Duplicate from unconst/Affine-5czsc2fc98-r496-sbsv5-offline-dpo-hialpha-midrank-lobeta-extrasteps-merged
Browse files- .gitattributes +36 -0
- README.md +11 -0
- chat_template.jinja +185 -0
- config.json +122 -0
- generation_config.json +13 -0
- merge_meta.json +161 -0
- model-00001-of-00016.safetensors +3 -0
- model-00002-of-00016.safetensors +3 -0
- model-00003-of-00016.safetensors +3 -0
- model-00004-of-00016.safetensors +3 -0
- model-00005-of-00016.safetensors +3 -0
- model-00006-of-00016.safetensors +3 -0
- model-00007-of-00016.safetensors +3 -0
- model-00008-of-00016.safetensors +3 -0
- model-00009-of-00016.safetensors +3 -0
- model-00010-of-00016.safetensors +3 -0
- model-00011-of-00016.safetensors +3 -0
- model-00012-of-00016.safetensors +3 -0
- model-00013-of-00016.safetensors +3 -0
- model-00014-of-00016.safetensors +3 -0
- model-00015-of-00016.safetensors +3 -0
- model-00016-of-00016.safetensors +3 -0
- model-visual-restored.safetensors +3 -0
- model.safetensors.index.json +1034 -0
- preprocessor_config.json +26 -0
- processor_config.json +60 -0
- tokenizer.json +3 -0
- tokenizer_config.json +37 -0
- video_preprocessor_config.json +21 -0
.gitattributes
ADDED
|
@@ -0,0 +1,36 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
*.7z filter=lfs diff=lfs merge=lfs -text
|
| 2 |
+
*.arrow filter=lfs diff=lfs merge=lfs -text
|
| 3 |
+
*.bin filter=lfs diff=lfs merge=lfs -text
|
| 4 |
+
*.bz2 filter=lfs diff=lfs merge=lfs -text
|
| 5 |
+
*.ckpt filter=lfs diff=lfs merge=lfs -text
|
| 6 |
+
*.ftz filter=lfs diff=lfs merge=lfs -text
|
| 7 |
+
*.gz filter=lfs diff=lfs merge=lfs -text
|
| 8 |
+
*.h5 filter=lfs diff=lfs merge=lfs -text
|
| 9 |
+
*.joblib filter=lfs diff=lfs merge=lfs -text
|
| 10 |
+
*.lfs.* filter=lfs diff=lfs merge=lfs -text
|
| 11 |
+
*.mlmodel filter=lfs diff=lfs merge=lfs -text
|
| 12 |
+
*.model filter=lfs diff=lfs merge=lfs -text
|
| 13 |
+
*.msgpack filter=lfs diff=lfs merge=lfs -text
|
| 14 |
+
*.npy filter=lfs diff=lfs merge=lfs -text
|
| 15 |
+
*.npz filter=lfs diff=lfs merge=lfs -text
|
| 16 |
+
*.onnx filter=lfs diff=lfs merge=lfs -text
|
| 17 |
+
*.ot filter=lfs diff=lfs merge=lfs -text
|
| 18 |
+
*.parquet filter=lfs diff=lfs merge=lfs -text
|
| 19 |
+
*.pb filter=lfs diff=lfs merge=lfs -text
|
| 20 |
+
*.pickle filter=lfs diff=lfs merge=lfs -text
|
| 21 |
+
*.pkl filter=lfs diff=lfs merge=lfs -text
|
| 22 |
+
*.pt filter=lfs diff=lfs merge=lfs -text
|
| 23 |
+
*.pth filter=lfs diff=lfs merge=lfs -text
|
| 24 |
+
*.rar filter=lfs diff=lfs merge=lfs -text
|
| 25 |
+
*.safetensors filter=lfs diff=lfs merge=lfs -text
|
| 26 |
+
saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
| 27 |
+
*.tar.* filter=lfs diff=lfs merge=lfs -text
|
| 28 |
+
*.tar filter=lfs diff=lfs merge=lfs -text
|
| 29 |
+
*.tflite filter=lfs diff=lfs merge=lfs -text
|
| 30 |
+
*.tgz filter=lfs diff=lfs merge=lfs -text
|
| 31 |
+
*.wasm filter=lfs diff=lfs merge=lfs -text
|
| 32 |
+
*.xz filter=lfs diff=lfs merge=lfs -text
|
| 33 |
+
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
+
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
+
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
+
tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
README.md
ADDED
|
@@ -0,0 +1,11 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: kevin954/Affine-5dfqbbh8ev-sft
|
| 3 |
+
library_name: transformers
|
| 4 |
+
pipeline_tag: text-generation
|
| 5 |
+
tags:
|
| 6 |
+
- affine-h1-merged-salvage
|
| 7 |
+
---
|
| 8 |
+
|
| 9 |
+
# H1 merged checkpoint salvage
|
| 10 |
+
|
| 11 |
+
LoRA-merged from `kevin954/Affine-5dfqbbh8ev-sft`. Private TTL insurance; not a submission until Stage-5 gate clears.
|
chat_template.jinja
ADDED
|
@@ -0,0 +1,185 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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 |
+
{#- Strict single-action output contract. Control tokens are described, never
|
| 43 |
+
written literally: split_special_tokens is false, so a literal role marker
|
| 44 |
+
here would encode as a real control token and break the prompt. #}
|
| 45 |
+
{%- set format_contract %}
|
| 46 |
+
# Output contract (strict)
|
| 47 |
+
|
| 48 |
+
Your reply is generated inside an already-open reasoning block. Emit exactly the following, in order, and nothing else:
|
| 49 |
+
|
| 50 |
+
1. Your private reasoning, as plain prose.
|
| 51 |
+
2. The closing think tag `</think>`, written exactly once.
|
| 52 |
+
3. A single line starting with `THOUGHT: ` giving one concise plan of at least 80 characters. State it once.
|
| 53 |
+
4. One blank line.
|
| 54 |
+
5. Exactly one closed ```bash fence containing the single next shell command.
|
| 55 |
+
6. Nothing at all after the closing fence. End the turn there.
|
| 56 |
+
|
| 57 |
+
Hard rules:
|
| 58 |
+
- Never write chat control tokens or bare role names (system, user, assistant).
|
| 59 |
+
- Never open a second reasoning block, and never repeat the closing think tag.
|
| 60 |
+
- Never repeat a sentence you have already written.
|
| 61 |
+
- Never write filler or announcements such as "The analysis is complete" or "Next command:".
|
| 62 |
+
- Never write any text after the closing bash fence.
