.visual.blocks.6.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 712 |
+
"model.visual.blocks.6.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 713 |
+
"model.visual.blocks.6.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 714 |
+
"model.visual.blocks.6.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 715 |
+
"model.visual.blocks.6.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 716 |
+
"model.visual.blocks.6.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 717 |
+
"model.visual.blocks.6.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 718 |
+
"model.visual.blocks.6.norm1.bias": "model-00004-of-00004.safetensors",
|
| 719 |
+
"model.visual.blocks.6.norm1.weight": "model-00004-of-00004.safetensors",
|
| 720 |
+
"model.visual.blocks.6.norm2.bias": "model-00004-of-00004.safetensors",
|
| 721 |
+
"model.visual.blocks.6.norm2.weight": "model-00004-of-00004.safetensors",
|
| 722 |
+
"model.visual.blocks.7.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 723 |
+
"model.visual.blocks.7.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 724 |
+
"model.visual.blocks.7.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 725 |
+
"model.visual.blocks.7.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 726 |
+
"model.visual.blocks.7.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 727 |
+
"model.visual.blocks.7.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 728 |
+
"model.visual.blocks.7.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 729 |
+
"model.visual.blocks.7.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 730 |
+
"model.visual.blocks.7.norm1.bias": "model-00004-of-00004.safetensors",
|
| 731 |
+
"model.visual.blocks.7.norm1.weight": "model-00004-of-00004.safetensors",
|
| 732 |
+
"model.visual.blocks.7.norm2.bias": "model-00004-of-00004.safetensors",
|
| 733 |
+
"model.visual.blocks.7.norm2.weight": "model-00004-of-00004.safetensors",
|
| 734 |
+
"model.visual.blocks.8.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 735 |
+
"model.visual.blocks.8.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 736 |
+
"model.visual.blocks.8.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 737 |
+
"model.visual.blocks.8.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 738 |
+
"model.visual.blocks.8.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 739 |
+
"model.visual.blocks.8.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 740 |
+
"model.visual.blocks.8.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 741 |
+
"model.visual.blocks.8.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 742 |
+
"model.visual.blocks.8.norm1.bias": "model-00004-of-00004.safetensors",
|
| 743 |
+
"model.visual.blocks.8.norm1.weight": "model-00004-of-00004.safetensors",
|
| 744 |
+
"model.visual.blocks.8.norm2.bias": "model-00004-of-00004.safetensors",
|
| 745 |
+
"model.visual.blocks.8.norm2.weight": "model-00004-of-00004.safetensors",
|
| 746 |
+
"model.visual.blocks.9.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 747 |
+
"model.visual.blocks.9.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 748 |
+
"model.visual.blocks.9.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 749 |
+
"model.visual.blocks.9.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 750 |
+
"model.visual.blocks.9.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 751 |
+
"model.visual.blocks.9.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 752 |
+
"model.visual.blocks.9.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 753 |
+
"model.visual.blocks.9.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 754 |
+
"model.visual.blocks.9.norm1.bias": "model-00004-of-00004.safetensors",
|
| 755 |
+
"model.visual.blocks.9.norm1.weight": "model-00004-of-00004.safetensors",
|
| 756 |
+
"model.visual.blocks.9.norm2.bias": "model-00004-of-00004.safetensors",
|
| 757 |
+
"model.visual.blocks.9.norm2.weight": "model-00004-of-00004.safetensors",
|
| 758 |
+
"model.visual.merger.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 759 |
+
"model.visual.merger.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 760 |
+
"model.visual.merger.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 761 |
+
"model.visual.merger.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 762 |
+
"model.visual.merger.norm.bias": "model-00004-of-00004.safetensors",
|
| 763 |
+
"model.visual.merger.norm.weight": "model-00004-of-00004.safetensors",
|
| 764 |
+
"model.visual.patch_embed.proj.bias": "model-00004-of-00004.safetensors",
|
| 765 |
+
"model.visual.patch_embed.proj.weight": "model-00004-of-00004.safetensors",
|
| 766 |
+
"model.visual.pos_embed.weight": "model-00004-of-00004.safetensors"
|
| 767 |
+
}
|
| 768 |
+
}
|
c1_no_history/round_3/preprocessor_config.json
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"size": {
|
| 3 |
+
"longest_edge": 16777216,
|
| 4 |
+
"shortest_edge": 65536
|
| 5 |
+
},
|
| 6 |
+
"patch_size": 16,
|
| 7 |
+
"temporal_patch_size": 2,
|
| 8 |
+
"merge_size": 2,
|
| 9 |
+
"image_mean": [
|
| 10 |
+
0.5,
|
| 11 |
+
0.5,
|
| 12 |
+
0.5
|
| 13 |
+
],
|
| 14 |
+
"image_std": [
|
| 15 |
+
0.5,
|
| 16 |
+
0.5,
|
| 17 |
+
0.5
|
| 18 |
+
],
|
| 19 |
+
"processor_class": "Qwen3VLProcessor",
|
| 20 |
+
"image_processor_type": "Qwen2VLImageProcessorFast"
|
| 21 |
+
}
|
c1_no_history/round_3/processor_config.json
ADDED
|
@@ -0,0 +1,60 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"image_processor": {
|
| 3 |
+
"do_convert_rgb": true,
|
| 4 |
+
"do_normalize": true,
|
| 5 |
+
"do_rescale": true,
|
| 6 |
+
"do_resize": true,
|
| 7 |
+
"image_mean": [
|
| 8 |
+
0.5,
|
| 9 |
+
0.5,
|
| 10 |
+
0.5
|
| 11 |
+
],
|
| 12 |
+
"image_processor_type": "Qwen2VLImageProcessor",
|
| 13 |
+
"image_std": [
|
| 14 |
+
0.5,
|
| 15 |
+
0.5,
|
| 16 |
+
0.5
|
| 17 |
+
],
|
| 18 |
+
"merge_size": 2,
|
| 19 |
+
"patch_size": 16,
|
| 20 |
+
"resample": 3,
|
| 21 |
+
"rescale_factor": 0.00392156862745098,
|
| 22 |
+
"size": {
|
| 23 |
+
"longest_edge": 16777216,
|
| 24 |
+
"shortest_edge": 65536
|
| 25 |
+
},
|
| 26 |
+
"temporal_patch_size": 2
|
| 27 |
+
},
|
| 28 |
+
"processor_class": "Qwen3VLProcessor",
|
| 29 |
+
"video_processor": {
|
| 30 |
+
"do_convert_rgb": true,
|
| 31 |
+
"do_normalize": true,
|
| 32 |
+
"do_rescale": true,
|
| 33 |
+
"do_resize": true,
|
| 34 |
+
"do_sample_frames": true,
|
| 35 |
+
"fps": 2,
|
| 36 |
+
"image_mean": [
|
| 37 |
+
0.5,
|
| 38 |
+
0.5,
|
| 39 |
+
0.5
|
| 40 |
+
],
|
| 41 |
+
"image_std": [
|
| 42 |
+
0.5,
|
| 43 |
+
0.5,
|
| 44 |
+
0.5
|
| 45 |
+
],
|
| 46 |
+
"max_frames": 768,
|
| 47 |
+
"merge_size": 2,
|
| 48 |
+
"min_frames": 4,
|
| 49 |
+
"patch_size": 16,
|
| 50 |
+
"resample": 3,
|
| 51 |
+
"rescale_factor": 0.00392156862745098,
|
| 52 |
+
"return_metadata": false,
|
| 53 |
+
"size": {
|
| 54 |
+
"longest_edge": 25165824,
|
| 55 |
+
"shortest_edge": 4096
|
| 56 |
+
},
|
| 57 |
+
"temporal_patch_size": 2,
|
| 58 |
+
"video_processor_type": "Qwen3VLVideoProcessor"
|
| 59 |
+
}
|
| 60 |
+
}
|
c1_no_history/round_3/tokenizer_config.json
ADDED
|
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"audio_bos_token": "<|audio_start|>",
|
| 4 |
+
"audio_eos_token": "<|audio_end|>",
|
| 5 |
+
"audio_token": "<|audio_pad|>",
|
| 6 |
+
"backend": "tokenizers",
|
| 7 |
+
"bos_token": null,
|
| 8 |
+
"clean_up_tokenization_spaces": false,
|
| 9 |
+
"eos_token": "<|im_end|>",
|
| 10 |
+
"errors": "replace",
|
| 11 |
+
"image_token": "<|image_pad|>",
|
| 12 |
+
"is_local": true,
|
| 13 |
+
"local_files_only": true,
|
| 14 |
+
"model_max_length": 262144,
|
| 15 |
+
"model_specific_special_tokens": {
|
| 16 |
+
"audio_bos_token": "<|audio_start|>",
|
| 17 |
+
"audio_eos_token": "<|audio_end|>",
|
| 18 |
+
"audio_token": "<|audio_pad|>",
|
| 19 |
+
"image_token": "<|image_pad|>",
|
| 20 |
+
"video_token": "<|video_pad|>",
|
| 21 |
+
"vision_bos_token": "<|vision_start|>",
|
| 22 |
+
"vision_eos_token": "<|vision_end|>"
|
| 23 |
+
},
|
| 24 |
+
"pad_token": "<|endoftext|>",
|
| 25 |
+
"pretokenize_regex": "(?i:'s|'t|'re|'ve|'m|'ll|'d)|[^\\r\\n\\p{L}\\p{N}]?[\\p{L}\\p{M}]+|\\p{N}| ?[^\\s\\p{L}\\p{M}\\p{N}]+[\\r\\n]*|\\s*[\\r\\n]+|\\s+(?!\\S)|\\s+",
|
| 26 |
+
"processor_class": "Qwen3VLProcessor",
|
| 27 |
+
"split_special_tokens": false,
|
| 28 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 29 |
+
"unk_token": null,
|
| 30 |
+
"video_token": "<|video_pad|>",
|
| 31 |
+
"vision_bos_token": "<|vision_start|>",
|
| 32 |
+
"vision_eos_token": "<|vision_end|>"
|
| 33 |
+
}
|
c1_no_history/round_3/trainer_state.json
ADDED
|
@@ -0,0 +1,2454 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 1.0,
|
| 6 |
+
"eval_steps": 80.0,
|
| 7 |
+
"global_step": 160,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"clip_ratio/high_max": 0.0,
|
| 14 |
+
"clip_ratio/high_mean": 0.0,
|
| 15 |
+
"clip_ratio/low_mean": 0.0,
|
| 16 |
+
"clip_ratio/low_min": 0.0,
|
| 17 |
+
"clip_ratio/region_mean": 0.0,
|
| 18 |
+
"completions/clipped_ratio": 0.390625,
|
| 19 |
+
"completions/max_length": 2048.0,
|
| 20 |
+
"completions/mean_length": 1319.82421875,
|
| 21 |
+
"completions/min_length": 98.0,
|
| 22 |
+
"epoch": 0.00625,
|
| 23 |
+
"frac_reward_zero_std": 0.3125,
|
| 24 |
+
"grad_norm": 3.3124016610321494,
|
| 25 |
+
"kl": 0.0,
|
| 26 |
+
"learning_rate": 2e-07,
|
| 27 |
+
"loss": 1.862645149230957e-09,
|
| 28 |
+
"reward": 0.37890625,
|
| 29 |
+
"reward_std": 0.33732348680496216,
|
| 30 |
+
"rewards/MajorityAccuracy/mean": 0.37890625,
|
| 31 |
+
"rewards/MajorityAccuracy/std": 0.48606500029563904,
|
| 32 |
+
"step": 1,
|
| 33 |
+
"step_time": 586.4041166603565
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"clip_ratio/high_max": 0.0,
|
| 37 |
+
"clip_ratio/high_mean": 0.0,
|
| 38 |
+
"clip_ratio/low_mean": 0.0,
|
| 39 |
+
"clip_ratio/low_min": 0.0,
|
| 40 |
+
"clip_ratio/region_mean": 0.0,
|
| 41 |
+
"epoch": 0.0125,
|
| 42 |
+
"grad_norm": 1.6292490215829085,
|
| 43 |
+
"kl": 0.0,
|
| 44 |
+
"learning_rate": 4e-07,
|
| 45 |
+
"loss": 2.7939677238464355e-08,
|
| 46 |
+
"step": 2,
|
| 47 |
+
"step_time": 72.92081013321877
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"clip_ratio/high_max": 0.00214298942591995,
|
| 51 |
+
"clip_ratio/high_mean": 0.0006315856189758051,
|
| 52 |
+
"clip_ratio/low_mean": 0.0004362593754194677,
|
| 53 |
+
"clip_ratio/low_min": 0.0,
|
| 54 |
+
"clip_ratio/region_mean": 0.0010678449980332516,
|
| 55 |
+
"epoch": 0.01875,
|
| 56 |
+
"grad_norm": 2.440509120457657,
|
| 57 |
+
"kl": 0.0001982461617444642,
|
| 58 |
+
"learning_rate": 6e-07,
|
| 59 |
+
"loss": 0.00013128481805324554,
|
| 60 |
+
"step": 3,
|
| 61 |
+
"step_time": 74.51723110675812
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"clip_ratio/high_max": 0.0006477759598055854,
|
| 65 |
+
"clip_ratio/high_mean": 0.00016194398995139636,
|
| 66 |
+
"clip_ratio/low_mean": 7.136935164453462e-05,
|
| 67 |
+
"clip_ratio/low_min": 0.0,
|
| 68 |
+
"clip_ratio/region_mean": 0.00023331334159593098,
|
| 69 |
+
"epoch": 0.025,
|
| 70 |
+
"grad_norm": 0.7974355735623617,
|
| 71 |
+
"kl": 0.00017149051564047113,
|
| 72 |
+
"learning_rate": 8e-07,
|
| 73 |
+
"loss": -0.00011302108032396063,
|
| 74 |
+
"step": 4,
|
| 75 |
+
"step_time": 75.21928973123431
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"clip_ratio/high_max": 0.0012759945529978722,
|
| 79 |
+
"clip_ratio/high_mean": 0.00031899863824946806,
|
| 80 |
+
"clip_ratio/low_mean": 0.0006407655419025104,
|
| 81 |
+
"clip_ratio/low_min": 0.0,
|
| 82 |
+
"clip_ratio/region_mean": 0.0009597642056178302,
|
| 83 |
+
"epoch": 0.03125,
|
| 84 |
+
"grad_norm": 2.7891620133942023,
|
| 85 |
+
"kl": 0.0002315386609552661,
|
| 86 |
+
"learning_rate": 1e-06,
|
| 87 |
+
"loss": 6.393343210220337e-05,
|
| 88 |
+
"step": 5,
|
| 89 |
+
"step_time": 73.71967608854175
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"clip_ratio/high_max": 0.0007007908134255558,
|
| 93 |
+
"clip_ratio/high_mean": 0.00017519770335638896,
|
| 94 |
+
"clip_ratio/low_mean": 0.0,
|
| 95 |
+
"clip_ratio/low_min": 0.0,
|
| 96 |
+
"clip_ratio/region_mean": 0.00017519770335638896,
|
| 97 |
+
"epoch": 0.0375,
|
| 98 |
+
"grad_norm": 0.8182009686114051,
|
| 99 |
+
"kl": 0.00015106919454410672,
|
| 100 |
+
"learning_rate": 9.998973021172564e-07,
|
| 101 |
+
"loss": -5.189677722228225e-06,
|
| 102 |
+
"step": 6,
|
| 103 |
+
"step_time": 73.155940502882
|
| 104 |
+
},
|
| 105 |
+
{
|
| 106 |
+
"clip_ratio/high_max": 0.002320280429557897,
|
| 107 |
+
"clip_ratio/high_mean": 0.0006703477884002496,
|
| 108 |
+
"clip_ratio/low_mean": 0.0003775782242882997,
|
| 109 |
+
"clip_ratio/low_min": 0.0,
|
| 110 |
+
"clip_ratio/region_mean": 0.0010479260090505704,
|
| 111 |
+
"epoch": 0.04375,
|
| 112 |
+
"grad_norm": 2.283473895799828,
|
| 113 |
+
"kl": 0.00023436340234184172,
|
| 114 |
+
"learning_rate": 9.99589250656446e-07,
|
| 115 |
+
"loss": -9.730452438816428e-05,
|
| 116 |
+
"step": 7,
|
| 117 |
+
"step_time": 74.37848389148712
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"clip_ratio/high_max": 0.0007673259096918628,
|
| 121 |
+
"clip_ratio/high_mean": 0.0001918314774229657,
|
| 122 |
+
"clip_ratio/low_mean": 0.00036848391027888283,
|
| 123 |
+
"clip_ratio/low_min": 0.0,
|
| 124 |
+
"clip_ratio/region_mean": 0.0005603153913398273,
|
| 125 |
+
"epoch": 0.05,
|
| 126 |
+
"grad_norm": 1.4211362507689693,
|
| 127 |
+
"kl": 0.00034328183755860664,
|
| 128 |
+
"learning_rate": 9.990759721625005e-07,
|
| 129 |
+
"loss": -1.703626367088873e-06,
|
| 130 |
+
"step": 8,
|
| 131 |
+
"step_time": 73.04301379993558
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"clip_ratio/high_max": 0.0,
|
| 135 |
+
"clip_ratio/high_mean": 0.0,
|
| 136 |
+
"clip_ratio/low_mean": 0.0,
|
| 137 |
+
"clip_ratio/low_min": 0.0,
|
| 138 |
+
"clip_ratio/region_mean": 0.0,
|
| 139 |
+
"completions/clipped_ratio": 0.37109375,
|
| 140 |
+
"completions/max_length": 2048.0,
|
| 141 |
+
"completions/mean_length": 1386.46875,
|
| 142 |
+
"completions/min_length": 215.0,
|
| 143 |
+
"epoch": 0.05625,
|
| 144 |
+
"frac_reward_zero_std": 0.3125,
|
| 145 |
+
"grad_norm": 0.9858118082790378,
|
| 146 |
+
"kl": 0.0002932602546934504,
|
| 147 |
+
"learning_rate": 9.983576774858775e-07,
|
| 148 |
+
"loss": 2.9423536034300923e-06,
|
| 149 |
+
"reward": 0.375,
|
| 150 |
+
"reward_std": 0.3333781063556671,
|
| 151 |
+
"rewards/MajorityAccuracy/mean": 0.375,
|
| 152 |
+
"rewards/MajorityAccuracy/std": 0.48507124185562134,
|
| 153 |
+
"step": 9,
|
| 154 |
+
"step_time": 515.6027593202889
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"clip_ratio/high_max": 0.005913959117606282,
|
| 158 |
+
"clip_ratio/high_mean": 0.0017524479480925947,
|
| 159 |
+
"clip_ratio/low_mean": 0.0003318049330118811,
|
| 160 |
+
"clip_ratio/low_min": 0.0,
|
| 161 |
+
"clip_ratio/region_mean": 0.002084252890199423,
|
| 162 |
+
"epoch": 0.0625,
|
| 163 |
+
"grad_norm": 2.341417682215076,
|
| 164 |
+
"kl": 0.0006325502690742724,
|
| 165 |
+
"learning_rate": 9.974346616959475e-07,
|
| 166 |
+
"loss": 0.00027754122857004404,
|
| 167 |
+
"step": 10,
|
| 168 |
+
"step_time": 72.93235791102052
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"clip_ratio/high_max": 0.0067783140984829515,
|
| 172 |
+
"clip_ratio/high_mean": 0.002098588694934733,
|
| 173 |
+
"clip_ratio/low_mean": 0.0012251827338332077,
|
| 174 |
+
"clip_ratio/low_min": 0.0,
|
| 175 |
+
"clip_ratio/region_mean": 0.0033237715033465065,
|
| 176 |
+
"epoch": 0.06875,
|
| 177 |
+
"grad_norm": 3.785764230022898,
|
| 178 |
+
"kl": 0.0011346856190357357,
|
| 179 |
+
"learning_rate": 9.963073039597796e-07,
|
| 180 |
+
"loss": 0.00048323068767786026,
|
| 181 |
+
"step": 11,
|
| 182 |
+
"step_time": 72.69160894304514
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"clip_ratio/high_max": 0.004737215393106453,
|
| 186 |
+
"clip_ratio/high_mean": 0.0014784955819777679,
|
| 187 |
+
"clip_ratio/low_mean": 0.0010722489896579646,
|
| 188 |
+
"clip_ratio/low_min": 0.0,
|
| 189 |
+
"clip_ratio/region_mean": 0.0025507445534458384,
|
| 190 |
+
"epoch": 0.075,
|
| 191 |
+
"grad_norm": 2.0688960199977333,
|
| 192 |
+
"kl": 0.0009786229857127182,
|
| 193 |
+
"learning_rate": 9.949760673863846e-07,
|
| 194 |
+
"loss": -0.00015864601300563663,
|
| 195 |
+
"step": 12,
|
| 196 |
+
"step_time": 75.4453185275197
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"clip_ratio/high_max": 0.005467217357363552,
|
| 200 |
+
"clip_ratio/high_mean": 0.0020121466950513422,
|
| 201 |
+
"clip_ratio/low_mean": 0.000903989675862249,
|
| 202 |
+
"clip_ratio/low_min": 0.0,
|
| 203 |
+
"clip_ratio/region_mean": 0.0029161363636376336,
|
| 204 |
+
"epoch": 0.08125,
|
| 205 |
+
"grad_norm": 6.051121805921291,
|
| 206 |
+
"kl": 0.001339213689789176,
|
| 207 |
+
"learning_rate": 9.934414988364722e-07,
|
| 208 |
+
"loss": -8.597718260716647e-05,
|
| 209 |
+
"step": 13,
|
| 210 |
+
"step_time": 72.90294267609715
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"clip_ratio/high_max": 0.0,
|
| 214 |
+
"clip_ratio/high_mean": 0.0,
|
| 215 |
+
"clip_ratio/low_mean": 0.0,
|
| 216 |
+
"clip_ratio/low_min": 0.0,
|
| 217 |
+
"clip_ratio/region_mean": 0.0,
|
| 218 |
+
"epoch": 0.0875,
|
| 219 |
+
"grad_norm": 0.008399019240633064,
|
| 220 |
+
"kl": 0.0010136385899386369,
|
| 221 |
+
"learning_rate": 9.917042286978063e-07,
|
| 222 |
+
"loss": 1.0136163837159984e-05,
|
| 223 |
+
"step": 14,
|
| 224 |
+
"step_time": 75.18222297355533
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"clip_ratio/high_max": 0.007346438011154532,
|
| 228 |
+
"clip_ratio/high_mean": 0.0032330497779184952,
|
| 229 |
+
"clip_ratio/low_mean": 0.0016643770795781165,
|
| 230 |
+
"clip_ratio/low_min": 0.0,
|
| 231 |
+
"clip_ratio/region_mean": 0.004897426930256188,
|
| 232 |
+
"epoch": 0.09375,
|
| 233 |
+
"grad_norm": 2.6423699316638882,
|
| 234 |
+
"kl": 0.001551278430270031,
|
| 235 |
+
"learning_rate": 9.897649706262473e-07,
|
| 236 |
+
"loss": 0.0005176542326807976,
|
| 237 |
+
"step": 15,
|
| 238 |
+
"step_time": 73.13420938327909
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"clip_ratio/high_max": 0.006438411015551537,
|
| 242 |
+
"clip_ratio/high_mean": 0.0023626570764463395,
|
| 243 |
+
"clip_ratio/low_mean": 0.001736044796416536,
|
| 244 |
+
"clip_ratio/low_min": 0.0,
|
| 245 |
+
"clip_ratio/region_mean": 0.004098701843759045,
|
| 246 |
+
"epoch": 0.1,
|
| 247 |
+
"grad_norm": 1.7832937835557001,
|
| 248 |
+
"kl": 0.0013892145798308775,
|
| 249 |
+
"learning_rate": 9.87624521252587e-07,
|
| 250 |
+
"loss": 0.00031106790993362665,
|
| 251 |
+
"step": 16,
|
| 252 |
+
"step_time": 75.06360037252307
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"clip_ratio/high_max": 0.0,
|
| 256 |
+
"clip_ratio/high_mean": 0.0,
|
| 257 |
+
"clip_ratio/low_mean": 0.0,
|
| 258 |
+
"clip_ratio/low_min": 0.0,
|
| 259 |
+
"clip_ratio/region_mean": 0.0,
|
| 260 |
+
"completions/clipped_ratio": 0.421875,
|
| 261 |
+
"completions/max_length": 2048.0,
|
| 262 |
+
"completions/mean_length": 1401.21875,
|
| 263 |
+
"completions/min_length": 115.0,
|
| 264 |
+
"epoch": 0.10625,
|
| 265 |
+
"frac_reward_zero_std": 0.28125,
|
| 266 |
+
"grad_norm": 2.56280393688364,
|
| 267 |
+
"kl": 0.002017713792156428,
|
| 268 |
+
"learning_rate": 9.852837598553008e-07,
|
| 269 |
+
"loss": 2.0175109966658056e-05,
|
| 270 |
+
"reward": 0.375,
|
| 271 |
+
"reward_std": 0.32746079564094543,
|
| 272 |
+
"rewards/MajorityAccuracy/mean": 0.375,
|
| 273 |
+
"rewards/MajorityAccuracy/std": 0.48507124185562134,
|
| 274 |
+
"step": 17,
|
| 275 |
+
"step_time": 518.1101262643933
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"clip_ratio/high_max": 0.0022012528716004454,
|
| 279 |
+
"clip_ratio/high_mean": 0.0005844475726917153,
|
| 280 |
+
"clip_ratio/low_mean": 0.0006446011539082974,
|
| 281 |
+
"clip_ratio/low_min": 0.0,
|
| 282 |
+
"clip_ratio/region_mean": 0.001229048692039214,
|
| 283 |
+
"epoch": 0.1125,
|
| 284 |
+
"grad_norm": 2.198369472625112,
|
| 285 |
+
"kl": 0.0014259621093515307,
|
| 286 |
+
"learning_rate": 9.827436479993468e-07,
|
| 287 |
+
"loss": -0.0001350492238998413,
|
| 288 |
+
"step": 18,
|
| 289 |
+
"step_time": 74.96455423533916
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"clip_ratio/high_max": 0.0016282985743600875,
|
| 293 |
+
"clip_ratio/high_mean": 0.00042883647984126583,
|
| 294 |
+
"clip_ratio/low_mean": 0.00039396983083861414,
|
| 295 |
+
"clip_ratio/low_min": 0.0,
|
| 296 |
+
"clip_ratio/region_mean": 0.0008228063161368482,
|
| 297 |
+
"epoch": 0.11875,
|
| 298 |
+
"grad_norm": 1.1793255879420863,
|
| 299 |
+
"kl": 0.0023846924741519615,
|
| 300 |
+
"learning_rate": 9.800052291411629e-07,
|
| 301 |
+
"loss": 0.0001620049588382244,
|
| 302 |
+
"step": 19,
|
| 303 |
+
"step_time": 73.56986336410046
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"clip_ratio/high_max": 0.003697735723108053,
|
| 307 |
+
"clip_ratio/high_mean": 0.0011400715302443132,
|
| 308 |
+
"clip_ratio/low_mean": 0.000674051814712584,
|
| 309 |
+
"clip_ratio/low_min": 0.0,
|
| 310 |
+
"clip_ratio/region_mean": 0.0018141233449568972,
|
| 311 |
+
"epoch": 0.125,
|
| 312 |
+
"grad_norm": 2.8447328606886297,
|
| 313 |
+
"kl": 0.0020955775980837643,
|
| 314 |
+
"learning_rate": 9.770696282000244e-07,
|
| 315 |
+
"loss": -1.6328536730725318e-05,
|
| 316 |
+
"step": 20,
|
| 317 |
+
"step_time": 72.71022105589509
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"clip_ratio/high_max": 0.0008477360679535195,
|
| 321 |
+
"clip_ratio/high_mean": 0.00021193401698837988,
|
| 322 |
+
"clip_ratio/low_mean": 0.0009658966300776228,
|
| 323 |
+
"clip_ratio/low_min": 0.0,
|
| 324 |
+
"clip_ratio/region_mean": 0.0011778306543419603,
|
| 325 |
+
"epoch": 0.13125,
|
| 326 |
+
"grad_norm": 1.2370131538353744,
|
| 327 |
+
"kl": 0.0019031072006328031,
|
| 328 |
+
"learning_rate": 9.739380510959364e-07,
|
| 329 |
+
"loss": 5.728652831749059e-05,
|
| 330 |
+
"step": 21,
|
| 331 |
+
"step_time": 74.71207629516721
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"clip_ratio/high_max": 0.00251423692679964,
|
| 335 |
+
"clip_ratio/high_mean": 0.0007658075701328926,
|
| 336 |
+
"clip_ratio/low_mean": 0.0022237766243051738,
|
| 337 |
+
"clip_ratio/low_min": 0.00018310546875,
|
| 338 |
+
"clip_ratio/region_mean": 0.0029895841580582783,
|
| 339 |
+
"epoch": 0.1375,
|
| 340 |
+
"grad_norm": 2.2160357392986954,
|
| 341 |
+
"kl": 0.0023773615539539605,
|
| 342 |
+
"learning_rate": 9.706117842542516e-07,
|
| 343 |
+
"loss": 0.00045497622340917587,
|
| 344 |
+
"step": 22,
|
| 345 |
+
"step_time": 73.99089928343892
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"clip_ratio/high_max": 0.003281255776528269,
|
| 349 |
+
"clip_ratio/high_mean": 0.0009775685612112284,
|
| 350 |
+
"clip_ratio/low_mean": 0.001737851620418951,
|
| 351 |
+
"clip_ratio/low_min": 0.0,
|
| 352 |
+
"clip_ratio/region_mean": 0.0027154201816301793,
|
| 353 |
+
"epoch": 0.14375,
|
| 354 |
+
"grad_norm": 1.674314885557786,
|
| 355 |
+
"kl": 0.0029124286957085133,
|
| 356 |
+
"learning_rate": 9.670921940772186e-07,
|
| 357 |
+
"loss": 0.00036634824937209487,
|
| 358 |
+
"step": 23,
|
| 359 |
+
"step_time": 73.94821800664067
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"clip_ratio/high_max": 0.004995768817025237,
|
| 363 |
+
"clip_ratio/high_mean": 0.0019029384675377514,
|
| 364 |
+
"clip_ratio/low_mean": 0.0004160483949817717,
|
| 365 |
+
"clip_ratio/low_min": 0.0,
|
| 366 |
+
"clip_ratio/region_mean": 0.0023189868697954807,
|
| 367 |
+
"epoch": 0.15,
|
| 368 |
+
"grad_norm": 1.6651666292794627,
|
| 369 |
+
"kl": 0.0022268321772571653,
|
| 370 |
+
"learning_rate": 9.633807263826744e-07,
|
| 371 |
+
"loss": 0.00043372801155783236,
|
| 372 |
+
"step": 24,
|
| 373 |
+
"step_time": 73.58986315876245
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"clip_ratio/high_max": 0.0,
|
| 377 |
+
"clip_ratio/high_mean": 0.0,
|
| 378 |
+
"clip_ratio/low_mean": 0.0,
|
| 379 |
+
"clip_ratio/low_min": 0.0,
|
| 380 |
+
"clip_ratio/region_mean": 0.0,
|
| 381 |
+
"completions/clipped_ratio": 0.28515625,
|
| 382 |
+
"completions/max_length": 2048.0,
|
| 383 |
+
"completions/mean_length": 1143.31640625,
|
| 384 |
+
"completions/min_length": 94.0,
|
| 385 |
+
"epoch": 0.15625,
|
| 386 |
+
"frac_reward_zero_std": 0.375,
|
| 387 |
+
"grad_norm": 2.3199870578102018,
|
| 388 |
+
"kl": 0.003094922110904008,
|
| 389 |
+
"learning_rate": 9.594789058101153e-07,
|
| 390 |
+
"loss": 3.09719835058786e-05,
|
| 391 |
+
"reward": 0.421875,
|
| 392 |
+
"reward_std": 0.28694993257522583,
|
| 393 |
+
"rewards/MajorityAccuracy/mean": 0.421875,
|
| 394 |
+
"rewards/MajorityAccuracy/std": 0.49482616782188416,
|
| 395 |
+
"step": 25,
|
| 396 |
+
"step_time": 514.9251414425671
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"clip_ratio/high_max": 0.0017982537974603474,
|
| 400 |
+
"clip_ratio/high_mean": 0.0006250398073461838,
|
| 401 |
+
"clip_ratio/low_mean": 8.306754170916975e-05,
|
| 402 |
+
"clip_ratio/low_min": 0.0,
|
| 403 |
+
"clip_ratio/region_mean": 0.0007081073490553536,
|
| 404 |
+
"epoch": 0.1625,
|
| 405 |
+
"grad_norm": 1.5064821472443921,
|
| 406 |
+
"kl": 0.008192298235371709,
|
| 407 |
+
"learning_rate": 9.55388335194388e-07,
|
| 408 |
+
"loss": -3.426106195547618e-05,
|
| 409 |
+
"step": 26,
|
| 410 |
+
"step_time": 73.49575755000114
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"clip_ratio/high_max": 0.001089246099581942,
|
| 414 |
+
"clip_ratio/high_mean": 0.0003241357335355133,
|
| 415 |
+
"clip_ratio/low_mean": 0.0002288818359375,
|
| 416 |
+
"clip_ratio/low_min": 0.0,
|
| 417 |
+
"clip_ratio/region_mean": 0.0005530175694730133,
|
| 418 |
+
"epoch": 0.16875,
|
| 419 |
+
"grad_norm": 1.2281313840258714,
|
| 420 |
+
"kl": 0.0033776113705243915,
|
| 421 |
+
"learning_rate": 9.511106949072587e-07,
|
| 422 |
+
"loss": -2.6553427233011462e-05,
|
| 423 |
+
"step": 27,
|
| 424 |
+
"step_time": 74.08307282626629
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"clip_ratio/high_max": 0.0018721480737440288,
|
| 428 |
+
"clip_ratio/high_mean": 0.0004680370184360072,
|
| 429 |
+
"clip_ratio/low_mean": 0.0011625949846347794,
|
| 430 |
+
"clip_ratio/low_min": 0.0,
|
| 431 |
+
"clip_ratio/region_mean": 0.0016306320103467442,
|
| 432 |
+
"epoch": 0.175,
|
| 433 |
+
"grad_norm": 2.593417607277329,
|
| 434 |
+
"kl": 0.003781731444178149,
|
| 435 |
+
"learning_rate": 9.466477421671295e-07,
|
| 436 |
+
"loss": 0.00030680285999551415,
|
| 437 |
+
"step": 28,
|
| 438 |
+
"step_time": 75.46486095711589
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"clip_ratio/high_max": 0.006725825893227011,
|
| 442 |
+
"clip_ratio/high_mean": 0.0017959852557396516,
|
| 443 |
+
"clip_ratio/low_mean": 0.0020974867802578956,
|
| 444 |
+
"clip_ratio/low_min": 0.0,
|
| 445 |
+
"clip_ratio/region_mean": 0.003893472021445632,
|
| 446 |
+
"epoch": 0.18125,
|
| 447 |
+
"grad_norm": 2.917668314992049,
|
| 448 |
+
"kl": 0.003236143820686266,
|
| 449 |
+
"learning_rate": 9.420013103171891e-07,
|
| 450 |
+
"loss": -1.0288320481777191e-05,
|
| 451 |
+
"step": 29,
|
| 452 |
+
"step_time": 72.97414231300354
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"clip_ratio/high_max": 0.0038061040177126415,
|
| 456 |
+
"clip_ratio/high_mean": 0.001309101107835886,
|
| 457 |
+
"clip_ratio/low_mean": 0.000929413792619016,
|
| 458 |
+
"clip_ratio/low_min": 0.0,
|
| 459 |
+
"clip_ratio/region_mean": 0.0022385149059118703,
|
| 460 |
+
"epoch": 0.1875,
|
| 461 |
+
"grad_norm": 2.4505190317343386,
|
| 462 |
+
"kl": 0.010554156091529876,
|
| 463 |
+
"learning_rate": 9.37173308072291e-07,
|
| 464 |
+
"loss": 0.0004595278587657958,
|
| 465 |
+
"step": 30,
|
| 466 |
+
"step_time": 74.38417251780629
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"clip_ratio/high_max": 0.0010262579016853124,
|
| 470 |
+
"clip_ratio/high_mean": 0.0002565644754213281,
|
| 471 |
+
"clip_ratio/low_mean": 0.00040915260979090817,
|
| 472 |
+
"clip_ratio/low_min": 0.0,
|
| 473 |
+
"clip_ratio/region_mean": 0.0006657170852122363,
|
| 474 |
+
"epoch": 0.19375,
|
| 475 |
+
"grad_norm": 1.3086782963154897,
|
| 476 |
+
"kl": 0.004154697380727157,
|
| 477 |
+
"learning_rate": 9.321657187348688e-07,
|
| 478 |
+
"loss": 0.0001478793565183878,
|
| 479 |
+
"step": 31,
|
| 480 |
+
"step_time": 73.63949031382799
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"clip_ratio/high_max": 0.0042785714322235435,
|
| 484 |
+
"clip_ratio/high_mean": 0.001399276741722133,
|
| 485 |
+
"clip_ratio/low_mean": 0.0002857588442566339,
|
| 486 |
+
"clip_ratio/low_min": 0.0,
|
| 487 |
+
"clip_ratio/region_mean": 0.0016850355896167457,
|
| 488 |
+
"epoch": 0.2,
|
| 489 |
+
"grad_norm": 1.2418317466642328,
|
| 490 |
+
"kl": 0.0028759884589817375,
|
| 491 |
+
"learning_rate": 9.269805993802128e-07,
|
| 492 |
+
"loss": 0.00023308521485887468,
|
| 493 |
+
"step": 32,
|
| 494 |
+
"step_time": 73.9297489002347
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"clip_ratio/high_max": 0.0,
|
| 498 |
+
"clip_ratio/high_mean": 0.0,
|
| 499 |
+
"clip_ratio/low_mean": 0.0,
|
| 500 |
+
"clip_ratio/low_min": 0.0,
|
| 501 |
+
"clip_ratio/region_mean": 0.0,
|
| 502 |
+
"completions/clipped_ratio": 0.3046875,
|
| 503 |
+
"completions/max_length": 2048.0,
|
| 504 |
+
"completions/mean_length": 1252.59765625,
|
| 505 |
+
"completions/min_length": 141.0,
|
| 506 |
+
"epoch": 0.20625,
|
| 507 |
+
"frac_reward_zero_std": 0.28125,
|
| 508 |
+
"grad_norm": 0.8788420429225156,
|
| 509 |
+
"kl": 0.004429537599207833,
|
| 510 |
+
"learning_rate": 9.216200800114411e-07,
|
| 511 |
+
"loss": 4.431261550053023e-05,
|
| 512 |
+
"reward": 0.37109375,
|
| 513 |
+
"reward_std": 0.3172135353088379,
|
| 514 |
+
"rewards/MajorityAccuracy/mean": 0.37109375,
|
| 515 |
+
"rewards/MajorityAccuracy/std": 0.48404380679130554,
|
| 516 |
+
"step": 33,
|
| 517 |
+
"step_time": 512.9330595731735
|
| 518 |
+
},
|
| 519 |
+
{
|
| 520 |
+
"clip_ratio/high_max": 0.000255623715929687,
|
| 521 |
+
"clip_ratio/high_mean": 6.390592898242176e-05,
|
| 522 |
+
"clip_ratio/low_mean": 0.0006227210251381621,
|
| 523 |
+
"clip_ratio/low_min": 0.0,
|
| 524 |
+
"clip_ratio/region_mean": 0.0006866269613965414,
|
| 525 |
+
"epoch": 0.2125,
|
| 526 |
+
"grad_norm": 1.7477503906590532,
|
| 527 |
+
"kl": 0.004020913649583235,
|
| 528 |
+
"learning_rate": 9.160863626845119e-07,
|
| 529 |
+
"loss": 0.00020554146612994373,
|
| 530 |
+
"step": 34,
|
| 531 |
+
"step_time": 74.78994200751185
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"clip_ratio/high_max": 0.002513440362235997,
|
| 535 |
+
"clip_ratio/high_mean": 0.0008788590585027123,
|
| 536 |
+
"clip_ratio/low_mean": 0.0005334548259270377,
|
| 537 |
+
"clip_ratio/low_min": 0.00012468827480915934,
|
| 538 |
+
"clip_ratio/region_mean": 0.0014123138862487394,
|
| 539 |
+
"epoch": 0.21875,
|
| 540 |
+
"grad_norm": 2.491581724754678,
|
| 541 |
+
"kl": 0.0037185786350164562,
|
| 542 |
+
"learning_rate": 9.103817206036382e-07,
|
| 543 |
+
"loss": -3.9536505937576294e-05,
|
| 544 |
+
"step": 35,
|
| 545 |
+
"step_time": 74.76116229221225
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"clip_ratio/high_max": 0.0012443893065210432,
|
| 549 |
+
"clip_ratio/high_mean": 0.0004113221039006021,
|
| 550 |
+
"clip_ratio/low_mean": 0.0004026170863653533,
|
| 551 |
+
"clip_ratio/low_min": 0.0,
|
| 552 |
+
"clip_ratio/region_mean": 0.0008139391866279766,
|
| 553 |
+
"epoch": 0.225,
|
| 554 |
+
"grad_norm": 1.730411844905007,
|
| 555 |
+
"kl": 0.0040513665007893,
|
| 556 |
+
"learning_rate": 9.045084971874737e-07,
|
| 557 |
+
"loss": 9.110011160373688e-05,
|
| 558 |
+
"step": 36,
|
| 559 |
+
"step_time": 73.85503848642111
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"clip_ratio/high_max": 0.005585263861576095,
|
| 563 |
+
"clip_ratio/high_mean": 0.002121661920682527,
|
| 564 |
+
"clip_ratio/low_mean": 0.0010152966315217782,
|
| 565 |
+
"clip_ratio/low_min": 0.0,
|
| 566 |
+
"clip_ratio/region_mean": 0.0031369585340144113,
|
| 567 |
+
"epoch": 0.23125,
|
| 568 |
+
"grad_norm": 2.750499866125593,
|
| 569 |
+
"kl": 0.004834469815250486,
|
| 570 |
+
"learning_rate": 8.984691051064574e-07,
|
| 571 |
+
"loss": 0.00019920508202631027,
|
| 572 |
+
"step": 37,
|
| 573 |
+
"step_time": 73.86414471641183
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"clip_ratio/high_max": 0.0038032257580198348,
|
| 577 |
+
"clip_ratio/high_mean": 0.0009508064395049587,
|
| 578 |
+
"clip_ratio/low_mean": 0.0007924861820356455,
|
| 579 |
+
"clip_ratio/low_min": 0.0,
|
| 580 |
+
"clip_ratio/region_mean": 0.0017432925960747525,
|
| 581 |
+
"epoch": 0.2375,
|
| 582 |
+
"grad_norm": 1.8474589140071782,
|
| 583 |
+
"kl": 0.00502755114575848,
|
| 584 |
+
"learning_rate": 8.922660252917087e-07,
|
| 585 |
+
"loss": 0.00010558079520706087,
|
| 586 |
+
"step": 38,
|
| 587 |
+
"step_time": 73.25661151111126
|
| 588 |
+
},
|
| 589 |
+
{
|
| 590 |
+
"clip_ratio/high_max": 0.013264735243865289,
|
| 591 |
+
"clip_ratio/high_mean": 0.003706509614858078,
|
| 592 |
+
"clip_ratio/low_mean": 0.0010669792391126975,
|
| 593 |
+
"clip_ratio/low_min": 0.0,
|
| 594 |
+
"clip_ratio/region_mean": 0.0047734889085404575,
|
| 595 |
+
"epoch": 0.24375,
|
| 596 |
+
"grad_norm": 2.925943887399924,
|
| 597 |
+
"kl": 0.007424190582241863,
|
| 598 |
+
"learning_rate": 8.859018059158809e-07,
|
| 599 |
+
"loss": 0.000978577882051468,
|
| 600 |
+
"step": 39,
|
| 601 |
+
"step_time": 71.86062966287136
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"clip_ratio/high_max": 0.0018555941351223737,
|
| 605 |
+
"clip_ratio/high_mean": 0.000807820339105092,
|
| 606 |
+
"clip_ratio/low_mean": 0.0023019425570964813,
|
| 607 |
+
"clip_ratio/low_min": 0.00010575295891612768,
|
| 608 |
+
"clip_ratio/region_mean": 0.003109762939857319,
|
| 609 |
+
"epoch": 0.25,
|
| 610 |
+
"grad_norm": 1.6701974840086564,
|
| 611 |
+
"kl": 0.004774957254994661,
|
| 612 |
+
"learning_rate": 8.793790613463954e-07,
|
| 613 |
+
"loss": 0.0006031543016433716,
|
| 614 |
+
"step": 40,
|
| 615 |
+
"step_time": 75.07580172643065
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"clip_ratio/high_max": 0.0,
|
| 619 |
+
"clip_ratio/high_mean": 0.0,
|
| 620 |
+
"clip_ratio/low_mean": 0.0,
|
| 621 |
+
"clip_ratio/low_min": 0.0,
|
| 622 |
+
"clip_ratio/region_mean": 0.0,
|
| 623 |
+
"completions/clipped_ratio": 0.4453125,
|
| 624 |
+
"completions/max_length": 2048.0,
|
| 625 |
+
"completions/mean_length": 1380.3046875,
|
| 626 |
+
"completions/min_length": 95.0,
|
| 627 |
+
"epoch": 0.25625,
|
| 628 |
+
"frac_reward_zero_std": 0.34375,
|
| 629 |
+
"grad_norm": 2.828737423579618,
|
| 630 |
+
"kl": 0.009824214444961399,
|
| 631 |
+
"learning_rate": 8.727004710714798e-07,
|
| 632 |
+
"loss": 9.823143773246557e-05,
|
| 633 |
+
"reward": 0.328125,
|
| 634 |
+
"reward_std": 0.29222697019577026,
|
| 635 |
+
"rewards/MajorityAccuracy/mean": 0.328125,
|
| 636 |
+
"rewards/MajorityAccuracy/std": 0.47045034170150757,
|
| 637 |
+
"step": 41,
|
| 638 |
+
"step_time": 518.8405158221722
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"clip_ratio/high_max": 0.0017947363521670923,
|
| 642 |
+
"clip_ratio/high_mean": 0.0006024986905686092,
|
| 643 |
+
"clip_ratio/low_mean": 0.0004931776275043376,
|
| 644 |
+
"clip_ratio/low_min": 0.0,
|
| 645 |
+
"clip_ratio/region_mean": 0.0010956763180729467,
|
| 646 |
+
"epoch": 0.2625,
|
| 647 |
+
"grad_norm": 1.6453372400624748,
|
| 648 |
+
"kl": 0.005062073818407953,
|
| 649 |
+
"learning_rate": 8.658687785994578e-07,
|
| 650 |
+
"loss": 0.0002101637946907431,
|
| 651 |
+
"step": 42,
|
| 652 |
+
"step_time": 75.14040485396981
|
| 653 |
+
},
|
| 654 |
+
{
|
| 655 |
+
"clip_ratio/high_max": 0.00012400794366840273,
|
| 656 |
+
"clip_ratio/high_mean": 3.100198591710068e-05,
|
| 657 |
+
"clip_ratio/low_mean": 0.0001373291015625,
|
| 658 |
+
"clip_ratio/low_min": 0.0,
|
| 659 |
+
"clip_ratio/region_mean": 0.00016833108747960068,
|
| 660 |
+
"epoch": 0.26875,
|
| 661 |
+
"grad_norm": 0.6179262138336553,
|
| 662 |
+
"kl": 0.0056806173524819314,
|
| 663 |
+
"learning_rate": 8.588867903317394e-07,
|
| 664 |
+
"loss": 9.155097359325737e-05,
|
| 665 |
+
"step": 43,
|
| 666 |
+
"step_time": 75.06620420143008
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"clip_ratio/high_max": 0.002774362001218833,
|
| 670 |
+
"clip_ratio/high_mean": 0.0007586770698253531,
|
| 671 |
+
"clip_ratio/low_mean": 0.00048828125,
|
| 672 |
+
"clip_ratio/low_min": 0.0,
|
| 673 |
+
"clip_ratio/region_mean": 0.0012469583161873743,
|
| 674 |
+
"epoch": 0.275,
|
| 675 |
+
"grad_norm": 1.8873123764520008,
|
| 676 |
+
"kl": 0.004963705723639578,
|
| 677 |
+
"learning_rate": 8.51757374409974e-07,
|
| 678 |
+
"loss": 0.00021060417930129915,
|
| 679 |
+
"step": 44,
|
| 680 |
+
"step_time": 75.13627487793565
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"clip_ratio/high_max": 0.004831871337955818,
|
| 684 |
+
"clip_ratio/high_mean": 0.0015659693235647865,
|
| 685 |
+
"clip_ratio/low_mean": 0.0006511137835332192,
|
| 686 |
+
"clip_ratio/low_min": 0.0,
|
| 687 |
+
"clip_ratio/region_mean": 0.002217083099822048,
|
| 688 |
+
"epoch": 0.28125,
|
| 689 |
+
"grad_norm": 1.5916141558055459,
|
| 690 |
+
"kl": 0.003686970885610208,
|
| 691 |
+
"learning_rate": 8.444834595378433e-07,
|
| 692 |
+
"loss": 0.00021767133148387074,
|
| 693 |
+
"step": 45,
|
| 694 |
+
"step_time": 73.49450017139316
|
| 695 |
+
},
|
| 696 |
+
{
|
| 697 |
+
"clip_ratio/high_max": 0.0056103784299921244,
|
| 698 |
+
"clip_ratio/high_mean": 0.001666085321630817,
|
| 699 |
+
"clip_ratio/low_mean": 0.001673196988122072,
|
| 700 |
+
"clip_ratio/low_min": 0.0,
|
| 701 |
+
"clip_ratio/region_mean": 0.003339282309752889,
|
| 702 |
+
"epoch": 0.2875,
|
| 703 |
+
"grad_norm": 3.693430607113529,
|
| 704 |
+
"kl": 0.005812845134641975,
|
| 705 |
+
"learning_rate": 8.370680337779736e-07,
|
| 706 |
+
"loss": 0.0009153332794085145,
|
| 707 |
+
"step": 46,
|
| 708 |
+
"step_time": 75.09204019606113
|
| 709 |
+
},
|
| 710 |
+
{
|
| 711 |
+
"clip_ratio/high_max": 0.00792945345165208,
|
| 712 |
+
"clip_ratio/high_mean": 0.003121035755611956,
|
| 713 |
+
"clip_ratio/low_mean": 0.002722196077229455,
|
| 714 |
+
"clip_ratio/low_min": 0.0001220703125,
|
| 715 |
+
"clip_ratio/region_mean": 0.005843231861945242,
|
| 716 |
+
"epoch": 0.29375,
|
| 717 |
+
"grad_norm": 2.8896408449260345,
|
| 718 |
+
"kl": 0.006805578421335667,
|
| 719 |
+
"learning_rate": 8.295141433244659e-07,
|
| 720 |
+
"loss": 0.00036280136555433273,
|
| 721 |
+
"step": 47,
|
| 722 |
+
"step_time": 74.82099534198642
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"clip_ratio/high_max": 0.008426151005551219,
|
| 726 |
+
"clip_ratio/high_mean": 0.002601568659883924,
|
| 727 |
+
"clip_ratio/low_mean": 0.002368030087382067,
|
| 728 |
+
"clip_ratio/low_min": 0.0,
|
| 729 |
+
"clip_ratio/region_mean": 0.004969598725438118,
|
| 730 |
+
"epoch": 0.3,
|
| 731 |
+
"grad_norm": 2.1630840040071195,
|
| 732 |
+
"kl": 0.005316657712683082,
|
| 733 |
+
"learning_rate": 8.218248912515442e-07,
|
| 734 |
+
"loss": -1.0576099157333374e-05,
|
| 735 |
+
"step": 48,
|
| 736 |
+
"step_time": 73.67506120353937
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"clip_ratio/high_max": 0.0,
|
| 740 |
+
"clip_ratio/high_mean": 0.0,
|
| 741 |
+
"clip_ratio/low_mean": 0.0,
|
| 742 |
+
"clip_ratio/low_min": 0.0,
|
| 743 |
+
"clip_ratio/region_mean": 0.0,
|
| 744 |
+
"completions/clipped_ratio": 0.37890625,
|
| 745 |
+
"completions/max_length": 2048.0,
|
| 746 |
+
"completions/mean_length": 1412.4375,
|
| 747 |
+
"completions/min_length": 133.0,
|
| 748 |
+
"epoch": 0.30625,
|
| 749 |
+
"frac_reward_zero_std": 0.28125,
|
| 750 |
+
"grad_norm": 2.92957336338197,
|
| 751 |
+
"kl": 0.007733477978035808,
|
| 752 |
+
"learning_rate": 8.140034362388397e-07,
|
| 753 |
+
"loss": 7.730829383945093e-05,
|
| 754 |
+
"reward": 0.30859375,
|
| 755 |
+
"reward_std": 0.31220486760139465,
|
| 756 |
+
"rewards/MajorityAccuracy/mean": 0.30859375,
|
| 757 |
+
"rewards/MajorityAccuracy/std": 0.46281784772872925,
|
| 758 |
+
"step": 49,
|
| 759 |
+
"step_time": 517.853542804718
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"clip_ratio/high_max": 0.0029977149970363826,
|
| 763 |
+
"clip_ratio/high_mean": 0.0008371976728085428,
|
| 764 |
+
"clip_ratio/low_mean": 0.0007322602286876645,
|
| 765 |
+
"clip_ratio/low_min": 0.0,
|
| 766 |
+
"clip_ratio/region_mean": 0.0015694579051341861,
|
| 767 |
+
"epoch": 0.3125,
|
| 768 |
+
"grad_norm": 2.552064954040967,
|
| 769 |
+
"kl": 0.007801662897691131,
|
| 770 |
+
"learning_rate": 8.060529912738314e-07,
|
| 771 |
+
"loss": 0.00014112915960140526,
|
| 772 |
+
"step": 50,
|
| 773 |
+
"step_time": 73.47678277269006
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"clip_ratio/high_max": 0.0020184632448945194,
|
| 777 |
+
"clip_ratio/high_mean": 0.0008545542732463218,
|
| 778 |
+
"clip_ratio/low_mean": 0.0006283475668169558,
|
| 779 |
+
"clip_ratio/low_min": 0.0001220703125,
|
| 780 |
+
"clip_ratio/region_mean": 0.00148290183278732,
|
| 781 |
+
"epoch": 0.31875,
|
| 782 |
+
"grad_norm": 2.259968232432451,
|
| 783 |
+
"kl": 0.0078028778079897165,
|
| 784 |
+
"learning_rate": 7.979768223319785e-07,
|
| 785 |
+
"loss": 0.00019369274377822876,
|
| 786 |
+
"step": 51,
|
| 787 |
+
"step_time": 75.28132904320955
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"clip_ratio/high_max": 0.0002020474203163758,
|
| 791 |
+
"clip_ratio/high_mean": 5.051185507909395e-05,
|
| 792 |
+
"clip_ratio/low_mean": 0.00015772347796882968,
|
| 793 |
+
"clip_ratio/low_min": 0.0,
|
| 794 |
+
"clip_ratio/region_mean": 0.00020823533304792363,
|
| 795 |
+
"epoch": 0.325,
|
| 796 |
+
"grad_norm": 1.0535734786096125,
|
| 797 |
+
"kl": 0.006850913632661104,
|
| 798 |
+
"learning_rate": 7.897782470350849e-07,
|
| 799 |
+
"loss": 7.483300578314811e-05,
|
| 800 |
+
"step": 52,
|
| 801 |
+
"step_time": 75.30334085598588
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"clip_ratio/high_max": 0.002083065250189975,
|
| 805 |
+
"clip_ratio/high_mean": 0.0006889648357173428,
|
| 806 |
+
"clip_ratio/low_mean": 0.0006200198986334726,
|
| 807 |
+
"clip_ratio/low_min": 0.0,
|
| 808 |
+
"clip_ratio/region_mean": 0.0013089847343508154,
|
| 809 |
+
"epoch": 0.33125,
|
| 810 |
+
"grad_norm": 2.0009991490661028,
|
| 811 |
+
"kl": 0.010516568319872022,
|
| 812 |
+
"learning_rate": 7.814606332884488e-07,
|
| 813 |
+
"loss": 0.0005573503440245986,
|
| 814 |
+
"step": 53,
|
| 815 |
+
"step_time": 73.14710393175483
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"clip_ratio/high_max": 0.005557261785725132,
|
| 819 |
+
"clip_ratio/high_mean": 0.002773577769403346,
|
| 820 |
+
"clip_ratio/low_mean": 0.0007075631910993252,
|
| 821 |
+
"clip_ratio/low_min": 0.0,
|
| 822 |
+
"clip_ratio/region_mean": 0.0034811409132089466,
|
| 823 |
+
"epoch": 0.3375,
|
| 824 |
+
"grad_norm": 1.8903515646535547,
|
| 825 |
+
"kl": 0.00706085265846923,
|
| 826 |
+
"learning_rate": 7.730273978973552e-07,
|
| 827 |
+
"loss": 0.00010735448449850082,
|
| 828 |
+
"step": 54,
|
| 829 |
+
"step_time": 73.77231953293085
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"clip_ratio/high_max": 0.003849053377052769,
|
| 833 |
+
"clip_ratio/high_mean": 0.0009622633442631923,
|
| 834 |
+
"clip_ratio/low_mean": 0.0005817745204694802,
|
| 835 |
+
"clip_ratio/low_min": 0.0,
|
| 836 |
+
"clip_ratio/region_mean": 0.001544037862913683,
|
| 837 |
+
"epoch": 0.34375,
|
| 838 |
+
"grad_norm": 1.2832983757226855,
|
| 839 |
+
"kl": 0.005780904495622963,
|
| 840 |
+
"learning_rate": 7.644820051634812e-07,
|
| 841 |
+
"loss": 0.00011970217747148126,
|
| 842 |
+
"step": 55,
|
| 843 |
+
"step_time": 75.30044076219201
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"clip_ratio/high_max": 0.0028245318681001663,
|
| 847 |
+
"clip_ratio/high_mean": 0.0007841117621865124,
|
| 848 |
+
"clip_ratio/low_mean": 0.001531955917016603,
|
| 849 |
+
"clip_ratio/low_min": 0.0,
|
| 850 |
+
"clip_ratio/region_mean": 0.0023160676792031154,
|
| 851 |
+
"epoch": 0.35,
|
| 852 |
+
"grad_norm": 3.9698743535722585,
|
| 853 |
+
"kl": 0.005560451711062342,
|
| 854 |
+
"learning_rate": 7.558279654617912e-07,
|
| 855 |
+
"loss": 0.0004266318283043802,
|
| 856 |
+
"step": 56,
|
| 857 |
+
"step_time": 73.596709754318
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"clip_ratio/high_max": 0.0,
|
| 861 |
+
"clip_ratio/high_mean": 0.0,
|
| 862 |
+
"clip_ratio/low_mean": 0.0,
|
| 863 |
+
"clip_ratio/low_min": 0.0,
|
| 864 |
+
"clip_ratio/region_mean": 0.0,
|
| 865 |
+
"completions/clipped_ratio": 0.390625,
|
| 866 |
+
"completions/max_length": 2048.0,
|
| 867 |
+
"completions/mean_length": 1257.0703125,
|
| 868 |
+
"completions/min_length": 111.0,
|
| 869 |
+
"epoch": 0.35625,
|
| 870 |
+
"frac_reward_zero_std": 0.25,
|
| 871 |
+
"grad_norm": 2.5499357726800618,
|
| 872 |
+
"kl": 0.021476697991602123,
|
| 873 |
+
"learning_rate": 7.470688337985029e-07,
|
| 874 |
+
"loss": 0.00021470517094712704,
|
| 875 |
+
"reward": 0.34375,
|
| 876 |
+
"reward_std": 0.3186565041542053,
|
| 877 |
+
"rewards/MajorityAccuracy/mean": 0.34375,
|
| 878 |
+
"rewards/MajorityAccuracy/std": 0.47588926553726196,
|
| 879 |
+
"step": 57,
|
| 880 |
+
"step_time": 514.5246491618454
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"clip_ratio/high_max": 0.0,
|
| 884 |
+
"clip_ratio/high_mean": 0.0,
|
| 885 |
+
"clip_ratio/low_mean": 0.00015441894356627017,
|
| 886 |
+
"clip_ratio/low_min": 0.0,
|
| 887 |
+
"clip_ratio/region_mean": 0.00015441894356627017,
|
| 888 |
+
"epoch": 0.3625,
|
| 889 |
+
"grad_norm": 0.8552717533687713,
|
| 890 |
+
"kl": 0.006395676638931036,
|
| 891 |
+
"learning_rate": 7.382082083507225e-07,
|
| 892 |
+
"loss": 0.00013803716865368187,
|
| 893 |
+
"step": 58,
|
| 894 |
+
"step_time": 75.2453901655972
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"clip_ratio/high_max": 0.002808668461511843,
|
| 898 |
+
"clip_ratio/high_mean": 0.0007346853199123871,
|
| 899 |
+
"clip_ratio/low_mean": 0.0008893508493201807,
|
| 900 |
+
"clip_ratio/low_min": 0.0,
|
| 901 |
+
"clip_ratio/region_mean": 0.0016240361728705466,
|
| 902 |
+
"epoch": 0.36875,
|
| 903 |
+
"grad_norm": 2.0016864737810116,
|
| 904 |
+
"kl": 0.00903841812396422,
|
| 905 |
+
"learning_rate": 7.292497289883432e-07,
|
| 906 |
+
"loss": 0.00012952275574207306,
|
| 907 |
+
"step": 59,
|
| 908 |
+
"step_time": 73.19410736858845
|
| 909 |
+
},
|
| 910 |
+
{
|
| 911 |
+
"clip_ratio/high_max": 0.0050339013687334955,
|
| 912 |
+