|
| 63 |
+
{%- endset %}
|
| 64 |
+
{%- if not messages %}
|
| 65 |
+
{{- raise_exception('No messages provided.') }}
|
| 66 |
+
{%- endif %}
|
| 67 |
+
{%- if tools and tools is iterable and tools is not mapping %}
|
| 68 |
+
{{- '<|im_start|>system\n' }}
|
| 69 |
+
{{- "# Tools\n\nYou have access to the following functions:\n\n<tools>" }}
|
| 70 |
+
{%- for tool in tools %}
|
| 71 |
+
{{- "\n" }}
|
| 72 |
+
{{- tool | tojson }}
|
| 73 |
+
{%- endfor %}
|
| 74 |
+
{{- "\n</tools>" }}
|
| 75 |
+
{{- '\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>' }}
|
| 76 |
+
{%- if messages[0].role == 'system' %}
|
| 77 |
+
{%- set content = render_content(messages[0].content, false, true)|trim %}
|
| 78 |
+
{%- if content %}
|
| 79 |
+
{{- '\n\n' + content }}
|
| 80 |
+
{%- endif %}
|
| 81 |
+
{%- endif %}
|
| 82 |
+
{{- '\n\n' + format_contract | trim }}
|
| 83 |
+
{{- '<|im_end|>\n' }}
|
| 84 |
+
{%- else %}
|
| 85 |
+
{%- set sys_content = '' %}
|
| 86 |
+
{%- if messages[0].role == 'system' %}
|
| 87 |
+
{%- set sys_content = render_content(messages[0].content, false, true)|trim %}
|
| 88 |
+
{%- endif %}
|
| 89 |
+
{%- if sys_content %}
|
| 90 |
+
{{- '<|im_start|>system\n' + sys_content + '\n\n' + format_contract | trim + '<|im_end|>\n' }}
|
| 91 |
+
{%- else %}
|
| 92 |
+
{{- '<|im_start|>system\n' + format_contract | trim + '<|im_end|>\n' }}
|
| 93 |
+
{%- endif %}
|
| 94 |
+
{%- endif %}
|
| 95 |
+
{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
|
| 96 |
+
{%- for message in messages[::-1] %}
|
| 97 |
+
{%- set index = (messages|length - 1) - loop.index0 %}
|
| 98 |
+
{%- if ns.multi_step_tool and message.role == "user" %}
|
| 99 |
+
{%- set content = render_content(message.content, false)|trim %}
|
| 100 |
+
{%- if not(content.startswith('<tool_response>') and content.endswith('</tool_response>')) %}
|
| 101 |
+
{%- set ns.multi_step_tool = false %}
|
| 102 |
+
{%- set ns.last_query_index = index %}
|
| 103 |
+
{%- endif %}
|
| 104 |
+
{%- endif %}
|
| 105 |
+
{%- endfor %}
|
| 106 |
+
{%- if ns.multi_step_tool %}
|
| 107 |
+
{{- raise_exception('No user query found in messages.') }}
|
| 108 |
+
{%- endif %}
|
| 109 |
+
{%- for message in messages %}
|
| 110 |
+
{%- set content = render_content(message.content, true)|trim %}
|
| 111 |
+
{%- if message.role == "system" %}
|
| 112 |
+
{%- if not loop.first %}
|
| 113 |
+
{{- raise_exception('System message must be at the beginning.') }}
|
| 114 |
+
{%- endif %}
|
| 115 |
+
{%- elif message.role == "user" %}
|
| 116 |
+
{{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
|
| 117 |
+
{%- elif message.role == "assistant" %}
|
| 118 |
+
{%- set reasoning_content = '' %}
|
| 119 |
+
{%- if message.reasoning_content is string %}
|
| 120 |
+
{%- set reasoning_content = message.reasoning_content %}
|
| 121 |
+
{%- else %}
|
| 122 |
+
{%- if '</think>' in content %}
|
| 123 |
+
{%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
|
| 124 |
+
{%- set content = content.split('</think>')[-1].lstrip('\n') %}
|
| 125 |
+
{%- endif %}
|
| 126 |
+
{%- endif %}
|
| 127 |
+
{#- Residual tags survive when a turn carried more than one close tag;
|
| 128 |
+
leaving them in would demonstrate the malformed pattern in-context. #}
|
| 129 |
+
{%- set reasoning_content = reasoning_content.replace('<think>', '').replace('</think>', '')|trim %}
|
| 130 |
+
{%- set content = content.replace('<think>', '').replace('</think>', '')|trim %}
|
| 131 |
+
{%- if (preserve_thinking is defined and preserve_thinking is true) or (loop.index0 > ns.last_query_index) %}
|
| 132 |
+
{{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content + '\n</think>\n\n' + content }}
|
| 133 |
+
{%- else %}
|
| 134 |
+
{{- '<|im_start|>' + message.role + '\n' + content }}
|
| 135 |
+
{%- endif %}
|
| 136 |
+
{%- if message.tool_calls and message.tool_calls is iterable and message.tool_calls is not mapping %}
|
| 137 |
+
{%- for tool_call in message.tool_calls %}
|
| 138 |
+
{%- if tool_call.function is defined %}
|
| 139 |
+
{%- set tool_call = tool_call.function %}
|
| 140 |
+
{%- endif %}
|
| 141 |
+
{%- if loop.first %}
|
| 142 |
+
{%- if content|trim %}
|
| 143 |
+
{{- '\n\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
|
| 144 |
+
{%- else %}
|
| 145 |
+
{{- '<tool_call>\n<function=' + tool_call.name + '>\n' }}
|
| 146 |
+
{%- endif %}
|
| 147 |
+
{%- else %}
|
| 148 |
+
{{- '\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
|
| 149 |
+
{%- endif %}
|
| 150 |
+
{%- if tool_call.arguments is defined %}
|
| 151 |
+
{%- for args_name, args_value in tool_call.arguments|items %}
|
| 152 |
+
{{- '<parameter=' + args_name + '>\n' }}
|
| 153 |
+
{%- set args_value = args_value | string if args_value is string else args_value | tojson | safe %}
|
| 154 |
+
{{- args_value }}
|
| 155 |
+
{{- '\n</parameter>\n' }}
|
| 156 |
+
{%- endfor %}
|
| 157 |
+
{%- endif %}
|
| 158 |
+
{{- '</function>\n</tool_call>' }}
|
| 159 |
+
{%- endfor %}
|
| 160 |
+
{%- endif %}
|
| 161 |
+
{{- '<|im_end|>\n' }}
|
| 162 |
+
{%- elif message.role == "tool" %}
|
| 163 |
+
{%- if loop.previtem and loop.previtem.role != "tool" %}
|
| 164 |
+
{{- '<|im_start|>user' }}
|
| 165 |
+
{%- endif %}
|
| 166 |
+
{{- '\n<tool_response>\n' }}
|
| 167 |
+
{{- content }}
|
| 168 |
+
{{- '\n</tool_response>' }}
|
| 169 |
+
{%- if not loop.last and loop.nextitem.role != "tool" %}
|
| 170 |
+
{{- '<|im_end|>\n' }}
|
| 171 |
+
{%- elif loop.last %}
|
| 172 |
+
{{- '<|im_end|>\n' }}
|
| 173 |
+
{%- endif %}
|
| 174 |
+
{%- else %}
|
| 175 |
+
{{- raise_exception('Unexpected message role.') }}
|
| 176 |
+
{%- endif %}
|
| 177 |
+
{%- endfor %}
|
| 178 |
+
{%- if add_generation_prompt %}
|
| 179 |
+
{{- '<|im_start|>assistant\n' }}
|
| 180 |
+
{%- if enable_thinking is defined and enable_thinking is false %}
|
| 181 |
+
{{- '<think>\n\n</think>\n\n' }}
|
| 182 |
+
{%- else %}
|
| 183 |
+
{{- '<think>\n' }}
|
| 184 |
+
{%- endif %}
|
| 185 |
+
{%- endif %}
|
config.json
ADDED
|
@@ -0,0 +1,122 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"Qwen3_5MoeForConditionalGeneration"
|
| 4 |
+
],
|
| 5 |
+
"dtype": "bfloat16",
|
| 6 |
+
"hidden_size": 2048,
|
| 7 |
+
"image_token_id": 248056,
|
| 8 |
+
"model_type": "qwen3_5_moe",
|
| 9 |
+
"text_config": {
|
| 10 |
+
"attention_bias": false,
|
| 11 |
+
"attention_dropout": 0.0,
|
| 12 |
+
"attn_output_gate": true,
|
| 13 |
+
"bos_token_id": 248044,
|
| 14 |
+
"dtype": "bfloat16",
|
| 15 |
+
"eos_token_id": 248044,
|
| 16 |
+
"full_attention_interval": 4,
|
| 17 |
+
"head_dim": 256,
|
| 18 |
+
"hidden_act": "silu",
|
| 19 |
+
"hidden_size": 2048,
|
| 20 |
+
"initializer_range": 0.02,
|
| 21 |
+
"layer_types": [
|
| 22 |
+
"linear_attention",
|
| 23 |
+
"linear_attention",
|
| 24 |
+
"linear_attention",
|
| 25 |
+
"full_attention",
|
| 26 |
+
"linear_attention",
|
| 27 |
+
"linear_attention",
|
| 28 |
+
"linear_attention",
|
| 29 |
+
"full_attention",
|
| 30 |
+
"linear_attention",
|
| 31 |
+
"linear_attention",
|
| 32 |
+
"linear_attention",
|
| 33 |
+
"full_attention",
|
| 34 |
+
"linear_attention",
|
| 35 |
+
"linear_attention",
|
| 36 |
+
"linear_attention",
|
| 37 |
+
"full_attention",
|
| 38 |
+
"linear_attention",
|
| 39 |
+
"linear_attention",
|
| 40 |
+
"linear_attention",
|
| 41 |
+
"full_attention",
|
| 42 |
+
"linear_attention",
|
| 43 |
+
"linear_attention",
|
| 44 |
+
"linear_attention",
|
| 45 |
+
"full_attention",
|
| 46 |
+
"linear_attention",
|
| 47 |
+
"linear_attention",
|
| 48 |
+
"linear_attention",
|
| 49 |
+
"full_attention",
|
| 50 |
+
"linear_attention",
|
| 51 |
+
"linear_attention",
|
| 52 |
+
"linear_attention",
|
| 53 |
+
"full_attention",
|
| 54 |
+
"linear_attention",
|
| 55 |
+
"linear_attention",
|
| 56 |
+
"linear_attention",
|
| 57 |
+
"full_attention",
|
| 58 |
+
"linear_attention",
|
| 59 |
+
"linear_attention",
|
| 60 |
+
"linear_attention",
|
| 61 |
+
"full_attention"
|
| 62 |
+
],
|
| 63 |
+
"linear_conv_kernel_dim": 4,
|
| 64 |
+
"linear_key_head_dim": 128,
|
| 65 |
+
"linear_num_key_heads": 16,
|
| 66 |
+
"linear_num_value_heads": 32,
|
| 67 |
+
"linear_value_head_dim": 128,
|
| 68 |
+
"mamba_ssm_dtype": "float32",
|
| 69 |
+
"max_position_embeddings": 262144,
|
| 70 |
+
"model_type": "qwen3_5_moe_text",
|
| 71 |
+
"moe_intermediate_size": 512,
|
| 72 |
+
"mtp_num_hidden_layers": 1,
|
| 73 |
+
"mtp_use_dedicated_embeddings": false,
|
| 74 |
+
"num_attention_heads": 16,
|
| 75 |
+
"num_experts": 256,
|
| 76 |
+
"num_experts_per_tok": 8,
|
| 77 |
+
"num_hidden_layers": 40,
|
| 78 |
+
"num_key_value_heads": 2,
|
| 79 |
+
"output_router_logits": false,
|
| 80 |
+
"pad_token_id": 248044,
|
| 81 |
+
"partial_rotary_factor": 0.25,
|
| 82 |
+
"rms_norm_eps": 1e-06,
|
| 83 |
+
"rope_parameters": {
|
| 84 |
+
"mrope_interleaved": true,
|
| 85 |
+
"mrope_section": [
|
| 86 |
+
11,
|
| 87 |
+
11,
|
| 88 |
+
10
|
| 89 |
+
],
|
| 90 |
+
"partial_rotary_factor": 0.25,
|
| 91 |
+
"rope_theta": 10000000,
|
| 92 |
+
"rope_type": "default"
|
| 93 |
+
},
|
| 94 |
+
"router_aux_loss_coef": 0.001,
|
| 95 |
+
"shared_expert_intermediate_size": 512,
|
| 96 |
+
"tie_word_embeddings": false,
|
| 97 |
+
"use_cache": true,
|
| 98 |
+
"vocab_size": 248320
|
| 99 |
+
},
|
| 100 |
+
"tie_word_embeddings": false,
|
| 101 |
+
"transformers_version": "5.15.0",
|
| 102 |
+
"video_token_id": 248057,
|
| 103 |
+
"vision_config": {
|
| 104 |
+
"deepstack_visual_indexes": [],
|
| 105 |
+
"depth": 27,
|
| 106 |
+
"dtype": "bfloat16",
|
| 107 |
+
"hidden_act": "gelu_pytorch_tanh",
|
| 108 |
+
"hidden_size": 1152,
|
| 109 |
+
"in_channels": 3,
|
| 110 |
+
"initializer_range": 0.02,
|
| 111 |
+
"intermediate_size": 4304,
|
| 112 |
+
"model_type": "qwen3_5_moe_vision",
|
| 113 |
+
"num_heads": 16,
|
| 114 |
+
"num_position_embeddings": 2304,
|
| 115 |
+
"out_hidden_size": 2048,
|
| 116 |
+
"patch_size": 16,
|
| 117 |
+
"spatial_merge_size": 2,
|
| 118 |
+
"temporal_patch_size": 2
|
| 119 |
+
},
|
| 120 |
+
"vision_end_token_id": 248054,
|
| 121 |
+
"vision_start_token_id": 248053
|
| 122 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 248044,
|
| 3 |
+
"do_sample": true,
|
| 4 |
+
"eos_token_id": [
|
| 5 |
+
248046,
|
| 6 |
+
248044
|
| 7 |
+
],
|
| 8 |
+
"pad_token_id": 248044,
|
| 9 |
+
"temperature": 1.0,
|
| 10 |
+
"top_k": 20,
|
| 11 |
+
"top_p": 0.95,
|
| 12 |
+
"transformers_version": "5.14.1"
|
| 13 |
+
}
|
merge_meta.json
ADDED
|
@@ -0,0 +1,161 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"base": "/root/hf/hub/models--ammazon--Affine-5dvqtektxx-sbs-v5/snapshots/64d83242312798db6eb82a3c9ee842097933a8ad",
|
| 3 |
+
"adapter": "/root/r496/train/adapter",
|
| 4 |
+
"out": "/tmp/r496_merged",
|
| 5 |
+
"device_map": "auto",
|
| 6 |
+
"cuda_visible": "0,1",
|
| 7 |
+
"base_first_shard": "/root/hf/hub/models--ammazon--Affine-5dvqtektxx-sbs-v5/snapshots/64d83242312798db6eb82a3c9ee842097933a8ad/model-00001-of-00016.safetensors",
|
| 8 |
+
"out_first_shard": "/tmp/r496_merged/model-00001-of-00016.safetensors",
|
| 9 |
+
"base_first_1MiB_sha256": "fb3c02c9f1f078baf274ecd4808796619eee1788ab71ec24b424bc39de87eecf",
|
| 10 |
+
"out_first_1MiB_sha256": "ddfbc2f3dd9e53957aed108f23b33efa1628bd4a1825f8b569cd1fc1b2a5d123",
|
| 11 |
+
"first_1MiB_identical": false,
|
| 12 |
+
"shard_windows": {
|
| 13 |
+
"model-00001-of-00016.safetensors": {
|
| 14 |
+
"size_base": 7108340968,
|
| 15 |
+
"size_out": 4323955448,
|
| 16 |
+
"equal": {
|
| 17 |
+
"head": false,
|
| 18 |
+
"mid": false,
|
| 19 |
+
"tail": false
|
| 20 |
+
}
|
| 21 |
+
},
|
| 22 |
+
"model-00002-of-00016.safetensors": {
|
| 23 |
+
"size_base": 2234834952,
|
| 24 |
+
"size_out": 4506431768,
|
| 25 |
+
"equal": {
|
| 26 |
+
"head": false,
|
| 27 |
+
"mid": false,
|
| 28 |
+
"tail": false
|
| 29 |
+
}
|
| 30 |
+
},
|
| 31 |
+
"model-00003-of-00016.safetensors": {
|
| 32 |
+
"size_base": 2980573480,
|
| 33 |
+
"size_out": 4988775056,
|
| 34 |
+
"equal": {
|
| 35 |
+
"head": false,
|
| 36 |
+
"mid": false,
|
| 37 |
+
"tail": false
|
| 38 |
+
}
|
| 39 |
+
},
|
| 40 |
+
"model-00004-of-00016.safetensors": {
|
| 41 |
+
"size_base": 5328721104,
|
| 42 |
+
"size_out": 3962207584,
|
| 43 |
+
"equal": {
|
| 44 |
+
"head": false,
|
| 45 |
+
"mid": false,
|
| 46 |
+
"tail": false
|
| 47 |
+
}
|
| 48 |
+
},
|
| 49 |
+
"model-00005-of-00016.safetensors": {
|
| 50 |
+
"size_base": 6832103168,
|
| 51 |
+
"size_out": 4506431736,
|
| 52 |
+
"equal": {
|
| 53 |
+
"head": false,
|
| 54 |
+
"mid": false,
|
| 55 |
+
"tail": false
|
| 56 |
+
}
|
| 57 |
+
},
|
| 58 |
+
"model-00006-of-00016.safetensors": {
|
| 59 |
+
"size_base": 3625335496,
|
| 60 |
+
"size_out": 4988775104,
|
| 61 |
+
"equal": {
|
| 62 |
+
"head": false,
|
| 63 |
+
"mid": false,
|
| 64 |
+
"tail": false
|
| 65 |
+
}
|
| 66 |
+
},
|
| 67 |
+
"model-00007-of-00016.safetensors": {
|
| 68 |
+
"size_base": 2389439944,
|
| 69 |
+
"size_out": 3962207632,
|
| 70 |
+
"equal": {
|
| 71 |
+
"head": false,
|
| 72 |
+
"mid": false,
|
| 73 |
+
"tail": false
|
| 74 |
+
}
|
| 75 |
+
},
|
| 76 |
+
"model-00008-of-00016.safetensors": {
|
| 77 |
+
"size_base": 6696032728,
|
| 78 |
+
"size_out": 4506431816,
|
| 79 |
+
"equal": {
|
| 80 |
+
"head": false,
|
| 81 |
+
"mid": false,
|
| 82 |
+
"tail": false
|
| 83 |
+
}
|
| 84 |
+
},
|
| 85 |
+
"model-00009-of-00016.safetensors": {
|
| 86 |
+
"size_base": 4266722672,
|
| 87 |
+
"size_out": 4988775104,
|
| 88 |
+
"equal": {
|
| 89 |
+
"head": false,
|
| 90 |
+
"mid": false,
|
| 91 |
+
"tail": false
|
| 92 |
+
}
|
| 93 |
+
},
|
| 94 |
+
"model-00010-of-00016.safetensors": {
|
| 95 |
+
"size_base": 7701193024,
|
| 96 |
+
"size_out": 3962207632,
|
| 97 |
+
"equal": {
|
| 98 |
+
"head": false,
|
| 99 |
+
"mid": false,
|
| 100 |
+
"tail": false
|
| 101 |
+
}
|
| 102 |
+
},
|
| 103 |
+
"model-00011-of-00016.safetensors": {
|
| 104 |
+
"size_base": 1699989616,
|
| 105 |
+
"size_out": 4506431816,
|
| 106 |