"clip_ratio/high_mean": 0.0017250011587748304,
|
| 913 |
+
"clip_ratio/low_mean": 0.0005358964845072478,
|
| 914 |
+
"clip_ratio/low_min": 0.0,
|
| 915 |
+
"clip_ratio/region_mean": 0.002260897643282078,
|
| 916 |
+
"epoch": 0.375,
|
| 917 |
+
"grad_norm": 3.707914335358473,
|
| 918 |
+
"kl": 0.009968466591089964,
|
| 919 |
+
"learning_rate": 7.201970757788171e-07,
|
| 920 |
+
"loss": 0.0002133939415216446,
|
| 921 |
+
"step": 60,
|
| 922 |
+
"step_time": 73.92276175320148
|
| 923 |
+
},
|
| 924 |
+
{
|
| 925 |
+
"clip_ratio/high_max": 0.013084159669233486,
|
| 926 |
+
"clip_ratio/high_mean": 0.003508057205181103,
|
| 927 |
+
"clip_ratio/low_mean": 0.001705861459413427,
|
| 928 |
+
"clip_ratio/low_min": 6.453278183471411e-05,
|
| 929 |
+
"clip_ratio/region_mean": 0.005213918622757774,
|
| 930 |
+
"epoch": 0.38125,
|
| 931 |
+
"grad_norm": 4.243014717816484,
|
| 932 |
+
"kl": 0.011330398498103023,
|
| 933 |
+
"learning_rate": 7.110539674754159e-07,
|
| 934 |
+
"loss": 0.0005903635174036026,
|
| 935 |
+
"step": 61,
|
| 936 |
+
"step_time": 72.85362226888537
|
| 937 |
+
},
|
| 938 |
+
{
|
| 939 |
+
"clip_ratio/high_max": 0.0026300970348529518,
|
| 940 |
+
"clip_ratio/high_mean": 0.0007853108836570755,
|
| 941 |
+
"clip_ratio/low_mean": 0.003347352030687034,
|
| 942 |
+
"clip_ratio/low_min": 0.0,
|
| 943 |
+
"clip_ratio/region_mean": 0.004132662928896025,
|
| 944 |
+
"epoch": 0.3875,
|
| 945 |
+
"grad_norm": 2.6428799574068877,
|
| 946 |
+
"kl": 0.012021031230688095,
|
| 947 |
+
"learning_rate": 7.018241599895973e-07,
|
| 948 |
+
"loss": 0.0003894465626217425,
|
| 949 |
+
"step": 62,
|
| 950 |
+
"step_time": 75.34997637197375
|
| 951 |
+
},
|
| 952 |
+
{
|
| 953 |
+
"clip_ratio/high_max": 0.00802093103993684,
|
| 954 |
+
"clip_ratio/high_mean": 0.00302723579807207,
|
| 955 |
+
"clip_ratio/low_mean": 0.001769591195625253,
|
| 956 |
+
"clip_ratio/low_min": 0.0,
|
| 957 |
+
"clip_ratio/region_mean": 0.004796827037353069,
|
| 958 |
+
"epoch": 0.39375,
|
| 959 |
+
"grad_norm": 4.598317997576748,
|
| 960 |
+
"kl": 0.023752863286063075,
|
| 961 |
+
"learning_rate": 6.925114448481088e-07,
|
| 962 |
+
"loss": 0.0005888612940907478,
|
| 963 |
+
"step": 63,
|
| 964 |
+
"step_time": 74.51600427180529
|
| 965 |
+
},
|
| 966 |
+
{
|
| 967 |
+
"clip_ratio/high_max": 0.0032654129900038242,
|
| 968 |
+
"clip_ratio/high_mean": 0.0008163532475009561,
|
| 969 |
+
"clip_ratio/low_mean": 0.0011590927751967683,
|
| 970 |
+
"clip_ratio/low_min": 0.0,
|
| 971 |
+
"clip_ratio/region_mean": 0.0019754460226977244,
|
| 972 |
+
"epoch": 0.4,
|
| 973 |
+
"grad_norm": 1.311604089251808,
|
| 974 |
+
"kl": 0.006660247745458037,
|
| 975 |
+
"learning_rate": 6.831196476354614e-07,
|
| 976 |
+
"loss": 0.00022772533702664077,
|
| 977 |
+
"step": 64,
|
| 978 |
+
"step_time": 75.31076156720519
|
| 979 |
+
},
|
| 980 |
+
{
|
| 981 |
+
"clip_ratio/high_max": 0.0,
|
| 982 |
+
"clip_ratio/high_mean": 0.0,
|
| 983 |
+
"clip_ratio/low_mean": 0.0,
|
| 984 |
+
"clip_ratio/low_min": 0.0,
|
| 985 |
+
"clip_ratio/region_mean": 0.0,
|
| 986 |
+
"completions/clipped_ratio": 0.32421875,
|
| 987 |
+
"completions/max_length": 2048.0,
|
| 988 |
+
"completions/mean_length": 1189.484375,
|
| 989 |
+
"completions/min_length": 84.0,
|
| 990 |
+
"epoch": 0.40625,
|
| 991 |
+
"frac_reward_zero_std": 0.15625,
|
| 992 |
+
"grad_norm": 4.2426698751537435,
|
| 993 |
+
"kl": 0.01012384460773319,
|
| 994 |
+
"learning_rate": 6.7365262642241e-07,
|
| 995 |
+
"loss": 0.00010122917592525482,
|
| 996 |
+
"reward": 0.40234375,
|
| 997 |
+
"reward_std": 0.38874194025993347,
|
| 998 |
+
"rewards/MajorityAccuracy/mean": 0.40234375,
|
| 999 |
+
"rewards/MajorityAccuracy/std": 0.4913311004638672,
|
| 1000 |
+
"step": 65,
|
| 1001 |
+
"step_time": 512.1485653743148
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"clip_ratio/high_max": 0.0068315342359710485,
|
| 1005 |
+
"clip_ratio/high_mean": 0.0017078835589927621,
|
| 1006 |
+
"clip_ratio/low_mean": 0.0007184024652815424,
|
| 1007 |
+
"clip_ratio/low_min": 0.0,
|
| 1008 |
+
"clip_ratio/region_mean": 0.002426286031550262,
|
| 1009 |
+
"epoch": 0.4125,
|
| 1010 |
+
"grad_norm": 2.3880781424733106,
|
| 1011 |
+
"kl": 0.017061628750525415,
|
| 1012 |
+
"learning_rate": 6.641142701810931e-07,
|
| 1013 |
+
"loss": -6.796215893700719e-05,
|
| 1014 |
+
"step": 66,
|
| 1015 |
+
"step_time": 74.98267927765846
|
| 1016 |
+
},
|
| 1017 |
+
{
|
| 1018 |
+
"clip_ratio/high_max": 0.00327561168523971,
|
| 1019 |
+
"clip_ratio/high_mean": 0.001282911853195401,
|
| 1020 |
+
"clip_ratio/low_mean": 0.00037773739313706756,
|
| 1021 |
+
"clip_ratio/low_min": 6.103515625e-05,
|
| 1022 |
+
"clip_ratio/region_mean": 0.0016606492499704473,
|
| 1023 |
+
"epoch": 0.41875,
|
| 1024 |
+
"grad_norm": 2.352714240220581,
|
| 1025 |
+
"kl": 0.01152385922614485,
|
| 1026 |
+
"learning_rate": 6.545084971874736e-07,
|
| 1027 |
+
"loss": 0.00020941905677318573,
|
| 1028 |
+
"step": 67,
|
| 1029 |
+
"step_time": 74.65032090246677
|
| 1030 |
+
},
|
| 1031 |
+
{
|
| 1032 |
+
"clip_ratio/high_max": 0.001841912031522952,
|
| 1033 |
+
"clip_ratio/high_mean": 0.000572920442209579,
|
| 1034 |
+
"clip_ratio/low_mean": 0.00046006681077415124,
|
| 1035 |
+
"clip_ratio/low_min": 0.0,
|
| 1036 |
+
"clip_ratio/region_mean": 0.0010329872311558574,
|
| 1037 |
+
"epoch": 0.425,
|
| 1038 |
+
"grad_norm": 2.541797753700781,
|
| 1039 |
+
"kl": 0.01541670341975987,
|
| 1040 |
+
"learning_rate": 6.448392534117461e-07,
|
| 1041 |
+
"loss": 0.0002972546499222517,
|
| 1042 |
+
"step": 68,
|
| 1043 |
+
"step_time": 74.94772301614285
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"clip_ratio/high_max": 0.0041237107798224315,
|
| 1047 |
+
"clip_ratio/high_mean": 0.0015053579627419822,
|
| 1048 |
+
"clip_ratio/low_mean": 0.0010605527786538005,
|
| 1049 |
+
"clip_ratio/low_min": 0.0,
|
| 1050 |
+
"clip_ratio/region_mean": 0.0025659107632236555,
|
| 1051 |
+
"epoch": 0.43125,
|
| 1052 |
+
"grad_norm": 4.013450339042411,
|
| 1053 |
+
"kl": 0.017797512060496956,
|
| 1054 |
+
"learning_rate": 6.351105108973644e-07,
|
| 1055 |
+
"loss": 0.00015508197247982025,
|
| 1056 |
+
"step": 69,
|
| 1057 |
+
"step_time": 73.01652951911092
|
| 1058 |
+
},
|
| 1059 |
+
{
|
| 1060 |
+
"clip_ratio/high_max": 0.001588877486938145,
|
| 1061 |
+
"clip_ratio/high_mean": 0.0005112563212605892,
|
| 1062 |
+
"clip_ratio/low_mean": 0.0008655746987642488,
|
| 1063 |
+
"clip_ratio/low_min": 0.0,
|
| 1064 |
+
"clip_ratio/region_mean": 0.0013768310091109015,
|
| 1065 |
+
"epoch": 0.4375,
|
| 1066 |
+
"grad_norm": 1.9559328780204837,
|
| 1067 |
+
"kl": 0.009341413737274706,
|
| 1068 |
+
"learning_rate": 6.253262661293602e-07,
|
| 1069 |
+
"loss": 0.00023261923342943192,
|
| 1070 |
+
"step": 70,
|
| 1071 |
+
"step_time": 74.50909152254462
|
| 1072 |
+
},
|
| 1073 |
+
{
|
| 1074 |
+
"clip_ratio/high_max": 0.009060770622454584,
|
| 1075 |
+
"clip_ratio/high_mean": 0.0031427216163137928,
|
| 1076 |
+
"clip_ratio/low_mean": 0.00023078540107235312,
|
| 1077 |
+
"clip_ratio/low_min": 0.0,
|
| 1078 |
+
"clip_ratio/region_mean": 0.0033735069882823154,
|
| 1079 |
+
"epoch": 0.44375,
|
| 1080 |
+
"grad_norm": 5.3073901145423035,
|
| 1081 |
+
"kl": 0.01722423790488392,
|
| 1082 |
+
"learning_rate": 6.154905383926216e-07,
|
| 1083 |
+
"loss": 0.00039194809505715966,
|
| 1084 |
+
"step": 71,
|
| 1085 |
+
"step_time": 73.19182716310024
|
| 1086 |
+
},
|
| 1087 |
+
{
|
| 1088 |
+
"clip_ratio/high_max": 0.0036734248424181715,
|
| 1089 |
+
"clip_ratio/high_mean": 0.001240803630935261,
|
| 1090 |
+
"clip_ratio/low_mean": 0.0015967423387337476,
|
| 1091 |
+
"clip_ratio/low_min": 0.0,
|
| 1092 |
+
"clip_ratio/region_mean": 0.00283754596603103,
|
| 1093 |
+
"epoch": 0.45,
|
| 1094 |
+
"grad_norm": 1.8292998779080083,
|
| 1095 |
+
"kl": 0.009431483689695597,
|
| 1096 |
+
"learning_rate": 6.056073681208037e-07,
|
| 1097 |
+
"loss": 0.0001969898585230112,
|
| 1098 |
+
"step": 72,
|
| 1099 |
+
"step_time": 75.23610006645322
|
| 1100 |
+
},
|
| 1101 |
+
{
|
| 1102 |
+
"clip_ratio/high_max": 0.0,
|
| 1103 |
+
"clip_ratio/high_mean": 0.0,
|
| 1104 |
+
"clip_ratio/low_mean": 0.0,
|
| 1105 |
+
"clip_ratio/low_min": 0.0,
|
| 1106 |
+
"clip_ratio/region_mean": 0.0,
|
| 1107 |
+
"completions/clipped_ratio": 0.25390625,
|
| 1108 |
+
"completions/max_length": 2048.0,
|
| 1109 |
+
"completions/mean_length": 1094.90234375,
|
| 1110 |
+
"completions/min_length": 109.0,
|
| 1111 |
+
"epoch": 0.45625,
|
| 1112 |
+
"frac_reward_zero_std": 0.1875,
|
| 1113 |
+
"grad_norm": 2.03186909495495,
|
| 1114 |
+
"kl": 0.008239429269451648,
|
| 1115 |
+
"learning_rate": 5.956808152365532e-07,
|
| 1116 |
+
"loss": 8.240239549195394e-05,
|
| 1117 |
+
"reward": 0.40234375,
|
| 1118 |
+
"reward_std": 0.34876301884651184,
|
| 1119 |
+
"rewards/MajorityAccuracy/mean": 0.40234375,
|
| 1120 |
+
"rewards/MajorityAccuracy/std": 0.4913311004638672,
|
| 1121 |
+
"step": 73,
|
| 1122 |
+
"step_time": 513.6517122983932
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"clip_ratio/high_max": 0.006689010362606496,
|
| 1126 |
+
"clip_ratio/high_mean": 0.001853848312748596,
|
| 1127 |
+
"clip_ratio/low_mean": 0.00035237788688391447,
|
| 1128 |
+
"clip_ratio/low_min": 0.0,
|
| 1129 |
+
"clip_ratio/region_mean": 0.0022062261996325105,
|
| 1130 |
+
"epoch": 0.4625,
|
| 1131 |
+
"grad_norm": 2.8106567222355574,
|
| 1132 |
+
"kl": 0.01241054991260171,
|
| 1133 |
+
"learning_rate": 5.857149574837268e-07,
|
| 1134 |
+
"loss": 0.0006982441991567612,
|
| 1135 |
+
"step": 74,
|
| 1136 |
+
"step_time": 73.13822842016816
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"clip_ratio/high_max": 0.002344781009014696,
|
| 1140 |
+
"clip_ratio/high_mean": 0.0007026938837952912,
|
| 1141 |
+
"clip_ratio/low_mean": 0.000901579885976389,
|
| 1142 |
+
"clip_ratio/low_min": 0.0,
|
| 1143 |
+
"clip_ratio/region_mean": 0.0016042737479438074,
|
| 1144 |
+
"epoch": 0.46875,
|
| 1145 |
+
"grad_norm": 2.5402922378359425,
|
| 1146 |
+
"kl": 0.012390228570438921,
|
| 1147 |
+
"learning_rate": 5.757138887522883e-07,
|
| 1148 |
+
"loss": 0.00030984642216935754,
|
| 1149 |
+
"step": 75,
|
| 1150 |
+
"step_time": 75.0773378238082
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"clip_ratio/high_max": 0.006753532812581398,
|
| 1154 |
+
"clip_ratio/high_mean": 0.002149270923837321,
|
| 1155 |
+
"clip_ratio/low_mean": 0.0009080338568310253,
|
| 1156 |
+
"clip_ratio/low_min": 0.0,
|
| 1157 |
+
"clip_ratio/region_mean": 0.003057304762478452,
|
| 1158 |
+
"epoch": 0.475,
|
| 1159 |
+
"grad_norm": 3.5447365623585143,
|
| 1160 |
+
"kl": 0.015119111980311573,
|
| 1161 |
+
"learning_rate": 5.656817173965732e-07,
|
| 1162 |
+
"loss": 0.0005974220111966133,
|
| 1163 |
+
"step": 76,
|
| 1164 |
+
"step_time": 74.41445146128535
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"clip_ratio/high_max": 0.0017539991240482777,
|
| 1168 |
+
"clip_ratio/high_mean": 0.0005082542411400937,
|
| 1169 |
+
"clip_ratio/low_mean": 0.0017884014232549816,
|
| 1170 |
+
"clip_ratio/low_min": 0.0,
|
| 1171 |
+
"clip_ratio/region_mean": 0.0022966556753090117,
|
| 1172 |
+
"epoch": 0.48125,
|
| 1173 |
+
"grad_norm": 2.365799490249296,
|
| 1174 |
+
"kl": 0.01375851349439472,
|
| 1175 |
+
"learning_rate": 5.556225645476118e-07,
|
| 1176 |
+
"loss": 0.00037313203210942447,
|
| 1177 |
+
"step": 77,
|
| 1178 |
+
"step_time": 74.46988702565432
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"clip_ratio/high_max": 0.0033521994191687554,
|
| 1182 |
+
"clip_ratio/high_mean": 0.0008875744169927202,
|
| 1183 |
+
"clip_ratio/low_mean": 0.001098724296753062,
|
| 1184 |
+
"clip_ratio/low_min": 0.0,
|
| 1185 |
+
"clip_ratio/region_mean": 0.001986298681003973,
|
| 1186 |
+
"epoch": 0.4875,
|
| 1187 |
+
"grad_norm": 3.128873369580921,
|
| 1188 |
+
"kl": 0.012375441961921751,
|
| 1189 |
+
"learning_rate": 5.455405624202032e-07,
|
| 1190 |
+
"loss": 0.0005378859350457788,
|
| 1191 |
+
"step": 78,
|
| 1192 |
+
"step_time": 73.28777541220188
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"clip_ratio/high_max": 0.0010606571740936488,
|
| 1196 |
+
"clip_ratio/high_mean": 0.00032560917316004634,
|
| 1197 |
+
"clip_ratio/low_mean": 0.0005907298836973496,
|
| 1198 |
+
"clip_ratio/low_min": 0.0,
|
| 1199 |
+
"clip_ratio/region_mean": 0.0009163390568573959,
|
| 1200 |
+
"epoch": 0.49375,
|
| 1201 |
+
"grad_norm": 1.5631906835443126,
|
| 1202 |
+
"kl": 0.009991050465032458,
|
| 1203 |
+
"learning_rate": 5.354398526154365e-07,
|
| 1204 |
+
"loss": 7.550208829343319e-05,
|
| 1205 |
+
"step": 79,
|
| 1206 |
+
"step_time": 74.0536216571927
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"clip_ratio/high_max": 0.005936803790973499,
|
| 1210 |
+
"clip_ratio/high_mean": 0.0019922254941775464,
|
| 1211 |
+
"clip_ratio/low_mean": 0.0009891458503261674,
|
| 1212 |
+
"clip_ratio/low_min": 0.0,
|
| 1213 |
+
"clip_ratio/region_mean": 0.0029813713917974383,
|
| 1214 |
+
"epoch": 0.5,
|
| 1215 |
+
"grad_norm": 2.3522244705151514,
|
| 1216 |
+
"kl": 0.011429251404479146,
|
| 1217 |
+
"learning_rate": 5.253245844193564e-07,
|
| 1218 |
+
"loss": 0.0004219347611069679,
|
| 1219 |
+
"step": 80,
|
| 1220 |
+
"step_time": 74.15648037567735
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"clip_ratio/high_max": 0.0,
|
| 1224 |
+
"clip_ratio/high_mean": 0.0,
|
| 1225 |
+
"clip_ratio/low_mean": 0.0,
|
| 1226 |
+
"clip_ratio/low_min": 0.0,
|
| 1227 |
+
"clip_ratio/region_mean": 0.0,
|
| 1228 |
+
"completions/clipped_ratio": 0.34765625,
|
| 1229 |
+
"completions/max_length": 2048.0,
|
| 1230 |
+
"completions/mean_length": 1236.078125,
|
| 1231 |
+
"completions/min_length": 87.0,
|
| 1232 |
+
"epoch": 0.50625,
|
| 1233 |
+
"frac_reward_zero_std": 0.34375,
|
| 1234 |
+
"grad_norm": 2.622189322771234,
|
| 1235 |
+
"kl": 0.012369139352813363,
|
| 1236 |
+
"learning_rate": 5.151989130984714e-07,
|
| 1237 |
+
"loss": 0.0001236768439412117,
|
| 1238 |
+
"reward": 0.3359375,
|
| 1239 |
+
"reward_std": 0.3031209111213684,
|
| 1240 |
+
"rewards/MajorityAccuracy/mean": 0.3359375,
|
| 1241 |
+
"rewards/MajorityAccuracy/std": 0.4732423424720764,
|
| 1242 |
+
"step": 81,
|
| 1243 |
+
"step_time": 518.1001953482628
|
| 1244 |
+
},
|
| 1245 |
+
{
|
| 1246 |
+
"clip_ratio/high_max": 0.0014928212622180581,
|
| 1247 |
+
"clip_ratio/high_mean": 0.00037320531555451453,
|
| 1248 |
+
"clip_ratio/low_mean": 0.00029844341770512983,
|
| 1249 |
+
"clip_ratio/low_min": 0.0,
|
| 1250 |
+
"clip_ratio/region_mean": 0.0006716487332596444,
|
| 1251 |
+
"epoch": 0.5125,
|
| 1252 |
+
"grad_norm": 2.8437754933575246,
|
| 1253 |
+
"kl": 0.009719027671962976,
|
| 1254 |
+
"learning_rate": 5.050669981928055e-07,
|
| 1255 |
+
"loss": 0.00033772338065318763,
|
| 1256 |
+
"step": 82,
|
| 1257 |
+
"step_time": 73.51018841937184
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"clip_ratio/high_max": 0.006956722369068302,
|
| 1261 |
+
"clip_ratio/high_mean": 0.002116990250215167,
|
| 1262 |
+
"clip_ratio/low_mean": 0.0012615501036634669,
|
| 1263 |
+
"clip_ratio/low_min": 0.0,
|
| 1264 |
+
"clip_ratio/region_mean": 0.0033785403065849096,
|
| 1265 |
+
"epoch": 0.51875,
|
| 1266 |
+
"grad_norm": 2.620775782305953,
|
| 1267 |
+
"kl": 0.01387023413553834,
|
| 1268 |
+
"learning_rate": 4.949330018071946e-07,
|
| 1269 |
+
"loss": 0.0005010068416595459,
|
| 1270 |
+
"step": 83,
|
| 1271 |
+
"step_time": 73.16471273824573
|
| 1272 |
+
},
|
| 1273 |
+
{
|
| 1274 |
+
"clip_ratio/high_max": 0.0005922343698330224,
|
| 1275 |
+
"clip_ratio/high_mean": 0.0001480585924582556,
|
| 1276 |
+
"clip_ratio/low_mean": 0.00014802419536863454,
|
| 1277 |
+
"clip_ratio/low_min": 0.0,
|
| 1278 |
+
"clip_ratio/region_mean": 0.0002960827878268901,
|
| 1279 |
+
"epoch": 0.525,
|
| 1280 |
+
"grad_norm": 1.3551036768181701,
|
| 1281 |
+
"kl": 0.008979964768514037,
|
| 1282 |
+
"learning_rate": 4.848010869015287e-07,
|
| 1283 |
+
"loss": 0.000133507011923939,
|
| 1284 |
+
"step": 84,
|
| 1285 |
+
"step_time": 75.17519514262676
|
| 1286 |
+
},
|
| 1287 |
+
{
|
| 1288 |
+
"clip_ratio/high_max": 0.0017287676528212614,
|
| 1289 |
+
"clip_ratio/high_mean": 0.00043219191320531536,
|
| 1290 |
+
"clip_ratio/low_mean": 0.0003032898639503401,
|
| 1291 |
+
"clip_ratio/low_min": 0.0,
|
| 1292 |
+
"clip_ratio/region_mean": 0.0007354817862506025,
|
| 1293 |
+
"epoch": 0.53125,
|
| 1294 |
+
"grad_norm": 1.418648050397273,
|
| 1295 |
+
"kl": 0.01661685877479613,
|
| 1296 |
+
"learning_rate": 4.746754155806437e-07,
|
| 1297 |
+
"loss": 0.0004885003436356783,
|
| 1298 |
+
"step": 85,
|
| 1299 |
+
"step_time": 75.43796879053116
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"clip_ratio/high_max": 0.0025570314246579073,
|
| 1303 |
+
"clip_ratio/high_mean": 0.0009420207479706733,
|
| 1304 |
+
"clip_ratio/low_mean": 0.001576508664584253,
|
| 1305 |
+
"clip_ratio/low_min": 0.0,
|
| 1306 |
+
"clip_ratio/region_mean": 0.0025185294507537037,
|
| 1307 |
+
"epoch": 0.5375,
|
| 1308 |
+
"grad_norm": 1.7508936555602743,
|
| 1309 |
+
"kl": 0.009869842207990587,
|
| 1310 |
+
"learning_rate": 4.645601473845635e-07,
|
| 1311 |
+
"loss": 0.0003307981532998383,
|
| 1312 |
+
"step": 86,
|
| 1313 |
+
"step_time": 73.44037295877934
|
| 1314 |
+
},
|
| 1315 |
+
{
|
| 1316 |
+
"clip_ratio/high_max": 0.004895088350167498,
|
| 1317 |
+
"clip_ratio/high_mean": 0.001456345482438337,
|
| 1318 |
+
"clip_ratio/low_mean": 0.001624355987587478,
|
| 1319 |
+
"clip_ratio/low_min": 0.0,
|
| 1320 |
+
"clip_ratio/region_mean": 0.003080701470025815,
|
| 1321 |
+
"epoch": 0.54375,
|
| 1322 |
+
"grad_norm": 3.4681549826644007,
|
| 1323 |
+
"kl": 0.024933831533417106,
|
| 1324 |
+
"learning_rate": 4.544594375797968e-07,
|
| 1325 |
+
"loss": 0.0008041909895837307,
|
| 1326 |
+
"step": 87,
|
| 1327 |
+
"step_time": 73.22020553052425
|
| 1328 |
+
},
|
| 1329 |
+
{
|
| 1330 |
+
"clip_ratio/high_max": 0.003208774433005601,
|
| 1331 |
+
"clip_ratio/high_mean": 0.0010142463579541072,
|
| 1332 |
+
"clip_ratio/low_mean": 0.002081554106553085,
|
| 1333 |
+
"clip_ratio/low_min": 0.0,
|
| 1334 |
+
"clip_ratio/region_mean": 0.0030958004790591076,
|
| 1335 |
+
"epoch": 0.55,
|
| 1336 |
+
"grad_norm": 4.131907486882405,
|
| 1337 |
+
"kl": 0.016653883154504,
|
| 1338 |
+
"learning_rate": 4.443774354523882e-07,
|
| 1339 |
+
"loss": 0.000540225300937891,
|
| 1340 |
+
"step": 88,
|
| 1341 |
+
"step_time": 74.37207113578916
|
| 1342 |
+
},
|
| 1343 |
+
{
|
| 1344 |
+
"clip_ratio/high_max": 0.0,
|
| 1345 |
+
"clip_ratio/high_mean": 0.0,
|
| 1346 |
+
"clip_ratio/low_mean": 0.0,
|
| 1347 |
+
"clip_ratio/low_min": 0.0,
|
| 1348 |
+
"clip_ratio/region_mean": 0.0,
|
| 1349 |
+
"completions/clipped_ratio": 0.3671875,
|
| 1350 |
+
"completions/max_length": 2048.0,
|
| 1351 |
+
"completions/mean_length": 1302.16015625,
|
| 1352 |
+
"completions/min_length": 137.0,
|
| 1353 |
+
"epoch": 0.55625,
|
| 1354 |
+
"frac_reward_zero_std": 0.34375,
|
| 1355 |
+
"grad_norm": 1.6890636517934459,
|
| 1356 |
+
"kl": 0.010946551687084138,
|
| 1357 |
+
"learning_rate": 4.3431828260342675e-07,
|
| 1358 |
+
"loss": 0.00010939687490463257,
|
| 1359 |
+
"reward": 0.3984375,
|
| 1360 |
+
"reward_std": 0.27473074197769165,
|
| 1361 |
+
"rewards/MajorityAccuracy/mean": 0.3984375,
|
| 1362 |
+
"rewards/MajorityAccuracy/std": 0.4905354082584381,
|
| 1363 |
+
"step": 89,
|
| 1364 |
+
"step_time": 515.7279123142362
|
| 1365 |
+
},
|
| 1366 |
+
{
|
| 1367 |
+
"clip_ratio/high_max": 0.0013784931506961584,
|
| 1368 |
+
"clip_ratio/high_mean": 0.0003446232876740396,
|
| 1369 |
+
"clip_ratio/low_mean": 6.72984060656745e-05,
|
| 1370 |
+
"clip_ratio/low_min": 0.0,
|
| 1371 |
+
"clip_ratio/region_mean": 0.0004119216937397141,
|
| 1372 |
+
"epoch": 0.5625,
|
| 1373 |
+
"grad_norm": 1.750612774235649,
|
| 1374 |
+
"kl": 0.008445076877251267,
|
| 1375 |
+
"learning_rate": 4.242861112477118e-07,
|
| 1376 |
+
"loss": 7.06603386788629e-05,
|
| 1377 |
+
"step": 90,
|
| 1378 |
+
"step_time": 73.98557273671031
|
| 1379 |
+
},
|
| 1380 |
+
{
|
| 1381 |
+
"clip_ratio/high_max": 0.00043226721027167514,
|
| 1382 |
+
"clip_ratio/high_mean": 0.00010806680256791878,
|
| 1383 |
+
"clip_ratio/low_mean": 0.0003528989036567509,
|
| 1384 |
+
"clip_ratio/low_min": 0.0,
|
| 1385 |
+
"clip_ratio/region_mean": 0.0004609657062246697,
|
| 1386 |
+
"epoch": 0.56875,
|
| 1387 |
+
"grad_norm": 1.276700991450794,
|
| 1388 |
+
"kl": 0.009478250576648861,
|
| 1389 |
+
"learning_rate": 4.1428504251627325e-07,
|
| 1390 |
+
"loss": 0.00014478375669568777,
|
| 1391 |
+
"step": 91,
|
| 1392 |
+
"step_time": 74.29446458816528
|
| 1393 |
+
},
|
| 1394 |
+
{
|
| 1395 |
+
"clip_ratio/high_max": 0.0007421274931402877,
|
| 1396 |
+
"clip_ratio/high_mean": 0.00022072331921663135,
|
| 1397 |
+
"clip_ratio/low_mean": 0.00020480176317505538,
|
| 1398 |
+
"clip_ratio/low_min": 0.0,
|
| 1399 |
+
"clip_ratio/region_mean": 0.0004255250751157291,
|
| 1400 |
+
"epoch": 0.575,
|
| 1401 |
+
"grad_norm": 1.8859919026999061,
|
| 1402 |
+
"kl": 0.014804138336330652,
|
| 1403 |
+
"learning_rate": 4.0431918476344685e-07,
|
| 1404 |
+
"loss": 0.00026248066569678485,
|
| 1405 |
+
"step": 92,
|
| 1406 |
+
"step_time": 72.84182591363788
|
| 1407 |
+
},
|
| 1408 |
+
{
|
| 1409 |
+
"clip_ratio/high_max": 0.002077397279208526,
|
| 1410 |
+
"clip_ratio/high_mean": 0.0007388126650766935,
|
| 1411 |
+
"clip_ratio/low_mean": 0.0005591123008343857,
|
| 1412 |
+
"clip_ratio/low_min": 0.0,
|
| 1413 |
+
"clip_ratio/region_mean": 0.0012979249768250156,
|
| 1414 |
+
"epoch": 0.58125,
|
| 1415 |
+
"grad_norm": 2.502773606113119,
|
| 1416 |
+
"kl": 0.012178544304333627,
|
| 1417 |
+
"learning_rate": 3.943926318791963e-07,
|
| 1418 |
+
"loss": 0.0003373315557837486,
|
| 1419 |
+
"step": 93,
|
| 1420 |
+
"step_time": 73.97821111232042
|
| 1421 |
+
},
|
| 1422 |
+
{
|
| 1423 |
+
"clip_ratio/high_max": 0.00435410367208533,
|
| 1424 |
+
"clip_ratio/high_mean": 0.001478655765822623,
|
| 1425 |
+
"clip_ratio/low_mean": 0.0003833878618024755,
|
| 1426 |
+
"clip_ratio/low_min": 0.0,
|
| 1427 |
+
"clip_ratio/region_mean": 0.0018620436458149925,
|
| 1428 |
+
"epoch": 0.5875,
|
| 1429 |
+
"grad_norm": 1.460053677657597,
|
| 1430 |
+
"kl": 0.015562864835374057,
|
| 1431 |
+
"learning_rate": 3.845094616073783e-07,
|
| 1432 |
+
"loss": 0.0003096975851804018,
|
| 1433 |
+
"step": 94,
|
| 1434 |
+
"step_time": 72.95763423666358
|
| 1435 |
+
},
|
| 1436 |
+
{
|
| 1437 |
+
"clip_ratio/high_max": 0.00045289855916053057,
|
| 1438 |
+
"clip_ratio/high_mean": 0.00011322463979013264,
|
| 1439 |
+
"clip_ratio/low_mean": 0.0005038403141952585,
|
| 1440 |
+
"clip_ratio/low_min": 0.0,
|
| 1441 |
+
"clip_ratio/region_mean": 0.0006170649539853912,
|
| 1442 |
+
"epoch": 0.59375,
|
| 1443 |
+
"grad_norm": 0.9840836397434568,
|
| 1444 |
+
"kl": 0.011064775753766298,
|
| 1445 |
+
"learning_rate": 3.7467373387063964e-07,
|
| 1446 |
+
"loss": 0.00020056264474987984,
|
| 1447 |
+
"step": 95,
|
| 1448 |
+
"step_time": 73.68392396718264
|
| 1449 |
+
},
|
| 1450 |
+
{
|
| 1451 |
+
"clip_ratio/high_max": 0.0034367612388450652,
|
| 1452 |
+
"clip_ratio/high_mean": 0.0008591903097112663,
|
| 1453 |
+
"clip_ratio/low_mean": 0.0006376958954206202,
|
| 1454 |
+
"clip_ratio/low_min": 0.0,
|
| 1455 |
+
"clip_ratio/region_mean": 0.0014968862233217806,
|
| 1456 |
+
"epoch": 0.6,
|
| 1457 |
+
"grad_norm": 2.0888601529659634,
|
| 1458 |
+
"kl": 0.017328333691693842,
|
| 1459 |
+
"learning_rate": 3.648894891026358e-07,
|
| 1460 |
+
"loss": 0.000281934771919623,
|
| 1461 |
+
"step": 96,
|
| 1462 |
+
"step_time": 74.91581713780761
|
| 1463 |
+
},
|
| 1464 |
+
{
|
| 1465 |
+
"clip_ratio/high_max": 0.0,
|
| 1466 |
+
"clip_ratio/high_mean": 0.0,
|
| 1467 |
+
"clip_ratio/low_mean": 0.0,
|
| 1468 |
+
"clip_ratio/low_min": 0.0,
|
| 1469 |
+
"clip_ratio/region_mean": 0.0,
|
| 1470 |
+
"completions/clipped_ratio": 0.44140625,
|
| 1471 |
+
"completions/max_length": 2048.0,
|
| 1472 |
+
"completions/mean_length": 1375.79296875,
|
| 1473 |
+
"completions/min_length": 99.0,
|
| 1474 |
+
"epoch": 0.60625,
|
| 1475 |
+
"frac_reward_zero_std": 0.375,
|
| 1476 |
+
"grad_norm": 1.7390698926261747,
|
| 1477 |
+
"kl": 0.011745358002372086,
|
| 1478 |
+
"learning_rate": 3.5516074658825397e-07,
|
| 1479 |
+
"loss": 0.00011746026575565338,
|
| 1480 |
+
"reward": 0.375,
|
| 1481 |
+
"reward_std": 0.2710474133491516,
|
| 1482 |
+
"rewards/MajorityAccuracy/mean": 0.375,
|
| 1483 |
+
"rewards/MajorityAccuracy/std": 0.48507124185562134,
|
| 1484 |
+
"step": 97,
|
| 1485 |
+
"step_time": 515.2795390300453
|
| 1486 |
+
},
|
| 1487 |
+
{
|
| 1488 |
+
"clip_ratio/high_max": 0.002640025661094114,
|
| 1489 |
+
"clip_ratio/high_mean": 0.0006600064152735285,
|
| 1490 |
+
"clip_ratio/low_mean": 0.00021775994537165388,
|
| 1491 |
+
"clip_ratio/low_min": 0.0,
|
| 1492 |
+
"clip_ratio/region_mean": 0.0008777663606451824,
|
| 1493 |
+
"epoch": 0.6125,
|
| 1494 |
+
"grad_norm": 1.3676165718200044,
|
| 1495 |
+
"kl": 0.012280755094252527,
|
| 1496 |
+
"learning_rate": 3.454915028125263e-07,
|
| 1497 |
+
"loss": 0.00023002100351732224,
|
| 1498 |
+
"step": 98,
|
| 1499 |
+
"step_time": 75.09991871193051
|
| 1500 |
+
},
|
| 1501 |
+
{
|
| 1502 |
+
"clip_ratio/high_max": 0.0014687925577163696,
|
| 1503 |
+
"clip_ratio/high_mean": 0.0004383286432130262,
|
| 1504 |
+
"clip_ratio/low_mean": 0.0003424110618652776,
|
| 1505 |
+
"clip_ratio/low_min": 0.0,
|
| 1506 |
+
"clip_ratio/region_mean": 0.0007807397050783038,
|
| 1507 |
+
"epoch": 0.61875,
|
| 1508 |
+
"grad_norm": 1.6577299596580408,
|
| 1509 |
+
"kl": 0.011880950652994215,
|
| 1510 |
+
"learning_rate": 3.3588572981890684e-07,
|
| 1511 |
+
"loss": 0.0002540815621614456,
|
| 1512 |
+
"step": 99,
|
| 1513 |
+
"step_time": 73.346854288131
|
| 1514 |
+
},
|
| 1515 |
+
{
|
| 1516 |
+
"clip_ratio/high_max": 0.0020868265273747966,
|
| 1517 |
+
"clip_ratio/high_mean": 0.0005217066318436991,
|
| 1518 |
+
"clip_ratio/low_mean": 8.159269054885954e-05,
|
| 1519 |
+
"clip_ratio/low_min": 0.0,
|
| 1520 |
+
"clip_ratio/region_mean": 0.0006032993223925587,
|
| 1521 |
+
"epoch": 0.625,
|
| 1522 |
+
"grad_norm": 1.156158720671418,
|
| 1523 |
+
"kl": 0.011838753940537572,
|
| 1524 |
+
"learning_rate": 3.263473735775899e-07,
|
| 1525 |
+
"loss": 2.9482842364814132e-05,
|
| 1526 |
+
"step": 100,
|
| 1527 |
+
"step_time": 74.05724911019206
|
| 1528 |
+
},
|
| 1529 |
+
{
|
| 1530 |
+
"clip_ratio/high_max": 0.003061715862713754,
|
| 1531 |
+
"clip_ratio/high_mean": 0.0009274477342842147,
|
| 1532 |
+
"clip_ratio/low_mean": 0.0003268284854129888,
|
| 1533 |
+
"clip_ratio/low_min": 0.0,
|
| 1534 |
+
"clip_ratio/region_mean": 0.0012542762124212459,
|
| 1535 |
+
"epoch": 0.63125,
|
| 1536 |
+
"grad_norm": 1.81955314925597,
|
| 1537 |
+
"kl": 0.012183128274045885,
|
| 1538 |
+
"learning_rate": 3.1688035236453865e-07,
|
| 1539 |
+
"loss": 0.0004271415527909994,
|
| 1540 |
+
"step": 101,
|
| 1541 |
+
"step_time": 73.91903552040458
|
| 1542 |
+
},
|
| 1543 |
+
{
|
| 1544 |
+
"clip_ratio/high_max": 0.0072146987658925354,
|
| 1545 |
+
"clip_ratio/high_mean": 0.0018450606585247442,
|
| 1546 |
+
"clip_ratio/low_mean": 0.0009235324905603193,
|
| 1547 |
+
"clip_ratio/low_min": 0.0,
|
| 1548 |
+
"clip_ratio/region_mean": 0.002768593174550915,
|
| 1549 |
+
"epoch": 0.6375,
|
| 1550 |
+
"grad_norm": 8.276150310995108,
|
| 1551 |
+
"kl": 0.02035100618377328,
|
| 1552 |
+
"learning_rate": 3.0748855515189096e-07,
|
| 1553 |
+
"loss": 5.1805654948111624e-05,
|
| 1554 |
+
"step": 102,
|
| 1555 |
+
"step_time": 75.08180133625865
|
| 1556 |
+
},
|
| 1557 |
+
{
|
| 1558 |
+
"clip_ratio/high_max": 0.003777011181227863,
|
| 1559 |
+
"clip_ratio/high_mean": 0.0012401808344293386,
|
| 1560 |
+
"clip_ratio/low_mean": 0.001006565504212631,
|
| 1561 |
+
"clip_ratio/low_min": 0.0,
|
| 1562 |
+
"clip_ratio/region_mean": 0.0022467463568318635,
|
| 1563 |
+
"epoch": 0.64375,
|
| 1564 |
+
"grad_norm": 3.5015048704892062,
|
| 1565 |
+
"kl": 0.025227537378668785,
|
| 1566 |
+
"learning_rate": 2.981758400104028e-07,
|
| 1567 |
+
"loss": 0.0004210706683807075,
|
| 1568 |
+
"step": 103,
|
| 1569 |
+
"step_time": 71.87072990834713
|
| 1570 |
+
},
|
| 1571 |
+
{
|
| 1572 |
+
"clip_ratio/high_max": 0.0012535296846181154,
|
| 1573 |
+
"clip_ratio/high_mean": 0.00031338242115452886,
|
| 1574 |
+
"clip_ratio/low_mean": 0.000886982299562078,
|
| 1575 |
+
"clip_ratio/low_min": 0.0,
|
| 1576 |
+
"clip_ratio/region_mean": 0.001200364698888734,
|
| 1577 |
+
"epoch": 0.65,
|
| 1578 |
+
"grad_norm": 2.2080973299387154,
|
| 1579 |
+
"kl": 0.01101521635428071,
|
| 1580 |
+
"learning_rate": 2.8894603252458403e-07,
|
| 1581 |
+
"loss": 2.7998758014291525e-05,
|
| 1582 |
+
"step": 104,
|
| 1583 |
+
"step_time": 73.61084084585309
|
| 1584 |
+
},
|
| 1585 |
+
{
|
| 1586 |
+
"clip_ratio/high_max": 0.0,
|
| 1587 |
+
"clip_ratio/high_mean": 0.0,
|
| 1588 |
+
"clip_ratio/low_mean": 0.0,
|
| 1589 |
+
"clip_ratio/low_min": 0.0,
|
| 1590 |
+
"clip_ratio/region_mean": 0.0,
|
| 1591 |
+
"completions/clipped_ratio": 0.28515625,
|
| 1592 |
+
"completions/max_length": 2048.0,
|
| 1593 |
+
"completions/mean_length": 1171.671875,
|
| 1594 |
+
"completions/min_length": 127.0,
|
| 1595 |
+
"epoch": 0.65625,
|
| 1596 |
+
"frac_reward_zero_std": 0.21875,
|
| 1597 |
+
"grad_norm": 1.7238105756454007,
|
| 1598 |
+
"kl": 0.011362664634361863,
|
| 1599 |
+
"learning_rate": 2.7980292422118277e-07,
|
| 1600 |
+
"loss": 0.00011365055979695171,
|
| 1601 |
+
"reward": 0.44140625,
|
| 1602 |
+
"reward_std": 0.360870897769928,
|
| 1603 |
+
"rewards/MajorityAccuracy/mean": 0.44140625,
|
| 1604 |
+
"rewards/MajorityAccuracy/std": 0.4975275695323944,
|
| 1605 |
+
"step": 105,
|
| 1606 |
+
"step_time": 514.9249229617417
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"clip_ratio/high_max": 0.0032288123256876133,
|
| 1610 |
+
"clip_ratio/high_mean": 0.0009273582054447616,
|
| 1611 |
+
"clip_ratio/low_mean": 0.0009988903475459665,
|
| 1612 |
+
"clip_ratio/low_min": 0.0,
|
| 1613 |
+
"clip_ratio/region_mean": 0.00192624851661094,
|
| 1614 |
+
"epoch": 0.6625,
|
| 1615 |
+
"grad_norm": 3.5324443135605637,
|
| 1616 |
+
"kl": 0.018030510284006596,
|
| 1617 |
+
"learning_rate": 2.70750271011657e-07,
|
| 1618 |
+
"loss": 9.713508188724518e-05,
|
| 1619 |
+
"step": 106,
|
| 1620 |
+
"step_time": 72.83120188489556
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"clip_ratio/high_max": 0.0030124174227239564,
|
| 1624 |
+
"clip_ratio/high_mean": 0.0010246479578199796,
|
| 1625 |
+
"clip_ratio/low_mean": 0.0011223115507164039,
|
| 1626 |
+
"clip_ratio/low_min": 0.0,
|
| 1627 |
+
"clip_ratio/region_mean": 0.0021469595230882987,
|
| 1628 |
+
"epoch": 0.66875,
|
| 1629 |
+
"grad_norm": 7.189781803241237,
|
| 1630 |
+
"kl": 0.01880924787838012,
|
| 1631 |
+
"learning_rate": 2.6179179164927754e-07,
|
| 1632 |
+
"loss": 0.000497031956911087,
|
| 1633 |
+
"step": 107,
|
| 1634 |
+
"step_time": 72.85796255990863
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"clip_ratio/high_max": 0.005722216534195468,
|
| 1638 |
+
"clip_ratio/high_mean": 0.001632646053622011,
|
| 1639 |
+
"clip_ratio/low_mean": 0.00035519834273145534,
|
| 1640 |
+
"clip_ratio/low_min": 0.0,
|
| 1641 |
+
"clip_ratio/region_mean": 0.0019878443781635724,
|
| 1642 |
+
"epoch": 0.675,
|
| 1643 |
+
"grad_norm": 2.966245995171975,
|
| 1644 |
+
"kl": 0.01314480195287615,
|
| 1645 |
+
"learning_rate": 2.529311662014972e-07,
|
| 1646 |
+
"loss": 0.0001007401806418784,
|
| 1647 |
+
"step": 108,
|
| 1648 |
+
"step_time": 75.1331765241921
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"clip_ratio/high_max": 0.0015820536500541493,
|
| 1652 |
+
"clip_ratio/high_mean": 0.0005858026597707067,
|
| 1653 |
+
"clip_ratio/low_mean": 0.00045593718823511153,
|
| 1654 |
+
"clip_ratio/low_min": 0.0,
|
| 1655 |
+
"clip_ratio/region_mean": 0.0010417398443678394,
|
| 1656 |
+
"epoch": 0.68125,
|
| 1657 |
+
"grad_norm": 1.7750685590418132,
|
| 1658 |
+
"kl": 0.008828479330986738,
|
| 1659 |
+
"learning_rate": 2.441720345382089e-07,
|
| 1660 |
+
"loss": 0.00026102495030499995,
|
| 1661 |
+
"step": 109,
|
| 1662 |
+
"step_time": 73.36835189163685
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"clip_ratio/high_max": 0.006236149150936399,
|
| 1666 |
+
"clip_ratio/high_mean": 0.001846955792643712,
|
| 1667 |
+
"clip_ratio/low_mean": 0.00038982358455541544,
|
| 1668 |
+
"clip_ratio/low_min": 0.0,
|
| 1669 |
+
"clip_ratio/region_mean": 0.0022367794008459896,
|
| 1670 |
+
"epoch": 0.6875,
|
| 1671 |
+
"grad_norm": 2.0417270214639665,
|
| 1672 |
+
"kl": 0.015330304624512792,
|
| 1673 |
+
"learning_rate": 2.355179948365189e-07,
|
| 1674 |
+
"loss": 0.00013756699627265334,
|
| 1675 |
+
"step": 110,
|
| 1676 |
+
"step_time": 72.68941617757082
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"clip_ratio/high_max": 0.0036506951437331736,
|
| 1680 |
+
"clip_ratio/high_mean": 0.0012910459045087919,
|
| 1681 |
+
"clip_ratio/low_mean": 0.0004657923273043707,
|
| 1682 |
+
"clip_ratio/low_min": 0.0,
|
| 1683 |
+
"clip_ratio/region_mean": 0.0017568382172612473,
|
| 1684 |
+
"epoch": 0.69375,
|
| 1685 |
+
"grad_norm": 3.51909142119646,
|
| 1686 |
+
"kl": 0.012806993792764843,
|
| 1687 |
+
"learning_rate": 2.26972602102645e-07,
|
| 1688 |
+
"loss": 0.0005704723298549652,
|
| 1689 |
+
"step": 111,
|
| 1690 |
+
"step_time": 72.50887331739068
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"clip_ratio/high_max": 0.003250174893764779,
|
| 1694 |
+
"clip_ratio/high_mean": 0.0011521263513714075,
|
| 1695 |
+
"clip_ratio/low_mean": 0.0003825086387223564,
|
| 1696 |
+
"clip_ratio/low_min": 0.0,
|
| 1697 |
+
"clip_ratio/region_mean": 0.0015346349864557851,
|
| 1698 |
+
"epoch": 0.7,
|
| 1699 |
+
"grad_norm": 1.8475829953341814,
|
| 1700 |
+
"kl": 0.010050091077573597,
|
| 1701 |
+
"learning_rate": 2.1853936671155127e-07,
|
| 1702 |
+
"loss": 0.00016183272236958146,
|
| 1703 |
+
"step": 112,
|
| 1704 |
+
"step_time": 75.24181795865297
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"clip_ratio/high_max": 0.0,
|
| 1708 |
+
"clip_ratio/high_mean": 0.0,
|
| 1709 |
+
"clip_ratio/low_mean": 0.0,
|
| 1710 |
+
"clip_ratio/low_min": 0.0,
|
| 1711 |
+
"clip_ratio/region_mean": 0.0,
|
| 1712 |
+
"completions/clipped_ratio": 0.359375,
|
| 1713 |
+
"completions/max_length": 2048.0,
|
| 1714 |
+
"completions/mean_length": 1303.921875,
|
| 1715 |
+
"completions/min_length": 115.0,
|
| 1716 |
+
"epoch": 0.70625,
|
| 1717 |
+
"frac_reward_zero_std": 0.21875,
|
| 1718 |
+
"grad_norm": 2.4810809834675225,
|
| 1719 |
+
"kl": 0.01335794455371797,
|
| 1720 |
+
"learning_rate": 2.1022175296491512e-07,
|
| 1721 |
+
"loss": 0.00013362988829612732,
|
| 1722 |
+
"reward": 0.44140625,
|
| 1723 |
+
"reward_std": 0.37377026677131653,
|
| 1724 |
+
"rewards/MajorityAccuracy/mean": 0.44140625,
|
| 1725 |
+
"rewards/MajorityAccuracy/std": 0.4975275695323944,
|
| 1726 |
+
"step": 113,
|
| 1727 |
+
"step_time": 517.2925749048591
|
| 1728 |
+
},
|
| 1729 |
+
{
|
| 1730 |
+
"clip_ratio/high_max": 0.0031772586662555113,
|
| 1731 |
+
"clip_ratio/high_mean": 0.00109138917105156,
|
| 1732 |
+
"clip_ratio/low_mean": 6.103515625e-05,
|
| 1733 |
+
"clip_ratio/low_min": 0.0,
|
| 1734 |
+
"clip_ratio/region_mean": 0.00115242432730156,
|
| 1735 |
+
"epoch": 0.7125,
|
| 1736 |
+
"grad_norm": 3.3350897393893284,
|
| 1737 |
+
"kl": 0.013246290269307792,
|
| 1738 |
+
"learning_rate": 2.0202317766802152e-07,
|
| 1739 |
+
"loss": 0.00044062454253435135,
|
| 1740 |
+
"step": 114,
|
| 1741 |
+
"step_time": 73.19391025602818
|
| 1742 |
+
},
|
| 1743 |
+
{
|
| 1744 |
+
"clip_ratio/high_max": 0.0007659874609089456,
|
| 1745 |
+
"clip_ratio/high_mean": 0.0001914968652272364,
|
| 1746 |
+
"clip_ratio/low_mean": 3.0517578125e-05,
|
| 1747 |
+
"clip_ratio/low_min": 0.0,
|
| 1748 |
+
"clip_ratio/region_mean": 0.0002220144433522364,
|
| 1749 |
+
"epoch": 0.71875,
|
| 1750 |
+
"grad_norm": 0.8530541529806249,
|
| 1751 |
+
"kl": 0.008775157621130347,
|
| 1752 |
+
"learning_rate": 1.9394700872616853e-07,
|
| 1753 |
+
"loss": -1.812746631912887e-06,
|
| 1754 |
+
"step": 115,
|
| 1755 |
+
"step_time": 75.27188849821687
|
| 1756 |
+
},
|
| 1757 |
+
{
|
| 1758 |
+
"clip_ratio/high_max": 0.001298512113862671,
|
| 1759 |
+
"clip_ratio/high_mean": 0.0003837762960756663,
|
| 1760 |
+
"clip_ratio/low_mean": 0.00029528792947530746,
|
| 1761 |
+
"clip_ratio/low_min": 0.0,
|
| 1762 |
+
"clip_ratio/region_mean": 0.0006790642182750162,
|
| 1763 |
+
"epoch": 0.725,
|
| 1764 |
+
"grad_norm": 1.5977334517031416,
|
| 1765 |
+
"kl": 0.008567865588702261,
|
| 1766 |
+
"learning_rate": 1.8599656376116024e-07,
|
| 1767 |
+
"loss": 0.00019357977726031095,
|
| 1768 |
+
"step": 116,
|
| 1769 |
+
"step_time": 74.1821870803833
|
| 1770 |
+
},
|
| 1771 |
+
{
|
| 1772 |
+
"clip_ratio/high_max": 0.0035068943980149925,
|
| 1773 |
+
"clip_ratio/high_mean": 0.0009114844433497638,
|
| 1774 |
+
"clip_ratio/low_mean": 0.0007376965513685718,
|
| 1775 |
+
"clip_ratio/low_min": 0.0,
|
| 1776 |
+
"clip_ratio/region_mean": 0.0016491809947183356,
|
| 1777 |
+
"epoch": 0.73125,
|
| 1778 |
+
"grad_norm": 3.3072322335542426,
|
| 1779 |
+
"kl": 0.011482867877930403,
|
| 1780 |
+
"learning_rate": 1.7817510874845582e-07,
|
| 1781 |
+
"loss": 0.00033505389001220465,
|
| 1782 |
+
"step": 117,
|
| 1783 |
+
"step_time": 75.24912935495377
|
| 1784 |
+
},
|
| 1785 |
+
{
|
| 1786 |
+
"clip_ratio/high_max": 0.0036398576921783388,
|
| 1787 |
+
"clip_ratio/high_mean": 0.0013193399354349822,
|
| 1788 |
+
"clip_ratio/low_mean": 0.0008404114887525793,
|
| 1789 |
+
"clip_ratio/low_min": 0.0,
|
| 1790 |
+
"clip_ratio/region_mean": 0.0021597514423774555,
|
| 1791 |
+
"epoch": 0.7375,
|
| 1792 |
+
"grad_norm": 2.940068351767873,
|
| 1793 |
+
"kl": 0.015434119966812432,
|
| 1794 |
+
"learning_rate": 1.7048585667553412e-07,
|
| 1795 |
+
"loss": 0.00035813357681035995,
|
| 1796 |
+
"step": 118,
|
| 1797 |
+
"step_time": 73.13601204380393
|
| 1798 |
+
},
|
| 1799 |
+
{
|
| 1800 |
+
"clip_ratio/high_max": 0.0032066218118416145,
|
| 1801 |
+
"clip_ratio/high_mean": 0.0011562564941414166,
|
| 1802 |
+
"clip_ratio/low_mean": 0.0005283398131723516,
|
| 1803 |
+
"clip_ratio/low_min": 0.0,
|
| 1804 |
+
"clip_ratio/region_mean": 0.0016845963018568,
|
| 1805 |
+
"epoch": 0.74375,
|
| 1806 |
+
"grad_norm": 5.4758540903031045,
|
| 1807 |
+
"kl": 0.01813096913974732,
|
| 1808 |
+
"learning_rate": 1.6293196622202632e-07,
|
| 1809 |
+
"loss": 0.0007301587611436844,
|
| 1810 |
+
"step": 119,
|
| 1811 |
+
"step_time": 73.82561327517033
|
| 1812 |
+
},
|
| 1813 |
+
{
|
| 1814 |
+
"clip_ratio/high_max": 0.0011147513141622767,
|
| 1815 |
+
"clip_ratio/high_mean": 0.0005462391891342122,
|
| 1816 |
+
"clip_ratio/low_mean": 0.00037287936538632493,
|
| 1817 |
+
"clip_ratio/low_min": 0.0,
|
| 1818 |
+
"clip_ratio/region_mean": 0.000919118558158516,
|
| 1819 |
+
"epoch": 0.75,
|
| 1820 |
+
"grad_norm": 1.3823651228710898,
|
| 1821 |
+
"kl": 0.02613177616149187,
|
| 1822 |
+
"learning_rate": 1.555165404621567e-07,
|
| 1823 |
+
"loss": 0.0004199602408334613,
|
| 1824 |
+
"step": 120,
|
| 1825 |
+
"step_time": 74.81176898255944
|
| 1826 |
+
},
|
| 1827 |
+
{
|
| 1828 |
+
"clip_ratio/high_max": 0.0,
|
| 1829 |
+
"clip_ratio/high_mean": 0.0,
|
| 1830 |
+
"clip_ratio/low_mean": 0.0,
|
| 1831 |
+
"clip_ratio/low_min": 0.0,
|
| 1832 |
+
"clip_ratio/region_mean": 0.0,
|
| 1833 |
+
"completions/clipped_ratio": 0.203125,
|
| 1834 |
+
"completions/max_length": 2048.0,
|
| 1835 |
+
"completions/mean_length": 1061.8125,
|
| 1836 |
+
"completions/min_length": 139.0,
|
| 1837 |
+
"epoch": 0.75625,
|
| 1838 |
+
"frac_reward_zero_std": 0.375,
|
| 1839 |
+
"grad_norm": 1.4043354330333686,
|
| 1840 |
+
"kl": 0.013537512277252972,
|
| 1841 |
+
"learning_rate": 1.4824262559002592e-07,
|
| 1842 |
+
"loss": 0.00013534000026993454,
|
| 1843 |
+
"reward": 0.48828125,
|
| 1844 |
+
"reward_std": 0.28182756900787354,
|
| 1845 |
+
"rewards/MajorityAccuracy/mean": 0.48828125,
|
| 1846 |
+
"rewards/MajorityAccuracy/std": 0.5008417963981628,
|
| 1847 |
+
"step": 121,
|
| 1848 |
+
"step_time": 508.2895901352167
|
| 1849 |
+
},
|
| 1850 |
+
{
|
| 1851 |
+
"clip_ratio/high_max": 0.002292546341777779,
|
| 1852 |
+
"clip_ratio/high_mean": 0.0007328741157834884,
|
| 1853 |
+
"clip_ratio/low_mean": 0.00047343446385639254,
|
| 1854 |
+
"clip_ratio/low_min": 0.0,
|
| 1855 |
+
"clip_ratio/region_mean": 0.0012063085850968491,
|
| 1856 |
+
"epoch": 0.7625,
|
| 1857 |
+
"grad_norm": 3.5091795086410764,
|
| 1858 |
+
"kl": 0.013237339677289128,
|
| 1859 |
+
"learning_rate": 1.4111320966826057e-07,
|
| 1860 |
+
"loss": 0.00034992489963769913,
|
| 1861 |
+
"step": 122,
|
| 1862 |
+
"step_time": 72.08983197808266
|
| 1863 |
+
},
|
| 1864 |
+
{
|
| 1865 |
+
"clip_ratio/high_max": 0.002044293549261056,
|
| 1866 |
+
"clip_ratio/high_mean": 0.000511073387315264,
|
| 1867 |
+
"clip_ratio/low_mean": 0.00020953363127773628,
|
| 1868 |
+
"clip_ratio/low_min": 0.0,
|
| 1869 |
+
"clip_ratio/region_mean": 0.0007206070258689579,
|
| 1870 |
+
"epoch": 0.76875,
|
| 1871 |
+
"grad_norm": 1.5028391455142824,
|
| 1872 |
+
"kl": 0.020657094195485115,
|
| 1873 |
+
"learning_rate": 1.3413122140054217e-07,
|
| 1874 |
+
"loss": 0.0003222337691113353,
|
| 1875 |
+
"step": 123,
|
| 1876 |
+
"step_time": 72.58411259949207
|
| 1877 |
+
},
|
| 1878 |
+
{
|
| 1879 |
+
"clip_ratio/high_max": 0.003064046672079712,
|
| 1880 |
+
"clip_ratio/high_mean": 0.0011300557052891236,
|
| 1881 |
+
"clip_ratio/low_mean": 0.0003669048164738342,
|
| 1882 |
+
"clip_ratio/low_min": 0.0,
|
| 1883 |
+
"clip_ratio/region_mean": 0.0014969605144870002,
|
| 1884 |
+
"epoch": 0.775,
|
| 1885 |
+
"grad_norm": 3.327586534858658,
|
| 1886 |
+
"kl": 0.01454687223304063,
|
| 1887 |
+
"learning_rate": 1.272995289285202e-07,
|
| 1888 |
+
"loss": 0.000593479722738266,
|
| 1889 |
+
"step": 124,
|
| 1890 |
+
"step_time": 73.5287130586803
|
| 1891 |
+
},
|
| 1892 |
+
{
|
| 1893 |
+
"clip_ratio/high_max": 0.0019552623998606578,
|
| 1894 |
+
"clip_ratio/high_mean": 0.0005907177583139855,
|
| 1895 |
+
"clip_ratio/low_mean": 0.0007302112535398919,
|
| 1896 |
+
"clip_ratio/low_min": 0.0,
|
| 1897 |
+
"clip_ratio/region_mean": 0.0013209290082158986,
|
| 1898 |
+
"epoch": 0.78125,
|
| 1899 |
+
"grad_norm": 1.8453488397518392,
|
| 1900 |
+
"kl": 0.011441939859651029,
|
| 1901 |
+
"learning_rate": 1.2062093865360457e-07,
|
| 1902 |
+
"loss": -0.00015172857092693448,
|
| 1903 |
+
"step": 125,
|
| 1904 |
+
"step_time": 73.56624042615294
|
| 1905 |
+
},
|
| 1906 |
+
{
|
| 1907 |
+
"clip_ratio/high_max": 0.002344116132007912,
|
| 1908 |
+
"clip_ratio/high_mean": 0.0008710784168215469,
|
| 1909 |
+
"clip_ratio/low_mean": 0.00044671280556940474,
|
| 1910 |
+
"clip_ratio/low_min": 0.0,
|
| 1911 |
+
"clip_ratio/region_mean": 0.0013177912187529728,
|
| 1912 |
+
"epoch": 0.7875,
|
| 1913 |
+
"grad_norm": 1.912455297236884,
|
| 1914 |
+
"kl": 0.01745795365422964,
|
| 1915 |
+
"learning_rate": 1.1409819408411897e-07,
|
| 1916 |
+
"loss": 0.00016700007836334407,
|
| 1917 |
+
"step": 126,
|
| 1918 |
+
"step_time": 74.653903901577
|
| 1919 |
+
},
|
| 1920 |
+
{
|
| 1921 |
+
"clip_ratio/high_max": 0.0016285306628560647,
|
| 1922 |
+
"clip_ratio/high_mean": 0.0005137808657309506,
|
| 1923 |
+
"clip_ratio/low_mean": 9.641111682867631e-05,
|
| 1924 |
+
"clip_ratio/low_min": 0.0,
|
| 1925 |
+
"clip_ratio/region_mean": 0.0006101919861976057,
|
| 1926 |
+
"epoch": 0.79375,
|
| 1927 |
+
"grad_norm": 1.3868575257639038,
|
| 1928 |
+
"kl": 0.013133186381310225,
|
| 1929 |
+
"learning_rate": 1.0773397470829143e-07,
|
| 1930 |
+
"loss": 0.00020790011330973357,
|
| 1931 |
+
"step": 127,
|
| 1932 |
+
"step_time": 73.78880555555224
|
| 1933 |
+
},
|
| 1934 |
+
{
|
| 1935 |
+
"clip_ratio/high_max": 0.001335989909421187,
|
| 1936 |
+
"clip_ratio/high_mean": 0.0003339974773552967,
|
| 1937 |
+
"clip_ratio/low_mean": 3.742515036719851e-05,
|
| 1938 |
+
"clip_ratio/low_min": 0.0,
|
| 1939 |
+
"clip_ratio/region_mean": 0.00037142262772249524,
|
| 1940 |
+
"epoch": 0.8,
|
| 1941 |
+
"grad_norm": 1.0679490195363799,
|
| 1942 |
+
"kl": 0.013708683662116528,
|
| 1943 |
+
"learning_rate": 1.0153089489354256e-07,
|
| 1944 |
+
"loss": 0.0001100408990168944,
|
| 1945 |
+
"step": 128,
|
| 1946 |
+
"step_time": 72.95284855738282
|
| 1947 |
+
},
|
| 1948 |
+
{
|
| 1949 |
+
"clip_ratio/high_max": 0.0,
|
| 1950 |
+
"clip_ratio/high_mean": 0.0,
|
| 1951 |
+
"clip_ratio/low_mean": 0.0,
|
| 1952 |
+
"clip_ratio/low_min": 0.0,
|
| 1953 |
+
"clip_ratio/region_mean": 0.0,
|
| 1954 |
+
"completions/clipped_ratio": 0.2890625,
|
| 1955 |
+
"completions/max_length": 2048.0,
|
| 1956 |
+
"completions/mean_length": 1171.8359375,
|
| 1957 |
+
"completions/min_length": 69.0,
|
| 1958 |
+
"epoch": 0.80625,
|
| 1959 |
+
"frac_reward_zero_std": 0.21875,
|
| 1960 |
+
"grad_norm": 1.4417164083669582,
|
| 1961 |
+
"kl": 0.021758307702839375,
|
| 1962 |
+
"learning_rate": 9.549150281252632e-08,
|
| 1963 |
+
"loss": 0.00021752234897576272,
|
| 1964 |
+
"reward": 0.44140625,
|
| 1965 |
+
"reward_std": 0.36428582668304443,
|
| 1966 |
+
"rewards/MajorityAccuracy/mean": 0.44140625,
|
| 1967 |
+
"rewards/MajorityAccuracy/std": 0.4975275695323944,
|
| 1968 |
+
"step": 129,
|
| 1969 |
+
"step_time": 514.9773829579353
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"clip_ratio/high_max": 0.0050052043225150555,
|
| 1973 |
+
"clip_ratio/high_mean": 0.001540335790195968,
|
| 1974 |
+
"clip_ratio/low_mean": 0.0012205747480038553,
|
| 1975 |
+
"clip_ratio/low_min": 0.0,
|
| 1976 |
+
"clip_ratio/region_mean": 0.0027609104799921624,
|
| 1977 |
+
"epoch": 0.8125,
|
| 1978 |
+
"grad_norm": 5.350896128612248,
|
| 1979 |
+
"kl": 0.022738856729120016,
|
| 1980 |
+
"learning_rate": 8.961827939636196e-08,
|
| 1981 |
+
"loss": -0.00017374567687511444,
|
| 1982 |
+
"step": 130,
|
| 1983 |
+
"step_time": 75.16066078096628
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"clip_ratio/high_max": 0.001555257651489228,
|
| 1987 |
+
"clip_ratio/high_mean": 0.0005310779088176787,
|
| 1988 |
+
"clip_ratio/low_mean": 0.0003485415472823661,
|
| 1989 |
+
"clip_ratio/low_min": 0.0,
|
| 1990 |
+
"clip_ratio/region_mean": 0.000879619452462066,
|
| 1991 |
+
"epoch": 0.81875,
|
| 1992 |
+
"grad_norm": 1.5048141308005336,
|
| 1993 |
+
"kl": 0.013211901998147368,
|
| 1994 |
+
"learning_rate": 8.391363731548811e-08,
|
| 1995 |
+
"loss": 0.00030215742299333215,
|
| 1996 |
+
"step": 131,
|
| 1997 |
+
"step_time": 73.97731839120388
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"clip_ratio/high_max": 0.002422504410787951,
|
| 2001 |
+