+
"equal": {
|
| 107 |
+
"head": false,
|
| 108 |
+
"mid": false,
|
| 109 |
+
"tail": false
|
| 110 |
+
}
|
| 111 |
+
},
|
| 112 |
+
"model-00012-of-00016.safetensors": {
|
| 113 |
+
"size_base": 1157632104,
|
| 114 |
+
"size_out": 4988775104,
|
| 115 |
+
"equal": {
|
| 116 |
+
"head": false,
|
| 117 |
+
"mid": false,
|
| 118 |
+
"tail": false
|
| 119 |
+
}
|
| 120 |
+
},
|
| 121 |
+
"model-00013-of-00016.safetensors": {
|
| 122 |
+
"size_base": 4332140784,
|
| 123 |
+
"size_out": 3962207632,
|
| 124 |
+
"equal": {
|
| 125 |
+
"head": false,
|
| 126 |
+
"mid": false,
|
| 127 |
+
"tail": false
|
| 128 |
+
}
|
| 129 |
+
},
|
| 130 |
+
"model-00014-of-00016.safetensors": {
|
| 131 |
+
"size_base": 3302671368,
|
| 132 |
+
"size_out": 4506431816,
|
| 133 |
+
"equal": {
|
| 134 |
+
"head": false,
|
| 135 |
+
"mid": false,
|
| 136 |
+
"tail": false
|
| 137 |
+
}
|
| 138 |
+
},
|
| 139 |
+
"model-00015-of-00016.safetensors": {
|
| 140 |
+
"size_base": 3342302960,
|
| 141 |
+
"size_out": 4988775104,
|
| 142 |
+
"equal": {
|
| 143 |
+
"head": false,
|
| 144 |
+
"mid": false,
|
| 145 |
+
"tail": false
|
| 146 |
+
}
|
| 147 |
+
},
|
| 148 |
+
"model-00016-of-00016.safetensors": {
|
| 149 |
+
"size_base": 7216459280,
|
| 150 |
+
"size_out": 1672494048,
|
| 151 |
+
"equal": {
|
| 152 |
+
"head": false,
|
| 153 |
+
"mid": false,
|
| 154 |
+
"tail": false
|
| 155 |
+
}
|
| 156 |
+
}
|
| 157 |
+
},
|
| 158 |
+
"weight_identical": false,
|
| 159 |
+
"elapsed_s": 136.1629137992859,
|
| 160 |
+
"finished_utc": "2026-08-16T06:10:16Z"
|
| 161 |
+
}
|
model-00001-of-00016.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e938b44f415e0f6a11e823ad8f2e9c92e8eba7c060a6a1ecf6909589e11443a3
|
| 3 |
+
size 4323955448
|
model-00002-of-00016.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:830bf54ba87b1cbd80edb56663b1f0981c7259047dbcb6b65d66a3eaef82c6e7
|
| 3 |
+
size 4506431768
|
model-00003-of-00016.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e3af827aa0a07c6d91dc4907f37fc63d17162c77de65530b31d4ec8c54d25ad0
|
| 3 |
+
size 4988775056
|
model-00004-of-00016.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3ba2adcecd010abaa9dba58a68021c65fe59ad14a86af7444a2fd3555ab25f73
|
| 3 |
+
size 3962207584
|
model-00005-of-00016.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:317d893bff3c1cce20c37190e900da0b553df7e715d45585b649d3a3b29d2d21
|
| 3 |
+
size 4506431736
|
model-00006-of-00016.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8fe35a6216693f2384eb97ef7b9440ba830a3260b6d709760078eee7399cd9d8
|
| 3 |
+
size 4988775104
|
model-00007-of-00016.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3f7bfb8a5776e303036c37d9e89fb23c158a27d9138b823bbc85b7ef1463e5c1
|
| 3 |
+
size 3962207632
|
model-00008-of-00016.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:55bf42325adcaa3cc03ac9af15b4c378296ac88de680a07b07b54b836fa6be02
|
| 3 |
+
size 4506431816
|
model-00009-of-00016.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b98aa6bbdd751994e07a67ad3f7454665b49c85e8b3a452f6555cb2f346b5f57
|
| 3 |
+
size 4988775104
|
model-00010-of-00016.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9d488786309eeceb66e2200a2666fb9f7bbcbe005cc96855956486b18dec0add
|
| 3 |
+
size 3962207632
|
model-00011-of-00016.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7414182436a3486f07fee135486e3f67d817a05a685c5afb26b46e9a8f82e3e1
|
| 3 |
+
size 4506431816
|
model-00012-of-00016.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1b8e7adcfbd049f5aec4d43a5284e2c980b775d07f0c7f25f9ec8ce2ab451e62
|
| 3 |
+
size 4988775104
|
model-00013-of-00016.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0eeb1f62f6a5dc4c40f9fd9aad160c3afa5986dee0eda6e431ff817b3582070e
|
| 3 |
+
size 3962207632
|
model-00014-of-00016.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c78512efc54e61be7433b3cc9ca950b14da487eb03bf870deac2faebed565221
|
| 3 |
+
size 4506431816
|
model-00015-of-00016.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dc090bac6875e5176a914e112c9ba9c59ede1b3ed4959f967fea569af0693428
|
| 3 |
+
size 4988775104
|
model-00016-of-00016.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e5d6a77de8713d18dd5c042c827d5975562f9227a7a8b18f1e485d766163d694
|
| 3 |
+
size 1672494048
|
model-visual-restored.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1e1b1d5f576d65f25427a6d8e89e2c76f4aea6a9383f17dd760dc420b233872a
|
| 3 |
+
size 893179504
|
model.safetensors.index.json
ADDED
|
@@ -0,0 +1,1034 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"metadata": {
|
| 3 |
+
"total_parameters": 35107181936,
|
| 4 |
+
"total_size": 70214493904
|
| 5 |
+
},
|
| 6 |
+
"weight_map": {
|
| 7 |
+
"lm_head.weight": "model-00001-of-00016.safetensors",
|
| 8 |
+
"model.language_model.embed_tokens.weight": "model-00001-of-00016.safetensors",
|
| 9 |
+
"model.language_model.layers.0.input_layernorm.weight": "model-00001-of-00016.safetensors",
|
| 10 |
+
"model.language_model.layers.0.linear_attn.A_log": "model-00001-of-00016.safetensors",
|
| 11 |
+
"model.language_model.layers.0.linear_attn.conv1d.weight": "model-00001-of-00016.safetensors",
|
| 12 |
+
"model.language_model.layers.0.linear_attn.dt_bias": "model-00001-of-00016.safetensors",
|
| 13 |
+
"model.language_model.layers.0.linear_attn.in_proj_a.weight": "model-00001-of-00016.safetensors",
|
| 14 |
+
"model.language_model.layers.0.linear_attn.in_proj_b.weight": "model-00001-of-00016.safetensors",
|
| 15 |
+
"model.language_model.layers.0.linear_attn.in_proj_qkv.weight": "model-00001-of-00016.safetensors",
|
| 16 |
+
"model.language_model.layers.0.linear_attn.in_proj_z.weight": "model-00001-of-00016.safetensors",
|
| 17 |
+
"model.language_model.layers.0.linear_attn.norm.weight": "model-00001-of-00016.safetensors",
|
| 18 |
+
"model.language_model.layers.0.linear_attn.out_proj.weight": "model-00001-of-00016.safetensors",
|
| 19 |
+
"model.language_model.layers.0.mlp.experts.down_proj": "model-00001-of-00016.safetensors",
|
| 20 |
+
"model.language_model.layers.0.mlp.experts.gate_up_proj": "model-00001-of-00016.safetensors",
|
| 21 |
+
"model.language_model.layers.0.mlp.gate.weight": "model-00001-of-00016.safetensors",
|
| 22 |
+
"model.language_model.layers.0.mlp.shared_expert.down_proj.weight": "model-00001-of-00016.safetensors",
|
| 23 |
+
"model.language_model.layers.0.mlp.shared_expert.gate_proj.weight": "model-00001-of-00016.safetensors",
|
| 24 |
+
"model.language_model.layers.0.mlp.shared_expert.up_proj.weight": "model-00001-of-00016.safetensors",
|
| 25 |
+
"model.language_model.layers.0.mlp.shared_expert_gate.weight": "model-00001-of-00016.safetensors",
|
| 26 |
+
"model.language_model.layers.0.post_attention_layernorm.weight": "model-00001-of-00016.safetensors",
|
| 27 |
+
"model.language_model.layers.1.input_layernorm.weight": "model-00001-of-00016.safetensors",
|
| 28 |
+
"model.language_model.layers.1.linear_attn.A_log": "model-00001-of-00016.safetensors",