"clip_ratio/high_mean": 0.0006056261026969878,
|
| 2002 |
+
"clip_ratio/low_mean": 0.0007379212947853375,
|
| 2003 |
+
"clip_ratio/low_min": 0.0,
|
| 2004 |
+
"clip_ratio/region_mean": 0.0013435473665595055,
|
| 2005 |
+
"epoch": 0.825,
|
| 2006 |
+
"grad_norm": 2.4156380184533326,
|
| 2007 |
+
"kl": 0.018357814522460103,
|
| 2008 |
+
"learning_rate": 7.837991998855897e-08,
|
| 2009 |
+
"loss": -1.9946834072470665e-05,
|
| 2010 |
+
"step": 132,
|
| 2011 |
+
"step_time": 73.03292130306363
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"clip_ratio/high_max": 0.0026058186776936054,
|
| 2015 |
+
"clip_ratio/high_mean": 0.001089577672246378,
|
| 2016 |
+
"clip_ratio/low_mean": 0.00019180944764229935,
|
| 2017 |
+
"clip_ratio/low_min": 0.0,
|
| 2018 |
+
"clip_ratio/region_mean": 0.0012813871217076667,
|
| 2019 |
+
"epoch": 0.83125,
|
| 2020 |
+
"grad_norm": 1.9268568373898995,
|
| 2021 |
+
"kl": 0.01177273690700531,
|
| 2022 |
+
"learning_rate": 7.301940061978722e-08,
|
| 2023 |
+
"loss": 6.651505827903748e-05,
|
| 2024 |
+
"step": 133,
|
| 2025 |
+
"step_time": 74.05381197854877
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"clip_ratio/high_max": 0.0008425677078776062,
|
| 2029 |
+
"clip_ratio/high_mean": 0.00022741594875697047,
|
| 2030 |
+
"clip_ratio/low_mean": 0.0002515294181648642,
|
| 2031 |
+
"clip_ratio/low_min": 0.0,
|
| 2032 |
+
"clip_ratio/region_mean": 0.00047894536692183465,
|
| 2033 |
+
"epoch": 0.8375,
|
| 2034 |
+
"grad_norm": 1.56654591144966,
|
| 2035 |
+
"kl": 0.012847630539909005,
|
| 2036 |
+
"learning_rate": 6.783428126513125e-08,
|
| 2037 |
+
"loss": 0.00036564498441293836,
|
| 2038 |
+
"step": 134,
|
| 2039 |
+
"step_time": 72.99300960823894
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"clip_ratio/high_max": 0.0036626184155466035,
|
| 2043 |
+
"clip_ratio/high_mean": 0.001394874318066286,
|
| 2044 |
+
"clip_ratio/low_mean": 0.0005731964884034824,
|
| 2045 |
+
"clip_ratio/low_min": 0.0,
|
| 2046 |
+
"clip_ratio/region_mean": 0.00196807083193562,
|
| 2047 |
+
"epoch": 0.84375,
|
| 2048 |
+
"grad_norm": 3.7176285086572105,
|
| 2049 |
+
"kl": 0.02578526095021516,
|
| 2050 |
+
"learning_rate": 6.282669192770895e-08,
|
| 2051 |
+
"loss": -5.2558258175849915e-05,
|
| 2052 |
+
"step": 135,
|
| 2053 |
+
"step_time": 72.6947511471808
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"clip_ratio/high_max": 0.0013524337409762666,
|
| 2057 |
+
"clip_ratio/high_mean": 0.0004198388123768382,
|
| 2058 |
+
"clip_ratio/low_mean": 0.00029960313986521214,
|
| 2059 |
+
"clip_ratio/low_min": 0.0,
|
| 2060 |
+
"clip_ratio/region_mean": 0.0007194419522420503,
|
| 2061 |
+
"epoch": 0.85,
|
| 2062 |
+
"grad_norm": 2.2351864327456963,
|
| 2063 |
+
"kl": 0.011891954927705228,
|
| 2064 |
+
"learning_rate": 5.799868968281074e-08,
|
| 2065 |
+
"loss": 0.0002445001155138016,
|
| 2066 |
+
"step": 136,
|
| 2067 |
+
"step_time": 75.12162413820624
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"clip_ratio/high_max": 0.0,
|
| 2071 |
+
"clip_ratio/high_mean": 0.0,
|
| 2072 |
+
"clip_ratio/low_mean": 0.0,
|
| 2073 |
+
"clip_ratio/low_min": 0.0,
|
| 2074 |
+
"clip_ratio/region_mean": 0.0,
|
| 2075 |
+
"completions/clipped_ratio": 0.3515625,
|
| 2076 |
+
"completions/max_length": 2048.0,
|
| 2077 |
+
"completions/mean_length": 1377.43359375,
|
| 2078 |
+
"completions/min_length": 180.0,
|
| 2079 |
+
"epoch": 0.85625,
|
| 2080 |
+
"frac_reward_zero_std": 0.34375,
|
| 2081 |
+
"grad_norm": 1.4910586277337923,
|
| 2082 |
+
"kl": 0.025006179115734994,
|
| 2083 |
+
"learning_rate": 5.335225783287051e-08,
|
| 2084 |
+
"loss": 0.0002499900874681771,
|
| 2085 |
+
"reward": 0.40625,
|
| 2086 |
+
"reward_std": 0.3071898818016052,
|
| 2087 |
+
"rewards/MajorityAccuracy/mean": 0.40625,
|
| 2088 |
+
"rewards/MajorityAccuracy/std": 0.49209436774253845,
|
| 2089 |
+
"step": 137,
|
| 2090 |
+
"step_time": 517.1175414547324
|
| 2091 |
+
},
|
| 2092 |
+
{
|
| 2093 |
+
"clip_ratio/high_max": 0.001622018899070099,
|
| 2094 |
+
"clip_ratio/high_mean": 0.0004538045759545639,
|
| 2095 |
+
"clip_ratio/low_mean": 0.0003204345703125,
|
| 2096 |
+
"clip_ratio/low_min": 0.0,
|
| 2097 |
+
"clip_ratio/region_mean": 0.0007742391462670639,
|
| 2098 |
+
"epoch": 0.8625,
|
| 2099 |
+
"grad_norm": 1.3464108614707406,
|
| 2100 |
+
"kl": 0.010792209883220494,
|
| 2101 |
+
"learning_rate": 4.888930509274125e-08,
|
| 2102 |
+
"loss": 0.00014851281594019383,
|
| 2103 |
+
"step": 138,
|
| 2104 |
+
"step_time": 75.12612386420369
|
| 2105 |
+
},
|
| 2106 |
+
{
|
| 2107 |
+
"clip_ratio/high_max": 0.0027745233091991395,
|
| 2108 |
+
"clip_ratio/high_mean": 0.0008188028878066689,
|
| 2109 |
+
"clip_ratio/low_mean": 4.57763671875e-05,
|
| 2110 |
+
"clip_ratio/low_min": 0.0,
|
| 2111 |
+
"clip_ratio/region_mean": 0.0008645792513561901,
|
| 2112 |
+
"epoch": 0.86875,
|
| 2113 |
+
"grad_norm": 1.2359076723863875,
|
| 2114 |
+
"kl": 0.013643871527165174,
|
| 2115 |
+
"learning_rate": 4.4611664805611794e-08,
|
| 2116 |
+
"loss": 0.00020019811927340925,
|
| 2117 |
+
"step": 139,
|
| 2118 |
+
"step_time": 73.08151721954346
|
| 2119 |
+
},
|
| 2120 |
+
{
|
| 2121 |
+
"clip_ratio/high_max": 0.002430364038445987,
|
| 2122 |
+
"clip_ratio/high_mean": 0.0006600183005502913,
|
| 2123 |
+
"clip_ratio/low_mean": 0.0002587194849184016,
|
| 2124 |
+
"clip_ratio/low_min": 0.0,
|
| 2125 |
+
"clip_ratio/region_mean": 0.0009187378018395975,
|
| 2126 |
+
"epoch": 0.875,
|
| 2127 |
+
"grad_norm": 1.9274414951413756,
|
| 2128 |
+
"kl": 0.012598582659848034,
|
| 2129 |
+
"learning_rate": 4.0521094189884696e-08,
|
| 2130 |
+
"loss": 0.0002846504794433713,
|
| 2131 |
+
"step": 140,
|
| 2132 |
+
"step_time": 73.32900973409414
|
| 2133 |
+
},
|
| 2134 |
+
{
|
| 2135 |
+
"clip_ratio/high_max": 0.004439846103196032,
|
| 2136 |
+
"clip_ratio/high_mean": 0.0013874533069611061,
|
| 2137 |
+
"clip_ratio/low_mean": 0.0017299874143645866,
|
| 2138 |
+
"clip_ratio/low_min": 0.0,
|
| 2139 |
+
"clip_ratio/region_mean": 0.0031174406976788305,
|
| 2140 |
+
"epoch": 0.88125,
|
| 2141 |
+
"grad_norm": 2.9247249124241526,
|
| 2142 |
+
"kl": 0.020254319068044424,
|
| 2143 |
+
"learning_rate": 3.6619273617325695e-08,
|
| 2144 |
+
"loss": -0.00014115209341980517,
|
| 2145 |
+
"step": 141,
|
| 2146 |
+
"step_time": 74.28333918377757
|
| 2147 |
+
},
|
| 2148 |
+
{
|
| 2149 |
+
"clip_ratio/high_max": 0.0003177559847244993,
|
| 2150 |
+
"clip_ratio/high_mean": 7.943899618112482e-05,
|
| 2151 |
+
"clip_ratio/low_mean": 4.8602069000480697e-05,
|
| 2152 |
+
"clip_ratio/low_min": 0.0,
|
| 2153 |
+
"clip_ratio/region_mean": 0.00012804106518160552,
|
| 2154 |
+
"epoch": 0.8875,
|
| 2155 |
+
"grad_norm": 0.8039791710377829,
|
| 2156 |
+
"kl": 0.011683915741741657,
|
| 2157 |
+
"learning_rate": 3.2907805922781476e-08,
|
| 2158 |
+
"loss": 0.00012935894483234733,
|
| 2159 |
+
"step": 142,
|
| 2160 |
+
"step_time": 74.63537713512778
|
| 2161 |
+
},
|
| 2162 |
+
{
|
| 2163 |
+
"clip_ratio/high_max": 0.0024360414754482917,
|
| 2164 |
+
"clip_ratio/high_mean": 0.0006704399256705074,
|
| 2165 |
+
"clip_ratio/low_mean": 0.00020207935995131265,
|
| 2166 |
+
"clip_ratio/low_min": 0.0,
|
| 2167 |
+
"clip_ratio/region_mean": 0.00087251928562182,
|
| 2168 |
+
"epoch": 0.89375,
|
| 2169 |
+
"grad_norm": 1.3509541946484116,
|
| 2170 |
+
"kl": 0.010289046098478138,
|
| 2171 |
+
"learning_rate": 2.9388215745748345e-08,
|
| 2172 |
+
"loss": 3.178737097186968e-05,
|
| 2173 |
+
"step": 143,
|
| 2174 |
+
"step_time": 73.63322289288044
|
| 2175 |
+
},
|
| 2176 |
+
{
|
| 2177 |
+
"clip_ratio/high_max": 0.0023877389321569353,
|
| 2178 |
+
"clip_ratio/high_mean": 0.0009364063807879575,
|
| 2179 |
+
"clip_ratio/low_mean": 0.0007192258926806971,
|
| 2180 |
+
"clip_ratio/low_min": 0.0001220703125,
|
| 2181 |
+
"clip_ratio/region_mean": 0.0016556322589167394,
|
| 2182 |
+
"epoch": 0.9,
|
| 2183 |
+
"grad_norm": 2.418125032671085,
|
| 2184 |
+
"kl": 0.013710160506889224,
|
| 2185 |
+
"learning_rate": 2.6061948904063658e-08,
|
| 2186 |
+
"loss": 0.00017416290938854218,
|
| 2187 |
+
"step": 144,
|
| 2188 |
+
"step_time": 72.88468716293573
|
| 2189 |
+
},
|
| 2190 |
+
{
|
| 2191 |
+
"clip_ratio/high_max": 0.0,
|
| 2192 |
+
"clip_ratio/high_mean": 0.0,
|
| 2193 |
+
"clip_ratio/low_mean": 0.0,
|
| 2194 |
+
"clip_ratio/low_min": 0.0,
|
| 2195 |
+
"clip_ratio/region_mean": 0.0,
|
| 2196 |
+
"completions/clipped_ratio": 0.3359375,
|
| 2197 |
+
"completions/max_length": 2048.0,
|
| 2198 |
+
"completions/mean_length": 1274.5078125,
|
| 2199 |
+
"completions/min_length": 105.0,
|
| 2200 |
+
"epoch": 0.90625,
|
| 2201 |
+
"frac_reward_zero_std": 0.34375,
|
| 2202 |
+
"grad_norm": 4.387439453260271,
|
| 2203 |
+
"kl": 0.010359042673371732,
|
| 2204 |
+
"learning_rate": 2.293037179997559e-08,
|
| 2205 |
+
"loss": 0.00010359704901929945,
|
| 2206 |
+
"reward": 0.3046875,
|
| 2207 |
+
"reward_std": 0.29287609457969666,
|
| 2208 |
+
"rewards/MajorityAccuracy/mean": 0.3046875,
|
| 2209 |
+
"rewards/MajorityAccuracy/std": 0.4611765742301941,
|
| 2210 |
+
"step": 145,
|
| 2211 |
+
"step_time": 517.3511182554066
|
| 2212 |
+
},
|
| 2213 |
+
{
|
| 2214 |
+
"clip_ratio/high_max": 0.005930699117016047,
|
| 2215 |
+
"clip_ratio/high_mean": 0.0020042620453750715,
|
| 2216 |
+
"clip_ratio/low_mean": 0.00018995662685483694,
|
| 2217 |
+
"clip_ratio/low_min": 0.0,
|
| 2218 |
+
"clip_ratio/region_mean": 0.0021942186722299084,
|
| 2219 |
+
"epoch": 0.9125,
|
| 2220 |
+
"grad_norm": 3.316800060081312,
|
| 2221 |
+
"kl": 0.015194247011095285,
|
| 2222 |
+
"learning_rate": 1.9994770858837107e-08,
|
| 2223 |
+
"loss": 0.0003903623146470636,
|
| 2224 |
+
"step": 146,
|
| 2225 |
+
"step_time": 75.11743240430951
|
| 2226 |
+
},
|
| 2227 |
+
{
|
| 2228 |
+
"clip_ratio/high_max": 0.0019946124375564978,
|
| 2229 |
+
"clip_ratio/high_mean": 0.0006634818601014558,
|
| 2230 |
+
"clip_ratio/low_mean": 0.0008884335311449831,
|
| 2231 |
+
"clip_ratio/low_min": 0.0,
|
| 2232 |
+
"clip_ratio/region_mean": 0.0015519153930654284,
|
| 2233 |
+
"epoch": 0.91875,
|
| 2234 |
+
"grad_norm": 3.9317798684823835,
|
| 2235 |
+
"kl": 0.015472290338948369,
|
| 2236 |
+
"learning_rate": 1.725635200065323e-08,
|
| 2237 |
+
"loss": 6.311998731689528e-06,
|
| 2238 |
+
"step": 147,
|
| 2239 |
+
"step_time": 74.05179583281279
|
| 2240 |
+
},
|
| 2241 |
+
{
|
| 2242 |
+
"clip_ratio/high_max": 0.0010895891464315355,
|
| 2243 |
+
"clip_ratio/high_mean": 0.00027239728660788387,
|
| 2244 |
+
"clip_ratio/low_mean": 0.0004549698642222211,
|
| 2245 |
+
"clip_ratio/low_min": 0.0,
|
| 2246 |
+
"clip_ratio/region_mean": 0.000727367150830105,
|
| 2247 |
+
"epoch": 0.925,
|
| 2248 |
+
"grad_norm": 1.7658786975860794,
|
| 2249 |
+
"kl": 0.010854134103283286,
|
| 2250 |
+
"learning_rate": 1.4716240144699187e-08,
|
| 2251 |
+
"loss": 0.00027612276608124375,
|
| 2252 |
+
"step": 148,
|
| 2253 |
+
"step_time": 74.79051082581282
|
| 2254 |
+
},
|
| 2255 |
+
{
|
| 2256 |
+
"clip_ratio/high_max": 0.001794461699319072,
|
| 2257 |
+
"clip_ratio/high_mean": 0.0005870545755897183,
|
| 2258 |
+
"clip_ratio/low_mean": 0.0007542552957602311,
|
| 2259 |
+
"clip_ratio/low_min": 0.0,
|
| 2260 |
+
"clip_ratio/region_mean": 0.001341309878625907,
|
| 2261 |
+
"epoch": 0.93125,
|
| 2262 |
+
"grad_norm": 2.4328771093860437,
|
| 2263 |
+
"kl": 0.014182504964992404,
|
| 2264 |
+
"learning_rate": 1.2375478747413015e-08,
|
| 2265 |
+
"loss": 0.00033027492463588715,
|
| 2266 |
+
"step": 149,
|
| 2267 |
+
"step_time": 74.20693530514836
|
| 2268 |
+
},
|
| 2269 |
+
{
|
| 2270 |
+
"clip_ratio/high_max": 0.0031610620790161192,
|
| 2271 |
+
"clip_ratio/high_mean": 0.0008389415306737646,
|
| 2272 |
+
"clip_ratio/low_mean": 0.0005106333119329065,
|
| 2273 |
+
"clip_ratio/low_min": 0.0,
|
| 2274 |
+
"clip_ratio/region_mean": 0.0013495748426066712,
|
| 2275 |
+
"epoch": 0.9375,
|
| 2276 |
+
"grad_norm": 3.357763548543685,
|
| 2277 |
+
"kl": 0.01612676098011434,
|
| 2278 |
+
"learning_rate": 1.0235029373752757e-08,
|
| 2279 |
+
"loss": 0.00010850631224457175,
|
| 2280 |
+
"step": 150,
|
| 2281 |
+
"step_time": 75.05084561556578
|
| 2282 |
+
},
|
| 2283 |
+
{
|
| 2284 |
+
"clip_ratio/high_max": 0.0011010176967829466,
|
| 2285 |
+
"clip_ratio/high_mean": 0.0002938777834060602,
|
| 2286 |
+
"clip_ratio/low_mean": 5.171164775674697e-05,
|
| 2287 |
+
"clip_ratio/low_min": 0.0,
|
| 2288 |
+
"clip_ratio/region_mean": 0.00034558943116280716,
|
| 2289 |
+
"epoch": 0.94375,
|
| 2290 |
+
"grad_norm": 1.228287008532962,
|
| 2291 |
+
"kl": 0.014011190854944289,
|
| 2292 |
+
"learning_rate": 8.295771302193721e-09,
|
| 2293 |
+
"loss": 0.00022285323939286172,
|
| 2294 |
+
"step": 151,
|
| 2295 |
+
"step_time": 74.3507028222084
|
| 2296 |
+
},
|
| 2297 |
+
{
|
| 2298 |
+
"clip_ratio/high_max": 0.0001524390245322138,
|
| 2299 |
+
"clip_ratio/high_mean": 3.810975613305345e-05,
|
| 2300 |
+
"clip_ratio/low_mean": 8.341082502738573e-05,
|
| 2301 |
+
"clip_ratio/low_min": 0.0,
|
| 2302 |
+
"clip_ratio/region_mean": 0.00012152058116043918,
|
| 2303 |
+
"epoch": 0.95,
|
| 2304 |
+
"grad_norm": 1.0940171130215217,
|
| 2305 |
+
"kl": 0.01205431460402906,
|
| 2306 |
+
"learning_rate": 6.558501163527963e-09,
|
| 2307 |
+
"loss": 0.00024235114688053727,
|
| 2308 |
+
"step": 152,
|
| 2309 |
+
"step_time": 73.7452283911407
|
| 2310 |
+
},
|
| 2311 |
+
{
|
| 2312 |
+
"clip_ratio/high_max": 0.0,
|
| 2313 |
+
"clip_ratio/high_mean": 0.0,
|
| 2314 |
+
"clip_ratio/low_mean": 0.0,
|
| 2315 |
+
"clip_ratio/low_min": 0.0,
|
| 2316 |
+
"clip_ratio/region_mean": 0.0,
|
| 2317 |
+
"completions/clipped_ratio": 0.27734375,
|
| 2318 |
+
"completions/max_length": 2048.0,
|
| 2319 |
+
"completions/mean_length": 1119.609375,
|
| 2320 |
+
"completions/min_length": 68.0,
|
| 2321 |
+
"epoch": 0.95625,
|
| 2322 |
+
"frac_reward_zero_std": 0.3125,
|
| 2323 |
+
"grad_norm": 1.8759180053007989,
|
| 2324 |
+
"kl": 0.014402540284208953,
|
| 2325 |
+
"learning_rate": 5.023932613615445e-09,
|
| 2326 |
+
"loss": 0.00014410493895411491,
|
| 2327 |
+
"reward": 0.453125,
|
| 2328 |
+
"reward_std": 0.31747061014175415,
|
| 2329 |
+
"rewards/MajorityAccuracy/mean": 0.453125,
|
| 2330 |
+
"rewards/MajorityAccuracy/std": 0.4987730085849762,
|
| 2331 |
+
"step": 153,
|
| 2332 |
+
"step_time": 516.458389442414
|
| 2333 |
+
},
|
| 2334 |
+
{
|
| 2335 |
+
"clip_ratio/high_max": 0.0030819087551208213,
|
| 2336 |
+
"clip_ratio/high_mean": 0.0009354328285553493,
|
| 2337 |
+
"clip_ratio/low_mean": 0.00019063470972469077,
|
| 2338 |
+
"clip_ratio/low_min": 0.0,
|
| 2339 |
+
"clip_ratio/region_mean": 0.0011260675455559976,
|
| 2340 |
+
"epoch": 0.9625,
|
| 2341 |
+
"grad_norm": 2.793842526122257,
|
| 2342 |
+
"kl": 0.012060714303515851,
|
| 2343 |
+
"learning_rate": 3.6926960402202666e-09,
|
| 2344 |
+
"loss": 0.00016674755897838622,
|
| 2345 |
+
"step": 154,
|
| 2346 |
+
"step_time": 75.2494084648788
|
| 2347 |
+
},
|
| 2348 |
+
{
|
| 2349 |
+
"clip_ratio/high_max": 0.004043390435981564,
|
| 2350 |
+
"clip_ratio/high_mean": 0.0012606568307091948,
|
| 2351 |
+
"clip_ratio/low_mean": 6.103515625e-05,
|
| 2352 |
+
"clip_ratio/low_min": 0.0,
|
| 2353 |
+
"clip_ratio/region_mean": 0.0013216919869591948,
|
| 2354 |
+
"epoch": 0.96875,
|
| 2355 |
+
"grad_norm": 3.8249412302579127,
|
| 2356 |
+
"kl": 0.02116782497614622,
|
| 2357 |
+
"learning_rate": 2.5653383040524224e-09,
|
| 2358 |
+
"loss": 0.0005570785724557936,
|
| 2359 |
+
"step": 155,
|
| 2360 |
+
"step_time": 75.4039362296462
|
| 2361 |
+
},
|
| 2362 |
+
{
|
| 2363 |
+
"clip_ratio/high_max": 0.0007394975691568106,
|
| 2364 |
+
"clip_ratio/high_mean": 0.00018487439228920266,
|
| 2365 |
+
"clip_ratio/low_mean": 0.0005462991539388895,
|
| 2366 |
+
"clip_ratio/low_min": 0.0,
|
| 2367 |
+
"clip_ratio/region_mean": 0.0007311735389521345,
|
| 2368 |
+
"epoch": 0.975,
|
| 2369 |
+
"grad_norm": 2.0292055647408236,
|
| 2370 |
+
"kl": 0.01149507483933121,
|
| 2371 |
+
"learning_rate": 1.642322514122385e-09,
|
| 2372 |
+
"loss": 0.0004784543998539448,
|
| 2373 |
+
"step": 156,
|
| 2374 |
+
"step_time": 74.40368612483144
|
| 2375 |
+
},
|
| 2376 |
+
{
|
| 2377 |
+
"clip_ratio/high_max": 0.0010022035130532458,
|
| 2378 |
+
"clip_ratio/high_mean": 0.00025055087826331146,
|
| 2379 |
+
"clip_ratio/low_mean": 0.0007107910350896418,
|
| 2380 |
+
"clip_ratio/low_min": 0.0,
|
| 2381 |
+
"clip_ratio/region_mean": 0.000961341898801038,
|
| 2382 |
+
"epoch": 0.98125,
|
| 2383 |
+
"grad_norm": 4.460227241602125,
|
| 2384 |
+
"kl": 0.01522372243925929,
|
| 2385 |
+
"learning_rate": 9.240278374995636e-10,
|
| 2386 |
+
"loss": 0.0008589210920035839,
|
| 2387 |
+
"step": 157,
|
| 2388 |
+
"step_time": 73.12140418589115
|
| 2389 |
+
},
|
| 2390 |
+
{
|
| 2391 |
+
"clip_ratio/high_max": 0.0029896866326453164,
|
| 2392 |
+
"clip_ratio/high_mean": 0.0009448904493183363,
|
| 2393 |
+
"clip_ratio/low_mean": 0.00015544603229500353,
|
| 2394 |
+
"clip_ratio/low_min": 0.0,
|
| 2395 |
+
"clip_ratio/region_mean": 0.0011003364816133399,
|
| 2396 |
+
"epoch": 0.9875,
|
| 2397 |
+
"grad_norm": 2.7462860672526745,
|
| 2398 |
+
"kl": 0.01321133365854621,
|
| 2399 |
+
"learning_rate": 4.1074934355384006e-10,
|
| 2400 |
+
"loss": 0.0002785533433780074,
|
| 2401 |
+
"step": 158,
|
| 2402 |
+
"step_time": 72.89160918816924
|
| 2403 |
+
},
|
| 2404 |
+
{
|
| 2405 |
+
"clip_ratio/high_max": 0.005339584895409644,
|
| 2406 |
+
"clip_ratio/high_mean": 0.0016664091672282666,
|
| 2407 |
+
"clip_ratio/low_mean": 0.00110359386599157,
|
| 2408 |
+
"clip_ratio/low_min": 7.35726862330921e-05,
|
| 2409 |
+
"clip_ratio/region_mean": 0.002770003047771752,
|
| 2410 |
+
"epoch": 0.99375,
|
| 2411 |
+
"grad_norm": 4.374632919127661,
|
| 2412 |
+
"kl": 0.01817822246812284,
|
| 2413 |
+
"learning_rate": 1.0269788274358537e-10,
|
| 2414 |
+
"loss": 0.0004222271963953972,
|
| 2415 |
+
"step": 159,
|
| 2416 |
+
"step_time": 72.9292745180428
|
| 2417 |
+
},
|
| 2418 |
+
{
|
| 2419 |
+
"clip_ratio/high_max": 0.003036498266737908,
|
| 2420 |
+
"clip_ratio/high_mean": 0.0009020652869367041,
|
| 2421 |
+
"clip_ratio/low_mean": 0.0001982546273211483,
|
| 2422 |
+
"clip_ratio/low_min": 0.0,
|
| 2423 |
+
"clip_ratio/region_mean": 0.0011003199106198736,
|
| 2424 |
+
"epoch": 1.0,
|
| 2425 |
+
"grad_norm": 1.59490769413064,
|
| 2426 |
+
"kl": 0.014776297379285097,
|
| 2427 |
+
"learning_rate": 0.0,
|
| 2428 |
+
"loss": 0.0001327590289292857,
|
| 2429 |
+
"step": 160,
|
| 2430 |
+
"step_time": 74.46356591209769
|
| 2431 |
+
}
|
| 2432 |
+
],
|
| 2433 |
+
"logging_steps": 1,
|
| 2434 |
+
"max_steps": 160,
|
| 2435 |
+
"num_input_tokens_seen": 0,
|
| 2436 |
+
"num_train_epochs": 1,
|
| 2437 |
+
"save_steps": 80,
|
| 2438 |
+
"stateful_callbacks": {
|
| 2439 |
+
"TrainerControl": {
|
| 2440 |
+
"args": {
|
| 2441 |
+
"should_epoch_stop": false,
|
| 2442 |
+
"should_evaluate": false,
|
| 2443 |
+
"should_log": false,
|
| 2444 |
+
"should_save": true,
|
| 2445 |
+
"should_training_stop": true
|
| 2446 |
+
},
|
| 2447 |
+
"attributes": {}
|
| 2448 |
+
}
|
| 2449 |
+
},
|
| 2450 |
+
"total_flos": 0.0,
|
| 2451 |
+
"train_batch_size": 1,
|
| 2452 |
+
"trial_name": null,
|
| 2453 |
+
"trial_params": null
|
| 2454 |
+
}
|
c1_no_history/round_4/chat_template.jinja
ADDED
|
@@ -0,0 +1,154 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{%- set image_count = namespace(value=0) %}
|
| 2 |
+
{%- set video_count = namespace(value=0) %}
|
| 3 |
+
{%- macro render_content(content, do_vision_count, is_system_content=false) %}
|
| 4 |
+
{%- if content is string %}
|
| 5 |
+
{{- content }}
|
| 6 |
+
{%- elif content is iterable and content is not mapping %}
|
| 7 |
+
{%- for item in content %}
|
| 8 |
+
{%- if 'image' in item or 'image_url' in item or item.type == 'image' %}
|
| 9 |
+
{%- if is_system_content %}
|
| 10 |
+
{{- raise_exception('System message cannot contain images.') }}
|
| 11 |
+
{%- endif %}
|
| 12 |
+
{%- if do_vision_count %}
|
| 13 |
+
{%- set image_count.value = image_count.value + 1 %}
|
| 14 |
+
{%- endif %}
|
| 15 |
+
{%- if add_vision_id %}
|
| 16 |
+
{{- 'Picture ' ~ image_count.value ~ ': ' }}
|
| 17 |
+
{%- endif %}
|
| 18 |
+
{{- '<|vision_start|><|image_pad|><|vision_end|>' }}
|
| 19 |
+
{%- elif 'video' in item or item.type == 'video' %}
|
| 20 |
+
{%- if is_system_content %}
|
| 21 |
+
{{- raise_exception('System message cannot contain videos.') }}
|
| 22 |
+
{%- endif %}
|
| 23 |
+
{%- if do_vision_count %}
|
| 24 |
+
{%- set video_count.value = video_count.value + 1 %}
|
| 25 |
+
{%- endif %}
|
| 26 |
+
{%- if add_vision_id %}
|
| 27 |
+
{{- 'Video ' ~ video_count.value ~ ': ' }}
|
| 28 |
+
{%- endif %}
|
| 29 |
+
{{- '<|vision_start|><|video_pad|><|vision_end|>' }}
|
| 30 |
+
{%- elif 'text' in item %}
|
| 31 |
+
{{- item.text }}
|
| 32 |
+
{%- else %}
|
| 33 |
+
{{- raise_exception('Unexpected item type in content.') }}
|
| 34 |
+
{%- endif %}
|
| 35 |
+
{%- endfor %}
|
| 36 |
+
{%- elif content is none or content is undefined %}
|
| 37 |
+
{{- '' }}
|
| 38 |
+
{%- else %}
|
| 39 |
+
{{- raise_exception('Unexpected content type.') }}
|
| 40 |
+
{%- endif %}
|
| 41 |
+
{%- endmacro %}
|
| 42 |
+
{%- if not messages %}
|
| 43 |
+
{{- raise_exception('No messages provided.') }}
|
| 44 |
+
{%- endif %}
|
| 45 |
+
{%- if tools and tools is iterable and tools is not mapping %}
|
| 46 |
+
{{- '<|im_start|>system\n' }}
|
| 47 |
+
{{- "# Tools\n\nYou have access to the following functions:\n\n<tools>" }}
|
| 48 |
+
{%- for tool in tools %}
|
| 49 |
+
{{- "\n" }}
|
| 50 |
+
{{- tool | tojson }}
|
| 51 |
+
{%- endfor %}
|
| 52 |
+
{{- "\n</tools>" }}
|
| 53 |
+
{{- '\n\nIf you choose to call a function ONLY reply in the following format with NO suffix:\n\n<tool_call>\n<function=example_function_name>\n<parameter=example_parameter_1>\nvalue_1\n</parameter>\n<parameter=example_parameter_2>\nThis is the value for the second parameter\nthat can span\nmultiple lines\n</parameter>\n</function>\n</tool_call>\n\n<IMPORTANT>\nReminder:\n- Function calls MUST follow the specified format: an inner <function=...></function> block must be nested within <tool_call></tool_call> XML tags\n- Required parameters MUST be specified\n- You may provide optional reasoning for your function call in natural language BEFORE the function call, but NOT after\n- If there is no function call available, answer the question like normal with your current knowledge and do not tell the user about function calls\n</IMPORTANT>' }}
|
| 54 |
+
{%- if messages[0].role == 'system' %}
|
| 55 |
+
{%- set content = render_content(messages[0].content, false, true)|trim %}
|
| 56 |
+
{%- if content %}
|
| 57 |
+
{{- '\n\n' + content }}
|
| 58 |
+
{%- endif %}
|
| 59 |
+
{%- endif %}
|
| 60 |
+
{{- '<|im_end|>\n' }}
|
| 61 |
+
{%- else %}
|
| 62 |
+
{%- if messages[0].role == 'system' %}
|
| 63 |
+
{%- set content = render_content(messages[0].content, false, true)|trim %}
|
| 64 |
+
{{- '<|im_start|>system\n' + content + '<|im_end|>\n' }}
|
| 65 |
+
{%- endif %}
|
| 66 |
+
{%- endif %}
|
| 67 |
+
{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
|
| 68 |
+
{%- for message in messages[::-1] %}
|
| 69 |
+
{%- set index = (messages|length - 1) - loop.index0 %}
|
| 70 |
+
{%- if ns.multi_step_tool and message.role == "user" %}
|
| 71 |
+
{%- set content = render_content(message.content, false)|trim %}
|
| 72 |
+
{%- if not(content.startswith('<tool_response>') and content.endswith('</tool_response>')) %}
|
| 73 |
+
{%- set ns.multi_step_tool = false %}
|
| 74 |
+
{%- set ns.last_query_index = index %}
|
| 75 |
+
{%- endif %}
|
| 76 |
+
{%- endif %}
|
| 77 |
+
{%- endfor %}
|
| 78 |
+
{%- if ns.multi_step_tool %}
|
| 79 |
+
{{- raise_exception('No user query found in messages.') }}
|
| 80 |
+
{%- endif %}
|
| 81 |
+
{%- for message in messages %}
|
| 82 |
+
{%- set content = render_content(message.content, true)|trim %}
|
| 83 |
+
{%- if message.role == "system" %}
|
| 84 |
+
{%- if not loop.first %}
|
| 85 |
+
{{- raise_exception('System message must be at the beginning.') }}
|
| 86 |
+
{%- endif %}
|
| 87 |
+
{%- elif message.role == "user" %}
|
| 88 |
+
{{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
|
| 89 |
+
{%- elif message.role == "assistant" %}
|
| 90 |
+
{%- set reasoning_content = '' %}
|
| 91 |
+
{%- if message.reasoning_content is string %}
|
| 92 |
+
{%- set reasoning_content = message.reasoning_content %}
|
| 93 |
+
{%- else %}
|
| 94 |
+
{%- if '</think>' in content %}
|
| 95 |
+
{%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
|
| 96 |
+
{%- set content = content.split('</think>')[-1].lstrip('\n') %}
|
| 97 |
+
{%- endif %}
|
| 98 |
+
{%- endif %}
|
| 99 |
+
{%- set reasoning_content = reasoning_content|trim %}
|
| 100 |
+
{%- if loop.index0 > ns.last_query_index %}
|
| 101 |
+
{{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content + '\n</think>\n\n' + content }}
|
| 102 |
+
{%- else %}
|
| 103 |
+
{{- '<|im_start|>' + message.role + '\n' + content }}
|
| 104 |
+
{%- endif %}
|
| 105 |
+
{%- if message.tool_calls and message.tool_calls is iterable and message.tool_calls is not mapping %}
|
| 106 |
+
{%- for tool_call in message.tool_calls %}
|
| 107 |
+
{%- if tool_call.function is defined %}
|
| 108 |
+
{%- set tool_call = tool_call.function %}
|
| 109 |
+
{%- endif %}
|
| 110 |
+
{%- if loop.first %}
|
| 111 |
+
{%- if content|trim %}
|
| 112 |
+
{{- '\n\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
|
| 113 |
+
{%- else %}
|
| 114 |
+
{{- '<tool_call>\n<function=' + tool_call.name + '>\n' }}
|
| 115 |
+
{%- endif %}
|
| 116 |
+
{%- else %}
|
| 117 |
+
{{- '\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
|
| 118 |
+
{%- endif %}
|
| 119 |
+
{%- if tool_call.arguments is defined %}
|
| 120 |
+
{%- for args_name, args_value in tool_call.arguments|items %}
|
| 121 |
+
{{- '<parameter=' + args_name + '>\n' }}
|
| 122 |
+
{%- set args_value = args_value | tojson | safe if args_value is mapping or (args_value is sequence and args_value is not string) else args_value | string %}
|
| 123 |
+
{{- args_value }}
|
| 124 |
+
{{- '\n</parameter>\n' }}
|
| 125 |
+
{%- endfor %}
|
| 126 |
+
{%- endif %}
|
| 127 |
+
{{- '</function>\n</tool_call>' }}
|
| 128 |
+
{%- endfor %}
|
| 129 |
+
{%- endif %}
|
| 130 |
+
{{- '<|im_end|>\n' }}
|
| 131 |
+
{%- elif message.role == "tool" %}
|
| 132 |
+
{%- if loop.previtem and loop.previtem.role != "tool" %}
|
| 133 |
+
{{- '<|im_start|>user' }}
|
| 134 |
+
{%- endif %}
|
| 135 |
+
{{- '\n<tool_response>\n' }}
|
| 136 |
+
{{- content }}
|
| 137 |
+
{{- '\n</tool_response>' }}
|
| 138 |
+
{%- if not loop.last and loop.nextitem.role != "tool" %}
|
| 139 |
+
{{- '<|im_end|>\n' }}
|
| 140 |
+
{%- elif loop.last %}
|
| 141 |
+
{{- '<|im_end|>\n' }}
|
| 142 |
+
{%- endif %}
|
| 143 |
+
{%- else %}
|
| 144 |
+
{{- raise_exception('Unexpected message role.') }}
|
| 145 |
+
{%- endif %}
|
| 146 |
+
{%- endfor %}
|
| 147 |
+
{%- if add_generation_prompt %}
|
| 148 |
+
{{- '<|im_start|>assistant\n' }}
|
| 149 |
+
{%- if enable_thinking is defined and enable_thinking is false %}
|
| 150 |
+
{{- '<think>\n\n</think>\n\n' }}
|
| 151 |
+
{%- else %}
|
| 152 |
+
{{- '<think>\n' }}
|
| 153 |
+
{%- endif %}
|
| 154 |
+
{%- endif %}
|
c1_no_history/round_4/config.json
ADDED
|
@@ -0,0 +1,113 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"Qwen3_5ForConditionalGeneration"
|
| 4 |
+
],
|
| 5 |
+
"dtype": "bfloat16",
|
| 6 |
+
"eos_token_id": 248046,
|
| 7 |
+
"hidden_size": 4096,
|
| 8 |
+
"image_token_id": 248056,
|
| 9 |
+
"model_type": "qwen3_5",
|
| 10 |
+
"pad_token_id": 248044,
|
| 11 |
+
"text_config": {
|
| 12 |
+
"attention_bias": false,
|
| 13 |
+
"attention_dropout": 0.0,
|
| 14 |
+
"attn_output_gate": true,
|
| 15 |
+
"bos_token_id": null,
|
| 16 |
+
"dtype": "bfloat16",
|
| 17 |
+
"eos_token_id": 248044,
|
| 18 |
+
"full_attention_interval": 4,
|
| 19 |
+
"head_dim": 256,
|
| 20 |
+
"hidden_act": "silu",
|
| 21 |
+
"hidden_size": 4096,
|
| 22 |
+
"initializer_range": 0.02,
|
| 23 |
+
"intermediate_size": 12288,
|
| 24 |
+
"layer_types": [
|
| 25 |
+
"linear_attention",
|
| 26 |
+
"linear_attention",
|
| 27 |
+
"linear_attention",
|
| 28 |
+
"full_attention",
|
| 29 |
+
"linear_attention",
|
| 30 |
+
"linear_attention",
|
| 31 |
+
"linear_attention",
|
| 32 |
+
"full_attention",
|
| 33 |
+
"linear_attention",
|
| 34 |
+
"linear_attention",
|
| 35 |
+
"linear_attention",
|
| 36 |
+
"full_attention",
|
| 37 |
+
"linear_attention",
|
| 38 |
+
"linear_attention",
|
| 39 |
+
"linear_attention",
|
| 40 |
+
"full_attention",
|
| 41 |
+
"linear_attention",
|
| 42 |
+
"linear_attention",
|
| 43 |
+
"linear_attention",
|
| 44 |
+
"full_attention",
|
| 45 |
+
"linear_attention",
|
| 46 |
+
"linear_attention",
|
| 47 |
+
"linear_attention",
|
| 48 |
+
"full_attention",
|
| 49 |
+
"linear_attention",
|
| 50 |
+
"linear_attention",
|
| 51 |
+
"linear_attention",
|
| 52 |
+
"full_attention",
|
| 53 |
+
"linear_attention",
|
| 54 |
+
"linear_attention",
|
| 55 |
+
"linear_attention",
|
| 56 |
+
"full_attention"
|
| 57 |
+
],
|
| 58 |
+
"linear_conv_kernel_dim": 4,
|
| 59 |
+
"linear_key_head_dim": 128,
|
| 60 |
+
"linear_num_key_heads": 16,
|
| 61 |
+
"linear_num_value_heads": 32,
|
| 62 |
+
"linear_value_head_dim": 128,
|
| 63 |
+
"mamba_ssm_dtype": "float32",
|
| 64 |
+
"max_position_embeddings": 262144,
|
| 65 |
+
"mlp_only_layers": [],
|
| 66 |
+
"model_type": "qwen3_5_text",
|
| 67 |
+
"mtp_num_hidden_layers": 1,
|
| 68 |
+
"mtp_use_dedicated_embeddings": false,
|
| 69 |
+
"num_attention_heads": 16,
|
| 70 |
+
"num_hidden_layers": 32,
|
| 71 |
+
"num_key_value_heads": 4,
|
| 72 |
+
"pad_token_id": 248044,
|
| 73 |
+
"partial_rotary_factor": 0.25,
|
| 74 |
+
"rms_norm_eps": 1e-06,
|
| 75 |
+
"rope_parameters": {
|
| 76 |
+
"mrope_interleaved": true,
|
| 77 |
+
"mrope_section": [
|
| 78 |
+
11,
|
| 79 |
+
11,
|
| 80 |
+
10
|
| 81 |
+
],
|
| 82 |
+
"partial_rotary_factor": 0.25,
|
| 83 |
+
"rope_theta": 10000000,
|
| 84 |
+
"rope_type": "default"
|
| 85 |
+
},
|
| 86 |
+
"tie_word_embeddings": false,
|
| 87 |
+
"use_cache": false,
|
| 88 |
+
"vocab_size": 248320
|
| 89 |
+
},
|
| 90 |
+
"tie_word_embeddings": false,
|
| 91 |
+
"transformers_version": "5.10.4",
|
| 92 |
+
"use_cache": false,
|
| 93 |
+
"video_token_id": 248057,
|
| 94 |
+
"vision_config": {
|
| 95 |
+
"deepstack_visual_indexes": [],
|
| 96 |
+
"depth": 27,
|
| 97 |
+
"dtype": "bfloat16",
|
| 98 |
+
"hidden_act": "gelu_pytorch_tanh",
|
| 99 |
+
"hidden_size": 1152,
|
| 100 |
+
"in_channels": 3,
|
| 101 |
+
"initializer_range": 0.02,
|
| 102 |
+
"intermediate_size": 4304,
|
| 103 |
+
"model_type": "qwen3_5_vision",
|
| 104 |
+
"num_heads": 16,
|
| 105 |
+
"num_position_embeddings": 2304,
|
| 106 |
+
"out_hidden_size": 4096,
|
| 107 |
+
"patch_size": 16,
|
| 108 |
+
"spatial_merge_size": 2,
|
| 109 |
+
"temporal_patch_size": 2
|
| 110 |
+
},
|
| 111 |
+
"vision_end_token_id": 248054,
|
| 112 |
+
"vision_start_token_id": 248053
|
| 113 |
+
}
|
c1_no_history/round_4/generation_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"eos_token_id": [
|
| 4 |
+
248044,
|
| 5 |
+
248046
|
| 6 |
+
],
|
| 7 |
+
"transformers_version": "5.10.4",
|
| 8 |
+
"use_cache": true
|
| 9 |
+
}
|
c1_no_history/round_4/model.safetensors.index.json
ADDED
|
@@ -0,0 +1,768 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"metadata": {
|
| 3 |
+
"total_parameters": 1469680,
|
| 4 |
+
"total_size": 18819627488
|
| 5 |
+
},
|
| 6 |
+
"weight_map": {
|
| 7 |
+
"lm_head.weight": "model-00001-of-00004.safetensors",
|
| 8 |
+
"model.language_model.embed_tokens.weight": "model-00001-of-00004.safetensors",
|
| 9 |
+
"model.language_model.layers.0.input_layernorm.weight": "model-00001-of-00004.safetensors",
|
| 10 |
+
"model.language_model.layers.0.linear_attn.A_log": "model-00001-of-00004.safetensors",
|
| 11 |
+
"model.language_model.layers.0.linear_attn.conv1d.weight": "model-00001-of-00004.safetensors",
|
| 12 |
+
"model.language_model.layers.0.linear_attn.dt_bias": "model-00001-of-00004.safetensors",
|
| 13 |
+
"model.language_model.layers.0.linear_attn.in_proj_a.weight": "model-00001-of-00004.safetensors",
|
| 14 |
+
"model.language_model.layers.0.linear_attn.in_proj_b.weight": "model-00001-of-00004.safetensors",
|
| 15 |
+
"model.language_model.layers.0.linear_attn.in_proj_qkv.weight": "model-00001-of-00004.safetensors",
|
| 16 |
+
"model.language_model.layers.0.linear_attn.in_proj_z.weight": "model-00001-of-00004.safetensors",
|
| 17 |
+
"model.language_model.layers.0.linear_attn.norm.weight": "model-00001-of-00004.safetensors",
|
| 18 |
+
"model.language_model.layers.0.linear_attn.out_proj.weight": "model-00001-of-00004.safetensors",
|
| 19 |
+
"model.language_model.layers.0.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
|
| 20 |
+
"model.language_model.layers.0.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
|
| 21 |
+
"model.language_model.layers.0.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
|
| 22 |
+
"model.language_model.layers.0.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
|
| 23 |
+
"model.language_model.layers.1.input_layernorm.weight": "model-00001-of-00004.safetensors",
|
| 24 |
+
"model.language_model.layers.1.linear_attn.A_log": "model-00001-of-00004.safetensors",
|
| 25 |
+
"model.language_model.layers.1.linear_attn.conv1d.weight": "model-00001-of-00004.safetensors",
|
| 26 |
+
"model.language_model.layers.1.linear_attn.dt_bias": "model-00001-of-00004.safetensors",
|
| 27 |
+
"model.language_model.layers.1.linear_attn.in_proj_a.weight": "model-00001-of-00004.safetensors",
|
| 28 |
+
"model.language_model.layers.1.linear_attn.in_proj_b.weight": "model-00001-of-00004.safetensors",
|
| 29 |
+
"model.language_model.layers.1.linear_attn.in_proj_qkv.weight": "model-00001-of-00004.safetensors",
|
| 30 |
+
"model.language_model.layers.1.linear_attn.in_proj_z.weight": "model-00001-of-00004.safetensors",
|
| 31 |
+
"model.language_model.layers.1.linear_attn.norm.weight": "model-00001-of-00004.safetensors",
|
| 32 |
+
"model.language_model.layers.1.linear_attn.out_proj.weight": "model-00001-of-00004.safetensors",
|
| 33 |
+
"model.language_model.layers.1.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
|
| 34 |
+
"model.language_model.layers.1.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
|
| 35 |
+
"model.language_model.layers.1.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
|
| 36 |
+
"model.language_model.layers.1.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
|
| 37 |
+
"model.language_model.layers.10.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 38 |
+
"model.language_model.layers.10.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 39 |
+
"model.language_model.layers.10.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 40 |
+
"model.language_model.layers.10.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 41 |
+
"model.language_model.layers.10.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 42 |
+
"model.language_model.layers.10.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 43 |
+
"model.language_model.layers.10.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 44 |
+
"model.language_model.layers.10.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 45 |
+
"model.language_model.layers.10.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 46 |
+
"model.language_model.layers.10.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 47 |
+
"model.language_model.layers.10.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 48 |
+
"model.language_model.layers.10.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 49 |
+
"model.language_model.layers.10.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 50 |
+
"model.language_model.layers.10.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 51 |
+
"model.language_model.layers.11.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 52 |
+
"model.language_model.layers.11.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 53 |
+
"model.language_model.layers.11.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 54 |
+
"model.language_model.layers.11.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 55 |
+
"model.language_model.layers.11.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 56 |
+
"model.language_model.layers.11.self_attn.k_norm.weight": "model-00002-of-00004.safetensors",
|
| 57 |
+
"model.language_model.layers.11.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
|
| 58 |
+
"model.language_model.layers.11.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
|
| 59 |
+
"model.language_model.layers.11.self_attn.q_norm.weight": "model-00002-of-00004.safetensors",
|
| 60 |
+
"model.language_model.layers.11.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
|
| 61 |
+
"model.language_model.layers.11.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
|
| 62 |
+
"model.language_model.layers.12.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 63 |
+
"model.language_model.layers.12.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 64 |
+
"model.language_model.layers.12.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 65 |
+
"model.language_model.layers.12.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 66 |
+
"model.language_model.layers.12.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 67 |
+
"model.language_model.layers.12.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 68 |
+
"model.language_model.layers.12.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 69 |
+
"model.language_model.layers.12.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 70 |
+
"model.language_model.layers.12.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 71 |
+
"model.language_model.layers.12.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 72 |
+
"model.language_model.layers.12.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 73 |
+
"model.language_model.layers.12.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 74 |
+
"model.language_model.layers.12.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 75 |
+
"model.language_model.layers.12.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 76 |
+
"model.language_model.layers.13.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 77 |
+
"model.language_model.layers.13.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 78 |
+
"model.language_model.layers.13.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 79 |
+
"model.language_model.layers.13.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 80 |
+
"model.language_model.layers.13.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 81 |
+
"model.language_model.layers.13.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 82 |
+
"model.language_model.layers.13.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 83 |
+
"model.language_model.layers.13.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 84 |
+
"model.language_model.layers.13.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 85 |
+
"model.language_model.layers.13.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 86 |
+
"model.language_model.layers.13.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 87 |
+
"model.language_model.layers.13.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 88 |
+
"model.language_model.layers.13.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 89 |
+
"model.language_model.layers.13.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 90 |
+
"model.language_model.layers.14.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 91 |
+
"model.language_model.layers.14.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 92 |
+
"model.language_model.layers.14.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 93 |
+
"model.language_model.layers.14.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 94 |
+
"model.language_model.layers.14.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 95 |
+
"model.language_model.layers.14.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 96 |
+
"model.language_model.layers.14.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 97 |
+
"model.language_model.layers.14.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 98 |
+
"model.language_model.layers.14.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 99 |
+
"model.language_model.layers.14.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 100 |
+
"model.language_model.layers.14.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 101 |
+
"model.language_model.layers.14.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 102 |
+
"model.language_model.layers.14.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 103 |
+
"model.language_model.layers.14.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 104 |
+
"model.language_model.layers.15.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 105 |
+
"model.language_model.layers.15.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 106 |
+
"model.language_model.layers.15.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 107 |
+
"model.language_model.layers.15.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 108 |
+
"model.language_model.layers.15.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 109 |
+
"model.language_model.layers.15.self_attn.k_norm.weight": "model-00003-of-00004.safetensors",
|
| 110 |
+
"model.language_model.layers.15.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
|
| 111 |
+
"model.language_model.layers.15.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
|
| 112 |
+
"model.language_model.layers.15.self_attn.q_norm.weight": "model-00003-of-00004.safetensors",
|
| 113 |
+
"model.language_model.layers.15.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
|
| 114 |
+
"model.language_model.layers.15.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
|
| 115 |
+
"model.language_model.layers.16.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 116 |
+
"model.language_model.layers.16.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 117 |
+
"model.language_model.layers.16.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 118 |
+
"model.language_model.layers.16.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 119 |
+
"model.language_model.layers.16.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 120 |
+
"model.language_model.layers.16.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 121 |
+
"model.language_model.layers.16.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 122 |
+
"model.language_model.layers.16.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 123 |
+
"model.language_model.layers.16.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 124 |
+
"model.language_model.layers.16.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 125 |
+
"model.language_model.layers.16.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 126 |
+
"model.language_model.layers.16.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 127 |
+
"model.language_model.layers.16.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 128 |
+
"model.language_model.layers.16.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 129 |
+
"model.language_model.layers.17.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 130 |
+
"model.language_model.layers.17.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 131 |
+
"model.language_model.layers.17.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 132 |
+
"model.language_model.layers.17.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 133 |
+
"model.language_model.layers.17.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 134 |
+
"model.language_model.layers.17.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 135 |
+
"model.language_model.layers.17.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 136 |
+
"model.language_model.layers.17.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 137 |
+
"model.language_model.layers.17.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 138 |
+
"model.language_model.layers.17.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 139 |
+
"model.language_model.layers.17.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 140 |
+
"model.language_model.layers.17.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 141 |
+
"model.language_model.layers.17.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 142 |
+
"model.language_model.layers.17.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 143 |
+
"model.language_model.layers.18.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 144 |
+
"model.language_model.layers.18.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 145 |
+
"model.language_model.layers.18.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 146 |
+
"model.language_model.layers.18.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 147 |
+
"model.language_model.layers.18.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 148 |
+
"model.language_model.layers.18.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 149 |
+
"model.language_model.layers.18.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 150 |
+
"model.language_model.layers.18.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 151 |
+
"model.language_model.layers.18.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 152 |
+
"model.language_model.layers.18.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 153 |
+
"model.language_model.layers.18.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 154 |
+
"model.language_model.layers.18.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 155 |
+
"model.language_model.layers.18.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 156 |
+
"model.language_model.layers.18.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 157 |
+
"model.language_model.layers.19.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 158 |
+
"model.language_model.layers.19.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 159 |
+
"model.language_model.layers.19.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 160 |
+
"model.language_model.layers.19.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 161 |
+
"model.language_model.layers.19.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 162 |
+
"model.language_model.layers.19.self_attn.k_norm.weight": "model-00003-of-00004.safetensors",
|
| 163 |
+
"model.language_model.layers.19.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
|
| 164 |
+
"model.language_model.layers.19.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
|
| 165 |
+
"model.language_model.layers.19.self_attn.q_norm.weight": "model-00003-of-00004.safetensors",
|
| 166 |
+
"model.language_model.layers.19.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
|
| 167 |
+
"model.language_model.layers.19.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
|
| 168 |
+
"model.language_model.layers.2.input_layernorm.weight": "model-00001-of-00004.safetensors",
|
| 169 |
+
"model.language_model.layers.2.linear_attn.A_log": "model-00001-of-00004.safetensors",
|
| 170 |
+
"model.language_model.layers.2.linear_attn.conv1d.weight": "model-00001-of-00004.safetensors",
|
| 171 |
+
"model.language_model.layers.2.linear_attn.dt_bias": "model-00001-of-00004.safetensors",
|
| 172 |
+
"model.language_model.layers.2.linear_attn.in_proj_a.weight": "model-00001-of-00004.safetensors",
|
| 173 |
+
"model.language_model.layers.2.linear_attn.in_proj_b.weight": "model-00001-of-00004.safetensors",
|
| 174 |
+
"model.language_model.layers.2.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 175 |
+
"model.language_model.layers.2.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 176 |
+
"model.language_model.layers.2.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 177 |
+
"model.language_model.layers.2.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 178 |
+
"model.language_model.layers.2.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 179 |
+
"model.language_model.layers.2.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 180 |
+
"model.language_model.layers.2.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 181 |
+
"model.language_model.layers.2.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 182 |
+
"model.language_model.layers.20.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 183 |
+
"model.language_model.layers.20.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 184 |
+
"model.language_model.layers.20.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 185 |
+
"model.language_model.layers.20.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 186 |
+
"model.language_model.layers.20.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 187 |
+
"model.language_model.layers.20.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 188 |
+
"model.language_model.layers.20.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 189 |
+
"model.language_model.layers.20.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 190 |
+
"model.language_model.layers.20.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 191 |
+
"model.language_model.layers.20.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 192 |
+
"model.language_model.layers.20.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 193 |
+
"model.language_model.layers.20.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 194 |
+
"model.language_model.layers.20.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 195 |
+
"model.language_model.layers.20.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 196 |
+
"model.language_model.layers.21.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 197 |
+
"model.language_model.layers.21.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 198 |
+
"model.language_model.layers.21.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 199 |
+
"model.language_model.layers.21.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 200 |
+
"model.language_model.layers.21.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 201 |
+
"model.language_model.layers.21.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 202 |
+
"model.language_model.layers.21.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 203 |
+
"model.language_model.layers.21.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 204 |
+
"model.language_model.layers.21.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 205 |
+
"model.language_model.layers.21.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 206 |
+
"model.language_model.layers.21.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 207 |
+
"model.language_model.layers.21.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 208 |
+
"model.language_model.layers.21.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 209 |
+
"model.language_model.layers.21.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 210 |
+
"model.language_model.layers.22.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 211 |
+
"model.language_model.layers.22.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 212 |
+
"model.language_model.layers.22.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 213 |
+
"model.language_model.layers.22.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 214 |
+
"model.language_model.layers.22.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 215 |
+
"model.language_model.layers.22.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 216 |
+
"model.language_model.layers.22.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 217 |
+
"model.language_model.layers.22.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 218 |
+
"model.language_model.layers.22.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 219 |
+
"model.language_model.layers.22.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 220 |
+
"model.language_model.layers.22.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 221 |
+
"model.language_model.layers.22.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 222 |
+
"model.language_model.layers.22.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 223 |
+
"model.language_model.layers.22.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 224 |
+
"model.language_model.layers.23.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 225 |
+
"model.language_model.layers.23.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 226 |
+
"model.language_model.layers.23.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 227 |
+
"model.language_model.layers.23.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 228 |
+
"model.language_model.layers.23.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 229 |
+
"model.language_model.layers.23.self_attn.k_norm.weight": "model-00003-of-00004.safetensors",
|
| 230 |
+
"model.language_model.layers.23.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
|
| 231 |
+
"model.language_model.layers.23.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
|
| 232 |
+
"model.language_model.layers.23.self_attn.q_norm.weight": "model-00003-of-00004.safetensors",
|
| 233 |
+
"model.language_model.layers.23.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
|
| 234 |
+
"model.language_model.layers.23.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
|
| 235 |
+
"model.language_model.layers.24.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 236 |
+
"model.language_model.layers.24.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 237 |
+
"model.language_model.layers.24.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 238 |
+
"model.language_model.layers.24.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 239 |
+
"model.language_model.layers.24.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 240 |
+
"model.language_model.layers.24.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 241 |
+
"model.language_model.layers.24.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 242 |
+
"model.language_model.layers.24.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 243 |
+
"model.language_model.layers.24.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 244 |
+
"model.language_model.layers.24.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 245 |
+
"model.language_model.layers.24.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 246 |
+
"model.language_model.layers.24.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 247 |
+
"model.language_model.layers.24.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 248 |
+
"model.language_model.layers.24.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 249 |
+
"model.language_model.layers.25.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 250 |
+
"model.language_model.layers.25.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 251 |
+
"model.language_model.layers.25.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 252 |
+
"model.language_model.layers.25.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 253 |
+
"model.language_model.layers.25.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 254 |
+
"model.language_model.layers.25.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 255 |
+
"model.language_model.layers.25.linear_attn.in_proj_qkv.weight": "model-00004-of-00004.safetensors",
|
| 256 |
+
"model.language_model.layers.25.linear_attn.in_proj_z.weight": "model-00004-of-00004.safetensors",
|
| 257 |
+
"model.language_model.layers.25.linear_attn.norm.weight": "model-00004-of-00004.safetensors",
|
| 258 |
+
"model.language_model.layers.25.linear_attn.out_proj.weight": "model-00004-of-00004.safetensors",
|
| 259 |
+
"model.language_model.layers.25.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 260 |
+
"model.language_model.layers.25.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 261 |
+
"model.language_model.layers.25.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 262 |
+
"model.language_model.layers.25.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 263 |
+
"model.language_model.layers.26.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 264 |
+
"model.language_model.layers.26.linear_attn.A_log": "model-00004-of-00004.safetensors",
|
| 265 |
+
"model.language_model.layers.26.linear_attn.conv1d.weight": "model-00004-of-00004.safetensors",
|
| 266 |
+
"model.language_model.layers.26.linear_attn.dt_bias": "model-00004-of-00004.safetensors",
|
| 267 |
+
"model.language_model.layers.26.linear_attn.in_proj_a.weight": "model-00004-of-00004.safetensors",
|
| 268 |
+
"model.language_model.layers.26.linear_attn.in_proj_b.weight": "model-00004-of-00004.safetensors",
|
| 269 |
+
"model.language_model.layers.26.linear_attn.in_proj_qkv.weight": "model-00004-of-00004.safetensors",
|
| 270 |
+
"model.language_model.layers.26.linear_attn.in_proj_z.weight": "model-00004-of-00004.safetensors",
|
| 271 |
+
"model.language_model.layers.26.linear_attn.norm.weight": "model-00004-of-00004.safetensors",
|
| 272 |
+
"model.language_model.layers.26.linear_attn.out_proj.weight": "model-00004-of-00004.safetensors",
|
| 273 |
+
"model.language_model.layers.26.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 274 |
+
"model.language_model.layers.26.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 275 |
+
"model.language_model.layers.26.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 276 |
+
"model.language_model.layers.26.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 277 |
+
"model.language_model.layers.27.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 278 |
+
"model.language_model.layers.27.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 279 |
+
"model.language_model.layers.27.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 280 |
+
"model.language_model.layers.27.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 281 |
+
"model.language_model.layers.27.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 282 |
+
"model.language_model.layers.27.self_attn.k_norm.weight": "model-00004-of-00004.safetensors",
|
| 283 |
+
"model.language_model.layers.27.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
|
| 284 |
+
"model.language_model.layers.27.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
|
| 285 |
+
"model.language_model.layers.27.self_attn.q_norm.weight": "model-00004-of-00004.safetensors",
|
| 286 |
+
"model.language_model.layers.27.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
|
| 287 |
+
"model.language_model.layers.27.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
|
| 288 |
+
"model.language_model.layers.28.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 289 |
+
"model.language_model.layers.28.linear_attn.A_log": "model-00004-of-00004.safetensors",
|
| 290 |
+
"model.language_model.layers.28.linear_attn.conv1d.weight": "model-00004-of-00004.safetensors",
|
| 291 |
+
"model.language_model.layers.28.linear_attn.dt_bias": "model-00004-of-00004.safetensors",
|
| 292 |
+
"model.language_model.layers.28.linear_attn.in_proj_a.weight": "model-00004-of-00004.safetensors",
|
| 293 |
+
"model.language_model.layers.28.linear_attn.in_proj_b.weight": "model-00004-of-00004.safetensors",
|
| 294 |
+
"model.language_model.layers.28.linear_attn.in_proj_qkv.weight": "model-00004-of-00004.safetensors",
|
| 295 |
+
"model.language_model.layers.28.linear_attn.in_proj_z.weight": "model-00004-of-00004.safetensors",
|
| 296 |
+
"model.language_model.layers.28.linear_attn.norm.weight": "model-00004-of-00004.safetensors",
|
| 297 |
+
"model.language_model.layers.28.linear_attn.out_proj.weight": "model-00004-of-00004.safetensors",
|
| 298 |
+
"model.language_model.layers.28.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 299 |
+
"model.language_model.layers.28.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 300 |
+
"model.language_model.layers.28.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 301 |
+
"model.language_model.layers.28.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 302 |
+
"model.language_model.layers.29.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 303 |
+
"model.language_model.layers.29.linear_attn.A_log": "model-00004-of-00004.safetensors",
|
| 304 |
+
"model.language_model.layers.29.linear_attn.conv1d.weight": "model-00004-of-00004.safetensors",
|
| 305 |
+
"model.language_model.layers.29.linear_attn.dt_bias": "model-00004-of-00004.safetensors",
|
| 306 |
+
"model.language_model.layers.29.linear_attn.in_proj_a.weight": "model-00004-of-00004.safetensors",
|
| 307 |
+
"model.language_model.layers.29.linear_attn.in_proj_b.weight": "model-00004-of-00004.safetensors",
|
| 308 |
+
"model.language_model.layers.29.linear_attn.in_proj_qkv.weight": "model-00004-of-00004.safetensors",
|
| 309 |
+
"model.language_model.layers.29.linear_attn.in_proj_z.weight": "model-00004-of-00004.safetensors",
|
| 310 |
+
"model.language_model.layers.29.linear_attn.norm.weight": "model-00004-of-00004.safetensors",
|
| 311 |
+
"model.language_model.layers.29.linear_attn.out_proj.weight": "model-00004-of-00004.safetensors",
|
| 312 |
+
"model.language_model.layers.29.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 313 |
+
"model.language_model.layers.29.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 314 |
+
"model.language_model.layers.29.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 315 |
+
"model.language_model.layers.29.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 316 |
+
"model.language_model.layers.3.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 317 |
+
"model.language_model.layers.3.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 318 |
+
"model.language_model.layers.3.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 319 |
+
"model.language_model.layers.3.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 320 |
+
"model.language_model.layers.3.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 321 |
+
"model.language_model.layers.3.self_attn.k_norm.weight": "model-00002-of-00004.safetensors",
|
| 322 |
+
"model.language_model.layers.3.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
|
| 323 |
+
"model.language_model.layers.3.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
|
| 324 |
+
"model.language_model.layers.3.self_attn.q_norm.weight": "model-00002-of-00004.safetensors",
|
| 325 |
+
"model.language_model.layers.3.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
|
| 326 |
+
"model.language_model.layers.3.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
|
| 327 |
+
"model.language_model.layers.30.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 328 |
+
"model.language_model.layers.30.linear_attn.A_log": "model-00004-of-00004.safetensors",
|
| 329 |
+
"model.language_model.layers.30.linear_attn.conv1d.weight": "model-00004-of-00004.safetensors",
|
| 330 |
+
"model.language_model.layers.30.linear_attn.dt_bias": "model-00004-of-00004.safetensors",
|
| 331 |
+
"model.language_model.layers.30.linear_attn.in_proj_a.weight": "model-00004-of-00004.safetensors",
|
| 332 |
+
"model.language_model.layers.30.linear_attn.in_proj_b.weight": "model-00004-of-00004.safetensors",
|
| 333 |
+
"model.language_model.layers.30.linear_attn.in_proj_qkv.weight": "model-00004-of-00004.safetensors",
|
| 334 |
+
"model.language_model.layers.30.linear_attn.in_proj_z.weight": "model-00004-of-00004.safetensors",
|
| 335 |
+
"model.language_model.layers.30.linear_attn.norm.weight": "model-00004-of-00004.safetensors",
|
| 336 |
+
"model.language_model.layers.30.linear_attn.out_proj.weight": "model-00004-of-00004.safetensors",
|
| 337 |
+
"model.language_model.layers.30.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 338 |
+
"model.language_model.layers.30.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 339 |
+
"model.language_model.layers.30.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 340 |
+
"model.language_model.layers.30.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 341 |
+
"model.language_model.layers.31.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 342 |
+
"model.language_model.layers.31.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 343 |
+
"model.language_model.layers.31.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 344 |
+
"model.language_model.layers.31.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 345 |
+
"model.language_model.layers.31.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 346 |
+
"model.language_model.layers.31.self_attn.k_norm.weight": "model-00004-of-00004.safetensors",
|
| 347 |
+
"model.language_model.layers.31.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
|
| 348 |
+
"model.language_model.layers.31.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
|
| 349 |
+
"model.language_model.layers.31.self_attn.q_norm.weight": "model-00004-of-00004.safetensors",
|
| 350 |
+
"model.language_model.layers.31.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
|
| 351 |
+
"model.language_model.layers.31.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
|
| 352 |
+
"model.language_model.layers.4.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 353 |
+
"model.language_model.layers.4.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 354 |
+
"model.language_model.layers.4.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 355 |
+
"model.language_model.layers.4.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 356 |
+
"model.language_model.layers.4.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 357 |
+
"model.language_model.layers.4.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 358 |
+
"model.language_model.layers.4.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 359 |
+
"model.language_model.layers.4.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 360 |
+
"model.language_model.layers.4.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 361 |
+
"model.language_model.layers.4.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 362 |
+
"model.language_model.layers.4.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 363 |
+
"model.language_model.layers.4.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 364 |
+
"model.language_model.layers.4.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 365 |
+
"model.language_model.layers.4.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 366 |
+
"model.language_model.layers.5.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 367 |
+
"model.language_model.layers.5.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 368 |
+
"model.language_model.layers.5.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 369 |
+
"model.language_model.layers.5.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 370 |
+
"model.language_model.layers.5.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 371 |
+
"model.language_model.layers.5.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 372 |
+
"model.language_model.layers.5.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 373 |
+
"model.language_model.layers.5.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 374 |
+
"model.language_model.layers.5.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 375 |
+
"model.language_model.layers.5.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 376 |
+
"model.language_model.layers.5.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 377 |
+
"model.language_model.layers.5.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 378 |
+
"model.language_model.layers.5.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 379 |
+
"model.language_model.layers.5.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 380 |
+
"model.language_model.layers.6.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 381 |
+
"model.language_model.layers.6.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 382 |
+
"model.language_model.layers.6.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 383 |
+
"model.language_model.layers.6.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 384 |
+
"model.language_model.layers.6.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 385 |
+
"model.language_model.layers.6.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 386 |
+
"model.language_model.layers.6.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 387 |
+
"model.language_model.layers.6.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 388 |
+
"model.language_model.layers.6.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 389 |
+
"model.language_model.layers.6.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 390 |
+
"model.language_model.layers.6.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 391 |
+
"model.language_model.layers.6.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 392 |
+
"model.language_model.layers.6.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 393 |
+
"model.language_model.layers.6.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 394 |
+
"model.language_model.layers.7.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 395 |
+
"model.language_model.layers.7.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 396 |
+
"model.language_model.layers.7.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 397 |
+
"model.language_model.layers.7.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 398 |
+
"model.language_model.layers.7.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 399 |
+
"model.language_model.layers.7.self_attn.k_norm.weight": "model-00002-of-00004.safetensors",
|
| 400 |
+
"model.language_model.layers.7.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
|
| 401 |
+
"model.language_model.layers.7.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
|
| 402 |
+
"model.language_model.layers.7.self_attn.q_norm.weight": "model-00002-of-00004.safetensors",
|
| 403 |
+
"model.language_model.layers.7.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
|
| 404 |
+
"model.language_model.layers.7.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
|
| 405 |
+
"model.language_model.layers.8.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 406 |
+
"model.language_model.layers.8.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 407 |
+
"model.language_model.layers.8.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 408 |
+
"model.language_model.layers.8.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 409 |
+
"model.language_model.layers.8.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 410 |
+
"model.language_model.layers.8.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 411 |
+
"model.language_model.layers.8.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 412 |
+
"model.language_model.layers.8.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 413 |
+
"model.language_model.layers.8.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 414 |
+
"model.language_model.layers.8.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 415 |
+
"model.language_model.layers.8.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 416 |
+
"model.language_model.layers.8.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 417 |
+
"model.language_model.layers.8.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 418 |
+
"model.language_model.layers.8.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 419 |
+
"model.language_model.layers.9.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 420 |
+
"model.language_model.layers.9.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 421 |
+
"model.language_model.layers.9.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 422 |
+
"model.language_model.layers.9.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 423 |
+
"model.language_model.layers.9.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 424 |
+
"model.language_model.layers.9.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 425 |
+
"model.language_model.layers.9.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 426 |
+
"model.language_model.layers.9.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 427 |
+
"model.language_model.layers.9.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 428 |
+
"model.language_model.layers.9.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 429 |
+
"model.language_model.layers.9.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 430 |
+
"model.language_model.layers.9.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 431 |
+
"model.language_model.layers.9.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 432 |
+
"model.language_model.layers.9.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 433 |
+
"model.language_model.norm.weight": "model-00004-of-00004.safetensors",
|
| 434 |
+
"model.visual.blocks.0.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 435 |
+
"model.visual.blocks.0.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 436 |
+
"model.visual.blocks.0.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 437 |
+
"model.visual.blocks.0.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 438 |
+
"model.visual.blocks.0.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 439 |
+
"model.visual.blocks.0.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 440 |
+
"model.visual.blocks.0.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 441 |
+
"model.visual.blocks.0.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 442 |
+
"model.visual.blocks.0.norm1.bias": "model-00004-of-00004.safetensors",
|
| 443 |
+
"model.visual.blocks.0.norm1.weight": "model-00004-of-00004.safetensors",
|
| 444 |
+
"model.visual.blocks.0.norm2.bias": "model-00004-of-00004.safetensors",
|
| 445 |
+
"model.visual.blocks.0.norm2.weight": "model-00004-of-00004.safetensors",
|
| 446 |
+
"model.visual.blocks.1.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 447 |
+
"model.visual.blocks.1.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 448 |
+
"model.visual.blocks.1.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 449 |
+
"model.visual.blocks.1.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 450 |
+
"model.visual.blocks.1.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 451 |
+
"model.visual.blocks.1.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 452 |
+
"model.visual.blocks.1.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 453 |
+
"model.visual.blocks.1.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 454 |
+
"model.visual.blocks.1.norm1.bias": "model-00004-of-00004.safetensors",
|
| 455 |
+
"model.visual.blocks.1.norm1.weight": "model-00004-of-00004.safetensors",
|
| 456 |
+
"model.visual.blocks.1.norm2.bias": "model-00004-of-00004.safetensors",
|
| 457 |
+
"model.visual.blocks.1.norm2.weight": "model-00004-of-00004.safetensors",
|
| 458 |
+
"model.visual.blocks.10.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 459 |
+
"model.visual.blocks.10.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 460 |
+
"model.visual.blocks.10.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 461 |
+
"model.visual.blocks.10.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 462 |
+
"model.visual.blocks.10.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 463 |
+
"model.visual.blocks.10.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 464 |
+
"model.visual.blocks.10.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 465 |
+
"model.visual.blocks.10.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 466 |
+
"model.visual.blocks.10.norm1.bias": "model-00004-of-00004.safetensors",
|
| 467 |
+
"model.visual.blocks.10.norm1.weight": "model-00004-of-00004.safetensors",
|
| 468 |
+
"model.visual.blocks.10.norm2.bias": "model-00004-of-00004.safetensors",
|
| 469 |
+
"model.visual.blocks.10.norm2.weight": "model-00004-of-00004.safetensors",
|
| 470 |
+
"model.visual.blocks.11.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 471 |
+
"model.visual.blocks.11.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 472 |
+
"model.visual.blocks.11.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 473 |
+
"model.visual.blocks.11.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 474 |
+
"model.visual.blocks.11.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 475 |
+
"model.visual.blocks.11.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 476 |
+
"model.visual.blocks.11.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 477 |
+
"model.visual.blocks.11.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 478 |
+
"model.visual.blocks.11.norm1.bias": "model-00004-of-00004.safetensors",
|
| 479 |
+
"model.visual.blocks.11.norm1.weight": "model-00004-of-00004.safetensors",
|
| 480 |
+
"model.visual.blocks.11.norm2.bias": "model-00004-of-00004.safetensors",
|
| 481 |
+
"model.visual.blocks.11.norm2.weight": "model-00004-of-00004.safetensors",
|
| 482 |
+
"model.visual.blocks.12.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 483 |
+
"model.visual.blocks.12.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 484 |
+
"model.visual.blocks.12.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 485 |
+
"model.visual.blocks.12.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 486 |
+
"model.visual.blocks.12.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 487 |
+
"model.visual.blocks.12.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 488 |
+
"model.visual.blocks.12.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 489 |
+
"model.visual.blocks.12.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 490 |
+
"model.visual.blocks.12.norm1.bias": "model-00004-of-00004.safetensors",
|
| 491 |
+
"model.visual.blocks.12.norm1.weight": "model-00004-of-00004.safetensors",
|
| 492 |
+
"model.visual.blocks.12.norm2.bias": "model-00004-of-00004.safetensors",
|
| 493 |
+
"model.visual.blocks.12.norm2.weight": "model-00004-of-00004.safetensors",
|
| 494 |
+
"model.visual.blocks.13.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 495 |
+
"model.visual.blocks.13.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 496 |
+
"model.visual.blocks.13.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 497 |
+
"model.visual.blocks.13.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 498 |
+
"model.visual.blocks.13.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 499 |
+
"model.visual.blocks.13.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 500 |
+
"model.visual.blocks.13.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 501 |
+
"model.visual.blocks.13.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 502 |
+
"model.visual.blocks.13.norm1.bias": "model-00004-of-00004.safetensors",
|
| 503 |
+
"model.visual.blocks.13.norm1.weight": "model-00004-of-00004.safetensors",
|
| 504 |
+
"model.visual.blocks.13.norm2.bias": "model-00004-of-00004.safetensors",
|
| 505 |
+
"model.visual.blocks.13.norm2.weight": "model-00004-of-00004.safetensors",
|
| 506 |
+
"model.visual.blocks.14.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 507 |
+
"model.visual.blocks.14.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 508 |
+
"model.visual.blocks.14.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 509 |
+
"model.visual.blocks.14.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 510 |
+
"model.visual.blocks.14.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 511 |
+
"model.visual.blocks.14.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 512 |
+
"model.visual.blocks.14.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 513 |
+
"model.visual.blocks.14.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 514 |
+
"model.visual.blocks.14.norm1.bias": "model-00004-of-00004.safetensors",
|
| 515 |
+
"model.visual.blocks.14.norm1.weight": "model-00004-of-00004.safetensors",
|
| 516 |
+
"model.visual.blocks.14.norm2.bias": "model-00004-of-00004.safetensors",
|
| 517 |
+
"model.visual.blocks.14.norm2.weight": "model-00004-of-00004.safetensors",
|
| 518 |
+
"model.visual.blocks.15.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 519 |
+
"model.visual.blocks.15.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 520 |
+
"model.visual.blocks.15.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 521 |
+
"model.visual.blocks.15.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 522 |
+
"model.visual.blocks.15.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 523 |
+
"model.visual.blocks.15.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 524 |
+
"model.visual.blocks.15.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 525 |
+
"model.visual.blocks.15.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 526 |
+
"model.visual.blocks.15.norm1.bias": "model-00004-of-00004.safetensors",
|
| 527 |
+
"model.visual.blocks.15.norm1.weight": "model-00004-of-00004.safetensors",
|
| 528 |
+
"model.visual.blocks.15.norm2.bias": "model-00004-of-00004.safetensors",
|
| 529 |
+
"model.visual.blocks.15.norm2.weight": "model-00004-of-00004.safetensors",
|
| 530 |
+
"model.visual.blocks.16.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 531 |
+
"model.visual.blocks.16.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 532 |
+
"model.visual.blocks.16.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 533 |
+
"model.visual.blocks.16.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 534 |
+
"model.visual.blocks.16.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 535 |
+
"model.visual.blocks.16.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 536 |
+
"model.visual.blocks.16.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 537 |
+
"model.visual.blocks.16.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 538 |
+
"model.visual.blocks.16.norm1.bias": "model-00004-of-00004.safetensors",
|
| 539 |
+
"model.visual.blocks.16.norm1.weight": "model-00004-of-00004.safetensors",
|
| 540 |
+
"model.visual.blocks.16.norm2.bias": "model-00004-of-00004.safetensors",
|
| 541 |
+
"model.visual.blocks.16.norm2.weight": "model-00004-of-00004.safetensors",
|
| 542 |
+
"model.visual.blocks.17.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 543 |
+
"model.visual.blocks.17.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 544 |
+
"model.visual.blocks.17.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 545 |
+
"model.visual.blocks.17.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 546 |
+
"model.visual.blocks.17.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 547 |
+
"model.visual.blocks.17.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 548 |
+
"model.visual.blocks.17.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 549 |
+
"model.visual.blocks.17.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 550 |
+
"model.visual.blocks.17.norm1.bias": "model-00004-of-00004.safetensors",
|
| 551 |
+
"model.visual.blocks.17.norm1.weight": "model-00004-of-00004.safetensors",
|
| 552 |
+
"model.visual.blocks.17.norm2.bias": "model-00004-of-00004.safetensors",
|
| 553 |
+
"model.visual.blocks.17.norm2.weight": "model-00004-of-00004.safetensors",
|
| 554 |
+
"model.visual.blocks.18.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 555 |
+
"model.visual.blocks.18.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 556 |
+
"model.visual.blocks.18.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 557 |
+
"model.visual.blocks.18.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 558 |
+
"model.visual.blocks.18.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 559 |
+
"model.visual.blocks.18.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 560 |
+
"model.visual.blocks.18.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 561 |
+
"model.visual.blocks.18.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 562 |
+
"model.visual.blocks.18.norm1.bias": "model-00004-of-00004.safetensors",
|
| 563 |
+
"model.visual.blocks.18.norm1.weight": "model-00004-of-00004.safetensors",
|
| 564 |
+
"model.visual.blocks.18.norm2.bias": "model-00004-of-00004.safetensors",
|
| 565 |
+
"model.visual.blocks.18.norm2.weight": "model-00004-of-00004.safetensors",
|
| 566 |
+
"model.visual.blocks.19.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 567 |
+
"model.visual.blocks.19.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 568 |
+
"model.visual.blocks.19.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 569 |
+
"model.visual.blocks.19.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 570 |
+
"model.visual.blocks.19.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 571 |
+
"model.visual.blocks.19.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 572 |
+
"model.visual.blocks.19.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 573 |
+
"model.visual.blocks.19.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 574 |
+
"model.visual.blocks.19.norm1.bias": "model-00004-of-00004.safetensors",
|
| 575 |
+
"model.visual.blocks.19.norm1.weight": "model-00004-of-00004.safetensors",
|
| 576 |
+
"model.visual.blocks.19.norm2.bias": "model-00004-of-00004.safetensors",
|
| 577 |
+
"model.visual.blocks.19.norm2.weight": "model-00004-of-00004.safetensors",
|
| 578 |
+
"model.visual.blocks.2.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 579 |
+
"model.visual.blocks.2.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 580 |
+
"model.visual.blocks.2.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 581 |
+
"model.visual.blocks.2.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 582 |
+
"model.visual.blocks.2.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 583 |
+
"model.visual.blocks.2.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 584 |
+
"model.visual.blocks.2.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 585 |
+
"model.visual.blocks.2.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 586 |
+
"model.visual.blocks.2.norm1.bias": "model-00004-of-00004.safetensors",
|
| 587 |
+
"model.visual.blocks.2.norm1.weight": "model-00004-of-00004.safetensors",
|
| 588 |
+
"model.visual.blocks.2.norm2.bias": "model-00004-of-00004.safetensors",
|
| 589 |
+
"model.visual.blocks.2.norm2.weight": "model-00004-of-00004.safetensors",
|
| 590 |
+
"model.visual.blocks.20.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 591 |
+
"model.visual.blocks.20.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 592 |
+
"model.visual.blocks.20.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 593 |
+
"model.visual.blocks.20.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 594 |
+
"model.visual.blocks.20.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 595 |
+
"model.visual.blocks.20.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 596 |
+
"model.visual.blocks.20.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 597 |
+
"model.visual.blocks.20.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 598 |
+
"model.visual.blocks.20.norm1.bias": "model-00004-of-00004.safetensors",
|
| 599 |
+
"model.visual.blocks.20.norm1.weight": "model-00004-of-00004.safetensors",
|
| 600 |
+
"model.visual.blocks.20.norm2.bias": "model-00004-of-00004.safetensors",
|
| 601 |
+
"model.visual.blocks.20.norm2.weight": "model-00004-of-00004.safetensors",
|
| 602 |
+
"model.visual.blocks.21.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 603 |
+
"model.visual.blocks.21.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 604 |
+
"model.visual.blocks.21.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 605 |
+
"model.visual.blocks.21.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 606 |
+
"model.visual.blocks.21.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 607 |
+
"model.visual.blocks.21.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 608 |
+
"model.visual.blocks.21.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 609 |
+
"model.visual.blocks.21.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 610 |
+
"model.visual.blocks.21.norm1.bias": "model-00004-of-00004.safetensors",
|
| 611 |
+
"model.visual.blocks.21.norm1.weight": "model-00004-of-00004.safetensors",
|
| 612 |
+
"model.visual.blocks.21.norm2.bias": "model-00004-of-00004.safetensors",
|
| 613 |
+
"model.visual.blocks.21.norm2.weight": "model-00004-of-00004.safetensors",
|
| 614 |
+
"model.visual.blocks.22.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 615 |
+
"model.visual.blocks.22.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 616 |
+
"model.visual.blocks.22.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 617 |
+
"model.visual.blocks.22.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 618 |
+
"model.visual.blocks.22.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 619 |
+
"model.visual.blocks.22.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 620 |
+
"model.visual.blocks.22.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 621 |
+
"model.visual.blocks.22.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 622 |
+
"model.visual.blocks.22.norm1.bias": "model-00004-of-00004.safetensors",
|
| 623 |
+
"model.visual.blocks.22.norm1.weight": "model-00004-of-00004.safetensors",
|
| 624 |
+
"model.visual.blocks.22.norm2.bias": "model-00004-of-00004.safetensors",
|
| 625 |
+
"model.visual.blocks.22.norm2.weight": "model-00004-of-00004.safetensors",
|
| 626 |
+
"model.visual.blocks.23.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 627 |
+
"model.visual.blocks.23.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 628 |
+
"model.visual.blocks.23.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 629 |
+
"model.visual.blocks.23.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 630 |
+
"model.visual.blocks.23.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 631 |
+
"model.visual.blocks.23.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 632 |
+
"model.visual.blocks.23.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 633 |
+
"model.visual.blocks.23.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 634 |
+
"model.visual.blocks.23.norm1.bias": "model-00004-of-00004.safetensors",
|
| 635 |
+
"model.visual.blocks.23.norm1.weight": "model-00004-of-00004.safetensors",
|
| 636 |
+
"model.visual.blocks.23.norm2.bias": "model-00004-of-00004.safetensors",
|
| 637 |
+
"model.visual.blocks.23.norm2.weight": "model-00004-of-00004.safetensors",
|
| 638 |
+
"model.visual.blocks.24.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 639 |
+
"model.visual.blocks.24.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 640 |
+
"model.visual.blocks.24.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 641 |
+
"model.visual.blocks.24.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 642 |
+
"model.visual.blocks.24.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 643 |
+
"model.visual.blocks.24.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 644 |
+
"model.visual.blocks.24.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 645 |
+
"model.visual.blocks.24.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 646 |
+
"model.visual.blocks.24.norm1.bias": "model-00004-of-00004.safetensors",
|
| 647 |
+
"model.visual.blocks.24.norm1.weight": "model-00004-of-00004.safetensors",
|
| 648 |
+
"model.visual.blocks.24.norm2.bias": "model-00004-of-00004.safetensors",
|
| 649 |
+
"model.visual.blocks.24.norm2.weight": "model-00004-of-00004.safetensors",
|
| 650 |
+
"model.visual.blocks.25.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 651 |
+
"model.visual.blocks.25.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 652 |
+
"model.visual.blocks.25.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 653 |
+
"model.visual.blocks.25.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 654 |
+
"model.visual.blocks.25.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 655 |
+
"model.visual.blocks.25.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 656 |
+
"model.visual.blocks.25.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 657 |
+
"model.visual.blocks.25.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 658 |
+
"model.visual.blocks.25.norm1.bias": "model-00004-of-00004.safetensors",
|
| 659 |
+
"model.visual.blocks.25.norm1.weight": "model-00004-of-00004.safetensors",
|
| 660 |
+
"model.visual.blocks.25.norm2.bias": "model-00004-of-00004.safetensors",
|
| 661 |
+
"model.visual.blocks.25.norm2.weight": "model-00004-of-00004.safetensors",
|
| 662 |
+
"model.visual.blocks.26.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 663 |
+
"model.visual.blocks.26.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 664 |
+
"model.visual.blocks.26.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 665 |
+
"model.visual.blocks.26.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 666 |
+
"model.visual.blocks.26.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 667 |
+
"model.visual.blocks.26.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 668 |
+