|
| 29 |
+
"model.language_model.layers.1.linear_attn.conv1d.weight": "model-00001-of-00016.safetensors",
|
| 30 |
+
"model.language_model.layers.1.linear_attn.dt_bias": "model-00001-of-00016.safetensors",
|
| 31 |
+
"model.language_model.layers.1.linear_attn.in_proj_a.weight": "model-00001-of-00016.safetensors",
|
| 32 |
+
"model.language_model.layers.1.linear_attn.in_proj_b.weight": "model-00001-of-00016.safetensors",
|
| 33 |
+
"model.language_model.layers.1.linear_attn.in_proj_qkv.weight": "model-00001-of-00016.safetensors",
|
| 34 |
+
"model.language_model.layers.1.linear_attn.in_proj_z.weight": "model-00001-of-00016.safetensors",
|
| 35 |
+
"model.language_model.layers.1.linear_attn.norm.weight": "model-00001-of-00016.safetensors",
|
| 36 |
+
"model.language_model.layers.1.linear_attn.out_proj.weight": "model-00001-of-00016.safetensors",
|
| 37 |
+
"model.language_model.layers.1.mlp.experts.down_proj": "model-00001-of-00016.safetensors",
|
| 38 |
+
"model.language_model.layers.1.mlp.experts.gate_up_proj": "model-00002-of-00016.safetensors",
|
| 39 |
+
"model.language_model.layers.1.mlp.gate.weight": "model-00002-of-00016.safetensors",
|
| 40 |
+
"model.language_model.layers.1.mlp.shared_expert.down_proj.weight": "model-00002-of-00016.safetensors",
|
| 41 |
+
"model.language_model.layers.1.mlp.shared_expert.gate_proj.weight": "model-00002-of-00016.safetensors",
|
| 42 |
+
"model.language_model.layers.1.mlp.shared_expert.up_proj.weight": "model-00002-of-00016.safetensors",
|
| 43 |
+
"model.language_model.layers.1.mlp.shared_expert_gate.weight": "model-00002-of-00016.safetensors",
|
| 44 |
+
"model.language_model.layers.1.post_attention_layernorm.weight": "model-00002-of-00016.safetensors",
|
| 45 |
+
"model.language_model.layers.10.input_layernorm.weight": "model-00005-of-00016.safetensors",
|
| 46 |
+
"model.language_model.layers.10.linear_attn.A_log": "model-00005-of-00016.safetensors",
|
| 47 |
+
"model.language_model.layers.10.linear_attn.conv1d.weight": "model-00005-of-00016.safetensors",
|
| 48 |
+
"model.language_model.layers.10.linear_attn.dt_bias": "model-00005-of-00016.safetensors",
|
| 49 |
+
"model.language_model.layers.10.linear_attn.in_proj_a.weight": "model-00005-of-00016.safetensors",
|
| 50 |
+
"model.language_model.layers.10.linear_attn.in_proj_b.weight": "model-00005-of-00016.safetensors",
|
| 51 |
+
"model.language_model.layers.10.linear_attn.in_proj_qkv.weight": "model-00005-of-00016.safetensors",
|
| 52 |
+
"model.language_model.layers.10.linear_attn.in_proj_z.weight": "model-00005-of-00016.safetensors",
|
| 53 |
+
"model.language_model.layers.10.linear_attn.norm.weight": "model-00005-of-00016.safetensors",
|
| 54 |
+
"model.language_model.layers.10.linear_attn.out_proj.weight": "model-00005-of-00016.safetensors",
|
| 55 |
+
"model.language_model.layers.10.mlp.experts.down_proj": "model-00005-of-00016.safetensors",
|
| 56 |
+
"model.language_model.layers.10.mlp.experts.gate_up_proj": "model-00005-of-00016.safetensors",
|
| 57 |
+
"model.language_model.layers.10.mlp.gate.weight": "model-00005-of-00016.safetensors",
|
| 58 |
+
"model.language_model.layers.10.mlp.shared_expert.down_proj.weight": "model-00005-of-00016.safetensors",
|
| 59 |
+
"model.language_model.layers.10.mlp.shared_expert.gate_proj.weight": "model-00005-of-00016.safetensors",
|
| 60 |
+
"model.language_model.layers.10.mlp.shared_expert.up_proj.weight": "model-00005-of-00016.safetensors",
|
| 61 |
+
"model.language_model.layers.10.mlp.shared_expert_gate.weight": "model-00005-of-00016.safetensors",
|
| 62 |
+
"model.language_model.layers.10.post_attention_layernorm.weight": "model-00005-of-00016.safetensors",
|
| 63 |
+
"model.language_model.layers.11.input_layernorm.weight": "model-00005-of-00016.safetensors",
|
| 64 |
+
"model.language_model.layers.11.mlp.experts.down_proj": "model-00005-of-00016.safetensors",
|
| 65 |
+
"model.language_model.layers.11.mlp.experts.gate_up_proj": "model-00005-of-00016.safetensors",
|
| 66 |
+
"model.language_model.layers.11.mlp.gate.weight": "model-00005-of-00016.safetensors",
|
| 67 |
+
"model.language_model.layers.11.mlp.shared_expert.down_proj.weight": "model-00005-of-00016.safetensors",
|
| 68 |
+
"model.language_model.layers.11.mlp.shared_expert.gate_proj.weight": "model-00005-of-00016.safetensors",
|
| 69 |
+
"model.language_model.layers.11.mlp.shared_expert.up_proj.weight": "model-00005-of-00016.safetensors",
|
| 70 |
+
"model.language_model.layers.11.mlp.shared_expert_gate.weight": "model-00005-of-00016.safetensors",
|
| 71 |
+
"model.language_model.layers.11.post_attention_layernorm.weight": "model-00005-of-00016.safetensors",
|
| 72 |
+
"model.language_model.layers.11.self_attn.k_norm.weight": "model-00005-of-00016.safetensors",
|
| 73 |
+
"model.language_model.layers.11.self_attn.k_proj.weight": "model-00005-of-00016.safetensors",
|
| 74 |
+
"model.language_model.layers.11.self_attn.o_proj.weight": "model-00005-of-00016.safetensors",
|
| 75 |
+
"model.language_model.layers.11.self_attn.q_norm.weight": "model-00005-of-00016.safetensors",
|
| 76 |
+
"model.language_model.layers.11.self_attn.q_proj.weight": "model-00005-of-00016.safetensors",
|
| 77 |
+
"model.language_model.layers.11.self_attn.v_proj.weight": "model-00005-of-00016.safetensors",
|
| 78 |
+
"model.language_model.layers.12.input_layernorm.weight": "model-00005-of-00016.safetensors",
|
| 79 |
+
"model.language_model.layers.12.linear_attn.A_log": "model-00005-of-00016.safetensors",
|
| 80 |
+
"model.language_model.layers.12.linear_attn.conv1d.weight": "model-00005-of-00016.safetensors",
|
| 81 |
+
"model.language_model.layers.12.linear_attn.dt_bias": "model-00005-of-00016.safetensors",
|
| 82 |
+
"model.language_model.layers.12.linear_attn.in_proj_a.weight": "model-00005-of-00016.safetensors",
|
| 83 |
+
"model.language_model.layers.12.linear_attn.in_proj_b.weight": "model-00005-of-00016.safetensors",
|
| 84 |
+
"model.language_model.layers.12.linear_attn.in_proj_qkv.weight": "model-00005-of-00016.safetensors",
|
| 85 |
+
"model.language_model.layers.12.linear_attn.in_proj_z.weight": "model-00005-of-00016.safetensors",
|
| 86 |
+
"model.language_model.layers.12.linear_attn.norm.weight": "model-00005-of-00016.safetensors",
|
| 87 |
+
"model.language_model.layers.12.linear_attn.out_proj.weight": "model-00005-of-00016.safetensors",
|
| 88 |
+
"model.language_model.layers.12.mlp.experts.down_proj": "model-00006-of-00016.safetensors",
|
| 89 |
+
"model.language_model.layers.12.mlp.experts.gate_up_proj": "model-00006-of-00016.safetensors",
|
| 90 |
+
"model.language_model.layers.12.mlp.gate.weight": "model-00006-of-00016.safetensors",
|
| 91 |
+
"model.language_model.layers.12.mlp.shared_expert.down_proj.weight": "model-00006-of-00016.safetensors",
|
| 92 |
+
"model.language_model.layers.12.mlp.shared_expert.gate_proj.weight": "model-00006-of-00016.safetensors",
|
| 93 |
+