"model.visual.blocks.26.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 669 |
+
"model.visual.blocks.26.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 670 |
+
"model.visual.blocks.26.norm1.bias": "model-00004-of-00004.safetensors",
|
| 671 |
+
"model.visual.blocks.26.norm1.weight": "model-00004-of-00004.safetensors",
|
| 672 |
+
"model.visual.blocks.26.norm2.bias": "model-00004-of-00004.safetensors",
|
| 673 |
+
"model.visual.blocks.26.norm2.weight": "model-00004-of-00004.safetensors",
|
| 674 |
+
"model.visual.blocks.3.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 675 |
+
"model.visual.blocks.3.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 676 |
+
"model.visual.blocks.3.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 677 |
+
"model.visual.blocks.3.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 678 |
+
"model.visual.blocks.3.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 679 |
+
"model.visual.blocks.3.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 680 |
+
"model.visual.blocks.3.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 681 |
+
"model.visual.blocks.3.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 682 |
+
"model.visual.blocks.3.norm1.bias": "model-00004-of-00004.safetensors",
|
| 683 |
+
"model.visual.blocks.3.norm1.weight": "model-00004-of-00004.safetensors",
|
| 684 |
+
"model.visual.blocks.3.norm2.bias": "model-00004-of-00004.safetensors",
|
| 685 |
+
"model.visual.blocks.3.norm2.weight": "model-00004-of-00004.safetensors",
|
| 686 |
+
"model.visual.blocks.4.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 687 |
+
"model.visual.blocks.4.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 688 |
+
"model.visual.blocks.4.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 689 |
+
"model.visual.blocks.4.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 690 |
+
"model.visual.blocks.4.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 691 |
+
"model.visual.blocks.4.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 692 |
+
"model.visual.blocks.4.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 693 |
+
"model.visual.blocks.4.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 694 |
+
"model.visual.blocks.4.norm1.bias": "model-00004-of-00004.safetensors",
|
| 695 |
+
"model.visual.blocks.4.norm1.weight": "model-00004-of-00004.safetensors",
|
| 696 |
+
"model.visual.blocks.4.norm2.bias": "model-00004-of-00004.safetensors",
|
| 697 |
+
"model.visual.blocks.4.norm2.weight": "model-00004-of-00004.safetensors",
|
| 698 |
+
"model.visual.blocks.5.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 699 |
+
"model.visual.blocks.5.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 700 |
+
"model.visual.blocks.5.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 701 |
+
"model.visual.blocks.5.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 702 |
+
"model.visual.blocks.5.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 703 |
+
"model.visual.blocks.5.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 704 |
+
"model.visual.blocks.5.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 705 |
+
"model.visual.blocks.5.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 706 |
+
"model.visual.blocks.5.norm1.bias": "model-00004-of-00004.safetensors",
|
| 707 |
+
"model.visual.blocks.5.norm1.weight": "model-00004-of-00004.safetensors",
|
| 708 |
+
"model.visual.blocks.5.norm2.bias": "model-00004-of-00004.safetensors",
|
| 709 |
+
"model.visual.blocks.5.norm2.weight": "model-00004-of-00004.safetensors",
|
| 710 |
+
"model.visual.blocks.6.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 711 |
+
"model.visual.blocks.6.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 712 |
+
"model.visual.blocks.6.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 713 |
+
"model.visual.blocks.6.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 714 |
+
"model.visual.blocks.6.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 715 |
+
"model.visual.blocks.6.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 716 |
+
"model.visual.blocks.6.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 717 |
+
"model.visual.blocks.6.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 718 |
+
"model.visual.blocks.6.norm1.bias": "model-00004-of-00004.safetensors",
|
| 719 |
+
"model.visual.blocks.6.norm1.weight": "model-00004-of-00004.safetensors",
|
| 720 |
+
"model.visual.blocks.6.norm2.bias": "model-00004-of-00004.safetensors",
|
| 721 |
+
"model.visual.blocks.6.norm2.weight": "model-00004-of-00004.safetensors",
|
| 722 |
+
"model.visual.blocks.7.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 723 |
+
"model.visual.blocks.7.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 724 |
+
"model.visual.blocks.7.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 725 |
+
"model.visual.blocks.7.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 726 |
+
"model.visual.blocks.7.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 727 |
+
"model.visual.blocks.7.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 728 |
+
"model.visual.blocks.7.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 729 |
+
"model.visual.blocks.7.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 730 |
+
"model.visual.blocks.7.norm1.bias": "model-00004-of-00004.safetensors",
|
| 731 |
+
"model.visual.blocks.7.norm1.weight": "model-00004-of-00004.safetensors",
|
| 732 |
+
"model.visual.blocks.7.norm2.bias": "model-00004-of-00004.safetensors",
|
| 733 |
+
"model.visual.blocks.7.norm2.weight": "model-00004-of-00004.safetensors",
|
| 734 |
+
"model.visual.blocks.8.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 735 |
+
"model.visual.blocks.8.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 736 |
+
"model.visual.blocks.8.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 737 |
+
"model.visual.blocks.8.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 738 |
+
"model.visual.blocks.8.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 739 |
+
"model.visual.blocks.8.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 740 |
+
"model.visual.blocks.8.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 741 |
+
"model.visual.blocks.8.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 742 |
+
"model.visual.blocks.8.norm1.bias": "model-00004-of-00004.safetensors",
|
| 743 |
+
"model.visual.blocks.8.norm1.weight": "model-00004-of-00004.safetensors",
|
| 744 |
+
"model.visual.blocks.8.norm2.bias": "model-00004-of-00004.safetensors",
|
| 745 |
+
"model.visual.blocks.8.norm2.weight": "model-00004-of-00004.safetensors",
|
| 746 |
+
"model.visual.blocks.9.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 747 |
+
"model.visual.blocks.9.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 748 |
+
"model.visual.blocks.9.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 749 |
+
"model.visual.blocks.9.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 750 |
+
"model.visual.blocks.9.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 751 |
+
"model.visual.blocks.9.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 752 |
+
"model.visual.blocks.9.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 753 |
+
"model.visual.blocks.9.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 754 |
+
"model.visual.blocks.9.norm1.bias": "model-00004-of-00004.safetensors",
|
| 755 |
+
"model.visual.blocks.9.norm1.weight": "model-00004-of-00004.safetensors",
|
| 756 |
+
"model.visual.blocks.9.norm2.bias": "model-00004-of-00004.safetensors",
|
| 757 |
+
"model.visual.blocks.9.norm2.weight": "model-00004-of-00004.safetensors",
|
| 758 |
+
"model.visual.merger.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 759 |
+
"model.visual.merger.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 760 |
+
"model.visual.merger.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 761 |
+
"model.visual.merger.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 762 |
+
"model.visual.merger.norm.bias": "model-00004-of-00004.safetensors",
|
| 763 |
+
"model.visual.merger.norm.weight": "model-00004-of-00004.safetensors",
|
| 764 |
+
"model.visual.patch_embed.proj.bias": "model-00004-of-00004.safetensors",
|
| 765 |
+
"model.visual.patch_embed.proj.weight": "model-00004-of-00004.safetensors",
|
| 766 |
+
"model.visual.pos_embed.weight": "model-00004-of-00004.safetensors"
|
| 767 |
+
}
|
| 768 |
+
}
|
c1_no_history/round_4/preprocessor_config.json
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"size": {
|
| 3 |
+
"longest_edge": 16777216,
|
| 4 |
+
"shortest_edge": 65536
|
| 5 |
+
},
|
| 6 |
+
"patch_size": 16,
|
| 7 |
+
"temporal_patch_size": 2,
|
| 8 |
+
"merge_size": 2,
|
| 9 |
+
"image_mean": [
|
| 10 |
+
0.5,
|
| 11 |
+
0.5,
|
| 12 |
+
0.5
|
| 13 |
+
],
|
| 14 |
+
"image_std": [
|
| 15 |
+
0.5,
|
| 16 |
+
0.5,
|
| 17 |
+
0.5
|
| 18 |
+
],
|
| 19 |
+
"processor_class": "Qwen3VLProcessor",
|
| 20 |
+
"image_processor_type": "Qwen2VLImageProcessorFast"
|
| 21 |
+
}
|
c1_no_history/round_4/processor_config.json
ADDED
|
@@ -0,0 +1,60 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"image_processor": {
|
| 3 |
+
"do_convert_rgb": true,
|
| 4 |
+
"do_normalize": true,
|
| 5 |
+
"do_rescale": true,
|
| 6 |
+
"do_resize": true,
|
| 7 |
+
"image_mean": [
|
| 8 |
+
0.5,
|
| 9 |
+
0.5,
|
| 10 |
+
0.5
|
| 11 |
+
],
|
| 12 |
+
"image_processor_type": "Qwen2VLImageProcessor",
|
| 13 |
+
"image_std": [
|
| 14 |
+
0.5,
|
| 15 |
+
0.5,
|
| 16 |
+
0.5
|
| 17 |
+
],
|
| 18 |
+
"merge_size": 2,
|
| 19 |
+
"patch_size": 16,
|
| 20 |
+
"resample": 3,
|
| 21 |
+
"rescale_factor": 0.00392156862745098,
|
| 22 |
+
"size": {
|
| 23 |
+
"longest_edge": 16777216,
|
| 24 |
+
"shortest_edge": 65536
|
| 25 |
+
},
|
| 26 |
+
"temporal_patch_size": 2
|
| 27 |
+
},
|
| 28 |
+
"processor_class": "Qwen3VLProcessor",
|
| 29 |
+
"video_processor": {
|
| 30 |
+
"do_convert_rgb": true,
|
| 31 |
+
"do_normalize": true,
|
| 32 |
+
"do_rescale": true,
|
| 33 |
+
"do_resize": true,
|
| 34 |
+
"do_sample_frames": true,
|
| 35 |
+
"fps": 2,
|
| 36 |
+
"image_mean": [
|
| 37 |
+
0.5,
|
| 38 |
+
0.5,
|
| 39 |
+
0.5
|
| 40 |
+
],
|
| 41 |
+
"image_std": [
|
| 42 |
+
0.5,
|
| 43 |
+
0.5,
|
| 44 |
+
0.5
|
| 45 |
+
],
|
| 46 |
+
"max_frames": 768,
|
| 47 |
+
"merge_size": 2,
|
| 48 |
+
"min_frames": 4,
|
| 49 |
+
"patch_size": 16,
|
| 50 |
+
"resample": 3,
|
| 51 |
+
"rescale_factor": 0.00392156862745098,
|
| 52 |
+
"return_metadata": false,
|
| 53 |
+
"size": {
|
| 54 |
+
"longest_edge": 25165824,
|
| 55 |
+
"shortest_edge": 4096
|
| 56 |
+
},
|
| 57 |
+
"temporal_patch_size": 2,
|
| 58 |
+
"video_processor_type": "Qwen3VLVideoProcessor"
|
| 59 |
+
}
|
| 60 |
+
}
|
c1_no_history/round_4/tokenizer_config.json
ADDED
|
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"audio_bos_token": "<|audio_start|>",
|
| 4 |
+
"audio_eos_token": "<|audio_end|>",
|
| 5 |
+
"audio_token": "<|audio_pad|>",
|
| 6 |
+
"backend": "tokenizers",
|
| 7 |
+
"bos_token": null,
|
| 8 |
+
"clean_up_tokenization_spaces": false,
|
| 9 |
+
"eos_token": "<|im_end|>",
|
| 10 |
+
"errors": "replace",
|
| 11 |
+
"image_token": "<|image_pad|>",
|
| 12 |
+
"is_local": true,
|
| 13 |
+
"local_files_only": true,
|
| 14 |
+
"model_max_length": 262144,
|
| 15 |
+
"model_specific_special_tokens": {
|
| 16 |
+
"audio_bos_token": "<|audio_start|>",
|
| 17 |
+
"audio_eos_token": "<|audio_end|>",
|
| 18 |
+
"audio_token": "<|audio_pad|>",
|
| 19 |
+
"image_token": "<|image_pad|>",
|
| 20 |
+
"video_token": "<|video_pad|>",
|
| 21 |
+
"vision_bos_token": "<|vision_start|>",
|
| 22 |
+
"vision_eos_token": "<|vision_end|>"
|
| 23 |
+
},
|
| 24 |
+
"pad_token": "<|endoftext|>",
|
| 25 |
+
"pretokenize_regex": "(?i:'s|'t|'re|'ve|'m|'ll|'d)|[^\\r\\n\\p{L}\\p{N}]?[\\p{L}\\p{M}]+|\\p{N}| ?[^\\s\\p{L}\\p{M}\\p{N}]+[\\r\\n]*|\\s*[\\r\\n]+|\\s+(?!\\S)|\\s+",
|
| 26 |
+
"processor_class": "Qwen3VLProcessor",
|
| 27 |
+
"split_special_tokens": false,
|
| 28 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 29 |
+
"unk_token": null,
|
| 30 |
+
"video_token": "<|video_pad|>",
|
| 31 |
+
"vision_bos_token": "<|vision_start|>",
|
| 32 |
+
"vision_eos_token": "<|vision_end|>"
|
| 33 |
+
}
|
c1_no_history/round_4/trainer_state.json
ADDED
|
@@ -0,0 +1,2454 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 1.0,
|
| 6 |
+
"eval_steps": 80.0,
|
| 7 |
+
"global_step": 160,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"clip_ratio/high_max": 0.0,
|
| 14 |
+
"clip_ratio/high_mean": 0.0,
|
| 15 |
+
"clip_ratio/low_mean": 0.0,
|
| 16 |
+
"clip_ratio/low_min": 0.0,
|
| 17 |
+
"clip_ratio/region_mean": 0.0,
|
| 18 |
+
"completions/clipped_ratio": 0.23828125,
|
| 19 |
+
"completions/max_length": 2048.0,
|
| 20 |
+
"completions/mean_length": 1101.41015625,
|
| 21 |
+
"completions/min_length": 177.0,
|
| 22 |
+
"epoch": 0.00625,
|
| 23 |
+
"frac_reward_zero_std": 0.28125,
|
| 24 |
+
"grad_norm": 2.986210827594357,
|
| 25 |
+
"kl": 0.0,
|
| 26 |
+
"learning_rate": 2e-07,
|
| 27 |
+
"loss": -7.450580596923828e-09,
|
| 28 |
+
"reward": 0.49609375,
|
| 29 |
+
"reward_std": 0.3240434229373932,
|
| 30 |
+
"rewards/MajorityAccuracy/mean": 0.49609375,
|
| 31 |
+
"rewards/MajorityAccuracy/std": 0.5009641051292419,
|
| 32 |
+
"step": 1,
|
| 33 |
+
"step_time": 583.2161430492997
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"clip_ratio/high_max": 0.0,
|
| 37 |
+
"clip_ratio/high_mean": 0.0,
|
| 38 |
+
"clip_ratio/low_mean": 0.0,
|
| 39 |
+
"clip_ratio/low_min": 0.0,
|
| 40 |
+
"clip_ratio/region_mean": 0.0,
|
| 41 |
+
"epoch": 0.0125,
|
| 42 |
+
"grad_norm": 0.7643909509061038,
|
| 43 |
+
"kl": 0.0,
|
| 44 |
+
"learning_rate": 4e-07,
|
| 45 |
+
"loss": 0.0,
|
| 46 |
+
"step": 2,
|
| 47 |
+
"step_time": 74.74499122053385
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"clip_ratio/high_max": 0.0009653237793827429,
|
| 51 |
+
"clip_ratio/high_mean": 0.00024133094484568574,
|
| 52 |
+
"clip_ratio/low_mean": 0.00018310546875,
|
| 53 |
+
"clip_ratio/low_min": 0.0,
|
| 54 |
+
"clip_ratio/region_mean": 0.00042443640995770693,
|
| 55 |
+
"epoch": 0.01875,
|
| 56 |
+
"grad_norm": 1.7791344428062719,
|
| 57 |
+
"kl": 0.00013221627068560338,
|
| 58 |
+
"learning_rate": 6e-07,
|
| 59 |
+
"loss": 0.00038584228605031967,
|
| 60 |
+
"step": 3,
|
| 61 |
+
"step_time": 73.19452557340264
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"clip_ratio/high_max": 0.00376238432363607,
|
| 65 |
+
"clip_ratio/high_mean": 0.001263119964278303,
|
| 66 |
+
"clip_ratio/low_mean": 0.0005730988614232047,
|
| 67 |
+
"clip_ratio/low_min": 0.0,
|
| 68 |
+
"clip_ratio/region_mean": 0.0018362187984166667,
|
| 69 |
+
"epoch": 0.025,
|
| 70 |
+
"grad_norm": 2.826552783574266,
|
| 71 |
+
"kl": 0.00031003495678305626,
|
| 72 |
+
"learning_rate": 8e-07,
|
| 73 |
+
"loss": -0.00011411858577048406,
|
| 74 |
+
"step": 4,
|
| 75 |
+
"step_time": 72.94428212940693
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"clip_ratio/high_max": 0.0022742489964002743,
|
| 79 |
+
"clip_ratio/high_mean": 0.0006753267007297836,
|
| 80 |
+
"clip_ratio/low_mean": 0.0005792393058072776,
|
| 81 |
+
"clip_ratio/low_min": 0.0,
|
| 82 |
+
"clip_ratio/region_mean": 0.0012545660065370612,
|
| 83 |
+
"epoch": 0.03125,
|
| 84 |
+
"grad_norm": 2.1093766103129172,
|
| 85 |
+
"kl": 0.000257398762187222,
|
| 86 |
+
"learning_rate": 1e-06,
|
| 87 |
+
"loss": 0.00017226410272996873,
|
| 88 |
+
"step": 5,
|
| 89 |
+
"step_time": 73.07078120857477
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"clip_ratio/high_max": 0.004188358609098941,
|
| 93 |
+
"clip_ratio/high_mean": 0.0012539835734060034,
|
| 94 |
+
"clip_ratio/low_mean": 0.0004177786504442338,
|
| 95 |
+
"clip_ratio/low_min": 0.0,
|
| 96 |
+
"clip_ratio/region_mean": 0.0016717621947464067,
|
| 97 |
+
"epoch": 0.0375,
|
| 98 |
+
"grad_norm": 2.212865774984777,
|
| 99 |
+
"kl": 0.00025537015608279034,
|
| 100 |
+
"learning_rate": 9.998973021172564e-07,
|
| 101 |
+
"loss": 0.0001360960304737091,
|
| 102 |
+
"step": 6,
|
| 103 |
+
"step_time": 73.52026772871614
|
| 104 |
+
},
|
| 105 |
+
{
|
| 106 |
+
"clip_ratio/high_max": 0.005162169662071392,
|
| 107 |
+
"clip_ratio/high_mean": 0.0017688537845970131,
|
| 108 |
+
"clip_ratio/low_mean": 0.0005609641739283688,
|
| 109 |
+
"clip_ratio/low_min": 0.0,
|
| 110 |
+
"clip_ratio/region_mean": 0.0023298179876292124,
|
| 111 |
+
"epoch": 0.04375,
|
| 112 |
+
"grad_norm": 2.3457719036847355,
|
| 113 |
+
"kl": 0.0005217589568928815,
|
| 114 |
+
"learning_rate": 9.99589250656446e-07,
|
| 115 |
+
"loss": 4.456589522305876e-05,
|
| 116 |
+
"step": 7,
|
| 117 |
+
"step_time": 72.96448488533497
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"clip_ratio/high_max": 0.0030574808915844187,
|
| 121 |
+
"clip_ratio/high_mean": 0.0009127279336098582,
|
| 122 |
+
"clip_ratio/low_mean": 0.00021585078502539545,
|
| 123 |
+
"clip_ratio/low_min": 0.0,
|
| 124 |
+
"clip_ratio/region_mean": 0.0011285787259112112,
|
| 125 |
+
"epoch": 0.05,
|
| 126 |
+
"grad_norm": 2.0729392758592233,
|
| 127 |
+
"kl": 0.0002522961021895753,
|
| 128 |
+
"learning_rate": 9.990759721625005e-07,
|
| 129 |
+
"loss": -1.8639431800693274e-05,
|
| 130 |
+
"step": 8,
|
| 131 |
+
"step_time": 74.48424570262432
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"clip_ratio/high_max": 0.0,
|
| 135 |
+
"clip_ratio/high_mean": 0.0,
|
| 136 |
+
"clip_ratio/low_mean": 0.0,
|
| 137 |
+
"clip_ratio/low_min": 0.0,
|
| 138 |
+
"clip_ratio/region_mean": 0.0,
|
| 139 |
+
"completions/clipped_ratio": 0.3359375,
|
| 140 |
+
"completions/max_length": 2048.0,
|
| 141 |
+
"completions/mean_length": 1317.28125,
|
| 142 |
+
"completions/min_length": 181.0,
|
| 143 |
+
"epoch": 0.05625,
|
| 144 |
+
"frac_reward_zero_std": 0.3125,
|
| 145 |
+
"grad_norm": 2.9122274091013556,
|
| 146 |
+
"kl": 0.00044586965304915793,
|
| 147 |
+
"learning_rate": 9.983576774858775e-07,
|
| 148 |
+
"loss": 4.460531727090711e-06,
|
| 149 |
+
"reward": 0.37890625,
|
| 150 |
+
"reward_std": 0.31811726093292236,
|
| 151 |
+
"rewards/MajorityAccuracy/mean": 0.37890625,
|
| 152 |
+
"rewards/MajorityAccuracy/std": 0.48606500029563904,
|
| 153 |
+
"step": 9,
|
| 154 |
+
"step_time": 515.6659286916256
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"clip_ratio/high_max": 0.0010244503209833056,
|
| 158 |
+
"clip_ratio/high_mean": 0.0002561125802458264,
|
| 159 |
+
"clip_ratio/low_mean": 0.0007505547255277634,
|
| 160 |
+
"clip_ratio/low_min": 0.0,
|
| 161 |
+
"clip_ratio/region_mean": 0.0010066673057735898,
|
| 162 |
+
"epoch": 0.0625,
|
| 163 |
+
"grad_norm": 1.7240976813837106,
|
| 164 |
+
"kl": 0.0005041500371589791,
|
| 165 |
+
"learning_rate": 9.974346616959475e-07,
|
| 166 |
+
"loss": -0.00017399847274646163,
|
| 167 |
+
"step": 10,
|
| 168 |
+
"step_time": 75.27954579517245
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"clip_ratio/high_max": 0.0032364657963626087,
|
| 172 |
+
"clip_ratio/high_mean": 0.0011235259553359356,
|
| 173 |
+
"clip_ratio/low_mean": 0.00012722507381113246,
|
| 174 |
+
"clip_ratio/low_min": 0.0,
|
| 175 |
+
"clip_ratio/region_mean": 0.0012507510255090892,
|
| 176 |
+
"epoch": 0.06875,
|
| 177 |
+
"grad_norm": 1.4696282305598591,
|
| 178 |
+
"kl": 0.0005940566225035582,
|
| 179 |
+
"learning_rate": 9.963073039597796e-07,
|
| 180 |
+
"loss": 1.89047132153064e-05,
|
| 181 |
+
"step": 11,
|
| 182 |
+
"step_time": 73.95050315186381
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"clip_ratio/high_max": 0.0030274216405814514,
|
| 186 |
+
"clip_ratio/high_mean": 0.0011397005255275872,
|
| 187 |
+
"clip_ratio/low_mean": 0.0008089735201792791,
|
| 188 |
+
"clip_ratio/low_min": 0.0,
|
| 189 |
+
"clip_ratio/region_mean": 0.0019486740347929299,
|
| 190 |
+
"epoch": 0.075,
|
| 191 |
+
"grad_norm": 2.0398137808617856,
|
| 192 |
+
"kl": 0.0009280411904910579,
|
| 193 |
+
"learning_rate": 9.949760673863846e-07,
|
| 194 |
+
"loss": 7.998965156730264e-05,
|
| 195 |
+
"step": 12,
|
| 196 |
+
"step_time": 75.18844247236848
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"clip_ratio/high_max": 0.004466986050829291,
|
| 200 |
+
"clip_ratio/high_mean": 0.0015386232116725296,
|
| 201 |
+
"clip_ratio/low_mean": 0.0016696544189471751,
|
| 202 |
+
"clip_ratio/low_min": 0.0,
|
| 203 |
+
"clip_ratio/region_mean": 0.0032082776306197047,
|
| 204 |
+
"epoch": 0.08125,
|
| 205 |
+
"grad_norm": 2.6455685218976446,
|
| 206 |
+
"kl": 0.0010770552980829962,
|
| 207 |
+
"learning_rate": 9.934414988364722e-07,
|
| 208 |
+
"loss": 0.00029361341148614883,
|
| 209 |
+
"step": 13,
|
| 210 |
+
"step_time": 73.42183781415224
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"clip_ratio/high_max": 0.0026307908992748708,
|
| 214 |
+
"clip_ratio/high_mean": 0.000844605230668094,
|
| 215 |
+
"clip_ratio/low_mean": 0.0007985231786733493,
|
| 216 |
+
"clip_ratio/low_min": 0.0,
|
| 217 |
+
"clip_ratio/region_mean": 0.001643128416617401,
|
| 218 |
+
"epoch": 0.0875,
|
| 219 |
+
"grad_norm": 1.312193321739416,
|
| 220 |
+
"kl": 0.0007943696618895046,
|
| 221 |
+
"learning_rate": 9.917042286978063e-07,
|
| 222 |
+
"loss": 0.00011120666022179648,
|
| 223 |
+
"step": 14,
|
| 224 |
+
"step_time": 75.3626418672502
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"clip_ratio/high_max": 0.0031872643885435537,
|
| 228 |
+
"clip_ratio/high_mean": 0.001191677092720056,
|
| 229 |
+
"clip_ratio/low_mean": 0.0008511603518854827,
|
| 230 |
+
"clip_ratio/low_min": 0.0,
|
| 231 |
+
"clip_ratio/region_mean": 0.002042837455519475,
|
| 232 |
+
"epoch": 0.09375,
|
| 233 |
+
"grad_norm": 1.094704558875748,
|
| 234 |
+
"kl": 0.0012269473518244922,
|
| 235 |
+
"learning_rate": 9.897649706262473e-07,
|
| 236 |
+
"loss": 0.00018617522437125444,
|
| 237 |
+
"step": 15,
|
| 238 |
+
"step_time": 72.9395806491375
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"clip_ratio/high_max": 0.0049534853314980865,
|
| 242 |
+
"clip_ratio/high_mean": 0.0014965567388571799,
|
| 243 |
+
"clip_ratio/low_mean": 0.0023673264076933265,
|
| 244 |
+
"clip_ratio/low_min": 0.0,
|
| 245 |
+
"clip_ratio/region_mean": 0.003863883117446676,
|
| 246 |
+
"epoch": 0.1,
|
| 247 |
+
"grad_norm": 1.991087001072465,
|
| 248 |
+
"kl": 0.0015714890323579311,
|
| 249 |
+
"learning_rate": 9.87624521252587e-07,
|
| 250 |
+
"loss": 9.724229312269017e-05,
|
| 251 |
+
"step": 16,
|
| 252 |
+
"step_time": 72.60190603137016
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"clip_ratio/high_max": 0.0,
|
| 256 |
+
"clip_ratio/high_mean": 0.0,
|
| 257 |
+
"clip_ratio/low_mean": 0.0,
|
| 258 |
+
"clip_ratio/low_min": 0.0,
|
| 259 |
+
"clip_ratio/region_mean": 0.0,
|
| 260 |
+
"completions/clipped_ratio": 0.3125,
|
| 261 |
+
"completions/max_length": 2048.0,
|
| 262 |
+
"completions/mean_length": 1205.0703125,
|
| 263 |
+
"completions/min_length": 105.0,
|
| 264 |
+
"epoch": 0.10625,
|
| 265 |
+
"frac_reward_zero_std": 0.34375,
|
| 266 |
+
"grad_norm": 1.1233260128662363,
|
| 267 |
+
"kl": 0.001433417113730684,
|
| 268 |
+
"learning_rate": 9.852837598553008e-07,
|
| 269 |
+
"loss": 1.432268982171081e-05,
|
| 270 |
+
"reward": 0.44140625,
|
| 271 |
+
"reward_std": 0.2872208058834076,
|
| 272 |
+
"rewards/MajorityAccuracy/mean": 0.44140625,
|
| 273 |
+
"rewards/MajorityAccuracy/std": 0.4975275695323944,
|
| 274 |
+
"step": 17,
|
| 275 |
+
"step_time": 515.4773605056107
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"clip_ratio/high_max": 0.0016977377817966044,
|
| 279 |
+
"clip_ratio/high_mean": 0.0005032490735175088,
|
| 280 |
+
"clip_ratio/low_mean": 5.540779966395348e-05,
|
| 281 |
+
"clip_ratio/low_min": 0.0,
|
| 282 |
+
"clip_ratio/region_mean": 0.0005586568731814623,
|
| 283 |
+
"epoch": 0.1125,
|
| 284 |
+
"grad_norm": 1.778797100818494,
|
| 285 |
+
"kl": 0.0016957134212134406,
|
| 286 |
+
"learning_rate": 9.827436479993468e-07,
|
| 287 |
+
"loss": 0.00044856476597487926,
|
| 288 |
+
"step": 18,
|
| 289 |
+
"step_time": 75.5112268589437
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"clip_ratio/high_max": 0.003549897635821253,
|
| 293 |
+
"clip_ratio/high_mean": 0.0010264404845656827,
|
| 294 |
+
"clip_ratio/low_mean": 0.0012105941204936244,
|
| 295 |
+
"clip_ratio/low_min": 0.0,
|
| 296 |
+
"clip_ratio/region_mean": 0.0022370345977833495,
|
| 297 |
+
"epoch": 0.11875,
|
| 298 |
+
"grad_norm": 2.3605899506927566,
|
| 299 |
+
"kl": 0.00181639488437213,
|
| 300 |
+
"learning_rate": 9.800052291411629e-07,
|
| 301 |
+
"loss": 0.0003306435828562826,
|
| 302 |
+
"step": 19,
|
| 303 |
+
"step_time": 74.99134537950158
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"clip_ratio/high_max": 0.0007003401988185942,
|
| 307 |
+
"clip_ratio/high_mean": 0.00017508504970464855,
|
| 308 |
+
"clip_ratio/low_mean": 0.0003706508578034118,
|
| 309 |
+
"clip_ratio/low_min": 0.0,
|
| 310 |
+
"clip_ratio/region_mean": 0.0005457359075080603,
|
| 311 |
+
"epoch": 0.125,
|
| 312 |
+
"grad_norm": 1.5057419366934186,
|
| 313 |
+
"kl": 0.002172395456000231,
|
| 314 |
+
"learning_rate": 9.770696282000244e-07,
|
| 315 |
+
"loss": 9.777593368198723e-05,
|
| 316 |
+
"step": 20,
|
| 317 |
+
"step_time": 75.02123565971851
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"clip_ratio/high_max": 0.004221810115268454,
|
| 321 |
+
"clip_ratio/high_mean": 0.0014001036688569002,
|
| 322 |
+
"clip_ratio/low_mean": 0.0023805425444152206,
|
| 323 |
+
"clip_ratio/low_min": 0.0,
|
| 324 |
+
"clip_ratio/region_mean": 0.0037806461623404175,
|
| 325 |
+
"epoch": 0.13125,
|
| 326 |
+
"grad_norm": 3.0332444712168813,
|
| 327 |
+
"kl": 0.002920472266850993,
|
| 328 |
+
"learning_rate": 9.739380510959364e-07,
|
| 329 |
+
"loss": 0.0002395128831267357,
|
| 330 |
+
"step": 21,
|
| 331 |
+
"step_time": 72.73501506447792
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"clip_ratio/high_max": 0.005972750281216577,
|
| 335 |
+
"clip_ratio/high_mean": 0.002521935326512903,
|
| 336 |
+
"clip_ratio/low_mean": 0.0001627087767701596,
|
| 337 |
+
"clip_ratio/low_min": 0.0,
|
| 338 |
+
"clip_ratio/region_mean": 0.0026846441032830626,
|
| 339 |
+
"epoch": 0.1375,
|
| 340 |
+
"grad_norm": 3.176277678469324,
|
| 341 |
+
"kl": 0.0036093950329814106,
|
| 342 |
+
"learning_rate": 9.706117842542516e-07,
|
| 343 |
+
"loss": 0.0002291882410645485,
|
| 344 |
+
"step": 22,
|
| 345 |
+
"step_time": 72.12555831298232
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"clip_ratio/high_max": 0.009282870625611395,
|
| 349 |
+
"clip_ratio/high_mean": 0.004265933443093672,
|
| 350 |
+
"clip_ratio/low_mean": 0.0011999526432191487,
|
| 351 |
+
"clip_ratio/low_min": 0.0,
|
| 352 |
+
"clip_ratio/region_mean": 0.00546588611905463,
|
| 353 |
+
"epoch": 0.14375,
|
| 354 |
+
"grad_norm": 2.7362530366656492,
|
| 355 |
+
"kl": 0.0032782613416202366,
|
| 356 |
+
"learning_rate": 9.670921940772186e-07,
|
| 357 |
+
"loss": -5.633849650621414e-05,
|
| 358 |
+
"step": 23,
|
| 359 |
+
"step_time": 72.39059096202254
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"clip_ratio/high_max": 0.007077173824654892,
|
| 363 |
+
"clip_ratio/high_mean": 0.002103563347191084,
|
| 364 |
+
"clip_ratio/low_mean": 0.0017432841832487611,
|
| 365 |
+
"clip_ratio/low_min": 0.0,
|
| 366 |
+
"clip_ratio/region_mean": 0.0038468476268462837,
|
| 367 |
+
"epoch": 0.15,
|
| 368 |
+
"grad_norm": 2.0859031271401456,
|
| 369 |
+
"kl": 0.0036649743269663304,
|
| 370 |
+
"learning_rate": 9.633807263826744e-07,
|
| 371 |
+
"loss": 0.00048707579844631255,
|
| 372 |
+
"step": 24,
|
| 373 |
+
"step_time": 74.9029706530273
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"clip_ratio/high_max": 0.0,
|
| 377 |
+
"clip_ratio/high_mean": 0.0,
|
| 378 |
+
"clip_ratio/low_mean": 0.0,
|
| 379 |
+
"clip_ratio/low_min": 0.0,
|
| 380 |
+
"clip_ratio/region_mean": 0.0,
|
| 381 |
+
"completions/clipped_ratio": 0.25390625,
|
| 382 |
+
"completions/max_length": 2048.0,
|
| 383 |
+
"completions/mean_length": 1098.59765625,
|
| 384 |
+
"completions/min_length": 172.0,
|
| 385 |
+
"epoch": 0.15625,
|
| 386 |
+
"frac_reward_zero_std": 0.25,
|
| 387 |
+
"grad_norm": 1.8860325909968372,
|
| 388 |
+
"kl": 0.002504447620594874,
|
| 389 |
+
"learning_rate": 9.594789058101153e-07,
|
| 390 |
+
"loss": 2.5072482458199374e-05,
|
| 391 |
+
"reward": 0.42578125,
|
| 392 |
+
"reward_std": 0.3362714648246765,
|
| 393 |
+
"rewards/MajorityAccuracy/mean": 0.42578125,
|
| 394 |
+
"rewards/MajorityAccuracy/std": 0.49542948603630066,
|
| 395 |
+
"step": 25,
|
| 396 |
+
"step_time": 508.8921617567539
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"clip_ratio/high_max": 0.0027294851461192593,
|
| 400 |
+
"clip_ratio/high_mean": 0.0009607041720300913,
|
| 401 |
+
"clip_ratio/low_mean": 0.00047694276872789487,
|
| 402 |
+
"clip_ratio/low_min": 0.0,
|
| 403 |
+
"clip_ratio/region_mean": 0.0014376469407579862,
|
| 404 |
+
"epoch": 0.1625,
|
| 405 |
+
"grad_norm": 2.76047527935676,
|
| 406 |
+
"kl": 0.0036508621415123343,
|
| 407 |
+
"learning_rate": 9.55388335194388e-07,
|
| 408 |
+
"loss": 1.9023194909095764e-05,
|
| 409 |
+
"step": 26,
|
| 410 |
+
"step_time": 72.8796560689807
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"clip_ratio/high_max": 0.0023239629808813334,
|
| 414 |
+
"clip_ratio/high_mean": 0.0006654476164840162,
|
| 415 |
+
"clip_ratio/low_mean": 0.0012706654160865583,
|
| 416 |
+
"clip_ratio/low_min": 0.0,
|
| 417 |
+
"clip_ratio/region_mean": 0.001936113039846532,
|
| 418 |
+
"epoch": 0.16875,
|
| 419 |
+
"grad_norm": 2.80461729799993,
|
| 420 |
+
"kl": 0.003567846171790734,
|
| 421 |
+
"learning_rate": 9.511106949072587e-07,
|
| 422 |
+
"loss": 0.0002724621444940567,
|
| 423 |
+
"step": 27,
|
| 424 |
+
"step_time": 72.72675117477775
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"clip_ratio/high_max": 0.002852880279533565,
|
| 428 |
+
"clip_ratio/high_mean": 0.0007774874102324247,
|
| 429 |
+
"clip_ratio/low_mean": 0.0011191796293132938,
|
| 430 |
+
"clip_ratio/low_min": 0.0,
|
| 431 |
+
"clip_ratio/region_mean": 0.0018966670395457186,
|
| 432 |
+
"epoch": 0.175,
|
| 433 |
+
"grad_norm": 1.9582341779158474,
|
| 434 |
+
"kl": 0.006552146282047033,
|
| 435 |
+
"learning_rate": 9.466477421671295e-07,
|
| 436 |
+
"loss": 0.00023710899404250085,
|
| 437 |
+
"step": 28,
|
| 438 |
+
"step_time": 73.03380630910397
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"clip_ratio/high_max": 0.002334020973648876,
|
| 442 |
+
"clip_ratio/high_mean": 0.000583505243412219,
|
| 443 |
+
"clip_ratio/low_mean": 0.0007142802787711844,
|
| 444 |
+
"clip_ratio/low_min": 0.0,
|
| 445 |
+
"clip_ratio/region_mean": 0.001297785493079573,
|
| 446 |
+
"epoch": 0.18125,
|
| 447 |
+
"grad_norm": 1.5608677350999125,
|
| 448 |
+
"kl": 0.0038447123079095036,
|
| 449 |
+
"learning_rate": 9.420013103171891e-07,
|
| 450 |
+
"loss": 4.633641947293654e-05,
|
| 451 |
+
"step": 29,
|
| 452 |
+
"step_time": 74.21080725640059
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"clip_ratio/high_max": 0.008956700650742278,
|
| 456 |
+
"clip_ratio/high_mean": 0.003435240956605412,
|
| 457 |
+
"clip_ratio/low_mean": 0.0012469125358620659,
|
| 458 |
+
"clip_ratio/low_min": 0.0,
|
| 459 |
+
"clip_ratio/region_mean": 0.004682153492467478,
|
| 460 |
+
"epoch": 0.1875,
|
| 461 |
+
"grad_norm": 2.6252763928988934,
|
| 462 |
+
"kl": 0.005018167168600485,
|
| 463 |
+
"learning_rate": 9.37173308072291e-07,
|
| 464 |
+
"loss": 0.000556175597012043,
|
| 465 |
+
"step": 30,
|
| 466 |
+
"step_time": 73.8274997472763
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"clip_ratio/high_max": 0.0032279720762744546,
|
| 470 |
+
"clip_ratio/high_mean": 0.0010905458184424788,
|
| 471 |
+
"clip_ratio/low_mean": 0.00180002494744258,
|
| 472 |
+
"clip_ratio/low_min": 0.0,
|
| 473 |
+
"clip_ratio/region_mean": 0.002890570744057186,
|
| 474 |
+
"epoch": 0.19375,
|
| 475 |
+
"grad_norm": 2.1284953441335035,
|
| 476 |
+
"kl": 0.004671758855693042,
|
| 477 |
+
"learning_rate": 9.321657187348688e-07,
|
| 478 |
+
"loss": 2.0641862647607923e-05,
|
| 479 |
+
"step": 31,
|
| 480 |
+
"step_time": 73.32230195775628
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"clip_ratio/high_max": 0.006049146148143336,
|
| 484 |
+
"clip_ratio/high_mean": 0.001512286537035834,
|
| 485 |
+
"clip_ratio/low_mean": 0.0025345211470266804,
|
| 486 |
+
"clip_ratio/low_min": 0.0,
|
| 487 |
+
"clip_ratio/region_mean": 0.004046807691338472,
|
| 488 |
+
"epoch": 0.2,
|
| 489 |
+
"grad_norm": 2.28421740562053,
|
| 490 |
+
"kl": 0.004929621005430818,
|
| 491 |
+
"learning_rate": 9.269805993802128e-07,
|
| 492 |
+
"loss": 0.0006588710239157081,
|
| 493 |
+
"step": 32,
|
| 494 |
+
"step_time": 75.21482723206282
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"clip_ratio/high_max": 0.0,
|
| 498 |
+
"clip_ratio/high_mean": 0.0,
|
| 499 |
+
"clip_ratio/low_mean": 0.0,
|
| 500 |
+
"clip_ratio/low_min": 0.0,
|
| 501 |
+
"clip_ratio/region_mean": 0.0,
|
| 502 |
+
"completions/clipped_ratio": 0.3359375,
|
| 503 |
+
"completions/max_length": 2048.0,
|
| 504 |
+
"completions/mean_length": 1270.203125,
|
| 505 |
+
"completions/min_length": 200.0,
|
| 506 |
+
"epoch": 0.20625,
|
| 507 |
+
"frac_reward_zero_std": 0.34375,
|
| 508 |
+
"grad_norm": 2.295130748260158,
|
| 509 |
+
"kl": 0.0044873123115394264,
|
| 510 |
+
"learning_rate": 9.216200800114411e-07,
|
| 511 |
+
"loss": 4.486540274228901e-05,
|
| 512 |
+
"reward": 0.375,
|
| 513 |
+
"reward_std": 0.29970598220825195,
|
| 514 |
+
"rewards/MajorityAccuracy/mean": 0.375,
|
| 515 |
+
"rewards/MajorityAccuracy/std": 0.48507124185562134,
|
| 516 |
+
"step": 33,
|
| 517 |
+
"step_time": 512.358951382339
|
| 518 |
+
},
|
| 519 |
+
{
|
| 520 |
+
"clip_ratio/high_max": 0.0034892777330242097,
|
| 521 |
+
"clip_ratio/high_mean": 0.0009414566011400893,
|
| 522 |
+
"clip_ratio/low_mean": 0.0003578683918021852,
|
| 523 |
+
"clip_ratio/low_min": 0.0,
|
| 524 |
+
"clip_ratio/region_mean": 0.0012993249983992428,
|
| 525 |
+
"epoch": 0.2125,
|
| 526 |
+
"grad_norm": 1.9415625504186904,
|
| 527 |
+
"kl": 0.004369023663457483,
|
| 528 |
+
"learning_rate": 9.160863626845119e-07,
|
| 529 |
+
"loss": 0.000429940817411989,
|
| 530 |
+
"step": 34,
|
| 531 |
+
"step_time": 73.61420370265841
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"clip_ratio/high_max": 0.0012698267819359899,
|
| 535 |
+
"clip_ratio/high_mean": 0.00031745669548399746,
|
| 536 |
+
"clip_ratio/low_mean": 0.0018699307547649369,
|
| 537 |
+
"clip_ratio/low_min": 0.0,
|
| 538 |
+
"clip_ratio/region_mean": 0.002187387472076807,
|
| 539 |
+
"epoch": 0.21875,
|
| 540 |
+
"grad_norm": 2.153499422260558,
|
| 541 |
+
"kl": 0.0067603670177049935,
|
| 542 |
+
"learning_rate": 9.103817206036382e-07,
|
| 543 |
+
"loss": 0.00038580060936510563,
|
| 544 |
+
"step": 35,
|
| 545 |
+
"step_time": 75.3569995649159
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"clip_ratio/high_max": 0.006042923807399347,
|
| 549 |
+
"clip_ratio/high_mean": 0.002284320340550039,
|
| 550 |
+
"clip_ratio/low_mean": 0.0011452227045083418,
|
| 551 |
+
"clip_ratio/low_min": 0.0,
|
| 552 |
+
"clip_ratio/region_mean": 0.003429543081438169,
|
| 553 |
+
"epoch": 0.225,
|
| 554 |
+
"grad_norm": 2.2267922365299433,
|
| 555 |
+
"kl": 0.005863867641892284,
|
| 556 |
+
"learning_rate": 9.045084971874737e-07,
|
| 557 |
+
"loss": 0.00032040663063526154,
|
| 558 |
+
"step": 36,
|
| 559 |
+
"step_time": 73.38930526003242
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"clip_ratio/high_max": 0.002643206637003459,
|
| 563 |
+
"clip_ratio/high_mean": 0.0006608016592508648,
|
| 564 |
+
"clip_ratio/low_mean": 5.3682886573369615e-05,
|
| 565 |
+
"clip_ratio/low_min": 0.0,
|
| 566 |
+
"clip_ratio/region_mean": 0.0007144845458242344,
|
| 567 |
+
"epoch": 0.23125,
|
| 568 |
+
"grad_norm": 0.7914225241001149,
|
| 569 |
+
"kl": 0.0041165190923493356,
|
| 570 |
+
"learning_rate": 8.984691051064574e-07,
|
| 571 |
+
"loss": 6.843218579888344e-05,
|
| 572 |
+
"step": 37,
|
| 573 |
+
"step_time": 73.37206369265914
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"clip_ratio/high_max": 0.0015008811169536784,
|
| 577 |
+
"clip_ratio/high_mean": 0.0003752202792384196,
|
| 578 |
+
"clip_ratio/low_mean": 0.0005831138696521521,
|
| 579 |
+
"clip_ratio/low_min": 0.0,
|
| 580 |
+
"clip_ratio/region_mean": 0.0009583341488905717,
|
| 581 |
+
"epoch": 0.2375,
|
| 582 |
+
"grad_norm": 1.1275713197881758,
|
| 583 |
+
"kl": 0.004531249054707587,
|
| 584 |
+
"learning_rate": 8.922660252917087e-07,
|
| 585 |
+
"loss": 0.00019019839237444103,
|
| 586 |
+
"step": 38,
|
| 587 |
+
"step_time": 73.32271432504058
|
| 588 |
+
},
|
| 589 |
+
{
|
| 590 |
+
"clip_ratio/high_max": 0.004429596185218543,
|
| 591 |
+
"clip_ratio/high_mean": 0.0021814644569531083,
|
| 592 |
+
"clip_ratio/low_mean": 0.00126208317669807,
|
| 593 |
+
"clip_ratio/low_min": 0.0,
|
| 594 |
+
"clip_ratio/region_mean": 0.0034435476118233055,
|
| 595 |
+
"epoch": 0.24375,
|
| 596 |
+
"grad_norm": 2.0939926819071184,
|
| 597 |
+
"kl": 0.007690560654737055,
|
| 598 |
+
"learning_rate": 8.859018059158809e-07,
|
| 599 |
+
"loss": 0.000765258155297488,
|
| 600 |
+
"step": 39,
|
| 601 |
+
"step_time": 72.71778002381325
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"clip_ratio/high_max": 0.003254579642089084,
|
| 605 |
+
"clip_ratio/high_mean": 0.0011665651618386619,
|
| 606 |
+
"clip_ratio/low_mean": 0.001501767779700458,
|
| 607 |
+
"clip_ratio/low_min": 0.0005744860682170838,
|
| 608 |
+
"clip_ratio/region_mean": 0.0026683329488150775,
|
| 609 |
+
"epoch": 0.25,
|
| 610 |
+
"grad_norm": 1.5398294268964459,
|
| 611 |
+
"kl": 0.00415677746059373,
|
| 612 |
+
"learning_rate": 8.793790613463954e-07,
|
| 613 |
+
"loss": 0.00025182031095027924,
|
| 614 |
+
"step": 40,
|
| 615 |
+
"step_time": 75.32824059203267
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"clip_ratio/high_max": 0.0,
|
| 619 |
+
"clip_ratio/high_mean": 0.0,
|
| 620 |
+
"clip_ratio/low_mean": 0.0,
|
| 621 |
+
"clip_ratio/low_min": 0.0,
|
| 622 |
+
"clip_ratio/region_mean": 0.0,
|
| 623 |
+
"completions/clipped_ratio": 0.3828125,
|
| 624 |
+
"completions/max_length": 2048.0,
|
| 625 |
+
"completions/mean_length": 1370.46484375,
|
| 626 |
+
"completions/min_length": 121.0,
|
| 627 |
+
"epoch": 0.25625,
|
| 628 |
+
"frac_reward_zero_std": 0.25,
|
| 629 |
+
"grad_norm": 2.801011279947724,
|
| 630 |
+
"kl": 0.006137209420558065,
|
| 631 |
+
"learning_rate": 8.727004710714798e-07,
|
| 632 |
+
"loss": 6.137228774605319e-05,
|
| 633 |
+
"reward": 0.38671875,
|
| 634 |
+
"reward_std": 0.35718268156051636,
|
| 635 |
+
"rewards/MajorityAccuracy/mean": 0.38671875,
|
| 636 |
+
"rewards/MajorityAccuracy/std": 0.4879522919654846,
|
| 637 |
+
"step": 41,
|
| 638 |
+
"step_time": 519.2862969636917
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"clip_ratio/high_max": 0.0012449617133825086,
|
| 642 |
+
"clip_ratio/high_mean": 0.00036838167761743534,
|
| 643 |
+
"clip_ratio/low_mean": 0.00029242513483040966,
|
| 644 |
+
"clip_ratio/low_min": 0.0,
|
| 645 |
+
"clip_ratio/region_mean": 0.0006608068106288556,
|
| 646 |
+
"epoch": 0.2625,
|
| 647 |
+
"grad_norm": 1.4982068597539515,
|
| 648 |
+
"kl": 0.006130625202786177,
|
| 649 |
+
"learning_rate": 8.658687785994578e-07,
|
| 650 |
+
"loss": 0.00024935364490374923,
|
| 651 |
+
"step": 42,
|
| 652 |
+
"step_time": 75.3356001637876
|
| 653 |
+
},
|
| 654 |
+
{
|
| 655 |
+
"clip_ratio/high_max": 0.002559020897024311,
|
| 656 |
+
"clip_ratio/high_mean": 0.0007488719675166067,
|
| 657 |
+
"clip_ratio/low_mean": 0.0002649934212968219,
|
| 658 |
+
"clip_ratio/low_min": 0.0,
|
| 659 |
+
"clip_ratio/region_mean": 0.0010138653924514074,
|
| 660 |
+
"epoch": 0.26875,
|
| 661 |
+
"grad_norm": 1.3571753780415543,
|
| 662 |
+
"kl": 0.005422542861197144,
|
| 663 |
+
"learning_rate": 8.588867903317394e-07,
|
| 664 |
+
"loss": 0.00010745729377958924,
|
| 665 |
+
"step": 43,
|
| 666 |
+
"step_time": 75.51394603773952
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"clip_ratio/high_max": 0.00523840356618166,
|
| 670 |
+
"clip_ratio/high_mean": 0.001777671481249854,
|
| 671 |
+
"clip_ratio/low_mean": 0.000694611546350643,
|
| 672 |
+
"clip_ratio/low_min": 0.0,
|
| 673 |
+
"clip_ratio/region_mean": 0.0024722830785322003,
|
| 674 |
+
"epoch": 0.275,
|
| 675 |
+
"grad_norm": 3.208490405017419,
|
| 676 |
+
"kl": 0.005703358561731875,
|
| 677 |
+
"learning_rate": 8.51757374409974e-07,
|
| 678 |
+
"loss": 0.000607617199420929,
|
| 679 |
+
"step": 44,
|
| 680 |
+
"step_time": 74.36214097216725
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"clip_ratio/high_max": 0.003623663302278146,
|
| 684 |
+
"clip_ratio/high_mean": 0.0010073418961837888,
|
| 685 |
+
"clip_ratio/low_mean": 0.0009745266652316786,
|
| 686 |
+
"clip_ratio/low_min": 0.0,
|
| 687 |
+
"clip_ratio/region_mean": 0.00198186855413951,
|
| 688 |
+
"epoch": 0.28125,
|
| 689 |
+
"grad_norm": 1.965628220783796,
|
| 690 |
+
"kl": 0.007460778055246919,
|
| 691 |
+
"learning_rate": 8.444834595378433e-07,
|
| 692 |
+
"loss": 0.00035774760181084275,
|
| 693 |
+
"step": 45,
|
| 694 |
+
"step_time": 74.56202270463109
|
| 695 |
+
},
|
| 696 |
+
{
|
| 697 |
+
"clip_ratio/high_max": 0.0030825579597149044,
|
| 698 |
+
"clip_ratio/high_mean": 0.0010653909703250974,
|
| 699 |
+
"clip_ratio/low_mean": 0.002226224567493773,
|
| 700 |
+
"clip_ratio/low_min": 0.0,
|
| 701 |
+
"clip_ratio/region_mean": 0.003291615590569563,
|
| 702 |
+
"epoch": 0.2875,
|
| 703 |
+
"grad_norm": 2.610364386147309,
|
| 704 |
+
"kl": 0.009423354058526456,
|
| 705 |
+
"learning_rate": 8.370680337779736e-07,
|
| 706 |
+
"loss": 0.00044197216629981995,
|
| 707 |
+
"step": 46,
|
| 708 |
+
"step_time": 72.78488124534488
|
| 709 |
+
},
|
| 710 |
+
{
|
| 711 |
+
"clip_ratio/high_max": 0.0015832487406441942,
|
| 712 |
+
"clip_ratio/high_mean": 0.0004805771059182007,
|
| 713 |
+
"clip_ratio/low_mean": 0.0014961584965931252,
|
| 714 |
+
"clip_ratio/low_min": 0.0,
|
| 715 |
+
"clip_ratio/region_mean": 0.001976735598873347,
|
| 716 |
+
"epoch": 0.29375,
|
| 717 |
+
"grad_norm": 1.9940658215554032,
|
| 718 |
+
"kl": 0.00576817145338282,
|
| 719 |
+
"learning_rate": 8.295141433244659e-07,
|
| 720 |
+
"loss": 0.0001441163767594844,
|
| 721 |
+
"step": 47,
|
| 722 |
+
"step_time": 74.32632047683
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"clip_ratio/high_max": 0.003462070890236646,
|
| 726 |
+
"clip_ratio/high_mean": 0.0013590175658464432,
|
| 727 |
+
"clip_ratio/low_mean": 0.001395915896864608,
|
| 728 |
+
"clip_ratio/low_min": 0.0,
|
| 729 |
+
"clip_ratio/region_mean": 0.0027549334918148816,
|
| 730 |
+
"epoch": 0.3,
|
| 731 |
+
"grad_norm": 1.380322063961704,
|
| 732 |
+
"kl": 0.0075088670128025115,
|
| 733 |
+
"learning_rate": 8.218248912515442e-07,
|
| 734 |
+
"loss": 0.00025261982227675617,
|
| 735 |
+
"step": 48,
|
| 736 |
+
"step_time": 73.32234900072217
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"clip_ratio/high_max": 0.0,
|
| 740 |
+
"clip_ratio/high_mean": 0.0,
|
| 741 |
+
"clip_ratio/low_mean": 0.0,
|
| 742 |
+
"clip_ratio/low_min": 0.0,
|
| 743 |
+
"clip_ratio/region_mean": 0.0,
|
| 744 |
+
"completions/clipped_ratio": 0.3203125,
|
| 745 |
+
"completions/max_length": 2048.0,
|
| 746 |
+
"completions/mean_length": 1313.8984375,
|
| 747 |
+
"completions/min_length": 140.0,
|
| 748 |
+
"epoch": 0.30625,
|
| 749 |
+
"frac_reward_zero_std": 0.25,
|
| 750 |
+
"grad_norm": 0.9349386209793862,
|
| 751 |
+
"kl": 0.007391649356577545,
|
| 752 |
+
"learning_rate": 8.140034362388397e-07,
|
| 753 |
+
"loss": 7.390060636680573e-05,
|
| 754 |
+
"reward": 0.4453125,
|
| 755 |
+
"reward_std": 0.3521764576435089,
|
| 756 |
+
"rewards/MajorityAccuracy/mean": 0.4453125,
|
| 757 |
+
"rewards/MajorityAccuracy/std": 0.49797382950782776,
|
| 758 |
+
"step": 49,
|
| 759 |
+
"step_time": 516.0815847478807
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"clip_ratio/high_max": 0.0026171093559241854,
|
| 763 |
+
"clip_ratio/high_mean": 0.0008478489162371261,
|
| 764 |
+
"clip_ratio/low_mean": 0.0013602433318737894,
|
| 765 |
+
"clip_ratio/low_min": 0.0,
|
| 766 |
+
"clip_ratio/region_mean": 0.0022080922353779897,
|
| 767 |
+
"epoch": 0.3125,
|
| 768 |
+
"grad_norm": 2.8171020177360306,
|
| 769 |
+
"kl": 0.007829381036572158,
|
| 770 |
+
"learning_rate": 8.060529912738314e-07,
|
| 771 |
+
"loss": 1.082196831703186e-06,
|
| 772 |
+
"step": 50,
|
| 773 |
+
"step_time": 75.03933786600828
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"clip_ratio/high_max": 0.0011442864270065911,
|
| 777 |
+
"clip_ratio/high_mean": 0.0004122871414438123,
|
| 778 |
+
"clip_ratio/low_mean": 0.0003731185133801773,
|
| 779 |
+
"clip_ratio/low_min": 0.0,
|
| 780 |
+
"clip_ratio/region_mean": 0.0007854056548239896,
|
| 781 |
+
"epoch": 0.31875,
|
| 782 |
+
"grad_norm": 1.3998620793825205,
|
| 783 |
+
"kl": 0.0081359957694076,
|
| 784 |
+
"learning_rate": 7.979768223319785e-07,
|
| 785 |
+
"loss": 0.0001542771206004545,
|
| 786 |
+
"step": 51,
|
| 787 |
+
"step_time": 74.69900735095143
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"clip_ratio/high_max": 0.001467001078708563,
|
| 791 |
+
"clip_ratio/high_mean": 0.00038433608460763935,
|
| 792 |
+
"clip_ratio/low_mean": 0.0004046112262585666,
|
| 793 |
+
"clip_ratio/low_min": 0.0,
|
| 794 |
+
"clip_ratio/region_mean": 0.0007889473072282271,
|
| 795 |
+
"epoch": 0.325,
|
| 796 |
+
"grad_norm": 2.5471393876331803,
|
| 797 |
+
"kl": 0.007960985065437853,
|
| 798 |
+
"learning_rate": 7.897782470350849e-07,
|
| 799 |
+
"loss": 0.0002810154401231557,
|
| 800 |
+
"step": 52,
|
| 801 |
+
"step_time": 73.79663929343224
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"clip_ratio/high_max": 0.00330408156150952,
|
| 805 |
+
"clip_ratio/high_mean": 0.001075653824955225,
|
| 806 |
+
"clip_ratio/low_mean": 0.0009418450827070046,
|
| 807 |
+
"clip_ratio/low_min": 0.0,
|
| 808 |
+
"clip_ratio/region_mean": 0.002017498904024251,
|
| 809 |
+
"epoch": 0.33125,
|
| 810 |
+
"grad_norm": 2.2923094874124974,
|
| 811 |
+
"kl": 0.009701867820695043,
|
| 812 |
+
"learning_rate": 7.814606332884488e-07,
|
| 813 |
+
"loss": 0.00031950633274391294,
|
| 814 |
+
"step": 53,
|
| 815 |
+
"step_time": 73.13905543461442
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"clip_ratio/high_max": 0.0013176402426324785,
|
| 819 |
+
"clip_ratio/high_mean": 0.00037186930421739817,
|
| 820 |
+
"clip_ratio/low_mean": 0.0014924706702004187,
|
| 821 |
+
"clip_ratio/low_min": 0.0,
|
| 822 |
+
"clip_ratio/region_mean": 0.0018643399962456897,
|
| 823 |
+
"epoch": 0.3375,
|
| 824 |
+
"grad_norm": 1.623325347602918,
|
| 825 |
+
"kl": 0.009092962951399386,
|
| 826 |
+
"learning_rate": 7.730273978973552e-07,
|
| 827 |
+
"loss": 0.0002578688436187804,
|
| 828 |
+
"step": 54,
|
| 829 |
+
"step_time": 73.43780096247792
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"clip_ratio/high_max": 0.007653981447219849,
|
| 833 |
+
"clip_ratio/high_mean": 0.002674676914466545,
|
| 834 |
+
"clip_ratio/low_mean": 0.000672074831527425,
|
| 835 |
+
"clip_ratio/low_min": 0.0,
|
| 836 |
+
"clip_ratio/region_mean": 0.003346751705976203,
|
| 837 |
+
"epoch": 0.34375,
|
| 838 |
+
"grad_norm": 2.106126126713996,
|
| 839 |
+
"kl": 0.006722021847963333,
|
| 840 |
+
"learning_rate": 7.644820051634812e-07,
|
| 841 |
+
"loss": 0.0004414509458001703,
|
| 842 |
+
"step": 55,
|
| 843 |
+
"step_time": 74.23114288598299
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"clip_ratio/high_max": 0.004459681382286362,
|
| 847 |
+
"clip_ratio/high_mean": 0.0014898704284860287,
|
| 848 |
+
"clip_ratio/low_mean": 0.0022329604835249484,
|
| 849 |
+
"clip_ratio/low_min": 0.0,
|
| 850 |
+
"clip_ratio/region_mean": 0.0037228309083729982,
|
| 851 |
+
"epoch": 0.35,
|
| 852 |
+
"grad_norm": 2.277914828160642,
|
| 853 |
+
"kl": 0.006279331108089536,
|
| 854 |
+
"learning_rate": 7.558279654617912e-07,
|
| 855 |
+
"loss": 0.0005022436380386353,
|
| 856 |
+
"step": 56,
|
| 857 |
+
"step_time": 74.43277643993497
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"clip_ratio/high_max": 0.0,
|
| 861 |
+
"clip_ratio/high_mean": 0.0,
|
| 862 |
+
"clip_ratio/low_mean": 0.0,
|
| 863 |
+
"clip_ratio/low_min": 0.0,
|
| 864 |
+
"clip_ratio/region_mean": 0.0,
|
| 865 |
+
"completions/clipped_ratio": 0.26171875,
|
| 866 |
+
"completions/max_length": 2048.0,
|
| 867 |
+
"completions/mean_length": 1189.99609375,
|
| 868 |
+
"completions/min_length": 101.0,
|
| 869 |
+
"epoch": 0.35625,
|
| 870 |
+
"frac_reward_zero_std": 0.25,
|
| 871 |
+
"grad_norm": 2.2303756522395437,
|
| 872 |
+
"kl": 0.007263063220307231,
|
| 873 |
+
"learning_rate": 7.470688337985029e-07,
|
| 874 |
+
"loss": 7.261335849761963e-05,
|
| 875 |
+
"reward": 0.453125,
|
| 876 |
+
"reward_std": 0.3441646099090576,
|
| 877 |
+
"rewards/MajorityAccuracy/mean": 0.453125,
|
| 878 |
+
"rewards/MajorityAccuracy/std": 0.4987730085849762,
|
| 879 |
+
"step": 57,
|
| 880 |
+
"step_time": 511.8526738025248
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"clip_ratio/high_max": 0.001533832291897852,
|
| 884 |
+
"clip_ratio/high_mean": 0.00043406462464190554,
|
| 885 |
+
"clip_ratio/low_mean": 0.00013662212950293906,
|
| 886 |
+
"clip_ratio/low_min": 0.0,
|
| 887 |
+
"clip_ratio/region_mean": 0.0005706867505068658,
|
| 888 |
+
"epoch": 0.3625,
|
| 889 |
+
"grad_norm": 1.1615091417080603,
|
| 890 |
+
"kl": 0.005804769403766841,
|
| 891 |
+
"learning_rate": 7.382082083507225e-07,
|
| 892 |
+
"loss": 0.00015143885684665293,
|
| 893 |
+
"step": 58,
|
| 894 |
+
"step_time": 74.09412939473987
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"clip_ratio/high_max": 0.002657092351000756,
|
| 898 |
+
"clip_ratio/high_mean": 0.0011151798389619216,
|
| 899 |
+
"clip_ratio/low_mean": 0.0004763772012665868,
|
| 900 |
+
"clip_ratio/low_min": 0.0,
|
| 901 |
+
"clip_ratio/region_mean": 0.0015915570184006356,
|
| 902 |
+
"epoch": 0.36875,
|
| 903 |
+
"grad_norm": 2.9994507716754164,
|
| 904 |
+
"kl": 0.0109400216024369,
|
| 905 |
+
"learning_rate": 7.292497289883432e-07,
|
| 906 |
+
"loss": 0.000593012198805809,
|
| 907 |
+
"step": 59,
|
| 908 |
+
"step_time": 72.97737488895655
|
| 909 |
+
},
|
| 910 |
+
{
|
| 911 |
+
"clip_ratio/high_max": 0.0,
|
| 912 |
+
"clip_ratio/high_mean": 0.0,
|
| 913 |
+
"clip_ratio/low_mean": 0.0,
|
| 914 |
+
"clip_ratio/low_min": 0.0,
|
| 915 |
+
"clip_ratio/region_mean": 0.0,
|
| 916 |
+
"epoch": 0.375,
|
| 917 |
+
"grad_norm": 0.021404946051071225,
|
| 918 |
+
"kl": 0.008182193792890757,
|
| 919 |
+
"learning_rate": 7.201970757788171e-07,
|
| 920 |
+
"loss": 8.18784610601142e-05,
|
| 921 |
+
"step": 60,
|
| 922 |
+
"step_time": 75.32762111350894
|
| 923 |
+
},
|
| 924 |
+
{
|
| 925 |
+
"clip_ratio/high_max": 0.004798955647856928,
|
| 926 |
+
"clip_ratio/high_mean": 0.0018523868457123172,
|
| 927 |
+
"clip_ratio/low_mean": 0.0007718609122093767,
|
| 928 |
+
"clip_ratio/low_min": 0.0,
|
| 929 |
+
"clip_ratio/region_mean": 0.002624247754283715,
|
| 930 |
+
"epoch": 0.38125,
|
| 931 |
+
"grad_norm": 2.706284687629399,
|
| 932 |
+
"kl": 0.009783167042769492,
|
| 933 |
+
"learning_rate": 7.110539674754159e-07,
|
| 934 |
+
"loss": 0.0004464881494641304,
|
| 935 |
+
"step": 61,
|
| 936 |
+
"step_time": 73.28571210801601
|
| 937 |
+
},
|
| 938 |
+
{
|
| 939 |
+
"clip_ratio/high_max": 0.003334568871650845,
|
| 940 |
+
"clip_ratio/high_mean": 0.0011105811572633684,
|
| 941 |
+
"clip_ratio/low_mean": 0.0017154762681457214,
|
| 942 |
+
"clip_ratio/low_min": 0.0,
|
| 943 |
+
"clip_ratio/region_mean": 0.002826057403581217,
|
| 944 |
+
"epoch": 0.3875,
|
| 945 |
+
"grad_norm": 2.728294972368696,
|
| 946 |
+
"kl": 0.011908081243745983,
|
| 947 |
+
"learning_rate": 7.018241599895973e-07,
|
| 948 |
+
"loss": 0.000224287185119465,
|
| 949 |
+
"step": 62,
|
| 950 |
+
"step_time": 74.52026894316077
|
| 951 |
+
},
|
| 952 |
+
{
|
| 953 |
+
"clip_ratio/high_max": 0.011124934593681246,
|
| 954 |
+
"clip_ratio/high_mean": 0.003255182091379538,
|
| 955 |
+
"clip_ratio/low_mean": 0.0014767848515475634,
|
| 956 |
+
"clip_ratio/low_min": 0.0,
|
| 957 |
+
"clip_ratio/region_mean": 0.004731966997496784,
|
| 958 |
+
"epoch": 0.39375,
|
| 959 |
+
"grad_norm": 3.1630455698667985,
|
| 960 |
+
"kl": 0.009619004325941205,
|
| 961 |
+
"learning_rate": 6.925114448481088e-07,
|
| 962 |
+
"loss": 0.00045295816380530596,
|
| 963 |
+
"step": 63,
|
| 964 |
+
"step_time": 73.34500496834517
|
| 965 |
+
},
|
| 966 |
+
{
|
| 967 |
+
"clip_ratio/high_max": 0.0033349687582813203,
|
| 968 |
+
"clip_ratio/high_mean": 0.0013413814740488306,
|
| 969 |
+
"clip_ratio/low_mean": 0.0016166098357643932,
|
| 970 |
+
"clip_ratio/low_min": 0.0004035793972434476,
|
| 971 |
+
"clip_ratio/region_mean": 0.0029579913825728,
|
| 972 |
+
"epoch": 0.4,
|
| 973 |
+
"grad_norm": 2.0150323901281375,
|
| 974 |
+
"kl": 0.0076264970120973885,
|
| 975 |
+
"learning_rate": 6.831196476354614e-07,
|
| 976 |
+
"loss": 0.00040422752499580383,
|
| 977 |
+
"step": 64,
|
| 978 |
+
"step_time": 74.19254837557673
|
| 979 |
+
},
|
| 980 |
+
{
|
| 981 |
+
"clip_ratio/high_max": 0.0,
|
| 982 |
+
"clip_ratio/high_mean": 0.0,
|
| 983 |
+
"clip_ratio/low_mean": 0.0,
|
| 984 |
+
"clip_ratio/low_min": 0.0,
|
| 985 |
+
"clip_ratio/region_mean": 0.0,
|
| 986 |
+
"completions/clipped_ratio": 0.26171875,
|
| 987 |
+
"completions/max_length": 2048.0,
|
| 988 |
+
"completions/mean_length": 1027.71484375,
|
| 989 |
+
"completions/min_length": 85.0,
|
| 990 |
+
"epoch": 0.40625,
|
| 991 |
+
"frac_reward_zero_std": 0.28125,
|
| 992 |
+
"grad_norm": 1.913043023779487,
|
| 993 |
+
"kl": 0.011851719347760081,
|
| 994 |
+
"learning_rate": 6.7365262642241e-07,
|
| 995 |
+
"loss": 0.00011846248526126146,
|
| 996 |
+
"reward": 0.43359375,
|
| 997 |
+
"reward_std": 0.33824095129966736,
|
| 998 |
+
"rewards/MajorityAccuracy/mean": 0.43359375,
|
| 999 |
+
"rewards/MajorityAccuracy/std": 0.4965413510799408,
|
| 1000 |
+
"step": 65,
|
| 1001 |
+
"step_time": 508.6424080841243
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"clip_ratio/high_max": 0.0017444547120248899,
|
| 1005 |
+
"clip_ratio/high_mean": 0.00043611367800622247,
|
| 1006 |
+
"clip_ratio/low_mean": 0.0009591371490387246,
|
| 1007 |
+
"clip_ratio/low_min": 0.0,
|
| 1008 |
+
"clip_ratio/region_mean": 0.0013952508343209047,
|
| 1009 |
+
"epoch": 0.4125,
|
| 1010 |
+
"grad_norm": 2.710579849305841,
|
| 1011 |
+
"kl": 0.016944863134995103,
|
| 1012 |
+
"learning_rate": 6.641142701810931e-07,
|
| 1013 |
+
"loss": 0.0003184548986610025,
|
| 1014 |
+
"step": 66,
|
| 1015 |
+
"step_time": 72.45338782295585
|
| 1016 |
+
},
|
| 1017 |
+
{
|
| 1018 |
+
"clip_ratio/high_max": 0.002428653446258977,
|
| 1019 |
+
"clip_ratio/high_mean": 0.0009222025400958955,
|
| 1020 |
+
"clip_ratio/low_mean": 0.0006293811602517962,
|
| 1021 |
+
"clip_ratio/low_min": 0.0,
|
| 1022 |
+
"clip_ratio/region_mean": 0.0015515836930717342,
|
| 1023 |
+
"epoch": 0.41875,
|
| 1024 |
+
"grad_norm": 1.802270888758818,
|
| 1025 |
+
"kl": 0.008259121561422944,
|
| 1026 |
+
"learning_rate": 6.545084971874736e-07,
|
| 1027 |
+
"loss": 0.00026009976863861084,
|
| 1028 |
+
"step": 67,
|
| 1029 |
+
"step_time": 74.40196067467332
|
| 1030 |
+
},
|
| 1031 |
+
{
|
| 1032 |
+
"clip_ratio/high_max": 0.0030373061599675566,
|
| 1033 |
+
"clip_ratio/high_mean": 0.0007593265399918891,
|
| 1034 |
+
"clip_ratio/low_mean": 0.0007969251128088217,
|
| 1035 |
+
"clip_ratio/low_min": 0.0,
|
| 1036 |
+
"clip_ratio/region_mean": 0.001556251649162732,
|
| 1037 |
+
"epoch": 0.425,
|
| 1038 |
+
"grad_norm": 2.429563005563351,
|
| 1039 |
+
"kl": 0.010638888110406697,
|
| 1040 |
+
"learning_rate": 6.448392534117461e-07,
|
| 1041 |
+
"loss": 0.0005287423846311867,
|
| 1042 |
+
"step": 68,
|
| 1043 |
+
"step_time": 75.51996095478535
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"clip_ratio/high_max": 0.0019085468957200646,
|
| 1047 |
+
"clip_ratio/high_mean": 0.0006161864221212454,
|
| 1048 |
+
"clip_ratio/low_mean": 0.0011205398914171383,
|
| 1049 |
+
"clip_ratio/low_min": 0.0001220703125,
|
| 1050 |
+
"clip_ratio/region_mean": 0.0017367263208143413,
|
| 1051 |
+
"epoch": 0.43125,
|
| 1052 |
+
"grad_norm": 2.6787279241137463,
|
| 1053 |
+
"kl": 0.010715971351601183,
|
| 1054 |
+
"learning_rate": 6.351105108973644e-07,
|
| 1055 |
+
"loss": 0.0004923101514577866,
|
| 1056 |
+
"step": 69,
|
| 1057 |
+
"step_time": 74.10161871090531
|
| 1058 |
+
},
|
| 1059 |
+
{
|
| 1060 |
+
"clip_ratio/high_max": 0.007202768581919372,
|
| 1061 |
+
"clip_ratio/high_mean": 0.0018539591692388058,
|
| 1062 |
+
"clip_ratio/low_mean": 0.0008657988801132888,
|
| 1063 |
+
"clip_ratio/low_min": 0.0,
|
| 1064 |
+
"clip_ratio/region_mean": 0.002719758020248264,
|
| 1065 |
+
"epoch": 0.4375,
|
| 1066 |
+
"grad_norm": 1.99122730685316,
|
| 1067 |
+
"kl": 0.02027798513881862,
|
| 1068 |
+
"learning_rate": 6.253262661293602e-07,
|
| 1069 |
+
"loss": 3.170860873069614e-05,
|
| 1070 |
+
"step": 70,
|
| 1071 |
+
"step_time": 72.81895193830132
|
| 1072 |
+
},
|
| 1073 |
+
{
|
| 1074 |
+
"clip_ratio/high_max": 0.006385865795891732,
|
| 1075 |
+
"clip_ratio/high_mean": 0.0020873981557087973,
|
| 1076 |
+
"clip_ratio/low_mean": 0.001506839176727226,
|
| 1077 |
+
"clip_ratio/low_min": 0.0,
|
| 1078 |
+
"clip_ratio/region_mean": 0.0035942374379374087,
|
| 1079 |
+
"epoch": 0.44375,
|
| 1080 |
+
"grad_norm": 2.790475205858305,
|
| 1081 |
+
"kl": 0.011305788182653487,
|
| 1082 |
+
"learning_rate": 6.154905383926216e-07,
|
| 1083 |
+
"loss": 2.1548941731452942e-05,
|
| 1084 |
+
"step": 71,
|
| 1085 |
+
"step_time": 74.20685712993145
|
| 1086 |
+
},
|
| 1087 |
+
{
|
| 1088 |
+
"clip_ratio/high_max": 0.0028665132413152605,
|
| 1089 |
+
"clip_ratio/high_mean": 0.0008875314451870508,
|
| 1090 |
+
"clip_ratio/low_mean": 0.002561279703513719,
|
| 1091 |
+
"clip_ratio/low_min": 0.0,
|
| 1092 |
+
"clip_ratio/region_mean": 0.003448811126872897,
|
| 1093 |
+
"epoch": 0.45,
|
| 1094 |
+
"grad_norm": 3.012741597791376,
|
| 1095 |
+
"kl": 0.013034800766035914,
|
| 1096 |
+
"learning_rate": 6.056073681208037e-07,
|
| 1097 |
+
"loss": 0.0004901915672235191,
|
| 1098 |
+
"step": 72,
|
| 1099 |
+
"step_time": 74.80292579904199
|
| 1100 |
+
},
|
| 1101 |
+
{
|
| 1102 |
+
"clip_ratio/high_max": 0.0,
|
| 1103 |
+
"clip_ratio/high_mean": 0.0,
|
| 1104 |
+
"clip_ratio/low_mean": 0.0,
|
| 1105 |
+
"clip_ratio/low_min": 0.0,
|
| 1106 |
+
"clip_ratio/region_mean": 0.0,
|
| 1107 |
+
"completions/clipped_ratio": 0.25390625,
|
| 1108 |
+
"completions/max_length": 2048.0,
|
| 1109 |
+
"completions/mean_length": 1104.64453125,
|
| 1110 |
+
"completions/min_length": 168.0,
|
| 1111 |
+
"epoch": 0.45625,
|
| 1112 |
+
"frac_reward_zero_std": 0.34375,
|
| 1113 |
+
"grad_norm": 1.7711935974966033,
|
| 1114 |
+
"kl": 0.011829601018689573,
|
| 1115 |
+
"learning_rate": 5.956808152365532e-07,
|
| 1116 |
+
"loss": 0.00011829112190753222,
|
| 1117 |
+
"reward": 0.5234375,
|
| 1118 |
+
"reward_std": 0.3019438683986664,
|
| 1119 |
+
"rewards/MajorityAccuracy/mean": 0.5234375,
|
| 1120 |
+
"rewards/MajorityAccuracy/std": 0.5004287362098694,
|
| 1121 |
+
"step": 73,
|
| 1122 |
+
"step_time": 513.0234688930213
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"clip_ratio/high_max": 0.003111783051281236,
|
| 1126 |
+
"clip_ratio/high_mean": 0.001058581819961546,
|
| 1127 |
+
"clip_ratio/low_mean": 0.00010648906754795462,
|
| 1128 |
+
"clip_ratio/low_min": 0.0,
|
| 1129 |
+
"clip_ratio/region_mean": 0.0011650708838715218,
|
| 1130 |
+
"epoch": 0.4625,
|
| 1131 |
+
"grad_norm": 2.1423299478004045,
|
| 1132 |
+
"kl": 0.010376534541137516,
|
| 1133 |
+
"learning_rate": 5.857149574837268e-07,
|
| 1134 |
+
"loss": 0.00031491974368691444,
|
| 1135 |
+
"step": 74,
|
| 1136 |
+
"step_time": 72.72536626085639
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"clip_ratio/high_max": 0.00359585796832107,
|
| 1140 |
+
"clip_ratio/high_mean": 0.0011286692824796773,
|
| 1141 |
+
"clip_ratio/low_mean": 0.0005291246925480664,
|
| 1142 |
+
"clip_ratio/low_min": 0.0,
|
| 1143 |
+
"clip_ratio/region_mean": 0.0016577939823037013,
|
| 1144 |
+
"epoch": 0.46875,
|
| 1145 |
+
"grad_norm": 2.828838175954941,
|
| 1146 |
+
"kl": 0.011379040544852614,
|
| 1147 |
+
"learning_rate": 5.757138887522883e-07,
|
| 1148 |
+
"loss": -0.0002879640960600227,
|
| 1149 |
+
"step": 75,
|
| 1150 |
+
"step_time": 74.70964882522821
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"clip_ratio/high_max": 0.0025426476495340466,
|
| 1154 |
+
"clip_ratio/high_mean": 0.0008464492639177479,
|
| 1155 |
+
"clip_ratio/low_mean": 0.0005352373973437352,
|
| 1156 |
+
"clip_ratio/low_min": 0.0,
|
| 1157 |
+
"clip_ratio/region_mean": 0.0013816866558045149,
|
| 1158 |
+
"epoch": 0.475,
|
| 1159 |
+
"grad_norm": 1.755370795336528,
|
| 1160 |
+
"kl": 0.0100885204738006,
|
| 1161 |
+
"learning_rate": 5.656817173965732e-07,
|
| 1162 |
+
"loss": 0.00019588571740314364,
|
| 1163 |
+
"step": 76,
|
| 1164 |
+
"step_time": 73.32339191809297
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"clip_ratio/high_max": 0.008715020376257598,
|
| 1168 |
+
"clip_ratio/high_mean": 0.0028472369303926826,
|
| 1169 |
+
"clip_ratio/low_mean": 0.0005957270404906012,
|
| 1170 |
+
"clip_ratio/low_min": 0.0,
|
| 1171 |
+
"clip_ratio/region_mean": 0.0034429639053996652,
|
| 1172 |
+
"epoch": 0.48125,
|
| 1173 |
+
"grad_norm": 1.7583193991324009,
|
| 1174 |
+
"kl": 0.012249218299984932,
|
| 1175 |
+
"learning_rate": 5.556225645476118e-07,
|
| 1176 |
+
"loss": 4.3854539399035275e-05,
|
| 1177 |
+
"step": 77,
|
| 1178 |
+
"step_time": 74.84596969559789
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"clip_ratio/high_max": 0.0010968863498419523,
|
| 1182 |
+
"clip_ratio/high_mean": 0.0002742215874604881,
|
| 1183 |
+
"clip_ratio/low_mean": 0.00018310546875,
|
| 1184 |
+
"clip_ratio/low_min": 0.0,
|
| 1185 |
+
"clip_ratio/region_mean": 0.0004573270562104881,
|
| 1186 |
+
"epoch": 0.4875,
|
| 1187 |
+
"grad_norm": 0.6791087125619685,
|
| 1188 |
+
"kl": 0.012874645064584911,
|
| 1189 |
+
"learning_rate": 5.455405624202032e-07,
|
| 1190 |
+
"loss": 0.00020579140982590616,
|
| 1191 |
+
"step": 78,
|
| 1192 |
+
"step_time": 72.68165834620595
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"clip_ratio/high_max": 0.0076710605644620955,
|
| 1196 |
+
"clip_ratio/high_mean": 0.0023161800418165512,
|
| 1197 |
+
"clip_ratio/low_mean": 0.0010354267797083594,
|
| 1198 |
+
"clip_ratio/low_min": 0.0,
|
| 1199 |
+
"clip_ratio/region_mean": 0.0033516068069729954,
|
| 1200 |
+
"epoch": 0.49375,
|
| 1201 |
+
"grad_norm": 1.6197171586589652,
|
| 1202 |
+
"kl": 0.012253022054210305,
|
| 1203 |
+
"learning_rate": 5.354398526154365e-07,
|
| 1204 |
+
"loss": 0.0005793888703919947,
|
| 1205 |
+
"step": 79,
|
| 1206 |
+
"step_time": 74.47651354596019
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"clip_ratio/high_max": 0.004948900153976865,
|
| 1210 |
+
"clip_ratio/high_mean": 0.0022943955009395722,
|
| 1211 |
+
"clip_ratio/low_mean": 0.002480681643646676,
|
| 1212 |
+
"clip_ratio/low_min": 0.0002420135569991544,
|
| 1213 |
+
"clip_ratio/region_mean": 0.004775077075464651,
|
| 1214 |
+
"epoch": 0.5,
|
| 1215 |
+
"grad_norm": 2.636476553387114,
|
| 1216 |
+
"kl": 0.009159992972854525,
|
| 1217 |
+
"learning_rate": 5.253245844193564e-07,
|
| 1218 |
+
"loss": 0.0004471801221370697,
|
| 1219 |
+
"step": 80,
|
| 1220 |
+
"step_time": 74.53143536299467
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"clip_ratio/high_max": 0.0,
|
| 1224 |
+
"clip_ratio/high_mean": 0.0,
|
| 1225 |
+
"clip_ratio/low_mean": 0.0,
|
| 1226 |
+
"clip_ratio/low_min": 0.0,
|
| 1227 |
+
"clip_ratio/region_mean": 0.0,
|
| 1228 |
+
"completions/clipped_ratio": 0.22265625,
|
| 1229 |
+
"completions/max_length": 2048.0,
|
| 1230 |
+
"completions/mean_length": 1113.5078125,
|
| 1231 |
+
"completions/min_length": 130.0,
|
| 1232 |
+
"epoch": 0.50625,
|
| 1233 |
+
"frac_reward_zero_std": 0.15625,
|
| 1234 |
+
"grad_norm": 2.437462663650732,
|
| 1235 |
+
"kl": 0.014007882680743933,
|
| 1236 |
+
"learning_rate": 5.151989130984714e-07,
|
| 1237 |
+
"loss": 0.0001402192428940907,
|
| 1238 |
+
"reward": 0.453125,
|
| 1239 |
+
"reward_std": 0.4154247045516968,
|
| 1240 |
+
"rewards/MajorityAccuracy/mean": 0.453125,
|
| 1241 |
+
"rewards/MajorityAccuracy/std": 0.4987730085849762,
|
| 1242 |
+
"step": 81,
|
| 1243 |
+
"step_time": 511.28541170433164
|
| 1244 |
+
},
|
| 1245 |
+
{
|
| 1246 |
+
"clip_ratio/high_max": 0.001999785687075928,
|
| 1247 |
+
"clip_ratio/high_mean": 0.000649039946438279,
|
| 1248 |
+
"clip_ratio/low_mean": 0.0004095527001481969,
|
| 1249 |
+
"clip_ratio/low_min": 0.0,
|
| 1250 |
+
"clip_ratio/region_mean": 0.001058592642948497,
|
| 1251 |
+
"epoch": 0.5125,
|
| 1252 |
+
"grad_norm": 2.767789706692989,
|
| 1253 |
+
"kl": 0.01138349308166653,
|
| 1254 |
+
"learning_rate": 5.050669981928055e-07,
|
| 1255 |
+
"loss": 0.00020510103786364198,
|
| 1256 |
+
"step": 82,
|
| 1257 |
+
"step_time": 72.86427146941423
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"clip_ratio/high_max": 0.0009888057466014288,
|
| 1261 |
+
"clip_ratio/high_mean": 0.00028390137777023483,
|
| 1262 |
+
"clip_ratio/low_mean": 0.00040138547046808526,
|
| 1263 |
+
"clip_ratio/low_min": 0.0,
|
| 1264 |
+
"clip_ratio/region_mean": 0.0006852868464193307,
|
| 1265 |
+
"epoch": 0.51875,
|
| 1266 |
+
"grad_norm": 1.2184109966911416,
|
| 1267 |
+
"kl": 0.00868189160246402,
|
| 1268 |
+
"learning_rate": 4.949330018071946e-07,
|
| 1269 |
+
"loss": -8.418422657996416e-05,
|
| 1270 |
+
"step": 83,
|
| 1271 |
+
"step_time": 74.83340260758996
|
| 1272 |
+
},
|
| 1273 |
+
{
|
| 1274 |
+
"clip_ratio/high_max": 0.002164612553315237,
|
| 1275 |
+
"clip_ratio/high_mean": 0.0008126823959173635,
|
| 1276 |
+
"clip_ratio/low_mean": 0.00097388479662186,
|
| 1277 |
+
"clip_ratio/low_min": 6.203474185895175e-05,
|
| 1278 |
+
"clip_ratio/region_mean": 0.0017865671834442765,
|
| 1279 |
+
"epoch": 0.525,
|
| 1280 |
+
"grad_norm": 2.995963014779342,
|
| 1281 |
+
"kl": 0.013743769726715982,
|
| 1282 |
+
"learning_rate": 4.848010869015287e-07,
|
| 1283 |
+
"loss": 0.00027044303715229034,
|
| 1284 |
+
"step": 84,
|
| 1285 |
+
"step_time": 73.61458361148834
|
| 1286 |
+
},
|
| 1287 |
+
{
|
| 1288 |
+
"clip_ratio/high_max": 0.0025311765930382535,
|
| 1289 |
+
"clip_ratio/high_mean": 0.0010570524136710446,
|
| 1290 |
+
"clip_ratio/low_mean": 0.0014499529788736254,
|
| 1291 |
+
"clip_ratio/low_min": 0.0,
|
| 1292 |
+
"clip_ratio/region_mean": 0.0025070053816307336,
|
| 1293 |
+
"epoch": 0.53125,
|
| 1294 |
+
"grad_norm": 2.321587028628179,
|
| 1295 |
+
"kl": 0.012635804247111082,
|
| 1296 |
+
"learning_rate": 4.746754155806437e-07,
|
| 1297 |
+
"loss": 0.00010993052273988724,
|
| 1298 |
+
"step": 85,
|
| 1299 |
+
"step_time": 73.99634441360831
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"clip_ratio/high_max": 0.0016447739180875942,
|
| 1303 |
+
"clip_ratio/high_mean": 0.0005325197707861662,
|
| 1304 |
+
"clip_ratio/low_mean": 0.0009895179537124932,
|
| 1305 |
+
"clip_ratio/low_min": 0.0,
|
| 1306 |
+
"clip_ratio/region_mean": 0.0015220377244986594,
|
| 1307 |
+
"epoch": 0.5375,
|
| 1308 |
+
"grad_norm": 2.0298588682797916,
|
| 1309 |
+
"kl": 0.013660596334375441,
|
| 1310 |
+
"learning_rate": 4.645601473845635e-07,
|
| 1311 |
+
"loss": 0.00010455396841280162,
|
| 1312 |
+
"step": 86,
|
| 1313 |
+
"step_time": 72.77361011505127
|
| 1314 |
+
},
|
| 1315 |
+
{
|
| 1316 |
+
"clip_ratio/high_max": 0.0038836976746097207,
|
| 1317 |
+
"clip_ratio/high_mean": 0.0011834766482934356,
|
| 1318 |
+
"clip_ratio/low_mean": 0.0017639883408264723,
|
| 1319 |
+
"clip_ratio/low_min": 0.0001220703125,
|
| 1320 |
+
"clip_ratio/region_mean": 0.0029474649927578866,
|
| 1321 |
+
"epoch": 0.54375,
|
| 1322 |
+
"grad_norm": 2.7290057438588664,
|
| 1323 |
+
"kl": 0.014111844822764397,
|
| 1324 |
+
"learning_rate": 4.544594375797968e-07,
|
| 1325 |
+
"loss": 8.19675624370575e-05,
|
| 1326 |
+
"step": 87,
|
| 1327 |
+
"step_time": 72.65440236777067
|
| 1328 |
+
},
|
| 1329 |
+
{
|
| 1330 |
+
"clip_ratio/high_max": 0.003319336748973001,
|
| 1331 |
+
"clip_ratio/high_mean": 0.0010932457098533632,
|
| 1332 |
+
"clip_ratio/low_mean": 0.000872844539117068,
|
| 1333 |
+
"clip_ratio/low_min": 0.0,
|
| 1334 |
+
"clip_ratio/region_mean": 0.001966090261703357,
|
| 1335 |
+
"epoch": 0.55,
|
| 1336 |
+
"grad_norm": 2.2571085135249995,
|
| 1337 |
+
"kl": 0.012561595300212502,
|
| 1338 |
+
"learning_rate": 4.443774354523882e-07,
|
| 1339 |
+
"loss": 9.269975998904556e-05,
|
| 1340 |
+
"step": 88,
|
| 1341 |
+
"step_time": 74.32562639191747
|
| 1342 |
+
},
|
| 1343 |
+
{
|
| 1344 |
+
"clip_ratio/high_max": 0.0,
|
| 1345 |
+
"clip_ratio/high_mean": 0.0,
|
| 1346 |
+
"clip_ratio/low_mean": 0.0,
|
| 1347 |
+
"clip_ratio/low_min": 0.0,
|
| 1348 |
+
"clip_ratio/region_mean": 0.0,
|
| 1349 |
+
"completions/clipped_ratio": 0.30078125,
|
| 1350 |
+
"completions/max_length": 2048.0,
|
| 1351 |
+
"completions/mean_length": 1118.33203125,
|
| 1352 |
+
"completions/min_length": 123.0,
|
| 1353 |
+
"epoch": 0.55625,
|
| 1354 |
+
"frac_reward_zero_std": 0.25,
|
| 1355 |
+
"grad_norm": 2.5098909460762617,
|
| 1356 |
+
"kl": 0.012415707809850574,
|
| 1357 |
+
"learning_rate": 4.3431828260342675e-07,
|
| 1358 |
+
"loss": 0.00012414250522851944,
|
| 1359 |
+
"reward": 0.41796875,
|
| 1360 |
+
"reward_std": 0.33903729915618896,
|
| 1361 |
+
"rewards/MajorityAccuracy/mean": 0.41796875,
|
| 1362 |
+
"rewards/MajorityAccuracy/std": 0.49419113993644714,
|
| 1363 |
+
"step": 89,
|
| 1364 |
+
"step_time": 509.8758964575827
|
| 1365 |
+
},
|
| 1366 |
+
{
|
| 1367 |
+
"clip_ratio/high_max": 0.000945723251788877,
|
| 1368 |
+
"clip_ratio/high_mean": 0.0003046227247978095,
|
| 1369 |
+
"clip_ratio/low_mean": 0.00043227803689660504,
|
| 1370 |
+
"clip_ratio/low_min": 0.0,
|
| 1371 |
+
"clip_ratio/region_mean": 0.0007369007325905841,
|
| 1372 |
+
"epoch": 0.5625,
|
| 1373 |
+
"grad_norm": 3.0602220969880003,
|
| 1374 |
+
"kl": 0.01795373868662864,
|
| 1375 |
+
"learning_rate": 4.242861112477118e-07,
|
| 1376 |
+
"loss": 0.00043129269033670425,
|
| 1377 |
+
"step": 90,
|
| 1378 |
+
"step_time": 72.513826623559
|
| 1379 |
+
},
|
| 1380 |
+
{
|
| 1381 |
+
"clip_ratio/high_max": 0.0005165289039723575,
|
| 1382 |
+
"clip_ratio/high_mean": 0.00012913222599308938,
|
| 1383 |
+
"clip_ratio/low_mean": 0.00078046911949059,
|
| 1384 |
+
"clip_ratio/low_min": 0.0,
|
| 1385 |
+
"clip_ratio/region_mean": 0.0009096013454836793,
|
| 1386 |
+
"epoch": 0.56875,
|
| 1387 |
+
"grad_norm": 4.00461301096923,
|
| 1388 |
+
"kl": 0.014442738029174507,
|
| 1389 |
+
"learning_rate": 4.1428504251627325e-07,
|
| 1390 |
+
"loss": 0.00046476908028125763,
|
| 1391 |
+
"step": 91,
|
| 1392 |
+
"step_time": 73.02188640832901
|
| 1393 |
+
},
|
| 1394 |
+
{
|
| 1395 |
+
"clip_ratio/high_max": 0.0030830453615635633,
|
| 1396 |
+
"clip_ratio/high_mean": 0.0011105355297331698,
|
| 1397 |
+
"clip_ratio/low_mean": 0.0008949626571848057,
|
| 1398 |
+
"clip_ratio/low_min": 0.00010254306835122406,
|
| 1399 |
+
"clip_ratio/region_mean": 0.0020054981869179755,
|
| 1400 |
+
"epoch": 0.575,
|
| 1401 |
+
"grad_norm": 3.7477755161462447,
|
| 1402 |
+
"kl": 0.015328466892242432,
|
| 1403 |
+
"learning_rate": 4.0431918476344685e-07,
|
| 1404 |
+
"loss": 0.0002448493614792824,
|
| 1405 |
+
"step": 92,
|
| 1406 |
+
"step_time": 73.59993440657854
|
| 1407 |
+
},
|
| 1408 |
+
{
|
| 1409 |
+
"clip_ratio/high_max": 0.003404190152650699,
|
| 1410 |
+
"clip_ratio/high_mean": 0.0010485929597052746,
|
| 1411 |
+
"clip_ratio/low_mean": 0.0012907164891657885,
|
| 1412 |
+
"clip_ratio/low_min": 0.0,
|
| 1413 |
+
"clip_ratio/region_mean": 0.0023393094452330843,
|
| 1414 |
+
"epoch": 0.58125,
|
| 1415 |
+
"grad_norm": 4.860452621936543,
|
| 1416 |
+
"kl": 0.012504537939094007,
|
| 1417 |
+
"learning_rate": 3.943926318791963e-07,
|
| 1418 |
+
"loss": 0.0006725070998072624,
|
| 1419 |
+
"step": 93,
|
| 1420 |
+
"step_time": 73.5823459289968
|
| 1421 |
+
},
|
| 1422 |
+
{
|
| 1423 |
+
"clip_ratio/high_max": 0.001166247355286032,
|
| 1424 |
+
"clip_ratio/high_mean": 0.00036392219772096723,
|
| 1425 |
+
"clip_ratio/low_mean": 0.0003926616773242131,
|
| 1426 |
+
"clip_ratio/low_min": 0.0,
|
| 1427 |
+
"clip_ratio/region_mean": 0.0007565838823211379,
|
| 1428 |
+
"epoch": 0.5875,
|
| 1429 |
+
"grad_norm": 1.0193150417745596,
|
| 1430 |
+
"kl": 0.009306605556048453,
|
| 1431 |
+
"learning_rate": 3.845094616073783e-07,
|
| 1432 |
+
"loss": 0.00020167470211163163,
|
| 1433 |
+
"step": 94,
|
| 1434 |
+
"step_time": 74.49038667604327
|
| 1435 |
+
},
|
| 1436 |
+
{
|
| 1437 |
+
"clip_ratio/high_max": 0.00282343928120099,
|
| 1438 |
+
"clip_ratio/high_mean": 0.0009135525251622312,
|
| 1439 |
+
"clip_ratio/low_mean": 0.001389655277307611,
|
| 1440 |
+
"clip_ratio/low_min": 0.0,
|
| 1441 |
+
"clip_ratio/region_mean": 0.002303207787917927,
|
| 1442 |
+
"epoch": 0.59375,
|
| 1443 |
+
"grad_norm": 2.9234852881480013,
|
| 1444 |
+
"kl": 0.010838062036782503,
|
| 1445 |
+
"learning_rate": 3.7467373387063964e-07,
|
| 1446 |
+
"loss": 9.109335951507092e-05,
|
| 1447 |
+
"step": 95,
|
| 1448 |
+
"step_time": 75.17482404783368
|
| 1449 |
+
},
|
| 1450 |
+
{
|
| 1451 |
+
"clip_ratio/high_max": 0.0001479478960391134,
|
| 1452 |
+
"clip_ratio/high_mean": 3.698697400977835e-05,
|
| 1453 |
+
"clip_ratio/low_mean": 0.0004987526044715196,
|
| 1454 |
+
"clip_ratio/low_min": 0.0,
|
| 1455 |
+
"clip_ratio/region_mean": 0.0005357395784812979,
|
| 1456 |
+
"epoch": 0.6,
|
| 1457 |
+
"grad_norm": 0.7315824993706517,
|
| 1458 |
+
"kl": 0.015260143205523491,
|
| 1459 |
+
"learning_rate": 3.648894891026358e-07,
|
| 1460 |
+
"loss": 0.000274527701549232,
|
| 1461 |
+
"step": 96,
|
| 1462 |
+
"step_time": 72.9783686362207
|
| 1463 |
+
},
|
| 1464 |
+
{
|
| 1465 |
+
"clip_ratio/high_max": 0.0,
|
| 1466 |
+
"clip_ratio/high_mean": 0.0,
|
| 1467 |
+
"clip_ratio/low_mean": 0.0,
|
| 1468 |
+
"clip_ratio/low_min": 0.0,
|
| 1469 |
+
"clip_ratio/region_mean": 0.0,
|
| 1470 |
+
"completions/clipped_ratio": 0.3515625,
|
| 1471 |
+
"completions/max_length": 2048.0,
|
| 1472 |
+
"completions/mean_length": 1331.6484375,
|
| 1473 |
+
"completions/min_length": 157.0,
|
| 1474 |
+
"epoch": 0.60625,
|
| 1475 |
+
"frac_reward_zero_std": 0.3125,
|
| 1476 |
+
"grad_norm": 2.024233579986159,
|
| 1477 |
+
"kl": 0.013446276308968663,
|
| 1478 |
+
"learning_rate": 3.5516074658825397e-07,
|
| 1479 |
+
"loss": 0.00013451337872538716,
|
| 1480 |
+
"reward": 0.37890625,
|
| 1481 |
+
"reward_std": 0.31865012645721436,
|
| 1482 |
+
"rewards/MajorityAccuracy/mean": 0.37890625,
|
| 1483 |
+
"rewards/MajorityAccuracy/std": 0.48606500029563904,
|
| 1484 |
+
"step": 97,
|
| 1485 |
+
"step_time": 513.4524273760617
|
| 1486 |
+
},
|
| 1487 |
+
{
|
| 1488 |
+
"clip_ratio/high_max": 0.0037205696426099166,
|
| 1489 |
+
"clip_ratio/high_mean": 0.0014697431288368534,
|
| 1490 |
+
"clip_ratio/low_mean": 0.00015838705257920083,
|
| 1491 |
+
"clip_ratio/low_min": 0.0,
|
| 1492 |
+
"clip_ratio/region_mean": 0.001628130185054033,
|
| 1493 |
+
"epoch": 0.6125,
|
| 1494 |
+
"grad_norm": 3.90698635616872,
|
| 1495 |
+
"kl": 0.01821469678543508,
|
| 1496 |
+
"learning_rate": 3.454915028125263e-07,
|
| 1497 |
+
"loss": 5.963072180747986e-05,
|
| 1498 |
+
"step": 98,
|
| 1499 |
+
"step_time": 74.3652068786323
|
| 1500 |
+
},
|
| 1501 |
+
{
|
| 1502 |
+
"clip_ratio/high_max": 0.0004335000048740767,
|
| 1503 |
+
"clip_ratio/high_mean": 0.00010837500121851917,
|
| 1504 |
+
"clip_ratio/low_mean": 6.103515625e-05,
|
| 1505 |
+
"clip_ratio/low_min": 0.0,
|
| 1506 |
+
"clip_ratio/region_mean": 0.00016941015746851917,
|
| 1507 |
+
"epoch": 0.61875,
|
| 1508 |
+
"grad_norm": 0.66679563224598,
|
| 1509 |
+
"kl": 0.008730458910577,
|
| 1510 |
+
"learning_rate": 3.3588572981890684e-07,
|
| 1511 |
+
"loss": 0.00011227793584112078,
|
| 1512 |
+
"step": 99,
|
| 1513 |
+
"step_time": 74.78353586420417
|
| 1514 |
+
},
|
| 1515 |
+
{
|
| 1516 |
+
"clip_ratio/high_max": 0.003887103513989132,
|
| 1517 |
+
"clip_ratio/high_mean": 0.001058211068084347,
|
| 1518 |
+
"clip_ratio/low_mean": 0.00043674557309714146,
|
| 1519 |
+
"clip_ratio/low_min": 0.0,
|
| 1520 |
+
"clip_ratio/region_mean": 0.0014949566229915945,
|
| 1521 |
+
"epoch": 0.625,
|
| 1522 |
+
"grad_norm": 2.103064145200142,
|
| 1523 |
+
"kl": 0.011321694357320666,
|
| 1524 |
+
"learning_rate": 3.263473735775899e-07,
|
| 1525 |
+
"loss": 0.0004523511743173003,
|
| 1526 |
+
"step": 100,
|
| 1527 |
+
"step_time": 73.49169047549367
|
| 1528 |
+
},
|
| 1529 |
+
{
|
| 1530 |
+
"clip_ratio/high_max": 0.0038711015731678344,
|
| 1531 |
+
"clip_ratio/high_mean": 0.0013209661865403177,
|
| 1532 |
+
"clip_ratio/low_mean": 0.0010835692228283733,
|
| 1533 |
+
"clip_ratio/low_min": 0.0,
|
| 1534 |
+
"clip_ratio/region_mean": 0.0024045353929977864,
|
| 1535 |
+
"epoch": 0.63125,
|
| 1536 |
+
"grad_norm": 2.9656880096450142,
|
| 1537 |
+
"kl": 0.017157551483251154,
|
| 1538 |
+
"learning_rate": 3.1688035236453865e-07,
|
| 1539 |
+
"loss": 0.00024260208010673523,
|
| 1540 |
+
"step": 101,
|
| 1541 |
+
"step_time": 73.03655776008964
|
| 1542 |
+
},
|
| 1543 |
+
{
|
| 1544 |
+
"clip_ratio/high_max": 0.004879164946032688,
|
| 1545 |
+
"clip_ratio/high_mean": 0.001343672840448562,
|
| 1546 |
+
"clip_ratio/low_mean": 0.0007238960533868521,
|
| 1547 |
+
"clip_ratio/low_min": 0.0,
|
| 1548 |
+
"clip_ratio/region_mean": 0.002067568901111372,
|
| 1549 |
+
"epoch": 0.6375,
|
| 1550 |
+
"grad_norm": 2.176730841688745,
|
| 1551 |
+
"kl": 0.013211242505349219,
|
| 1552 |
+
"learning_rate": 3.0748855515189096e-07,
|
| 1553 |
+
"loss": 6.035657861502841e-05,
|
| 1554 |
+
"step": 102,
|
| 1555 |
+
"step_time": 74.43861296400428
|
| 1556 |
+
},
|
| 1557 |
+
{
|
| 1558 |
+
"clip_ratio/high_max": 0.002294147401698865,
|
| 1559 |
+
"clip_ratio/high_mean": 0.0008955950834206305,
|
| 1560 |
+
"clip_ratio/low_mean": 0.0009236409168806858,
|
| 1561 |
+
"clip_ratio/low_min": 0.0,
|
| 1562 |
+
"clip_ratio/region_mean": 0.001819235971197486,
|
| 1563 |
+
"epoch": 0.64375,
|
| 1564 |
+
"grad_norm": 2.0139660282956715,
|
| 1565 |
+
"kl": 0.012527223327197134,
|
| 1566 |
+
"learning_rate": 2.981758400104028e-07,
|
| 1567 |
+
"loss": 0.0001851143315434456,
|
| 1568 |
+
"step": 103,
|
| 1569 |
+
"step_time": 72.99568805098534
|
| 1570 |
+
},
|
| 1571 |
+
{
|
| 1572 |
+
"clip_ratio/high_max": 0.0013545988185796887,
|
| 1573 |
+
"clip_ratio/high_mean": 0.00033864970464492217,
|
| 1574 |
+
"clip_ratio/low_mean": 6.103515625e-05,
|
| 1575 |
+
"clip_ratio/low_min": 0.0,
|
| 1576 |
+
"clip_ratio/region_mean": 0.00039968486089492217,
|
| 1577 |
+
"epoch": 0.65,
|
| 1578 |
+
"grad_norm": 0.7188046121067608,
|
| 1579 |
+
"kl": 0.010590940713882446,
|
| 1580 |
+
"learning_rate": 2.8894603252458403e-07,
|
| 1581 |
+
"loss": 0.00014627663767896593,
|
| 1582 |
+
"step": 104,
|
| 1583 |
+
"step_time": 74.32656337320805
|
| 1584 |
+
},
|
| 1585 |
+
{
|
| 1586 |
+
"clip_ratio/high_max": 0.0,
|
| 1587 |
+
"clip_ratio/high_mean": 0.0,
|
| 1588 |
+
"clip_ratio/low_mean": 0.0,
|
| 1589 |
+
"clip_ratio/low_min": 0.0,
|
| 1590 |
+
"clip_ratio/region_mean": 0.0,
|
| 1591 |
+
"completions/clipped_ratio": 0.30078125,
|
| 1592 |
+
"completions/max_length": 2048.0,
|
| 1593 |
+
"completions/mean_length": 1185.44140625,
|
| 1594 |
+
"completions/min_length": 101.0,
|
| 1595 |
+
"epoch": 0.65625,
|
| 1596 |
+
"frac_reward_zero_std": 0.28125,
|
| 1597 |
+
"grad_norm": 0.030039258155558534,
|
| 1598 |
+
"kl": 0.010324922681320459,
|
| 1599 |
+
"learning_rate": 2.7980292422118277e-07,
|
| 1600 |
+
"loss": 0.00010326277697458863,
|
| 1601 |
+
"reward": 0.46875,
|
| 1602 |
+
"reward_std": 0.3256346583366394,
|
| 1603 |
+
"rewards/MajorityAccuracy/mean": 0.46875,
|
| 1604 |
+
"rewards/MajorityAccuracy/std": 0.5,
|
| 1605 |
+
"step": 105,
|
| 1606 |
+
"step_time": 508.91598969325423
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"clip_ratio/high_max": 0.0023039314037305303,
|
| 1610 |
+
"clip_ratio/high_mean": 0.0007846390526538016,
|
| 1611 |
+
"clip_ratio/low_mean": 0.000358324236003682,
|
| 1612 |
+
"clip_ratio/low_min": 0.0,
|
| 1613 |
+
"clip_ratio/region_mean": 0.0011429633104853565,
|
| 1614 |
+
"epoch": 0.6625,
|
| 1615 |
+
"grad_norm": 3.035318376116158,
|
| 1616 |
+
"kl": 0.01831530430354178,
|
| 1617 |
+
"learning_rate": 2.70750271011657e-07,
|
| 1618 |
+
"loss": 0.0005411822348833084,
|
| 1619 |
+
"step": 106,
|
| 1620 |
+
"step_time": 72.49610098823905
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"clip_ratio/high_max": 0.0032874929602257907,
|
| 1624 |
+
"clip_ratio/high_mean": 0.0008765876373217907,
|
| 1625 |
+
"clip_ratio/low_mean": 0.00032476869819220155,
|
| 1626 |
+
"clip_ratio/low_min": 0.0,
|
| 1627 |
+
"clip_ratio/region_mean": 0.001201356339151971,
|
| 1628 |
+
"epoch": 0.66875,
|
| 1629 |
+
"grad_norm": 1.8648576894442905,
|
| 1630 |
+
"kl": 0.010542559495661408,
|
| 1631 |
+
"learning_rate": 2.6179179164927754e-07,
|
| 1632 |
+
"loss": 5.412474274635315e-05,
|
| 1633 |
+
"step": 107,
|
| 1634 |
+
"step_time": 73.30012211948633
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"clip_ratio/high_max": 0.0024597935553174466,
|
| 1638 |
+
"clip_ratio/high_mean": 0.0008725106708880048,
|
| 1639 |
+
"clip_ratio/low_mean": 0.0006123567291069776,
|
| 1640 |
+
"clip_ratio/low_min": 0.0,
|
| 1641 |
+
"clip_ratio/region_mean": 0.0014848673963570036,
|
| 1642 |
+
"epoch": 0.675,
|
| 1643 |
+
"grad_norm": 1.9015087942204416,
|
| 1644 |
+
"kl": 0.011379874078556895,
|
| 1645 |
+
"learning_rate": 2.529311662014972e-07,
|
| 1646 |
+
"loss": 0.0004289001226425171,
|
| 1647 |
+
"step": 108,
|
| 1648 |
+
"step_time": 74.5721581466496
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"clip_ratio/high_max": 0.0016439486207673326,
|
| 1652 |
+
"clip_ratio/high_mean": 0.0005178136234462727,
|
| 1653 |
+
"clip_ratio/low_mean": 0.00027134107585879974,
|
| 1654 |
+
"clip_ratio/low_min": 0.0,
|
| 1655 |
+
"clip_ratio/region_mean": 0.0007891546956670936,
|
| 1656 |
+
"epoch": 0.68125,
|
| 1657 |
+
"grad_norm": 2.0548652473561475,
|
| 1658 |
+
"kl": 0.011981348041445017,
|
| 1659 |
+
"learning_rate": 2.441720345382089e-07,
|
| 1660 |
+
"loss": 0.0002480585826560855,
|
| 1661 |
+
"step": 109,
|
| 1662 |
+
"step_time": 74.8028103671968
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"clip_ratio/high_max": 0.0020322695054346696,
|
| 1666 |
+
"clip_ratio/high_mean": 0.0006936919562576804,
|
| 1667 |
+
"clip_ratio/low_mean": 0.00044720154073729645,
|
| 1668 |
+
"clip_ratio/low_min": 0.0,
|
| 1669 |
+
"clip_ratio/region_mean": 0.0011408934969949769,
|
| 1670 |
+
"epoch": 0.6875,
|
| 1671 |
+
"grad_norm": 2.670185284791766,
|
| 1672 |
+
"kl": 0.015695515787228942,
|
| 1673 |
+
"learning_rate": 2.355179948365189e-07,
|
| 1674 |
+
"loss": 0.00015004145097918808,
|
| 1675 |
+
"step": 110,
|
| 1676 |
+
"step_time": 74.68548646569252
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"clip_ratio/high_max": 0.005265811036224477,
|
| 1680 |
+
"clip_ratio/high_mean": 0.0016542720368306618,
|
| 1681 |
+
"clip_ratio/low_mean": 0.0008743348698772024,
|
| 1682 |
+
"clip_ratio/low_min": 0.0,
|
| 1683 |
+
"clip_ratio/region_mean": 0.0025286069430876523,
|
| 1684 |
+
"epoch": 0.69375,
|
| 1685 |
+
"grad_norm": 2.7006331230690375,
|
| 1686 |
+
"kl": 0.02251513279043138,
|
| 1687 |
+
"learning_rate": 2.26972602102645e-07,
|
| 1688 |
+
"loss": -7.399119203910232e-05,
|
| 1689 |
+
"step": 111,
|
| 1690 |
+
"step_time": 72.76664720848203
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"clip_ratio/high_max": 0.0022779267164878547,
|
| 1694 |
+
"clip_ratio/high_mean": 0.0007306839033844881,
|
| 1695 |
+
"clip_ratio/low_mean": 0.0006510534331027884,
|
| 1696 |
+
"clip_ratio/low_min": 0.0,
|
| 1697 |
+
"clip_ratio/region_mean": 0.0013817373182973824,
|
| 1698 |
+
"epoch": 0.7,
|
| 1699 |
+
"grad_norm": 2.9074213472999855,
|
| 1700 |
+
"kl": 0.014699326362460852,
|
| 1701 |
+
"learning_rate": 2.1853936671155127e-07,
|
| 1702 |
+
"loss": 0.00039001123514026403,
|
| 1703 |
+
"step": 112,
|
| 1704 |
+
"step_time": 72.97640189900994
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"clip_ratio/high_max": 0.0,
|
| 1708 |
+
"clip_ratio/high_mean": 0.0,
|
| 1709 |
+
"clip_ratio/low_mean": 0.0,
|
| 1710 |
+
"clip_ratio/low_min": 0.0,
|
| 1711 |
+
"clip_ratio/region_mean": 0.0,
|
| 1712 |
+
"completions/clipped_ratio": 0.265625,
|
| 1713 |
+
"completions/max_length": 2048.0,
|
| 1714 |
+
"completions/mean_length": 1209.81640625,
|
| 1715 |
+
"completions/min_length": 116.0,
|
| 1716 |
+
"epoch": 0.70625,
|
| 1717 |
+
"frac_reward_zero_std": 0.28125,
|
| 1718 |
+
"grad_norm": 3.1117411176679775,
|
| 1719 |
+
"kl": 0.017801032168790698,
|
| 1720 |
+
"learning_rate": 2.1022175296491512e-07,
|
| 1721 |
+
"loss": 0.0001779504818841815,
|
| 1722 |
+
"reward": 0.46875,
|
| 1723 |
+
"reward_std": 0.3325856924057007,
|
| 1724 |
+
"rewards/MajorityAccuracy/mean": 0.46875,
|
| 1725 |
+
"rewards/MajorityAccuracy/std": 0.5,
|
| 1726 |
+
"step": 113,
|
| 1727 |
+
"step_time": 513.3052420243621
|
| 1728 |
+
},
|
| 1729 |
+
{
|
| 1730 |
+
"clip_ratio/high_max": 0.002949203480966389,
|
| 1731 |
+
"clip_ratio/high_mean": 0.0010433234492666088,
|
| 1732 |
+
"clip_ratio/low_mean": 0.0002535246112529421,
|
| 1733 |
+
"clip_ratio/low_min": 0.0,
|
| 1734 |
+
"clip_ratio/region_mean": 0.001296848051424604,
|
| 1735 |
+
"epoch": 0.7125,
|
| 1736 |
+
"grad_norm": 1.8837997787971237,
|
| 1737 |
+
"kl": 0.014758818899281323,
|
| 1738 |
+
"learning_rate": 2.0202317766802152e-07,
|
| 1739 |
+
"loss": 1.360886381007731e-05,
|
| 1740 |
+
"step": 114,
|
| 1741 |
+
"step_time": 72.32398896291852
|
| 1742 |
+
},
|
| 1743 |
+
{
|
| 1744 |
+
"clip_ratio/high_max": 0.0,
|
| 1745 |
+
"clip_ratio/high_mean": 0.0,
|
| 1746 |
+
"clip_ratio/low_mean": 0.0002162299606425222,
|
| 1747 |
+
"clip_ratio/low_min": 0.0,
|
| 1748 |
+
"clip_ratio/region_mean": 0.0002162299606425222,
|
| 1749 |
+
"epoch": 0.71875,
|
| 1750 |
+
"grad_norm": 1.384166856356004,
|
| 1751 |
+
"kl": 0.012360319844447076,
|
| 1752 |
+
"learning_rate": 1.9394700872616853e-07,
|
| 1753 |
+
"loss": 9.207347466144711e-05,
|
| 1754 |
+
"step": 115,
|
| 1755 |
+
"step_time": 73.95661000907421
|
| 1756 |
+
},
|
| 1757 |
+
{
|
| 1758 |
+
"clip_ratio/high_max": 0.0019464775687083602,
|
| 1759 |
+
"clip_ratio/high_mean": 0.0005213416297920048,
|
| 1760 |
+
"clip_ratio/low_mean": 0.0002997436131408904,
|
| 1761 |
+
"clip_ratio/low_min": 0.0,
|
| 1762 |
+
"clip_ratio/region_mean": 0.0008210852429328952,
|
| 1763 |
+
"epoch": 0.725,
|
| 1764 |
+
"grad_norm": 2.657221387982295,
|
| 1765 |
+
"kl": 0.013543186127208173,
|
| 1766 |
+
"learning_rate": 1.8599656376116024e-07,
|
| 1767 |
+
"loss": 0.0001850096305133775,
|
| 1768 |
+
"step": 116,
|
| 1769 |
+
"step_time": 74.85563575103879
|
| 1770 |
+
},
|
| 1771 |
+
{
|
| 1772 |
+
"clip_ratio/high_max": 0.0013928585685789585,
|
| 1773 |
+
"clip_ratio/high_mean": 0.000492317027237732,
|
| 1774 |
+
"clip_ratio/low_mean": 0.0003164000445394777,
|
| 1775 |
+
"clip_ratio/low_min": 0.0,
|
| 1776 |
+
"clip_ratio/region_mean": 0.0008087170572252944,
|
| 1777 |
+
"epoch": 0.73125,
|
| 1778 |
+
"grad_norm": 1.5113361828799334,
|
| 1779 |
+
"kl": 0.010296113265212625,
|
| 1780 |
+
"learning_rate": 1.7817510874845582e-07,
|
| 1781 |
+
"loss": -1.3626180589199066e-05,
|
| 1782 |
+
"step": 117,
|
| 1783 |
+
"step_time": 74.4402880743146
|
| 1784 |
+
},
|
| 1785 |
+
{
|
| 1786 |
+
"clip_ratio/high_max": 0.0027039795022574253,
|
| 1787 |
+
"clip_ratio/high_mean": 0.0009460828096052865,
|
| 1788 |
+
"clip_ratio/low_mean": 0.0005931829655310139,
|
| 1789 |
+
"clip_ratio/low_min": 0.0,
|
| 1790 |
+
"clip_ratio/region_mean": 0.0015392657878692262,
|
| 1791 |
+
"epoch": 0.7375,
|
| 1792 |
+
"grad_norm": 1.7104097606769009,
|
| 1793 |
+
"kl": 0.0108446863014251,
|
| 1794 |
+
"learning_rate": 1.7048585667553412e-07,
|
| 1795 |
+
"loss": 0.0002545565366744995,
|
| 1796 |
+
"step": 118,
|
| 1797 |
+
"step_time": 74.31127118319273
|
| 1798 |
+
},
|
| 1799 |
+
{
|
| 1800 |
+
"clip_ratio/high_max": 0.001953940351086203,
|
| 1801 |
+
"clip_ratio/high_mean": 0.000624574911853415,
|
| 1802 |
+
"clip_ratio/low_mean": 0.00035788821878668386,
|
| 1803 |
+
"clip_ratio/low_min": 0.0,
|
| 1804 |
+
"clip_ratio/region_mean": 0.0009824631342780776,
|
| 1805 |
+
"epoch": 0.74375,
|
| 1806 |
+
"grad_norm": 1.564614502738393,
|
| 1807 |
+
"kl": 0.010188602842390537,
|
| 1808 |
+
"learning_rate": 1.6293196622202632e-07,
|
| 1809 |
+
"loss": -4.070153954671696e-05,
|
| 1810 |
+
"step": 119,
|
| 1811 |
+
"step_time": 73.05638172104955
|
| 1812 |
+
},
|
| 1813 |
+
{
|
| 1814 |
+
"clip_ratio/high_max": 0.0025597490603104234,
|
| 1815 |
+
"clip_ratio/high_mean": 0.0008101043822534848,
|
| 1816 |
+
"clip_ratio/low_mean": 0.0006440655415644869,
|
| 1817 |
+
"clip_ratio/low_min": 0.0,
|
| 1818 |
+
"clip_ratio/region_mean": 0.0014541698910761625,
|
| 1819 |
+
"epoch": 0.75,
|
| 1820 |
+
"grad_norm": 3.0216204551653973,
|
| 1821 |
+
"kl": 0.013856111210770905,
|
| 1822 |
+
"learning_rate": 1.555165404621567e-07,
|
| 1823 |
+
"loss": 0.00041135604260489345,
|
| 1824 |
+
"step": 120,
|
| 1825 |
+
"step_time": 73.1470375135541
|
| 1826 |
+
},
|
| 1827 |
+
{
|
| 1828 |
+
"clip_ratio/high_max": 0.0,
|
| 1829 |
+
"clip_ratio/high_mean": 0.0,
|
| 1830 |
+
"clip_ratio/low_mean": 0.0,
|
| 1831 |
+
"clip_ratio/low_min": 0.0,
|
| 1832 |
+
"clip_ratio/region_mean": 0.0,
|
| 1833 |
+
"completions/clipped_ratio": 0.1796875,
|
| 1834 |
+
"completions/max_length": 2048.0,
|
| 1835 |
+
"completions/mean_length": 970.953125,
|
| 1836 |
+
"completions/min_length": 109.0,
|
| 1837 |
+
"epoch": 0.75625,
|
| 1838 |
+
"frac_reward_zero_std": 0.25,
|
| 1839 |
+
"grad_norm": 38.94070228477994,
|
| 1840 |
+
"kl": 0.017914578900672495,
|
| 1841 |
+
"learning_rate": 1.4824262559002592e-07,
|
| 1842 |
+
"loss": 0.00017904024571180344,
|
| 1843 |
+
"reward": 0.4140625,
|
| 1844 |
+
"reward_std": 0.3612491488456726,
|
| 1845 |
+
"rewards/MajorityAccuracy/mean": 0.4140625,
|
| 1846 |
+
"rewards/MajorityAccuracy/std": 0.4935242235660553,
|
| 1847 |
+
"step": 121,
|
| 1848 |
+
"step_time": 502.9487722180784
|
| 1849 |
+
},
|
| 1850 |
+
{
|
| 1851 |
+
"clip_ratio/high_max": 0.0021715334441978484,
|
| 1852 |
+
"clip_ratio/high_mean": 0.0005428833610494621,
|
| 1853 |
+
"clip_ratio/low_mean": 0.00026226834597764537,
|
| 1854 |
+
"clip_ratio/low_min": 0.0,
|
| 1855 |
+
"clip_ratio/region_mean": 0.0008051517070271075,
|
| 1856 |
+
"epoch": 0.7625,
|
| 1857 |
+
"grad_norm": 2.5629091792435483,
|
| 1858 |
+
"kl": 0.016575733199715614,
|
| 1859 |
+
"learning_rate": 1.4111320966826057e-07,
|
| 1860 |
+
"loss": 0.0006199295748956501,
|
| 1861 |
+
"step": 122,
|
| 1862 |
+
"step_time": 73.21493119373918
|
| 1863 |
+
},
|
| 1864 |
+
{
|
| 1865 |
+
"clip_ratio/high_max": 0.0019036461962969042,
|
| 1866 |
+
"clip_ratio/high_mean": 0.0005358659163903212,
|
| 1867 |
+
"clip_ratio/low_mean": 0.00042437368210812565,
|
| 1868 |
+
"clip_ratio/low_min": 0.0,
|
| 1869 |
+
"clip_ratio/region_mean": 0.0009602396075933939,
|
| 1870 |
+
"epoch": 0.76875,
|
| 1871 |
+
"grad_norm": 2.616146844713034,
|
| 1872 |
+
"kl": 0.017179179936647415,
|
| 1873 |
+
"learning_rate": 1.3413122140054217e-07,
|
| 1874 |
+
"loss": 0.00029046961572021246,
|
| 1875 |
+
"step": 123,
|
| 1876 |
+
"step_time": 74.28767842054367
|
| 1877 |
+
},
|
| 1878 |
+
{
|
| 1879 |
+
"clip_ratio/high_max": 0.0023136833624448627,
|
| 1880 |
+
"clip_ratio/high_mean": 0.0007207895832834765,
|
| 1881 |
+
"clip_ratio/low_mean": 0.00032264593392028473,
|
| 1882 |
+
"clip_ratio/low_min": 0.0,
|
| 1883 |
+
"clip_ratio/region_mean": 0.0010434355062898248,
|
| 1884 |
+
"epoch": 0.775,
|
| 1885 |
+
"grad_norm": 4.6867956068349255,
|
| 1886 |
+
"kl": 0.014364082831889391,
|
| 1887 |
+
"learning_rate": 1.272995289285202e-07,
|
| 1888 |
+
"loss": 0.0003849201602861285,
|
| 1889 |
+
"step": 124,
|
| 1890 |
+
"step_time": 73.07671456411481
|
| 1891 |
+
},
|
| 1892 |
+
{
|
| 1893 |
+
"clip_ratio/high_max": 0.0034383015299681574,
|
| 1894 |
+
"clip_ratio/high_mean": 0.001082753216905985,
|
| 1895 |
+
"clip_ratio/low_mean": 0.0012956284772371873,
|
| 1896 |
+
"clip_ratio/low_min": 0.0,
|
| 1897 |
+
"clip_ratio/region_mean": 0.0023783816723152995,
|
| 1898 |
+
"epoch": 0.78125,
|
| 1899 |
+
"grad_norm": 1.9455995566980075,
|
| 1900 |
+
"kl": 0.01443166087847203,
|
| 1901 |
+
"learning_rate": 1.2062093865360457e-07,
|
| 1902 |
+
"loss": -0.0001663451548665762,
|
| 1903 |
+
"step": 125,
|
| 1904 |
+
"step_time": 72.86572542786598
|
| 1905 |
+
},
|
| 1906 |
+
{
|
| 1907 |
+
"clip_ratio/high_max": 0.003039627663383726,
|
| 1908 |
+
"clip_ratio/high_mean": 0.0009150033820333192,
|
| 1909 |
+
"clip_ratio/low_mean": 0.00044116072422184516,
|
| 1910 |
+
"clip_ratio/low_min": 0.0,
|
| 1911 |
+
"clip_ratio/region_mean": 0.0013561641098931432,
|
| 1912 |
+
"epoch": 0.7875,
|
| 1913 |
+
"grad_norm": 3.4687569402971024,
|
| 1914 |
+
"kl": 0.01432532467879355,
|
| 1915 |
+