"model.language_model.layers.12.mlp.shared_expert.up_proj.weight": "model-00006-of-00016.safetensors",
|
| 94 |
+
"model.language_model.layers.12.mlp.shared_expert_gate.weight": "model-00006-of-00016.safetensors",
|
| 95 |
+
"model.language_model.layers.12.post_attention_layernorm.weight": "model-00006-of-00016.safetensors",
|
| 96 |
+
"model.language_model.layers.13.input_layernorm.weight": "model-00006-of-00016.safetensors",
|
| 97 |
+
"model.language_model.layers.13.linear_attn.A_log": "model-00006-of-00016.safetensors",
|
| 98 |
+
"model.language_model.layers.13.linear_attn.conv1d.weight": "model-00006-of-00016.safetensors",
|
| 99 |
+
"model.language_model.layers.13.linear_attn.dt_bias": "model-00006-of-00016.safetensors",
|
| 100 |
+
"model.language_model.layers.13.linear_attn.in_proj_a.weight": "model-00006-of-00016.safetensors",
|
| 101 |
+
"model.language_model.layers.13.linear_attn.in_proj_b.weight": "model-00006-of-00016.safetensors",
|
| 102 |
+
"model.language_model.layers.13.linear_attn.in_proj_qkv.weight": "model-00006-of-00016.safetensors",
|
| 103 |
+
"model.language_model.layers.13.linear_attn.in_proj_z.weight": "model-00006-of-00016.safetensors",
|
| 104 |
+
"model.language_model.layers.13.linear_attn.norm.weight": "model-00006-of-00016.safetensors",
|
| 105 |
+
"model.language_model.layers.13.linear_attn.out_proj.weight": "model-00006-of-00016.safetensors",
|
| 106 |
+
"model.language_model.layers.13.mlp.experts.down_proj": "model-00006-of-00016.safetensors",
|
| 107 |
+
"model.language_model.layers.13.mlp.experts.gate_up_proj": "model-00006-of-00016.safetensors",
|
| 108 |
+
"model.language_model.layers.13.mlp.gate.weight": "model-00006-of-00016.safetensors",
|
| 109 |
+
"model.language_model.layers.13.mlp.shared_expert.down_proj.weight": "model-00006-of-00016.safetensors",
|
| 110 |
+
"model.language_model.layers.13.mlp.shared_expert.gate_proj.weight": "model-00006-of-00016.safetensors",
|
| 111 |
+
"model.language_model.layers.13.mlp.shared_expert.up_proj.weight": "model-00006-of-00016.safetensors",
|
| 112 |
+
"model.language_model.layers.13.mlp.shared_expert_gate.weight": "model-00006-of-00016.safetensors",
|
| 113 |
+
"model.language_model.layers.13.post_attention_layernorm.weight": "model-00006-of-00016.safetensors",
|
| 114 |
+
"model.language_model.layers.14.input_layernorm.weight": "model-00006-of-00016.safetensors",
|
| 115 |
+
"model.language_model.layers.14.linear_attn.A_log": "model-00006-of-00016.safetensors",
|
| 116 |
+
"model.language_model.layers.14.linear_attn.conv1d.weight": "model-00006-of-00016.safetensors",
|
| 117 |
+
"model.language_model.layers.14.linear_attn.dt_bias": "model-00006-of-00016.safetensors",
|
| 118 |
+
"model.language_model.layers.14.linear_attn.in_proj_a.weight": "model-00006-of-00016.safetensors",
|
| 119 |
+
"model.language_model.layers.14.linear_attn.in_proj_b.weight": "model-00006-of-00016.safetensors",
|
| 120 |
+
"model.language_model.layers.14.linear_attn.in_proj_qkv.weight": "model-00006-of-00016.safetensors",
|
| 121 |
+
"model.language_model.layers.14.linear_attn.in_proj_z.weight": "model-00006-of-00016.safetensors",
|
| 122 |
+
"model.language_model.layers.14.linear_attn.norm.weight": "model-00006-of-00016.safetensors",
|
| 123 |
+
"model.language_model.layers.14.linear_attn.out_proj.weight": "model-00006-of-00016.safetensors",
|
| 124 |
+
"model.language_model.layers.14.mlp.experts.down_proj": "model-00006-of-00016.safetensors",
|
| 125 |
+
"model.language_model.layers.14.mlp.experts.gate_up_proj": "model-00006-of-00016.safetensors",
|
| 126 |
+
"model.language_model.layers.14.mlp.gate.weight": "model-00006-of-00016.safetensors",
|
| 127 |
+
"model.language_model.layers.14.mlp.shared_expert.down_proj.weight": "model-00006-of-00016.safetensors",
|
| 128 |
+
"model.language_model.layers.14.mlp.shared_expert.gate_proj.weight": "model-00006-of-00016.safetensors",
|
| 129 |
+
"model.language_model.layers.14.mlp.shared_expert.up_proj.weight": "model-00006-of-00016.safetensors",
|
| 130 |
+
"model.language_model.layers.14.mlp.shared_expert_gate.weight": "model-00006-of-00016.safetensors",
|
| 131 |
+
"model.language_model.layers.14.post_attention_layernorm.weight": "model-00006-of-00016.safetensors",
|
| 132 |
+
"model.language_model.layers.15.input_layernorm.weight": "model-00006-of-00016.safetensors",
|
| 133 |
+
"model.language_model.layers.15.mlp.experts.down_proj": "model-00007-of-00016.safetensors",
|
| 134 |
+
"model.language_model.layers.15.mlp.experts.gate_up_proj": "model-00007-of-00016.safetensors",
|
| 135 |
+
"model.language_model.layers.15.mlp.gate.weight": "model-00007-of-00016.safetensors",
|
| 136 |
+
"model.language_model.layers.15.mlp.shared_expert.down_proj.weight": "model-00007-of-00016.safetensors",
|
| 137 |
+
"model.language_model.layers.15.mlp.shared_expert.gate_proj.weight": "model-00007-of-00016.safetensors",
|
| 138 |
+
"model.language_model.layers.15.mlp.shared_expert.up_proj.weight": "model-00007-of-00016.safetensors",
|
| 139 |
+
"model.language_model.layers.15.mlp.shared_expert_gate.weight": "model-00007-of-00016.safetensors",
|
| 140 |
+
"model.language_model.layers.15.post_attention_layernorm.weight": "model-00007-of-00016.safetensors",
|
| 141 |
+
"model.language_model.layers.15.self_attn.k_norm.weight": "model-00007-of-00016.safetensors",
|
| 142 |
+
"model.language_model.layers.15.self_attn.k_proj.weight": "model-00007-of-00016.safetensors",
|
| 143 |
+
"model.language_model.layers.15.self_attn.o_proj.weight": "model-00007-of-00016.safetensors",
|
| 144 |
+
"model.language_model.layers.15.self_attn.q_norm.weight": "model-00007-of-00016.safetensors",
|
| 145 |
+
"model.language_model.layers.15.self_attn.q_proj.weight": "model-00007-of-00016.safetensors",
|
| 146 |
+
"model.language_model.layers.15.self_attn.v_proj.weight": "model-00007-of-00016.safetensors",
|
| 147 |
+
"model.language_model.layers.16.input_layernorm.weight": "model-00007-of-00016.safetensors",
|
| 148 |
+
"model.language_model.layers.16.linear_attn.A_log": "model-00007-of-00016.safetensors",
|
| 149 |
+
"model.language_model.layers.16.linear_attn.conv1d.weight": "model-00007-of-00016.safetensors",
|
| 150 |
+
"model.language_model.layers.16.linear_attn.dt_bias": "model-00007-of-00016.safetensors",
|
| 151 |
+
"model.language_model.layers.16.linear_attn.in_proj_a.weight": "model-00007-of-00016.safetensors",
|
| 152 |
+
"model.language_model.layers.16.linear_attn.in_proj_b.weight": "model-00007-of-00016.safetensors",
|
| 153 |
+
"model.language_model.layers.16.linear_attn.in_proj_qkv.weight": "model-00007-of-00016.safetensors",
|
| 154 |
+
"model.language_model.layers.16.linear_attn.in_proj_z.weight": "model-00007-of-00016.safetensors",
|
| 155 |
+
"model.language_model.layers.16.linear_attn.norm.weight": "model-00007-of-00016.safetensors",
|
| 156 |
+
"model.language_model.layers.16.linear_attn.out_proj.weight": "model-00007-of-00016.safetensors",
|
| 157 |
+
"model.language_model.layers.16.mlp.experts.down_proj": "model-00007-of-00016.safetensors",
|
| 158 |
+
"model.language_model.layers.16.mlp.experts.gate_up_proj": "model-00007-of-00016.safetensors",
|
| 159 |
+
"model.language_model.layers.16.mlp.gate.weight": "model-00007-of-00016.safetensors",
|
| 160 |
+
"model.language_model.layers.16.mlp.shared_expert.down_proj.weight": "model-00007-of-00016.safetensors",
|
| 161 |
+
"model.language_model.layers.16.mlp.shared_expert.gate_proj.weight": "model-00007-of-00016.safetensors",
|
| 162 |
+
"model.language_model.layers.16.mlp.shared_expert.up_proj.weight": "model-00007-of-00016.safetensors",
|
| 163 |
+
"model.language_model.layers.16.mlp.shared_expert_gate.weight": "model-00007-of-00016.safetensors",
|
| 164 |
+
"model.language_model.layers.16.post_attention_layernorm.weight": "model-00007-of-00016.safetensors",
|
| 165 |
+
"model.language_model.layers.17.input_layernorm.weight": "model-00007-of-00016.safetensors",
|
| 166 |
+
"model.language_model.layers.17.linear_attn.A_log": "model-00007-of-00016.safetensors",
|
| 167 |
+
"model.language_model.layers.17.linear_attn.conv1d.weight": "model-00007-of-00016.safetensors",
|
| 168 |
+
"model.language_model.layers.17.linear_attn.dt_bias": "model-00007-of-00016.safetensors",
|
| 169 |
+
"model.language_model.layers.17.linear_attn.in_proj_a.weight": "model-00007-of-00016.safetensors",
|
| 170 |
+
"model.language_model.layers.17.linear_attn.in_proj_b.weight": "model-00007-of-00016.safetensors",
|
| 171 |
+
"model.language_model.layers.17.linear_attn.in_proj_qkv.weight": "model-00007-of-00016.safetensors",
|
| 172 |
+
"model.language_model.layers.17.linear_attn.in_proj_z.weight": "model-00007-of-00016.safetensors",
|
| 173 |
+
"model.language_model.layers.17.linear_attn.norm.weight": "model-00007-of-00016.safetensors",
|
| 174 |
+
"model.language_model.layers.17.linear_attn.out_proj.weight": "model-00007-of-00016.safetensors",
|
| 175 |
+
"model.language_model.layers.17.mlp.experts.down_proj": "model-00007-of-00016.safetensors",
|
| 176 |
+
"model.language_model.layers.17.mlp.experts.gate_up_proj": "model-00008-of-00016.safetensors",
|
| 177 |
+
"model.language_model.layers.17.mlp.gate.weight": "model-00008-of-00016.safetensors",
|
| 178 |
+
"model.language_model.layers.17.mlp.shared_expert.down_proj.weight": "model-00008-of-00016.safetensors",
|
| 179 |
+
"model.language_model.layers.17.mlp.shared_expert.gate_proj.weight": "model-00008-of-00016.safetensors",
|
| 180 |
+
"model.language_model.layers.17.mlp.shared_expert.up_proj.weight": "model-00008-of-00016.safetensors",
|
| 181 |
+
"model.language_model.layers.17.mlp.shared_expert_gate.weight": "model-00008-of-00016.safetensors",
|
| 182 |
+
"model.language_model.layers.17.post_attention_layernorm.weight": "model-00008-of-00016.safetensors",
|
| 183 |
+
"model.language_model.layers.18.input_layernorm.weight": "model-00008-of-00016.safetensors",
|
| 184 |
+
"model.language_model.layers.18.linear_attn.A_log": "model-00008-of-00016.safetensors",
|
| 185 |
+
"model.language_model.layers.18.linear_attn.conv1d.weight": "model-00008-of-00016.safetensors",
|
| 186 |
+
"model.language_model.layers.18.linear_attn.dt_bias": "model-00008-of-00016.safetensors",
|
| 187 |
+
"model.language_model.layers.18.linear_attn.in_proj_a.weight": "model-00008-of-00016.safetensors",
|
| 188 |
+
"model.language_model.layers.18.linear_attn.in_proj_b.weight": "model-00008-of-00016.safetensors",
|
| 189 |
+
"model.language_model.layers.18.linear_attn.in_proj_qkv.weight": "model-00008-of-00016.safetensors",
|
| 190 |
+
"model.language_model.layers.18.linear_attn.in_proj_z.weight": "model-00008-of-00016.safetensors",
|
| 191 |
+
"model.language_model.layers.18.linear_attn.norm.weight": "model-00008-of-00016.safetensors",
|
| 192 |
+
"model.language_model.layers.18.linear_attn.out_proj.weight": "model-00008-of-00016.safetensors",
|
| 193 |
+
"model.language_model.layers.18.mlp.experts.down_proj": "model-00008-of-00016.safetensors",
|
| 194 |
+
"model.language_model.layers.18.mlp.experts.gate_up_proj": "model-00008-of-00016.safetensors",
|
| 195 |
+
"model.language_model.layers.18.mlp.gate.weight": "model-00008-of-00016.safetensors",
|
| 196 |
+
"model.language_model.layers.18.mlp.shared_expert.down_proj.weight": "model-00008-of-00016.safetensors",
|
| 197 |
+
"model.language_model.layers.18.mlp.shared_expert.gate_proj.weight": "model-00008-of-00016.safetensors",
|
| 198 |
+
"model.language_model.layers.18.mlp.shared_expert.up_proj.weight": "model-00008-of-00016.safetensors",
|
| 199 |
+
"model.language_model.layers.18.mlp.shared_expert_gate.weight": "model-00008-of-00016.safetensors",
|
| 200 |
+
"model.language_model.layers.18.post_attention_layernorm.weight": "model-00008-of-00016.safetensors",
|
| 201 |
+
"model.language_model.layers.19.input_layernorm.weight": "model-00008-of-00016.safetensors",
|
| 202 |
+
"model.language_model.layers.19.mlp.experts.down_proj": "model-00008-of-00016.safetensors",
|
| 203 |
+
"model.language_model.layers.19.mlp.experts.gate_up_proj": "model-00008-of-00016.safetensors",
|
| 204 |
+
"model.language_model.layers.19.mlp.gate.weight": "model-00008-of-00016.safetensors",
|
| 205 |
+
"model.language_model.layers.19.mlp.shared_expert.down_proj.weight": "model-00008-of-00016.safetensors",
|
| 206 |
+
"model.language_model.layers.19.mlp.shared_expert.gate_proj.weight": "model-00008-of-00016.safetensors",
|
| 207 |
+
"model.language_model.layers.19.mlp.shared_expert.up_proj.weight": "model-00008-of-00016.safetensors",
|
| 208 |
+
"model.language_model.layers.19.mlp.shared_expert_gate.weight": "model-00008-of-00016.safetensors",
|
| 209 |
+
"model.language_model.layers.19.post_attention_layernorm.weight": "model-00008-of-00016.safetensors",
|
| 210 |
+
"model.language_model.layers.19.self_attn.k_norm.weight": "model-00008-of-00016.safetensors",
|
| 211 |
+
"model.language_model.layers.19.self_attn.k_proj.weight": "model-00008-of-00016.safetensors",
|
| 212 |
+
"model.language_model.layers.19.self_attn.o_proj.weight": "model-00008-of-00016.safetensors",
|
| 213 |
+
"model.language_model.layers.19.self_attn.q_norm.weight": "model-00008-of-00016.safetensors",
|
| 214 |
+
"model.language_model.layers.19.self_attn.q_proj.weight": "model-00008-of-00016.safetensors",
|
| 215 |
+
"model.language_model.layers.19.self_attn.v_proj.weight": "model-00008-of-00016.safetensors",
|
| 216 |
+
"model.language_model.layers.2.input_layernorm.weight": "model-00002-of-00016.safetensors",
|
| 217 |
+
"model.language_model.layers.2.linear_attn.A_log": "model-00002-of-00016.safetensors",
|
| 218 |
+
"model.language_model.layers.2.linear_attn.conv1d.weight": "model-00002-of-00016.safetensors",
|
| 219 |
+
"model.language_model.layers.2.linear_attn.dt_bias": "model-00002-of-00016.safetensors",
|
| 220 |
+