"learning_rate": 1.1409819408411897e-07,
|
| 1916 |
+
"loss": 0.0003951581893488765,
|
| 1917 |
+
"step": 126,
|
| 1918 |
+
"step_time": 71.27972013130784
|
| 1919 |
+
},
|
| 1920 |
+
{
|
| 1921 |
+
"clip_ratio/high_max": 0.0051903360144933686,
|
| 1922 |
+
"clip_ratio/high_mean": 0.0017125617123383563,
|
| 1923 |
+
"clip_ratio/low_mean": 0.000602532072662143,
|
| 1924 |
+
"clip_ratio/low_min": 0.0,
|
| 1925 |
+
"clip_ratio/region_mean": 0.00231509375225869,
|
| 1926 |
+
"epoch": 0.79375,
|
| 1927 |
+
"grad_norm": 3.4850639858905423,
|
| 1928 |
+
"kl": 0.016185807180590928,
|
| 1929 |
+
"learning_rate": 1.0773397470829143e-07,
|
| 1930 |
+
"loss": 0.000527041032910347,
|
| 1931 |
+
"step": 127,
|
| 1932 |
+
"step_time": 72.36988684162498
|
| 1933 |
+
},
|
| 1934 |
+
{
|
| 1935 |
+
"clip_ratio/high_max": 0.002350392773223575,
|
| 1936 |
+
"clip_ratio/high_mean": 0.0006940717776160454,
|
| 1937 |
+
"clip_ratio/low_mean": 0.0005309218540787697,
|
| 1938 |
+
"clip_ratio/low_min": 0.0,
|
| 1939 |
+
"clip_ratio/region_mean": 0.0012249936335138045,
|
| 1940 |
+
"epoch": 0.8,
|
| 1941 |
+
"grad_norm": 1.349741695675087,
|
| 1942 |
+
"kl": 0.014131805626675487,
|
| 1943 |
+
"learning_rate": 1.0153089489354256e-07,
|
| 1944 |
+
"loss": 0.00014612237282563,
|
| 1945 |
+
"step": 128,
|
| 1946 |
+
"step_time": 75.02361815050244
|
| 1947 |
+
},
|
| 1948 |
+
{
|
| 1949 |
+
"clip_ratio/high_max": 0.0,
|
| 1950 |
+
"clip_ratio/high_mean": 0.0,
|
| 1951 |
+
"clip_ratio/low_mean": 0.0,
|
| 1952 |
+
"clip_ratio/low_min": 0.0,
|
| 1953 |
+
"clip_ratio/region_mean": 0.0,
|
| 1954 |
+
"completions/clipped_ratio": 0.234375,
|
| 1955 |
+
"completions/max_length": 2048.0,
|
| 1956 |
+
"completions/mean_length": 1048.62890625,
|
| 1957 |
+
"completions/min_length": 92.0,
|
| 1958 |
+
"epoch": 0.80625,
|
| 1959 |
+
"frac_reward_zero_std": 0.28125,
|
| 1960 |
+
"grad_norm": 1.5853295216787153,
|
| 1961 |
+
"kl": 0.01426891703158617,
|
| 1962 |
+
"learning_rate": 9.549150281252632e-08,
|
| 1963 |
+
"loss": 0.00014276121510192752,
|
| 1964 |
+
"reward": 0.4921875,
|
| 1965 |
+
"reward_std": 0.3320576846599579,
|
| 1966 |
+
"rewards/MajorityAccuracy/mean": 0.4921875,
|
| 1967 |
+
"rewards/MajorityAccuracy/std": 0.5009182691574097,
|
| 1968 |
+
"step": 129,
|
| 1969 |
+
"step_time": 509.16119331121445
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"clip_ratio/high_max": 0.0020524649880826473,
|
| 1973 |
+
"clip_ratio/high_mean": 0.0007414130013785325,
|
| 1974 |
+
"clip_ratio/low_mean": 0.00019824528135359287,
|
| 1975 |
+
"clip_ratio/low_min": 0.0,
|
| 1976 |
+
"clip_ratio/region_mean": 0.0009396582754561678,
|
| 1977 |
+
"epoch": 0.8125,
|
| 1978 |
+
"grad_norm": 1.9216472838743954,
|
| 1979 |
+
"kl": 0.009388849255628884,
|
| 1980 |
+
"learning_rate": 8.961827939636196e-08,
|
| 1981 |
+
"loss": 0.00019847185467369854,
|
| 1982 |
+
"step": 130,
|
| 1983 |
+
"step_time": 75.15510265529156
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"clip_ratio/high_max": 0.0028523287328425795,
|
| 1987 |
+
"clip_ratio/high_mean": 0.0008350911521119997,
|
| 1988 |
+
"clip_ratio/low_mean": 0.0002642755425767973,
|
| 1989 |
+
"clip_ratio/low_min": 0.0,
|
| 1990 |
+
"clip_ratio/region_mean": 0.0010993667092407122,
|
| 1991 |
+
"epoch": 0.81875,
|
| 1992 |
+
"grad_norm": 2.20180164534035,
|
| 1993 |
+
"kl": 0.015062682214193046,
|
| 1994 |
+
"learning_rate": 8.391363731548811e-08,
|
| 1995 |
+
"loss": 0.0004961859667673707,
|
| 1996 |
+
"step": 131,
|
| 1997 |
+
"step_time": 73.51469936594367
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"clip_ratio/high_max": 0.0020305332727730274,
|
| 2001 |
+
"clip_ratio/high_mean": 0.0007918234041426331,
|
| 2002 |
+
"clip_ratio/low_mean": 0.0011652825269266032,
|
| 2003 |
+
"clip_ratio/low_min": 0.0,
|
| 2004 |
+
"clip_ratio/region_mean": 0.0019571059237932786,
|
| 2005 |
+
"epoch": 0.825,
|
| 2006 |
+
"grad_norm": 2.1800682497492576,
|
| 2007 |
+
"kl": 0.01607281109318137,
|
| 2008 |
+
"learning_rate": 7.837991998855897e-08,
|
| 2009 |
+
"loss": 9.050378866959363e-05,
|
| 2010 |
+
"step": 132,
|
| 2011 |
+
"step_time": 73.16071617603302
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"clip_ratio/high_max": 0.005413196398876607,
|
| 2015 |
+
"clip_ratio/high_mean": 0.0018023425218416378,
|
| 2016 |
+
"clip_ratio/low_mean": 0.0005819578873342834,
|
| 2017 |
+
"clip_ratio/low_min": 0.0,
|
| 2018 |
+
"clip_ratio/region_mean": 0.0023843004018999636,
|
| 2019 |
+
"epoch": 0.83125,
|
| 2020 |
+
"grad_norm": 4.349667744810731,
|
| 2021 |
+
"kl": 0.020439567044377327,
|
| 2022 |
+
"learning_rate": 7.301940061978722e-08,
|
| 2023 |
+
"loss": 0.00031285732984542847,
|
| 2024 |
+
"step": 133,
|
| 2025 |
+
"step_time": 71.77879596501589
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"clip_ratio/high_max": 0.0016331294318661094,
|
| 2029 |
+
"clip_ratio/high_mean": 0.00045781125299981795,
|
| 2030 |
+
"clip_ratio/low_mean": 0.000129511648992775,
|
| 2031 |
+
"clip_ratio/low_min": 0.0,
|
| 2032 |
+
"clip_ratio/region_mean": 0.000587322901992593,
|
| 2033 |
+
"epoch": 0.8375,
|
| 2034 |
+
"grad_norm": 2.593288163912129,
|
| 2035 |
+
"kl": 0.014495007460936904,
|
| 2036 |
+
"learning_rate": 6.783428126513125e-08,
|
| 2037 |
+
"loss": 0.00011580558202695101,
|
| 2038 |
+
"step": 134,
|
| 2039 |
+
"step_time": 73.13445693999529
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"clip_ratio/high_max": 0.003605847123253625,
|
| 2043 |
+
"clip_ratio/high_mean": 0.001148435148934368,
|
| 2044 |
+
"clip_ratio/low_mean": 0.0003829055222013267,
|
| 2045 |
+
"clip_ratio/low_min": 0.0,
|
| 2046 |
+
"clip_ratio/region_mean": 0.0015313406620407477,
|
| 2047 |
+
"epoch": 0.84375,
|
| 2048 |
+
"grad_norm": 2.7202426195918052,
|
| 2049 |
+
"kl": 0.017668991815298796,
|
| 2050 |
+
"learning_rate": 6.282669192770895e-08,
|
| 2051 |
+
"loss": 0.00017234589904546738,
|
| 2052 |
+
"step": 135,
|
| 2053 |
+
"step_time": 73.37359900027514
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"clip_ratio/high_max": 0.00037202381645329297,
|
| 2057 |
+
"clip_ratio/high_mean": 9.300595411332324e-05,
|
| 2058 |
+
"clip_ratio/low_mean": 0.0001558603544253856,
|
| 2059 |
+
"clip_ratio/low_min": 0.0,
|
| 2060 |
+
"clip_ratio/region_mean": 0.00024886630853870884,
|
| 2061 |
+
"epoch": 0.85,
|
| 2062 |
+
"grad_norm": 1.2976238454060678,
|
| 2063 |
+
"kl": 0.01629375573247671,
|
| 2064 |
+
"learning_rate": 5.799868968281074e-08,
|
| 2065 |
+
"loss": 0.0002391697489656508,
|
| 2066 |
+
"step": 136,
|
| 2067 |
+
"step_time": 72.94468743354082
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"clip_ratio/high_max": 0.0,
|
| 2071 |
+
"clip_ratio/high_mean": 0.0,
|
| 2072 |
+
"clip_ratio/low_mean": 0.0,
|
| 2073 |
+
"clip_ratio/low_min": 0.0,
|
| 2074 |
+
"clip_ratio/region_mean": 0.0,
|
| 2075 |
+
"completions/clipped_ratio": 0.3359375,
|
| 2076 |
+
"completions/max_length": 2048.0,
|
| 2077 |
+
"completions/mean_length": 1186.95703125,
|
| 2078 |
+
"completions/min_length": 144.0,
|
| 2079 |
+
"epoch": 0.85625,
|
| 2080 |
+
"frac_reward_zero_std": 0.40625,
|
| 2081 |
+
"grad_norm": 1.4196860907660045,
|
| 2082 |
+
"kl": 0.012532974360510707,
|
| 2083 |
+
"learning_rate": 5.335225783287051e-08,
|
| 2084 |
+
"loss": 0.00012530735693871975,
|
| 2085 |
+
"reward": 0.40625,
|
| 2086 |
+
"reward_std": 0.28945231437683105,
|
| 2087 |
+
"rewards/MajorityAccuracy/mean": 0.40625,
|
| 2088 |
+
"rewards/MajorityAccuracy/std": 0.49209436774253845,
|
| 2089 |
+
"step": 137,
|
| 2090 |
+
"step_time": 511.77231634780765
|
| 2091 |
+
},
|
| 2092 |
+
{
|
| 2093 |
+
"clip_ratio/high_max": 0.003232624221709557,
|
| 2094 |
+
"clip_ratio/high_mean": 0.0014515863040287513,
|
| 2095 |
+
"clip_ratio/low_mean": 0.0006515702116303146,
|
| 2096 |
+
"clip_ratio/low_min": 0.0,
|
| 2097 |
+
"clip_ratio/region_mean": 0.002103156512021087,
|
| 2098 |
+
"epoch": 0.8625,
|
| 2099 |
+
"grad_norm": 2.5760998543824085,
|
| 2100 |
+
"kl": 0.015228840988129377,
|
| 2101 |
+
"learning_rate": 4.888930509274125e-08,
|
| 2102 |
+
"loss": 0.0002501225098967552,
|
| 2103 |
+
"step": 138,
|
| 2104 |
+
"step_time": 73.33046870678663
|
| 2105 |
+
},
|
| 2106 |
+
{
|
| 2107 |
+
"clip_ratio/high_max": 0.0024041714932536706,
|
| 2108 |
+
"clip_ratio/high_mean": 0.000671188779961085,
|
| 2109 |
+
"clip_ratio/low_mean": 0.000273559689958347,
|
| 2110 |
+
"clip_ratio/low_min": 0.0,
|
| 2111 |
+
"clip_ratio/region_mean": 0.000944748469919432,
|
| 2112 |
+
"epoch": 0.86875,
|
| 2113 |
+
"grad_norm": 1.5009003799479834,
|
| 2114 |
+
"kl": 0.009767578565515578,
|
| 2115 |
+
"learning_rate": 4.4611664805611794e-08,
|
| 2116 |
+
"loss": 0.00012098962906748056,
|
| 2117 |
+
"step": 139,
|
| 2118 |
+
"step_time": 74.82619997113943
|
| 2119 |
+
},
|
| 2120 |
+
{
|
| 2121 |
+
"clip_ratio/high_max": 0.0017365278545184992,
|
| 2122 |
+
"clip_ratio/high_mean": 0.0004745588576042792,
|
| 2123 |
+
"clip_ratio/low_mean": 0.0005758852203143761,
|
| 2124 |
+
"clip_ratio/low_min": 0.0,
|
| 2125 |
+
"clip_ratio/region_mean": 0.001050444072461687,
|
| 2126 |
+
"epoch": 0.875,
|
| 2127 |
+
"grad_norm": 3.1106898869292103,
|
| 2128 |
+
"kl": 0.01543115742970258,
|
| 2129 |
+
"learning_rate": 4.0521094189884696e-08,
|
| 2130 |
+
"loss": 0.00011425917909946293,
|
| 2131 |
+
"step": 140,
|
| 2132 |
+
"step_time": 72.8097133897245
|
| 2133 |
+
},
|
| 2134 |
+
{
|
| 2135 |
+
"clip_ratio/high_max": 0.0038706468549207784,
|
| 2136 |
+
"clip_ratio/high_mean": 0.0009676617137301946,
|
| 2137 |
+
"clip_ratio/low_mean": 9.469935866945889e-05,
|
| 2138 |
+
"clip_ratio/low_min": 0.0,
|
| 2139 |
+
"clip_ratio/region_mean": 0.001062361070580664,
|
| 2140 |
+
"epoch": 0.88125,
|
| 2141 |
+
"grad_norm": 2.3504809175045516,
|
| 2142 |
+
"kl": 0.01317990489769727,
|
| 2143 |
+
"learning_rate": 3.6619273617325695e-08,
|
| 2144 |
+
"loss": 0.00022322710719890893,
|
| 2145 |
+
"step": 141,
|
| 2146 |
+
"step_time": 75.31241586431861
|
| 2147 |
+
},
|
| 2148 |
+
{
|
| 2149 |
+
"clip_ratio/high_max": 0.0,
|
| 2150 |
+
"clip_ratio/high_mean": 0.0,
|
| 2151 |
+
"clip_ratio/low_mean": 3.591954009607434e-05,
|
| 2152 |
+
"clip_ratio/low_min": 0.0,
|
| 2153 |
+
"clip_ratio/region_mean": 3.591954009607434e-05,
|
| 2154 |
+
"epoch": 0.8875,
|
| 2155 |
+
"grad_norm": 0.9663943016958322,
|
| 2156 |
+
"kl": 0.009216298290994018,
|
| 2157 |
+
"learning_rate": 3.2907805922781476e-08,
|
| 2158 |
+
"loss": 0.00021239795023575425,
|
| 2159 |
+
"step": 142,
|
| 2160 |
+
"step_time": 73.2460791580379
|
| 2161 |
+
},
|
| 2162 |
+
{
|
| 2163 |
+
"clip_ratio/high_max": 0.00747315157786943,
|
| 2164 |
+
"clip_ratio/high_mean": 0.002556122220994439,
|
| 2165 |
+
"clip_ratio/low_mean": 0.0005522011379071046,
|
| 2166 |
+
"clip_ratio/low_min": 0.0,
|
| 2167 |
+
"clip_ratio/region_mean": 0.0031083233552635647,
|
| 2168 |
+
"epoch": 0.89375,
|
| 2169 |
+
"grad_norm": 4.566105993811372,
|
| 2170 |
+
"kl": 0.01367004297208041,
|
| 2171 |
+
"learning_rate": 2.9388215745748345e-08,
|
| 2172 |
+
"loss": 0.0001438697800040245,
|
| 2173 |
+
"step": 143,
|
| 2174 |
+
"step_time": 72.51130413636565
|
| 2175 |
+
},
|
| 2176 |
+
{
|
| 2177 |
+
"clip_ratio/high_max": 0.0009799722174648196,
|
| 2178 |
+
"clip_ratio/high_mean": 0.0003056726709473878,
|
| 2179 |
+
"clip_ratio/low_mean": 0.00024855698393366765,
|
| 2180 |
+
"clip_ratio/low_min": 0.0,
|
| 2181 |
+
"clip_ratio/region_mean": 0.0005542296548810555,
|
| 2182 |
+
"epoch": 0.9,
|
| 2183 |
+
"grad_norm": 2.4491252901487113,
|
| 2184 |
+
"kl": 0.014086044044233859,
|
| 2185 |
+
"learning_rate": 2.6061948904063658e-08,
|
| 2186 |
+
"loss": 0.00017750589177012444,
|
| 2187 |
+
"step": 144,
|
| 2188 |
+
"step_time": 72.8478989340365
|
| 2189 |
+
},
|
| 2190 |
+
{
|
| 2191 |
+
"clip_ratio/high_max": 0.0,
|
| 2192 |
+
"clip_ratio/high_mean": 0.0,
|
| 2193 |
+
"clip_ratio/low_mean": 0.0,
|
| 2194 |
+
"clip_ratio/low_min": 0.0,
|
| 2195 |
+
"clip_ratio/region_mean": 0.0,
|
| 2196 |
+
"completions/clipped_ratio": 0.27734375,
|
| 2197 |
+
"completions/max_length": 2048.0,
|
| 2198 |
+
"completions/mean_length": 1169.87890625,
|
| 2199 |
+
"completions/min_length": 85.0,
|
| 2200 |
+
"epoch": 0.90625,
|
| 2201 |
+
"frac_reward_zero_std": 0.15625,
|
| 2202 |
+
"grad_norm": 3.823051307838312,
|
| 2203 |
+
"kl": 0.01587809680495411,
|
| 2204 |
+
"learning_rate": 2.293037179997559e-08,
|
| 2205 |
+
"loss": 0.00015884824097156525,
|
| 2206 |
+
"reward": 0.4453125,
|
| 2207 |
+
"reward_std": 0.3813767433166504,
|
| 2208 |
+
"rewards/MajorityAccuracy/mean": 0.4453125,
|
| 2209 |
+
"rewards/MajorityAccuracy/std": 0.49797382950782776,
|
| 2210 |
+
"step": 145,
|
| 2211 |
+
"step_time": 514.3241759650409
|
| 2212 |
+
},
|
| 2213 |
+
{
|
| 2214 |
+
"clip_ratio/high_max": 0.002048870373982936,
|
| 2215 |
+
"clip_ratio/high_mean": 0.0005604429170489311,
|
| 2216 |
+
"clip_ratio/low_mean": 0.0003576864692149684,
|
| 2217 |
+
"clip_ratio/low_min": 0.0,
|
| 2218 |
+
"clip_ratio/region_mean": 0.0009181293862638995,
|
| 2219 |
+
"epoch": 0.9125,
|
| 2220 |
+
"grad_norm": 5.455060604679547,
|
| 2221 |
+
"kl": 0.01314276794437319,
|
| 2222 |
+
"learning_rate": 1.9994770858837107e-08,
|
| 2223 |
+
"loss": -3.9240305341081694e-05,
|
| 2224 |
+
"step": 146,
|
| 2225 |
+
"step_time": 74.53192065656185
|
| 2226 |
+
},
|
| 2227 |
+
{
|
| 2228 |
+
"clip_ratio/high_max": 0.001343083604297135,
|
| 2229 |
+
"clip_ratio/high_mean": 0.00036511363032332156,
|
| 2230 |
+
"clip_ratio/low_mean": 0.00037709655225626193,
|
| 2231 |
+
"clip_ratio/low_min": 0.0,
|
| 2232 |
+
"clip_ratio/region_mean": 0.0007422101825795835,
|
| 2233 |
+
"epoch": 0.91875,
|
| 2234 |
+
"grad_norm": 1.9542783991100507,
|
| 2235 |
+
"kl": 0.015989506617188454,
|
| 2236 |
+
"learning_rate": 1.725635200065323e-08,
|
| 2237 |
+
"loss": 0.00013546661648433656,
|
| 2238 |
+
"step": 147,
|
| 2239 |
+
"step_time": 73.53811445459723
|
| 2240 |
+
},
|
| 2241 |
+
{
|
| 2242 |
+
"clip_ratio/high_max": 0.0018197018944192678,
|
| 2243 |
+
"clip_ratio/high_mean": 0.0006235582695808262,
|
| 2244 |
+
"clip_ratio/low_mean": 0.000335693359375,
|
| 2245 |
+
"clip_ratio/low_min": 0.0,
|
| 2246 |
+
"clip_ratio/region_mean": 0.0009592516289558262,
|
| 2247 |
+
"epoch": 0.925,
|
| 2248 |
+
"grad_norm": 1.8546197857489226,
|
| 2249 |
+
"kl": 0.00966759049333632,
|
| 2250 |
+
"learning_rate": 1.4716240144699187e-08,
|
| 2251 |
+
"loss": 0.00013379215670283884,
|
| 2252 |
+
"step": 148,
|
| 2253 |
+
"step_time": 75.19388054311275
|
| 2254 |
+
},
|
| 2255 |
+
{
|
| 2256 |
+
"clip_ratio/high_max": 0.005560761725064367,
|
| 2257 |
+
"clip_ratio/high_mean": 0.0015944942497299053,
|
| 2258 |
+
"clip_ratio/low_mean": 0.0005555712050409056,
|
| 2259 |
+
"clip_ratio/low_min": 0.0,
|
| 2260 |
+
"clip_ratio/region_mean": 0.0021500654765986837,
|
| 2261 |
+
"epoch": 0.93125,
|
| 2262 |
+
"grad_norm": 1.8350333864018886,
|
| 2263 |
+
"kl": 0.01854721351992339,
|
| 2264 |
+
"learning_rate": 1.2375478747413015e-08,
|
| 2265 |
+
"loss": 2.697875606827438e-05,
|
| 2266 |
+
"step": 149,
|
| 2267 |
+
"step_time": 74.73031372204423
|
| 2268 |
+
},
|
| 2269 |
+
{
|
| 2270 |
+
"clip_ratio/high_max": 0.002603469547466375,
|
| 2271 |
+
"clip_ratio/high_mean": 0.0008127552318910602,
|
| 2272 |
+
"clip_ratio/low_mean": 0.0002840022098098416,
|
| 2273 |
+
"clip_ratio/low_min": 0.0,
|
| 2274 |
+
"clip_ratio/region_mean": 0.0010967574453388806,
|
| 2275 |
+
"epoch": 0.9375,
|
| 2276 |
+
"grad_norm": 1.7966231376921085,
|
| 2277 |
+
"kl": 0.013860831037163734,
|
| 2278 |
+
"learning_rate": 1.0235029373752757e-08,
|
| 2279 |
+
"loss": 0.0002549358760006726,
|
| 2280 |
+
"step": 150,
|
| 2281 |
+
"step_time": 73.12263548374176
|
| 2282 |
+
},
|
| 2283 |
+
{
|
| 2284 |
+
"clip_ratio/high_max": 0.004017262748675421,
|
| 2285 |
+
"clip_ratio/high_mean": 0.0011635359842330217,
|
| 2286 |
+
"clip_ratio/low_mean": 0.0008665119603392668,
|
| 2287 |
+
"clip_ratio/low_min": 0.0,
|
| 2288 |
+
"clip_ratio/region_mean": 0.002030047937296331,
|
| 2289 |
+
"epoch": 0.94375,
|
| 2290 |
+
"grad_norm": 4.755826819340509,
|
| 2291 |
+
"kl": 0.016803148202598095,
|
| 2292 |
+
"learning_rate": 8.295771302193721e-09,
|
| 2293 |
+
"loss": 0.00037591904401779175,
|
| 2294 |
+
"step": 151,
|
| 2295 |
+
"step_time": 74.43394039943814
|
| 2296 |
+
},
|
| 2297 |
+
{
|
| 2298 |
+
"clip_ratio/high_max": 0.001781867365934886,
|
| 2299 |
+
"clip_ratio/high_mean": 0.0004454668414837215,
|
| 2300 |
+
"clip_ratio/low_mean": 0.0005764926390838809,
|
| 2301 |
+
"clip_ratio/low_min": 0.0,
|
| 2302 |
+
"clip_ratio/region_mean": 0.001021959469653666,
|
| 2303 |
+
"epoch": 0.95,
|
| 2304 |
+
"grad_norm": 2.9105586185245715,
|
| 2305 |
+
"kl": 0.01546309795230627,
|
| 2306 |
+
"learning_rate": 6.558501163527963e-09,
|
| 2307 |
+
"loss": 0.00036095641553401947,
|
| 2308 |
+
"step": 152,
|
| 2309 |
+
"step_time": 72.89345175772905
|
| 2310 |
+
},
|
| 2311 |
+
{
|
| 2312 |
+
"clip_ratio/high_max": 0.0,
|
| 2313 |
+
"clip_ratio/high_mean": 0.0,
|
| 2314 |
+
"clip_ratio/low_mean": 0.0,
|
| 2315 |
+
"clip_ratio/low_min": 0.0,
|
| 2316 |
+
"clip_ratio/region_mean": 0.0,
|
| 2317 |
+
"completions/clipped_ratio": 0.2578125,
|
| 2318 |
+
"completions/max_length": 2048.0,
|
| 2319 |
+
"completions/mean_length": 1079.86328125,
|
| 2320 |
+
"completions/min_length": 127.0,
|
| 2321 |
+
"epoch": 0.95625,
|
| 2322 |
+
"frac_reward_zero_std": 0.3125,
|
| 2323 |
+
"grad_norm": 1.6899080056845546,
|
| 2324 |
+
"kl": 0.013396909344010055,
|
| 2325 |
+
"learning_rate": 5.023932613615445e-09,
|
| 2326 |
+
"loss": 0.0001339974842267111,
|
| 2327 |
+
"reward": 0.4375,
|
| 2328 |
+
"reward_std": 0.29078546166419983,
|
| 2329 |
+
"rewards/MajorityAccuracy/mean": 0.4375,
|
| 2330 |
+
"rewards/MajorityAccuracy/std": 0.49705013632774353,
|
| 2331 |
+
"step": 153,
|
| 2332 |
+
"step_time": 509.2396182715893
|
| 2333 |
+
},
|
| 2334 |
+
{
|
| 2335 |
+
"clip_ratio/high_max": 0.0013671347260242328,
|
| 2336 |
+
"clip_ratio/high_mean": 0.0005312471294018906,
|
| 2337 |
+
"clip_ratio/low_mean": 0.00033049228295567445,
|
| 2338 |
+
"clip_ratio/low_min": 0.0,
|
| 2339 |
+
"clip_ratio/region_mean": 0.0008617394159955438,
|
| 2340 |
+
"epoch": 0.9625,
|
| 2341 |
+
"grad_norm": 1.758910204231182,
|
| 2342 |
+
"kl": 0.011747804004698992,
|
| 2343 |
+
"learning_rate": 3.6926960402202666e-09,
|
| 2344 |
+
"loss": 7.152618491090834e-05,
|
| 2345 |
+
"step": 154,
|
| 2346 |
+
"step_time": 74.8022861033678
|
| 2347 |
+
},
|
| 2348 |
+
{
|
| 2349 |
+
"clip_ratio/high_max": 0.006100582308135927,
|
| 2350 |
+
"clip_ratio/high_mean": 0.0019987884370493703,
|
| 2351 |
+
"clip_ratio/low_mean": 0.001225780215463601,
|
| 2352 |
+
"clip_ratio/low_min": 0.0,
|
| 2353 |
+
"clip_ratio/region_mean": 0.0032245686452370137,
|
| 2354 |
+
"epoch": 0.96875,
|
| 2355 |
+
"grad_norm": 3.6084902321782413,
|
| 2356 |
+
"kl": 0.018516098614782095,
|
| 2357 |
+
"learning_rate": 2.5653383040524224e-09,
|
| 2358 |
+
"loss": 0.0004977874923497438,
|
| 2359 |
+
"step": 155,
|
| 2360 |
+
"step_time": 72.65597991645336
|
| 2361 |
+
},
|
| 2362 |
+
{
|
| 2363 |
+
"clip_ratio/high_max": 0.0028998178604524583,
|
| 2364 |
+
"clip_ratio/high_mean": 0.0007953770036692731,
|
| 2365 |
+
"clip_ratio/low_mean": 0.00025517528047203086,
|
| 2366 |
+
"clip_ratio/low_min": 0.0,
|
| 2367 |
+
"clip_ratio/region_mean": 0.001050552284141304,
|
| 2368 |
+
"epoch": 0.975,
|
| 2369 |
+
"grad_norm": 1.4212062368905087,
|
| 2370 |
+
"kl": 0.01714646548498422,
|
| 2371 |
+
"learning_rate": 1.642322514122385e-09,
|
| 2372 |
+
"loss": 0.0004000420740339905,
|
| 2373 |
+
"step": 156,
|
| 2374 |
+
"step_time": 74.3429843224585
|
| 2375 |
+
},
|
| 2376 |
+
{
|
| 2377 |
+
"clip_ratio/high_max": 0.001330314640654251,
|
| 2378 |
+
"clip_ratio/high_mean": 0.00039352496969513595,
|
| 2379 |
+
"clip_ratio/low_mean": 0.0005242356201051734,
|
| 2380 |
+
"clip_ratio/low_min": 0.0,
|
| 2381 |
+
"clip_ratio/region_mean": 0.000917760597076267,
|
| 2382 |
+
"epoch": 0.98125,
|
| 2383 |
+
"grad_norm": 1.9836929521847406,
|
| 2384 |
+
"kl": 0.01735136902425438,
|
| 2385 |
+
"learning_rate": 9.240278374995636e-10,
|
| 2386 |
+
"loss": 0.0004292658413760364,
|
| 2387 |
+
"step": 157,
|
| 2388 |
+
"step_time": 74.0317906215787
|
| 2389 |
+
},
|
| 2390 |
+
{
|
| 2391 |
+
"clip_ratio/high_max": 0.0048068553587654606,
|
| 2392 |
+
"clip_ratio/high_mean": 0.0013377792238316033,
|
| 2393 |
+
"clip_ratio/low_mean": 0.000706089282175526,
|
| 2394 |
+
"clip_ratio/low_min": 0.0,
|
| 2395 |
+
"clip_ratio/region_mean": 0.0020438685023691505,
|
| 2396 |
+
"epoch": 0.9875,
|
| 2397 |
+
"grad_norm": 2.229147523940593,
|
| 2398 |
+
"kl": 0.01734938146546483,
|
| 2399 |
+
"learning_rate": 4.1074934355384006e-10,
|
| 2400 |
+
"loss": -0.00023244787007570267,
|
| 2401 |
+
"step": 158,
|
| 2402 |
+
"step_time": 73.07154578343034
|
| 2403 |
+
},
|
| 2404 |
+
{
|
| 2405 |
+
"clip_ratio/high_max": 0.004428355488926172,
|
| 2406 |
+
"clip_ratio/high_mean": 0.0011960023184656166,
|
| 2407 |
+
"clip_ratio/low_mean": 0.0007353129367402289,
|
| 2408 |
+
"clip_ratio/low_min": 0.0,
|
| 2409 |
+
"clip_ratio/region_mean": 0.001931315284309676,
|
| 2410 |
+
"epoch": 0.99375,
|
| 2411 |
+
"grad_norm": 2.5915534771661872,
|
| 2412 |
+
"kl": 0.017426876234821975,
|
| 2413 |
+
"learning_rate": 1.0269788274358537e-10,
|
| 2414 |
+
"loss": 0.0001338826259598136,
|
| 2415 |
+
"step": 159,
|
| 2416 |
+
"step_time": 72.93239786848426
|
| 2417 |
+
},
|
| 2418 |
+
{
|
| 2419 |
+
"clip_ratio/high_max": 0.0008206523198168725,
|
| 2420 |
+
"clip_ratio/high_mean": 0.00020516307995421812,
|
| 2421 |
+
"clip_ratio/low_mean": 0.0001019929586618673,
|
| 2422 |
+
"clip_ratio/low_min": 0.0,
|
| 2423 |
+
"clip_ratio/region_mean": 0.0003071560349781066,
|
| 2424 |
+
"epoch": 1.0,
|
| 2425 |
+
"grad_norm": 1.96109937879624,
|
| 2426 |
+
"kl": 0.012856202200055122,
|
| 2427 |
+
"learning_rate": 0.0,
|
| 2428 |
+
"loss": 0.00017605732136871666,
|
| 2429 |
+
"step": 160,
|
| 2430 |
+
"step_time": 74.85312191024423
|
| 2431 |
+
}
|
| 2432 |
+
],
|
| 2433 |
+
"logging_steps": 1,
|
| 2434 |
+
"max_steps": 160,
|
| 2435 |
+
"num_input_tokens_seen": 0,
|
| 2436 |
+
"num_train_epochs": 1,
|
| 2437 |
+
"save_steps": 80,
|
| 2438 |
+
"stateful_callbacks": {
|
| 2439 |
+
"TrainerControl": {
|
| 2440 |
+
"args": {
|
| 2441 |
+
"should_epoch_stop": false,
|
| 2442 |
+
"should_evaluate": false,
|
| 2443 |
+
"should_log": false,
|
| 2444 |
+
"should_save": true,
|
| 2445 |
+
"should_training_stop": true
|
| 2446 |
+
},
|
| 2447 |
+
"attributes": {}
|
| 2448 |
+
}
|
| 2449 |
+
},
|
| 2450 |
+
"total_flos": 0.0,
|
| 2451 |
+
"train_batch_size": 1,
|
| 2452 |
+
"trial_name": null,
|
| 2453 |
+
"trial_params": null
|
| 2454 |
+
}
|
c2_uniform_difficulty/round_1/model.safetensors.index.json
ADDED
|
@@ -0,0 +1,768 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"metadata": {
|
| 3 |
+
"total_parameters": 1469680,
|
| 4 |
+
"total_size": 18819627488
|
| 5 |
+
},
|
| 6 |
+
"weight_map": {
|
| 7 |
+
"lm_head.weight": "model-00001-of-00004.safetensors",
|
| 8 |
+
"model.language_model.embed_tokens.weight": "model-00001-of-00004.safetensors",
|
| 9 |
+
"model.language_model.layers.0.input_layernorm.weight": "model-00001-of-00004.safetensors",
|
| 10 |
+
"model.language_model.layers.0.linear_attn.A_log": "model-00001-of-00004.safetensors",
|
| 11 |
+
"model.language_model.layers.0.linear_attn.conv1d.weight": "model-00001-of-00004.safetensors",
|
| 12 |
+
"model.language_model.layers.0.linear_attn.dt_bias": "model-00001-of-00004.safetensors",
|
| 13 |
+
"model.language_model.layers.0.linear_attn.in_proj_a.weight": "model-00001-of-00004.safetensors",
|
| 14 |
+
"model.language_model.layers.0.linear_attn.in_proj_b.weight": "model-00001-of-00004.safetensors",
|
| 15 |
+
"model.language_model.layers.0.linear_attn.in_proj_qkv.weight": "model-00001-of-00004.safetensors",
|
| 16 |
+
"model.language_model.layers.0.linear_attn.in_proj_z.weight": "model-00001-of-00004.safetensors",
|
| 17 |
+
"model.language_model.layers.0.linear_attn.norm.weight": "model-00001-of-00004.safetensors",
|
| 18 |
+
"model.language_model.layers.0.linear_attn.out_proj.weight": "model-00001-of-00004.safetensors",
|
| 19 |
+
"model.language_model.layers.0.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
|
| 20 |
+
"model.language_model.layers.0.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
|
| 21 |
+
"model.language_model.layers.0.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
|
| 22 |
+
"model.language_model.layers.0.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
|
| 23 |
+
"model.language_model.layers.1.input_layernorm.weight": "model-00001-of-00004.safetensors",
|
| 24 |
+
"model.language_model.layers.1.linear_attn.A_log": "model-00001-of-00004.safetensors",
|
| 25 |
+
"model.language_model.layers.1.linear_attn.conv1d.weight": "model-00001-of-00004.safetensors",
|
| 26 |
+
"model.language_model.layers.1.linear_attn.dt_bias": "model-00001-of-00004.safetensors",
|
| 27 |
+
"model.language_model.layers.1.linear_attn.in_proj_a.weight": "model-00001-of-00004.safetensors",
|
| 28 |
+
"model.language_model.layers.1.linear_attn.in_proj_b.weight": "model-00001-of-00004.safetensors",
|
| 29 |
+
"model.language_model.layers.1.linear_attn.in_proj_qkv.weight": "model-00001-of-00004.safetensors",
|
| 30 |
+
"model.language_model.layers.1.linear_attn.in_proj_z.weight": "model-00001-of-00004.safetensors",
|
| 31 |
+
"model.language_model.layers.1.linear_attn.norm.weight": "model-00001-of-00004.safetensors",
|
| 32 |
+
"model.language_model.layers.1.linear_attn.out_proj.weight": "model-00001-of-00004.safetensors",
|
| 33 |
+
"model.language_model.layers.1.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
|
| 34 |
+
"model.language_model.layers.1.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
|
| 35 |
+
"model.language_model.layers.1.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
|
| 36 |
+
"model.language_model.layers.1.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
|
| 37 |
+
"model.language_model.layers.10.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 38 |
+
"model.language_model.layers.10.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 39 |
+
"model.language_model.layers.10.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 40 |
+
"model.language_model.layers.10.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 41 |
+
"model.language_model.layers.10.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 42 |
+
"model.language_model.layers.10.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 43 |
+
"model.language_model.layers.10.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 44 |
+
"model.language_model.layers.10.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 45 |
+
"model.language_model.layers.10.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 46 |
+
"model.language_model.layers.10.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 47 |
+
"model.language_model.layers.10.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 48 |
+
"model.language_model.layers.10.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 49 |
+
"model.language_model.layers.10.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 50 |
+
"model.language_model.layers.10.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 51 |
+
"model.language_model.layers.11.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 52 |
+
"model.language_model.layers.11.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 53 |
+
"model.language_model.layers.11.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 54 |
+
"model.language_model.layers.11.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 55 |
+
"model.language_model.layers.11.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 56 |
+
"model.language_model.layers.11.self_attn.k_norm.weight": "model-00002-of-00004.safetensors",
|
| 57 |
+
"model.language_model.layers.11.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
|
| 58 |
+
"model.language_model.layers.11.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
|
| 59 |
+
"model.language_model.layers.11.self_attn.q_norm.weight": "model-00002-of-00004.safetensors",
|
| 60 |
+
"model.language_model.layers.11.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
|
| 61 |
+
"model.language_model.layers.11.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
|
| 62 |
+
"model.language_model.layers.12.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 63 |
+
"model.language_model.layers.12.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 64 |
+
"model.language_model.layers.12.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 65 |
+
"model.language_model.layers.12.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 66 |
+
"model.language_model.layers.12.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 67 |
+
"model.language_model.layers.12.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 68 |
+
"model.language_model.layers.12.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 69 |
+
"model.language_model.layers.12.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 70 |
+
"model.language_model.layers.12.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 71 |
+
"model.language_model.layers.12.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 72 |
+
"model.language_model.layers.12.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 73 |
+
"model.language_model.layers.12.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 74 |
+
"model.language_model.layers.12.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 75 |
+
"model.language_model.layers.12.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 76 |
+
"model.language_model.layers.13.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 77 |
+
"model.language_model.layers.13.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 78 |
+
"model.language_model.layers.13.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 79 |
+
"model.language_model.layers.13.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 80 |
+
"model.language_model.layers.13.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 81 |
+
"model.language_model.layers.13.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 82 |
+
"model.language_model.layers.13.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 83 |
+
"model.language_model.layers.13.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 84 |
+
"model.language_model.layers.13.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 85 |
+
"model.language_model.layers.13.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 86 |
+
"model.language_model.layers.13.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 87 |
+
"model.language_model.layers.13.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 88 |
+
"model.language_model.layers.13.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 89 |
+
"model.language_model.layers.13.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 90 |
+
"model.language_model.layers.14.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 91 |
+
"model.language_model.layers.14.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 92 |
+
"model.language_model.layers.14.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 93 |
+
"model.language_model.layers.14.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 94 |
+
"model.language_model.layers.14.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 95 |
+
"model.language_model.layers.14.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 96 |
+
"model.language_model.layers.14.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 97 |
+
"model.language_model.layers.14.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 98 |
+
"model.language_model.layers.14.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 99 |
+
"model.language_model.layers.14.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 100 |
+
"model.language_model.layers.14.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 101 |
+
"model.language_model.layers.14.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 102 |
+
"model.language_model.layers.14.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 103 |
+
"model.language_model.layers.14.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 104 |
+
"model.language_model.layers.15.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 105 |
+
"model.language_model.layers.15.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 106 |
+
"model.language_model.layers.15.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 107 |
+
"model.language_model.layers.15.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 108 |
+
"model.language_model.layers.15.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 109 |
+
"model.language_model.layers.15.self_attn.k_norm.weight": "model-00003-of-00004.safetensors",
|
| 110 |
+
"model.language_model.layers.15.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
|
| 111 |
+
"model.language_model.layers.15.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
|
| 112 |
+
"model.language_model.layers.15.self_attn.q_norm.weight": "model-00003-of-00004.safetensors",
|
| 113 |
+
"model.language_model.layers.15.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
|
| 114 |
+
"model.language_model.layers.15.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
|
| 115 |
+
"model.language_model.layers.16.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 116 |
+
"model.language_model.layers.16.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 117 |
+
"model.language_model.layers.16.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 118 |
+
"model.language_model.layers.16.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 119 |
+
"model.language_model.layers.16.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 120 |
+
"model.language_model.layers.16.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 121 |
+
"model.language_model.layers.16.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 122 |
+
"model.language_model.layers.16.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 123 |
+
"model.language_model.layers.16.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 124 |
+
"model.language_model.layers.16.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 125 |
+
"model.language_model.layers.16.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 126 |
+
"model.language_model.layers.16.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 127 |
+
"model.language_model.layers.16.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 128 |
+
"model.language_model.layers.16.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 129 |
+
"model.language_model.layers.17.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 130 |
+
"model.language_model.layers.17.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 131 |
+
"model.language_model.layers.17.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 132 |
+
"model.language_model.layers.17.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 133 |
+
"model.language_model.layers.17.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 134 |
+
"model.language_model.layers.17.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 135 |
+
"model.language_model.layers.17.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 136 |
+
"model.language_model.layers.17.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 137 |
+
"model.language_model.layers.17.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 138 |
+
"model.language_model.layers.17.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 139 |
+
"model.language_model.layers.17.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 140 |
+
"model.language_model.layers.17.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 141 |
+
"model.language_model.layers.17.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 142 |
+
"model.language_model.layers.17.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 143 |
+
"model.language_model.layers.18.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 144 |
+
"model.language_model.layers.18.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 145 |
+
"model.language_model.layers.18.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 146 |
+
"model.language_model.layers.18.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 147 |
+
"model.language_model.layers.18.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 148 |
+
"model.language_model.layers.18.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 149 |
+
"model.language_model.layers.18.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 150 |
+
"model.language_model.layers.18.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 151 |
+
"model.language_model.layers.18.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 152 |
+
"model.language_model.layers.18.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 153 |
+
"model.language_model.layers.18.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 154 |
+
"model.language_model.layers.18.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 155 |
+
"model.language_model.layers.18.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 156 |
+
"model.language_model.layers.18.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 157 |
+
"model.language_model.layers.19.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 158 |
+
"model.language_model.layers.19.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 159 |
+
"model.language_model.layers.19.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 160 |
+
"model.language_model.layers.19.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 161 |
+
"model.language_model.layers.19.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 162 |
+
"model.language_model.layers.19.self_attn.k_norm.weight": "model-00003-of-00004.safetensors",
|
| 163 |
+
"model.language_model.layers.19.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
|
| 164 |
+
"model.language_model.layers.19.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
|
| 165 |
+
"model.language_model.layers.19.self_attn.q_norm.weight": "model-00003-of-00004.safetensors",
|
| 166 |
+
"model.language_model.layers.19.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
|
| 167 |
+
"model.language_model.layers.19.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
|
| 168 |
+
"model.language_model.layers.2.input_layernorm.weight": "model-00001-of-00004.safetensors",
|
| 169 |
+
"model.language_model.layers.2.linear_attn.A_log": "model-00001-of-00004.safetensors",
|
| 170 |
+
"model.language_model.layers.2.linear_attn.conv1d.weight": "model-00001-of-00004.safetensors",
|
| 171 |
+
"model.language_model.layers.2.linear_attn.dt_bias": "model-00001-of-00004.safetensors",
|
| 172 |
+
"model.language_model.layers.2.linear_attn.in_proj_a.weight": "model-00001-of-00004.safetensors",
|
| 173 |
+
"model.language_model.layers.2.linear_attn.in_proj_b.weight": "model-00001-of-00004.safetensors",
|
| 174 |
+
"model.language_model.layers.2.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 175 |
+
"model.language_model.layers.2.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 176 |
+
"model.language_model.layers.2.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 177 |
+
"model.language_model.layers.2.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 178 |
+
"model.language_model.layers.2.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 179 |
+
"model.language_model.layers.2.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 180 |
+
"model.language_model.layers.2.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 181 |
+
"model.language_model.layers.2.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 182 |
+
"model.language_model.layers.20.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 183 |
+
"model.language_model.layers.20.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 184 |
+
"model.language_model.layers.20.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 185 |
+
"model.language_model.layers.20.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 186 |
+
"model.language_model.layers.20.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 187 |
+
"model.language_model.layers.20.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 188 |
+
"model.language_model.layers.20.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 189 |
+
"model.language_model.layers.20.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 190 |
+
"model.language_model.layers.20.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 191 |
+
"model.language_model.layers.20.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 192 |
+
"model.language_model.layers.20.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 193 |
+
"model.language_model.layers.20.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 194 |
+
"model.language_model.layers.20.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 195 |
+
"model.language_model.layers.20.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 196 |
+
"model.language_model.layers.21.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 197 |
+
"model.language_model.layers.21.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 198 |
+
"model.language_model.layers.21.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 199 |
+
"model.language_model.layers.21.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 200 |
+
"model.language_model.layers.21.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 201 |
+
"model.language_model.layers.21.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 202 |
+
"model.language_model.layers.21.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 203 |
+
"model.language_model.layers.21.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 204 |
+
"model.language_model.layers.21.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 205 |
+
"model.language_model.layers.21.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 206 |
+
"model.language_model.layers.21.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 207 |
+
"model.language_model.layers.21.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 208 |
+
"model.language_model.layers.21.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 209 |
+
"model.language_model.layers.21.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 210 |
+
"model.language_model.layers.22.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 211 |
+
"model.language_model.layers.22.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 212 |
+
"model.language_model.layers.22.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 213 |
+
"model.language_model.layers.22.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 214 |
+
"model.language_model.layers.22.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 215 |
+
"model.language_model.layers.22.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 216 |
+
"model.language_model.layers.22.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 217 |
+
"model.language_model.layers.22.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 218 |
+
"model.language_model.layers.22.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 219 |
+
"model.language_model.layers.22.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 220 |
+
"model.language_model.layers.22.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 221 |
+
"model.language_model.layers.22.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 222 |
+
"model.language_model.layers.22.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 223 |
+
"model.language_model.layers.22.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 224 |
+
"model.language_model.layers.23.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 225 |
+
"model.language_model.layers.23.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 226 |
+
"model.language_model.layers.23.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 227 |
+
"model.language_model.layers.23.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 228 |
+
"model.language_model.layers.23.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 229 |
+
"model.language_model.layers.23.self_attn.k_norm.weight": "model-00003-of-00004.safetensors",
|
| 230 |
+
"model.language_model.layers.23.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
|
| 231 |
+
"model.language_model.layers.23.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
|
| 232 |
+
"model.language_model.layers.23.self_attn.q_norm.weight": "model-00003-of-00004.safetensors",
|
| 233 |
+
"model.language_model.layers.23.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
|
| 234 |
+
"model.language_model.layers.23.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
|
| 235 |
+
"model.language_model.layers.24.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 236 |
+
"model.language_model.layers.24.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 237 |
+
"model.language_model.layers.24.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 238 |
+
"model.language_model.layers.24.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 239 |
+
"model.language_model.layers.24.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 240 |
+
"model.language_model.layers.24.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 241 |
+
"model.language_model.layers.24.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 242 |
+
"model.language_model.layers.24.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 243 |
+
"model.language_model.layers.24.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 244 |
+
"model.language_model.layers.24.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 245 |
+
"model.language_model.layers.24.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 246 |
+
"model.language_model.layers.24.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 247 |
+
"model.language_model.layers.24.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 248 |
+
"model.language_model.layers.24.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 249 |
+
"model.language_model.layers.25.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 250 |
+
"model.language_model.layers.25.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 251 |
+
"model.language_model.layers.25.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 252 |
+
"model.language_model.layers.25.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 253 |
+
"model.language_model.layers.25.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 254 |
+
"model.language_model.layers.25.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 255 |
+
"model.language_model.layers.25.linear_attn.in_proj_qkv.weight": "model-00004-of-00004.safetensors",
|
| 256 |
+
"model.language_model.layers.25.linear_attn.in_proj_z.weight": "model-00004-of-00004.safetensors",
|
| 257 |
+
"model.language_model.layers.25.linear_attn.norm.weight": "model-00004-of-00004.safetensors",
|
| 258 |
+
"model.language_model.layers.25.linear_attn.out_proj.weight": "model-00004-of-00004.safetensors",
|
| 259 |
+
"model.language_model.layers.25.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 260 |
+
"model.language_model.layers.25.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 261 |
+
"model.language_model.layers.25.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 262 |
+
"model.language_model.layers.25.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 263 |
+
"model.language_model.layers.26.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 264 |
+
"model.language_model.layers.26.linear_attn.A_log": "model-00004-of-00004.safetensors",
|
| 265 |
+
"model.language_model.layers.26.linear_attn.conv1d.weight": "model-00004-of-00004.safetensors",
|
| 266 |
+
"model.language_model.layers.26.linear_attn.dt_bias": "model-00004-of-00004.safetensors",
|
| 267 |
+
"model.language_model.layers.26.linear_attn.in_proj_a.weight": "model-00004-of-00004.safetensors",
|
| 268 |
+
"model.language_model.layers.26.linear_attn.in_proj_b.weight": "model-00004-of-00004.safetensors",
|
| 269 |
+
"model.language_model.layers.26.linear_attn.in_proj_qkv.weight": "model-00004-of-00004.safetensors",
|
| 270 |
+
"model.language_model.layers.26.linear_attn.in_proj_z.weight": "model-00004-of-00004.safetensors",
|
| 271 |
+
"model.language_model.layers.26.linear_attn.norm.weight": "model-00004-of-00004.safetensors",
|
| 272 |
+
"model.language_model.layers.26.linear_attn.out_proj.weight": "model-00004-of-00004.safetensors",
|
| 273 |
+
"model.language_model.layers.26.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 274 |
+
"model.language_model.layers.26.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 275 |
+
"model.language_model.layers.26.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 276 |
+
"model.language_model.layers.26.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 277 |
+
"model.language_model.layers.27.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 278 |
+
"model.language_model.layers.27.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 279 |
+
"model.language_model.layers.27.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 280 |
+
"model.language_model.layers.27.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 281 |
+
"model.language_model.layers.27.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 282 |
+
"model.language_model.layers.27.self_attn.k_norm.weight": "model-00004-of-00004.safetensors",
|
| 283 |
+
"model.language_model.layers.27.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
|
| 284 |
+
"model.language_model.layers.27.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
|
| 285 |
+
"model.language_model.layers.27.self_attn.q_norm.weight": "model-00004-of-00004.safetensors",
|
| 286 |
+
"model.language_model.layers.27.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
|
| 287 |
+
"model.language_model.layers.27.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
|
| 288 |
+
"model.language_model.layers.28.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 289 |
+
"model.language_model.layers.28.linear_attn.A_log": "model-00004-of-00004.safetensors",
|
| 290 |
+
"model.language_model.layers.28.linear_attn.conv1d.weight": "model-00004-of-00004.safetensors",
|
| 291 |
+
"model.language_model.layers.28.linear_attn.dt_bias": "model-00004-of-00004.safetensors",
|
| 292 |
+
"model.language_model.layers.28.linear_attn.in_proj_a.weight": "model-00004-of-00004.safetensors",
|
| 293 |
+
"model.language_model.layers.28.linear_attn.in_proj_b.weight": "model-00004-of-00004.safetensors",
|
| 294 |
+
"model.language_model.layers.28.linear_attn.in_proj_qkv.weight": "model-00004-of-00004.safetensors",
|
| 295 |
+
"model.language_model.layers.28.linear_attn.in_proj_z.weight": "model-00004-of-00004.safetensors",
|
| 296 |
+
"model.language_model.layers.28.linear_attn.norm.weight": "model-00004-of-00004.safetensors",
|
| 297 |
+
"model.language_model.layers.28.linear_attn.out_proj.weight": "model-00004-of-00004.safetensors",
|
| 298 |
+
"model.language_model.layers.28.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 299 |
+
"model.language_model.layers.28.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 300 |
+
"model.language_model.layers.28.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 301 |
+
"model.language_model.layers.28.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 302 |
+
"model.language_model.layers.29.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 303 |
+
"model.language_model.layers.29.linear_attn.A_log": "model-00004-of-00004.safetensors",
|
| 304 |
+
"model.language_model.layers.29.linear_attn.conv1d.weight": "model-00004-of-00004.safetensors",
|
| 305 |
+
"model.language_model.layers.29.linear_attn.dt_bias": "model-00004-of-00004.safetensors",
|
| 306 |
+
"model.language_model.layers.29.linear_attn.in_proj_a.weight": "model-00004-of-00004.safetensors",
|
| 307 |
+
"model.language_model.layers.29.linear_attn.in_proj_b.weight": "model-00004-of-00004.safetensors",
|
| 308 |
+
"model.language_model.layers.29.linear_attn.in_proj_qkv.weight": "model-00004-of-00004.safetensors",
|
| 309 |
+
"model.language_model.layers.29.linear_attn.in_proj_z.weight": "model-00004-of-00004.safetensors",
|
| 310 |
+
"model.language_model.layers.29.linear_attn.norm.weight": "model-00004-of-00004.safetensors",
|
| 311 |
+
"model.language_model.layers.29.linear_attn.out_proj.weight": "model-00004-of-00004.safetensors",
|
| 312 |
+
"model.language_model.layers.29.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 313 |
+
"model.language_model.layers.29.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 314 |
+
"model.language_model.layers.29.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 315 |
+
"model.language_model.layers.29.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 316 |
+
"model.language_model.layers.3.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 317 |
+
"model.language_model.layers.3.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 318 |
+
"model.language_model.layers.3.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 319 |
+
"model.language_model.layers.3.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 320 |
+
"model.language_model.layers.3.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 321 |
+
"model.language_model.layers.3.self_attn.k_norm.weight": "model-00002-of-00004.safetensors",
|
| 322 |
+
"model.language_model.layers.3.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
|
| 323 |
+
"model.language_model.layers.3.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
|
| 324 |
+
"model.language_model.layers.3.self_attn.q_norm.weight": "model-00002-of-00004.safetensors",
|
| 325 |
+
"model.language_model.layers.3.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
|
| 326 |
+
"model.language_model.layers.3.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
|
| 327 |
+
"model.language_model.layers.30.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 328 |
+
"model.language_model.layers.30.linear_attn.A_log": "model-00004-of-00004.safetensors",
|
| 329 |
+
"model.language_model.layers.30.linear_attn.conv1d.weight": "model-00004-of-00004.safetensors",
|
| 330 |
+
"model.language_model.layers.30.linear_attn.dt_bias": "model-00004-of-00004.safetensors",
|
| 331 |
+
"model.language_model.layers.30.linear_attn.in_proj_a.weight": "model-00004-of-00004.safetensors",
|
| 332 |
+
"model.language_model.layers.30.linear_attn.in_proj_b.weight": "model-00004-of-00004.safetensors",
|
| 333 |
+
"model.language_model.layers.30.linear_attn.in_proj_qkv.weight": "model-00004-of-00004.safetensors",
|
| 334 |
+
"model.language_model.layers.30.linear_attn.in_proj_z.weight": "model-00004-of-00004.safetensors",
|
| 335 |
+
"model.language_model.layers.30.linear_attn.norm.weight": "model-00004-of-00004.safetensors",
|
| 336 |
+
"model.language_model.layers.30.linear_attn.out_proj.weight": "model-00004-of-00004.safetensors",
|
| 337 |
+
"model.language_model.layers.30.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 338 |
+
"model.language_model.layers.30.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 339 |
+
"model.language_model.layers.30.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 340 |
+
"model.language_model.layers.30.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 341 |
+
"model.language_model.layers.31.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 342 |
+
"model.language_model.layers.31.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 343 |
+
"model.language_model.layers.31.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 344 |
+
"model.language_model.layers.31.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 345 |
+
"model.language_model.layers.31.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 346 |
+
"model.language_model.layers.31.self_attn.k_norm.weight": "model-00004-of-00004.safetensors",
|
| 347 |
+
"model.language_model.layers.31.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
|
| 348 |
+
"model.language_model.layers.31.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
|
| 349 |
+
"model.language_model.layers.31.self_attn.q_norm.weight": "model-00004-of-00004.safetensors",
|
| 350 |
+
"model.language_model.layers.31.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
|
| 351 |
+
"model.language_model.layers.31.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
|
| 352 |
+
"model.language_model.layers.4.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 353 |
+
"model.language_model.layers.4.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 354 |
+
"model.language_model.layers.4.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 355 |
+
"model.language_model.layers.4.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 356 |
+
"model.language_model.layers.4.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 357 |
+
"model.language_model.layers.4.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 358 |
+
"model.language_model.layers.4.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 359 |
+
"model.language_model.layers.4.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 360 |
+
"model.language_model.layers.4.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 361 |
+
"model.language_model.layers.4.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 362 |
+
"model.language_model.layers.4.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 363 |
+
"model.language_model.layers.4.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 364 |
+
"model.language_model.layers.4.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 365 |
+
"model.language_model.layers.4.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 366 |
+
"model.language_model.layers.5.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 367 |
+
"model.language_model.layers.5.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 368 |
+
"model.language_model.layers.5.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 369 |
+
"model.language_model.layers.5.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 370 |
+
"model.language_model.layers.5.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 371 |
+
"model.language_model.layers.5.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 372 |
+
"model.language_model.layers.5.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 373 |
+
"model.language_model.layers.5.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 374 |
+
"model.language_model.layers.5.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 375 |
+
"model.language_model.layers.5.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 376 |
+
"model.language_model.layers.5.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 377 |
+
"model.language_model.layers.5.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 378 |
+
"model.language_model.layers.5.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 379 |
+
"model.language_model.layers.5.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 380 |
+
"model.language_model.layers.6.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 381 |
+
"model.language_model.layers.6.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 382 |
+
"model.language_model.layers.6.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 383 |
+
"model.language_model.layers.6.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 384 |
+
"model.language_model.layers.6.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 385 |
+
"model.language_model.layers.6.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 386 |
+
"model.language_model.layers.6.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 387 |
+
"model.language_model.layers.6.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 388 |
+
"model.language_model.layers.6.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 389 |
+
"model.language_model.layers.6.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 390 |
+
"model.language_model.layers.6.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 391 |
+
"model.language_model.layers.6.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 392 |
+
"model.language_model.layers.6.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 393 |
+
"model.language_model.layers.6.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 394 |
+
"model.language_model.layers.7.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 395 |
+
"model.language_model.layers.7.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 396 |
+
"model.language_model.layers.7.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 397 |
+
"model.language_model.layers.7.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 398 |
+
"model.language_model.layers.7.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 399 |
+
"model.language_model.layers.7.self_attn.k_norm.weight": "model-00002-of-00004.safetensors",
|
| 400 |
+
"model.language_model.layers.7.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
|
| 401 |
+
"model.language_model.layers.7.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
|
| 402 |
+
"model.language_model.layers.7.self_attn.q_norm.weight": "model-00002-of-00004.safetensors",
|
| 403 |
+
"model.language_model.layers.7.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
|
| 404 |
+
"model.language_model.layers.7.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
|
| 405 |
+
"model.language_model.layers.8.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 406 |
+
"model.language_model.layers.8.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 407 |
+
"model.language_model.layers.8.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 408 |
+
"model.language_model.layers.8.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 409 |
+
"model.language_model.layers.8.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 410 |
+
"model.language_model.layers.8.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 411 |
+
"model.language_model.layers.8.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 412 |
+
"model.language_model.layers.8.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 413 |
+
"model.language_model.layers.8.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 414 |
+
"model.language_model.layers.8.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 415 |
+
"model.language_model.layers.8.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 416 |
+
"model.language_model.layers.8.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 417 |
+
"model.language_model.layers.8.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 418 |
+
"model.language_model.layers.8.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 419 |
+
"model.language_model.layers.9.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 420 |
+
"model.language_model.layers.9.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 421 |
+
"model.language_model.layers.9.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 422 |
+
"model.language_model.layers.9.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 423 |
+
"model.language_model.layers.9.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 424 |
+
"model.language_model.layers.9.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 425 |
+
"model.language_model.layers.9.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 426 |
+
"model.language_model.layers.9.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 427 |
+
"model.language_model.layers.9.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 428 |
+
"model.language_model.layers.9.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 429 |
+
"model.language_model.layers.9.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 430 |
+
"model.language_model.layers.9.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 431 |
+
"model.language_model.layers.9.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 432 |
+
"model.language_model.layers.9.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 433 |
+
"model.language_model.norm.weight": "model-00004-of-00004.safetensors",
|
| 434 |
+
"model.visual.blocks.0.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 435 |
+
"model.visual.blocks.0.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 436 |
+
"model.visual.blocks.0.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 437 |
+
"model.visual.blocks.0.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 438 |
+
"model.visual.blocks.0.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 439 |
+
"model.visual.blocks.0.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 440 |
+
"model.visual.blocks.0.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 441 |
+
"model.visual.blocks.0.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 442 |
+
"model.visual.blocks.0.norm1.bias": "model-00004-of-00004.safetensors",
|
| 443 |
+
"model.visual.blocks.0.norm1.weight": "model-00004-of-00004.safetensors",
|
| 444 |
+
"model.visual.blocks.0.norm2.bias": "model-00004-of-00004.safetensors",
|
| 445 |
+
"model.visual.blocks.0.norm2.weight": "model-00004-of-00004.safetensors",
|
| 446 |
+
"model.visual.blocks.1.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 447 |
+
"model.visual.blocks.1.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 448 |
+
"model.visual.blocks.1.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 449 |
+
"model.visual.blocks.1.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 450 |
+
"model.visual.blocks.1.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 451 |
+
"model.visual.blocks.1.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 452 |
+
"model.visual.blocks.1.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 453 |
+
"model.visual.blocks.1.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 454 |
+
"model.visual.blocks.1.norm1.bias": "model-00004-of-00004.safetensors",
|
| 455 |
+
"model.visual.blocks.1.norm1.weight": "model-00004-of-00004.safetensors",
|
| 456 |
+
"model.visual.blocks.1.norm2.bias": "model-00004-of-00004.safetensors",
|
| 457 |
+
"model.visual.blocks.1.norm2.weight": "model-00004-of-00004.safetensors",
|
| 458 |
+
"model.visual.blocks.10.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 459 |
+
"model.visual.blocks.10.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 460 |
+
"model.visual.blocks.10.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 461 |
+
"model.visual.blocks.10.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 462 |
+
"model.visual.blocks.10.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 463 |
+
"model.visual.blocks.10.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 464 |
+
"model.visual.blocks.10.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 465 |
+
"model.visual.blocks.10.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 466 |
+
"model.visual.blocks.10.norm1.bias": "model-00004-of-00004.safetensors",
|
| 467 |
+
"model.visual.blocks.10.norm1.weight": "model-00004-of-00004.safetensors",
|
| 468 |
+
"model.visual.blocks.10.norm2.bias": "model-00004-of-00004.safetensors",
|
| 469 |
+
"model.visual.blocks.10.norm2.weight": "model-00004-of-00004.safetensors",
|
| 470 |
+
"model.visual.blocks.11.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 471 |
+
"model.visual.blocks.11.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 472 |
+
"model.visual.blocks.11.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 473 |
+
"model.visual.blocks.11.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 474 |
+
"model.visual.blocks.11.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 475 |
+
"model.visual.blocks.11.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 476 |
+
"model.visual.blocks.11.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 477 |
+
"model.visual.blocks.11.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 478 |
+
"model.visual.blocks.11.norm1.bias": "model-00004-of-00004.safetensors",
|
| 479 |
+
"model.visual.blocks.11.norm1.weight": "model-00004-of-00004.safetensors",
|
| 480 |
+
"model.visual.blocks.11.norm2.bias": "model-00004-of-00004.safetensors",
|
| 481 |
+
"model.visual.blocks.11.norm2.weight": "model-00004-of-00004.safetensors",
|
| 482 |
+
"model.visual.blocks.12.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 483 |
+
"model.visual.blocks.12.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 484 |
+
"model.visual.blocks.12.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 485 |
+
"model.visual.blocks.12.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 486 |
+
"model.visual.blocks.12.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 487 |
+
"model.visual.blocks.12.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 488 |
+
"model.visual.blocks.12.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 489 |
+
"model.visual.blocks.12.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 490 |
+
"model.visual.blocks.12.norm1.bias": "model-00004-of-00004.safetensors",
|
| 491 |
+
"model.visual.blocks.12.norm1.weight": "model-00004-of-00004.safetensors",
|
| 492 |
+
"model.visual.blocks.12.norm2.bias": "model-00004-of-00004.safetensors",
|
| 493 |
+
"model.visual.blocks.12.norm2.weight": "model-00004-of-00004.safetensors",
|
| 494 |
+
"model.visual.blocks.13.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 495 |
+
"model.visual.blocks.13.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 496 |
+
"model.visual.blocks.13.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 497 |
+
"model.visual.blocks.13.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 498 |
+
"model.visual.blocks.13.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 499 |
+
"model.visual.blocks.13.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 500 |
+
"model.visual.blocks.13.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 501 |
+
"model.visual.blocks.13.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 502 |
+
"model.visual.blocks.13.norm1.bias": "model-00004-of-00004.safetensors",
|
| 503 |
+
"model.visual.blocks.13.norm1.weight": "model-00004-of-00004.safetensors",
|
| 504 |
+
"model.visual.blocks.13.norm2.bias": "model-00004-of-00004.safetensors",
|
| 505 |
+
"model.visual.blocks.13.norm2.weight": "model-00004-of-00004.safetensors",
|
| 506 |
+
"model.visual.blocks.14.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 507 |
+
"model.visual.blocks.14.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 508 |
+
"model.visual.blocks.14.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 509 |
+
"model.visual.blocks.14.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 510 |
+
"model.visual.blocks.14.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 511 |
+
"model.visual.blocks.14.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 512 |
+
"model.visual.blocks.14.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 513 |
+
"model.visual.blocks.14.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 514 |
+
"model.visual.blocks.14.norm1.bias": "model-00004-of-00004.safetensors",
|
| 515 |
+
"model.visual.blocks.14.norm1.weight": "model-00004-of-00004.safetensors",
|
| 516 |
+
"model.visual.blocks.14.norm2.bias": "model-00004-of-00004.safetensors",
|
| 517 |
+
"model.visual.blocks.14.norm2.weight": "model-00004-of-00004.safetensors",
|
| 518 |
+
"model.visual.blocks.15.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 519 |
+
"model.visual.blocks.15.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 520 |
+
"model.visual.blocks.15.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 521 |
+
"model.visual.blocks.15.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 522 |
+
"model.visual.blocks.15.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 523 |
+
"model.visual.blocks.15.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 524 |
+
"model.visual.blocks.15.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 525 |
+
"model.visual.blocks.15.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 526 |
+
"model.visual.blocks.15.norm1.bias": "model-00004-of-00004.safetensors",
|
| 527 |
+
"model.visual.blocks.15.norm1.weight": "model-00004-of-00004.safetensors",
|
| 528 |
+
"model.visual.blocks.15.norm2.bias": "model-00004-of-00004.safetensors",
|
| 529 |
+
"model.visual.blocks.15.norm2.weight": "model-00004-of-00004.safetensors",
|
| 530 |
+
"model.visual.blocks.16.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 531 |
+
"model.visual.blocks.16.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 532 |
+
"model.visual.blocks.16.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 533 |
+
"model.visual.blocks.16.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 534 |
+
"model.visual.blocks.16.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 535 |
+
"model.visual.blocks.16.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 536 |
+
"model.visual.blocks.16.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 537 |
+
"model.visual.blocks.16.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 538 |
+
"model.visual.blocks.16.norm1.bias": "model-00004-of-00004.safetensors",
|
| 539 |
+
"model.visual.blocks.16.norm1.weight": "model-00004-of-00004.safetensors",
|
| 540 |
+
"model.visual.blocks.16.norm2.bias": "model-00004-of-00004.safetensors",
|
| 541 |
+
"model.visual.blocks.16.norm2.weight": "model-00004-of-00004.safetensors",
|
| 542 |
+
"model.visual.blocks.17.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 543 |
+
"model.visual.blocks.17.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 544 |
+
"model.visual.blocks.17.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 545 |
+
"model.visual.blocks.17.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 546 |
+
"model.visual.blocks.17.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 547 |
+
"model.visual.blocks.17.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 548 |
+
"model.visual.blocks.17.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 549 |
+
"model.visual.blocks.17.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 550 |
+
"model.visual.blocks.17.norm1.bias": "model-00004-of-00004.safetensors",
|
| 551 |
+
"model.visual.blocks.17.norm1.weight": "model-00004-of-00004.safetensors",
|
| 552 |
+
"model.visual.blocks.17.norm2.bias": "model-00004-of-00004.safetensors",
|
| 553 |
+
"model.visual.blocks.17.norm2.weight": "model-00004-of-00004.safetensors",
|
| 554 |
+
"model.visual.blocks.18.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 555 |
+
"model.visual.blocks.18.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 556 |
+
"model.visual.blocks.18.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 557 |
+
"model.visual.blocks.18.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 558 |
+
"model.visual.blocks.18.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 559 |
+
"model.visual.blocks.18.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 560 |
+
"model.visual.blocks.18.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 561 |
+
"model.visual.blocks.18.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 562 |
+
"model.visual.blocks.18.norm1.bias": "model-00004-of-00004.safetensors",
|
| 563 |
+
"model.visual.blocks.18.norm1.weight": "model-00004-of-00004.safetensors",
|
| 564 |
+
"model.visual.blocks.18.norm2.bias": "model-00004-of-00004.safetensors",
|
| 565 |
+
"model.visual.blocks.18.norm2.weight": "model-00004-of-00004.safetensors",
|
| 566 |
+
"model.visual.blocks.19.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 567 |
+
"model.visual.blocks.19.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 568 |
+
"model.visual.blocks.19.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 569 |
+
"model.visual.blocks.19.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 570 |
+
"model.visual.blocks.19.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 571 |
+
"model.visual.blocks.19.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 572 |
+
"model.visual.blocks.19.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 573 |
+
"model.visual.blocks.19.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 574 |
+
"model.visual.blocks.19.norm1.bias": "model-00004-of-00004.safetensors",
|
| 575 |
+
"model.visual.blocks.19.norm1.weight": "model-00004-of-00004.safetensors",
|
| 576 |
+
"model.visual.blocks.19.norm2.bias": "model-00004-of-00004.safetensors",
|
| 577 |
+
"model.visual.blocks.19.norm2.weight": "model-00004-of-00004.safetensors",
|
| 578 |
+
"model.visual.blocks.2.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 579 |
+
"model.visual.blocks.2.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 580 |
+
"model.visual.blocks.2.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 581 |
+
"model.visual.blocks.2.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 582 |
+
"model.visual.blocks.2.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 583 |
+
"model.visual.blocks.2.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 584 |
+
"model.visual.blocks.2.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 585 |
+
"model.visual.blocks.2.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 586 |
+
"model.visual.blocks.2.norm1.bias": "model-00004-of-00004.safetensors",
|
| 587 |
+
"model.visual.blocks.2.norm1.weight": "model-00004-of-00004.safetensors",
|
| 588 |
+
"model.visual.blocks.2.norm2.bias": "model-00004-of-00004.safetensors",
|
| 589 |
+
"model.visual.blocks.2.norm2.weight": "model-00004-of-00004.safetensors",
|
| 590 |
+
"model.visual.blocks.20.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 591 |
+
"model.visual.blocks.20.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 592 |
+
"model.visual.blocks.20.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 593 |
+
"model.visual.blocks.20.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 594 |
+
"model.visual.blocks.20.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 595 |
+
"model.visual.blocks.20.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 596 |
+
"model.visual.blocks.20.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 597 |
+
"model.visual.blocks.20.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 598 |
+
"model.visual.blocks.20.norm1.bias": "model-00004-of-00004.safetensors",
|
| 599 |
+
"model.visual.blocks.20.norm1.weight": "model-00004-of-00004.safetensors",
|
| 600 |
+
"model.visual.blocks.20.norm2.bias": "model-00004-of-00004.safetensors",
|
| 601 |
+
"model.visual.blocks.20.norm2.weight": "model-00004-of-00004.safetensors",
|
| 602 |
+
"model.visual.blocks.21.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 603 |
+
"model.visual.blocks.21.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 604 |
+
"model.visual.blocks.21.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 605 |
+
"model.visual.blocks.21.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 606 |
+
"model.visual.blocks.21.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 607 |
+
"model.visual.blocks.21.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 608 |
+
"model.visual.blocks.21.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 609 |
+
"model.visual.blocks.21.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 610 |
+
"model.visual.blocks.21.norm1.bias": "model-00004-of-00004.safetensors",
|
| 611 |
+
"model.visual.blocks.21.norm1.weight": "model-00004-of-00004.safetensors",
|
| 612 |
+
"model.visual.blocks.21.norm2.bias": "model-00004-of-00004.safetensors",
|
| 613 |
+
"model.visual.blocks.21.norm2.weight": "model-00004-of-00004.safetensors",
|
| 614 |
+
"model.visual.blocks.22.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 615 |
+
"model.visual.blocks.22.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 616 |
+
"model.visual.blocks.22.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 617 |
+
"model.visual.blocks.22.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 618 |
+
"model.visual.blocks.22.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 619 |
+
"model.visual.blocks.22.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 620 |
+
"model.visual.blocks.22.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 621 |
+
"model.visual.blocks.22.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 622 |
+
"model.visual.blocks.22.norm1.bias": "model-00004-of-00004.safetensors",
|
| 623 |
+
"model.visual.blocks.22.norm1.weight": "model-00004-of-00004.safetensors",
|
| 624 |
+
"model.visual.blocks.22.norm2.bias": "model-00004-of-00004.safetensors",
|
| 625 |
+
"model.visual.blocks.22.norm2.weight": "model-00004-of-00004.safetensors",
|
| 626 |
+
"model.visual.blocks.23.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 627 |
+
"model.visual.blocks.23.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 628 |
+
"model.visual.blocks.23.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 629 |
+
"model.visual.blocks.23.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 630 |
+
"model.visual.blocks.23.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 631 |
+
"model.visual.blocks.23.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 632 |
+
"model.visual.blocks.23.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 633 |
+
"model.visual.blocks.23.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 634 |
+
"model.visual.blocks.23.norm1.bias": "model-00004-of-00004.safetensors",
|
| 635 |
+
"model.visual.blocks.23.norm1.weight": "model-00004-of-00004.safetensors",
|
| 636 |
+
"model.visual.blocks.23.norm2.bias": "model-00004-of-00004.safetensors",
|
| 637 |
+
"model.visual.blocks.23.norm2.weight": "model-00004-of-00004.safetensors",
|
| 638 |
+
"model.visual.blocks.24.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 639 |
+
"model.visual.blocks.24.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 640 |
+
"model.visual.blocks.24.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 641 |
+
"model.visual.blocks.24.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 642 |
+
"model.visual.blocks.24.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 643 |
+
"model.visual.blocks.24.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 644 |
+
"model.visual.blocks.24.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 645 |
+
"model.visual.blocks.24.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 646 |
+
"model.visual.blocks.24.norm1.bias": "model-00004-of-00004.safetensors",
|
| 647 |
+
"model.visual.blocks.24.norm1.weight": "model-00004-of-00004.safetensors",
|
| 648 |
+
"model.visual.blocks.24.norm2.bias": "model-00004-of-00004.safetensors",
|
| 649 |
+
"model.visual.blocks.24.norm2.weight": "model-00004-of-00004.safetensors",
|
| 650 |
+
"model.visual.blocks.25.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 651 |
+
"model.visual.blocks.25.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 652 |
+
"model.visual.blocks.25.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 653 |
+
"model.visual.blocks.25.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 654 |
+
"model.visual.blocks.25.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 655 |
+
"model.visual.blocks.25.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 656 |
+
"model.visual.blocks.25.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 657 |
+
"model.visual.blocks.25.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 658 |
+
"model.visual.blocks.25.norm1.bias": "model-00004-of-00004.safetensors",
|
| 659 |
+
"model.visual.blocks.25.norm1.weight": "model-00004-of-00004.safetensors",
|
| 660 |
+
"model.visual.blocks.25.norm2.bias": "model-00004-of-00004.safetensors",
|
| 661 |
+
"model.visual.blocks.25.norm2.weight": "model-00004-of-00004.safetensors",
|
| 662 |
+
"model.visual.blocks.26.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 663 |
+
"model.visual.blocks.26.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 664 |
+
"model.visual.blocks.26.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 665 |
+
"model.visual.blocks.26.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 666 |
+
"model.visual.blocks.26.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 667 |
+
"model.visual.blocks.26.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 668 |
+
"model.visual.blocks.26.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 669 |
+
"model.visual.blocks.26.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 670 |
+
"model.visual.blocks.26.norm1.bias": "model-00004-of-00004.safetensors",
|
| 671 |
+
"model.visual.blocks.26.norm1.weight": "model-00004-of-00004.safetensors",
|
| 672 |
+
"model.visual.blocks.26.norm2.bias": "model-00004-of-00004.safetensors",
|
| 673 |
+
"model.visual.blocks.26.norm2.weight": "model-00004-of-00004.safetensors",
|
| 674 |
+
"model.visual.blocks.3.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 675 |
+
"model.visual.blocks.3.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 676 |
+
"model.visual.blocks.3.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 677 |
+
"model.visual.blocks.3.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 678 |
+
"model.visual.blocks.3.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 679 |
+
"model.visual.blocks.3.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 680 |
+
"model.visual.blocks.3.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 681 |
+
"model.visual.blocks.3.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 682 |
+
"model.visual.blocks.3.norm1.bias": "model-00004-of-00004.safetensors",
|
| 683 |
+
"model.visual.blocks.3.norm1.weight": "model-00004-of-00004.safetensors",
|
| 684 |
+
"model.visual.blocks.3.norm2.bias": "model-00004-of-00004.safetensors",
|
| 685 |
+
"model.visual.blocks.3.norm2.weight": "model-00004-of-00004.safetensors",
|
| 686 |
+
"model.visual.blocks.4.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 687 |
+
"model.visual.blocks.4.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 688 |
+
"model.visual.blocks.4.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 689 |
+
"model.visual.blocks.4.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 690 |
+
"model.visual.blocks.4.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 691 |
+
"model.visual.blocks.4.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 692 |
+
"model.visual.blocks.4.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 693 |
+
"model.visual.blocks.4.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 694 |
+
"model.visual.blocks.4.norm1.bias": "model-00004-of-00004.safetensors",
|
| 695 |
+
"model.visual.blocks.4.norm1.weight": "model-00004-of-00004.safetensors",
|
| 696 |
+
"model.visual.blocks.4.norm2.bias": "model-00004-of-00004.safetensors",
|
| 697 |
+
"model.visual.blocks.4.norm2.weight": "model-00004-of-00004.safetensors",
|
| 698 |
+
"model.visual.blocks.5.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 699 |
+
"model.visual.blocks.5.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 700 |
+
"model.visual.blocks.5.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 701 |
+
"model.visual.blocks.5.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 702 |
+
"model.visual.blocks.5.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 703 |
+
"model.visual.blocks.5.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 704 |
+
"model.visual.blocks.5.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 705 |
+
"model.visual.blocks.5.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 706 |
+
"model.visual.blocks.5.norm1.bias": "model-00004-of-00004.safetensors",
|
| 707 |
+
"model.visual.blocks.5.norm1.weight": "model-00004-of-00004.safetensors",
|
| 708 |
+
"model.visual.blocks.5.norm2.bias": "model-00004-of-00004.safetensors",
|
| 709 |
+
"model.visual.blocks.5.norm2.weight": "model-00004-of-00004.safetensors",
|
| 710 |
+
"model.visual.blocks.6.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 711 |
+
"model.visual.blocks.6.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 712 |
+
"model.visual.blocks.6.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 713 |
+
"model.visual.blocks.6.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 714 |
+
"model.visual.blocks.6.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 715 |
+
"model.visual.blocks.6.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 716 |
+
"model.visual.blocks.6.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 717 |
+
"model.visual.blocks.6.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 718 |
+
"model.visual.blocks.6.norm1.bias": "model-00004-of-00004.safetensors",
|
| 719 |
+
"model.visual.blocks.6.norm1.weight": "model-00004-of-00004.safetensors",
|
| 720 |
+
"model.visual.blocks.6.norm2.bias": "model-00004-of-00004.safetensors",
|
| 721 |
+
"model.visual.blocks.6.norm2.weight": "model-00004-of-00004.safetensors",
|
| 722 |
+
"model.visual.blocks.7.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 723 |
+
"model.visual.blocks.7.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 724 |
+
"model.visual.blocks.7.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 725 |
+
"model.visual.blocks.7.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 726 |
+
"model.visual.blocks.7.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 727 |
+
"model.visual.blocks.7.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 728 |
+
"model.visual.blocks.7.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 729 |
+
"model.visual.blocks.7.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 730 |
+
"model.visual.blocks.7.norm1.bias": "model-00004-of-00004.safetensors",
|
| 731 |
+
"model.visual.blocks.7.norm1.weight": "model-00004-of-00004.safetensors",
|
| 732 |
+
"model.visual.blocks.7.norm2.bias": "model-00004-of-00004.safetensors",
|
| 733 |
+
"model.visual.blocks.7.norm2.weight": "model-00004-of-00004.safetensors",
|
| 734 |
+
"model.visual.blocks.8.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 735 |
+
"model.visual.blocks.8.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 736 |
+
"model.visual.blocks.8.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 737 |
+
"model.visual.blocks.8.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 738 |
+
"model.visual.blocks.8.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 739 |
+
"model.visual.blocks.8.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 740 |
+
"model.visual.blocks.8.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 741 |
+
"model.visual.blocks.8.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 742 |
+
"model.visual.blocks.8.norm1.bias": "model-00004-of-00004.safetensors",
|
| 743 |
+
"model.visual.blocks.8.norm1.weight": "model-00004-of-00004.safetensors",
|
| 744 |
+
"model.visual.blocks.8.norm2.bias": "model-00004-of-00004.safetensors",
|
| 745 |
+
"model.visual.blocks.8.norm2.weight": "model-00004-of-00004.safetensors",
|
| 746 |
+
"model.visual.blocks.9.attn.proj.bias": "model-00004-of-00004.safetensors",
|
| 747 |
+
"model.visual.blocks.9.attn.proj.weight": "model-00004-of-00004.safetensors",
|
| 748 |
+
"model.visual.blocks.9.attn.qkv.bias": "model-00004-of-00004.safetensors",
|
| 749 |
+
"model.visual.blocks.9.attn.qkv.weight": "model-00004-of-00004.safetensors",
|
| 750 |
+
"model.visual.blocks.9.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 751 |
+
"model.visual.blocks.9.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 752 |
+
"model.visual.blocks.9.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 753 |
+
"model.visual.blocks.9.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 754 |
+
"model.visual.blocks.9.norm1.bias": "model-00004-of-00004.safetensors",
|
| 755 |
+
"model.visual.blocks.9.norm1.weight": "model-00004-of-00004.safetensors",
|
| 756 |
+
"model.visual.blocks.9.norm2.bias": "model-00004-of-00004.safetensors",
|
| 757 |
+
"model.visual.blocks.9.norm2.weight": "model-00004-of-00004.safetensors",
|
| 758 |
+
"model.visual.merger.linear_fc1.bias": "model-00004-of-00004.safetensors",
|
| 759 |
+
"model.visual.merger.linear_fc1.weight": "model-00004-of-00004.safetensors",
|
| 760 |
+
"model.visual.merger.linear_fc2.bias": "model-00004-of-00004.safetensors",
|
| 761 |
+
"model.visual.merger.linear_fc2.weight": "model-00004-of-00004.safetensors",
|
| 762 |
+
"model.visual.merger.norm.bias": "model-00004-of-00004.safetensors",
|
| 763 |
+
"model.visual.merger.norm.weight": "model-00004-of-00004.safetensors",
|
| 764 |
+
"model.visual.patch_embed.proj.bias": "model-00004-of-00004.safetensors",
|
| 765 |
+
"model.visual.patch_embed.proj.weight": "model-00004-of-00004.safetensors",
|
| 766 |
+
"model.visual.pos_embed.weight": "model-00004-of-00004.safetensors"
|
| 767 |
+
}
|
| 768 |
+
}
|