"model.language_model.layers.2.linear_attn.in_proj_a.weight": "model-00002-of-00016.safetensors",
|
| 221 |
+
"model.language_model.layers.2.linear_attn.in_proj_b.weight": "model-00002-of-00016.safetensors",
|
| 222 |
+
"model.language_model.layers.2.linear_attn.in_proj_qkv.weight": "model-00002-of-00016.safetensors",
|
| 223 |
+
"model.language_model.layers.2.linear_attn.in_proj_z.weight": "model-00002-of-00016.safetensors",
|
| 224 |
+
"model.language_model.layers.2.linear_attn.norm.weight": "model-00002-of-00016.safetensors",
|
| 225 |
+
"model.language_model.layers.2.linear_attn.out_proj.weight": "model-00002-of-00016.safetensors",
|
| 226 |
+
"model.language_model.layers.2.mlp.experts.down_proj": "model-00002-of-00016.safetensors",
|
| 227 |
+
"model.language_model.layers.2.mlp.experts.gate_up_proj": "model-00002-of-00016.safetensors",
|
| 228 |
+
"model.language_model.layers.2.mlp.gate.weight": "model-00002-of-00016.safetensors",
|
| 229 |
+
"model.language_model.layers.2.mlp.shared_expert.down_proj.weight": "model-00002-of-00016.safetensors",
|
| 230 |
+
"model.language_model.layers.2.mlp.shared_expert.gate_proj.weight": "model-00002-of-00016.safetensors",
|
| 231 |
+
"model.language_model.layers.2.mlp.shared_expert.up_proj.weight": "model-00002-of-00016.safetensors",
|
| 232 |
+
"model.language_model.layers.2.mlp.shared_expert_gate.weight": "model-00002-of-00016.safetensors",
|
| 233 |
+
"model.language_model.layers.2.post_attention_layernorm.weight": "model-00002-of-00016.safetensors",
|
| 234 |
+
"model.language_model.layers.20.input_layernorm.weight": "model-00008-of-00016.safetensors",
|
| 235 |
+
"model.language_model.layers.20.linear_attn.A_log": "model-00008-of-00016.safetensors",
|
| 236 |
+
"model.language_model.layers.20.linear_attn.conv1d.weight": "model-00008-of-00016.safetensors",
|
| 237 |
+
"model.language_model.layers.20.linear_attn.dt_bias": "model-00008-of-00016.safetensors",
|
| 238 |
+
"model.language_model.layers.20.linear_attn.in_proj_a.weight": "model-00008-of-00016.safetensors",
|
| 239 |
+
"model.language_model.layers.20.linear_attn.in_proj_b.weight": "model-00008-of-00016.safetensors",
|
| 240 |
+
"model.language_model.layers.20.linear_attn.in_proj_qkv.weight": "model-00008-of-00016.safetensors",
|
| 241 |
+
"model.language_model.layers.20.linear_attn.in_proj_z.weight": "model-00008-of-00016.safetensors",
|
| 242 |
+
"model.language_model.layers.20.linear_attn.norm.weight": "model-00008-of-00016.safetensors",
|
| 243 |
+
"model.language_model.layers.20.linear_attn.out_proj.weight": "model-00008-of-00016.safetensors",
|
| 244 |
+
"model.language_model.layers.20.mlp.experts.down_proj": "model-00009-of-00016.safetensors",
|
| 245 |
+
"model.language_model.layers.20.mlp.experts.gate_up_proj": "model-00009-of-00016.safetensors",
|
| 246 |
+
"model.language_model.layers.20.mlp.gate.weight": "model-00009-of-00016.safetensors",
|
| 247 |
+
"model.language_model.layers.20.mlp.shared_expert.down_proj.weight": "model-00009-of-00016.safetensors",
|
| 248 |
+
"model.language_model.layers.20.mlp.shared_expert.gate_proj.weight": "model-00009-of-00016.safetensors",
|
| 249 |
+
"model.language_model.layers.20.mlp.shared_expert.up_proj.weight": "model-00009-of-00016.safetensors",
|
| 250 |
+
"model.language_model.layers.20.mlp.shared_expert_gate.weight": "model-00009-of-00016.safetensors",
|
| 251 |
+
"model.language_model.layers.20.post_attention_layernorm.weight": "model-00009-of-00016.safetensors",
|
| 252 |
+
"model.language_model.layers.21.input_layernorm.weight": "model-00009-of-00016.safetensors",
|
| 253 |
+
"model.language_model.layers.21.linear_attn.A_log": "model-00009-of-00016.safetensors",
|
| 254 |
+
"model.language_model.layers.21.linear_attn.conv1d.weight": "model-00009-of-00016.safetensors",
|
| 255 |
+
"model.language_model.layers.21.linear_attn.dt_bias": "model-00009-of-00016.safetensors",
|
| 256 |
+
"model.language_model.layers.21.linear_attn.in_proj_a.weight": "model-00009-of-00016.safetensors",
|
| 257 |
+
"model.language_model.layers.21.linear_attn.in_proj_b.weight": "model-00009-of-00016.safetensors",
|
| 258 |
+
"model.language_model.layers.21.linear_attn.in_proj_qkv.weight": "model-00009-of-00016.safetensors",
|
| 259 |
+
"model.language_model.layers.21.linear_attn.in_proj_z.weight": "model-00009-of-00016.safetensors",
|
| 260 |
+
"model.language_model.layers.21.linear_attn.norm.weight": "model-00009-of-00016.safetensors",
|
| 261 |
+
"model.language_model.layers.21.linear_attn.out_proj.weight": "model-00009-of-00016.safetensors",
|
| 262 |
+
"model.language_model.layers.21.mlp.experts.down_proj": "model-00009-of-00016.safetensors",
|
| 263 |
+
"model.language_model.layers.21.mlp.experts.gate_up_proj": "model-00009-of-00016.safetensors",
|
| 264 |
+
"model.language_model.layers.21.mlp.gate.weight": "model-00009-of-00016.safetensors",
|
| 265 |
+
"model.language_model.layers.21.mlp.shared_expert.down_proj.weight": "model-00009-of-00016.safetensors",
|
| 266 |
+
"model.language_model.layers.21.mlp.shared_expert.gate_proj.weight": "model-00009-of-00016.safetensors",
|
| 267 |
+
"model.language_model.layers.21.mlp.shared_expert.up_proj.weight": "model-00009-of-00016.safetensors",
|
| 268 |
+
"model.language_model.layers.21.mlp.shared_expert_gate.weight": "model-00009-of-00016.safetensors",
|
| 269 |
+
"model.language_model.layers.21.post_attention_layernorm.weight": "model-00009-of-00016.safetensors",
|
| 270 |
+
"model.language_model.layers.22.input_layernorm.weight": "model-00009-of-00016.safetensors",
|
| 271 |
+
"model.language_model.layers.22.linear_attn.A_log": "model-00009-of-00016.safetensors",
|
| 272 |
+
"model.language_model.layers.22.linear_attn.conv1d.weight": "model-00009-of-00016.safetensors",
|
| 273 |
+
"model.language_model.layers.22.linear_attn.dt_bias": "model-00009-of-00016.safetensors",
|
| 274 |
+
"model.language_model.layers.22.linear_attn.in_proj_a.weight": "model-00009-of-00016.safetensors",
|
| 275 |
+
"model.language_model.layers.22.linear_attn.in_proj_b.weight": "model-00009-of-00016.safetensors",
|
| 276 |
+
"model.language_model.layers.22.linear_attn.in_proj_qkv.weight": "model-00009-of-00016.safetensors",
|
| 277 |
+
"model.language_model.layers.22.linear_attn.in_proj_z.weight": "model-00009-of-00016.safetensors",
|
| 278 |
+
"model.language_model.layers.22.linear_attn.norm.weight": "model-00009-of-00016.safetensors",
|
| 279 |
+
"model.language_model.layers.22.linear_attn.out_proj.weight": "model-00009-of-00016.safetensors",
|
| 280 |
+
"model.language_model.layers.22.mlp.experts.down_proj": "model-00009-of-00016.safetensors",
|
| 281 |
+
"model.language_model.layers.22.mlp.experts.gate_up_proj": "model-00009-of-00016.safetensors",
|
| 282 |
+
"model.language_model.layers.22.mlp.gate.weight": "model-00009-of-00016.safetensors",
|
| 283 |
+
"model.language_model.layers.22.mlp.shared_expert.down_proj.weight": "model-00009-of-00016.safetensors",
|
| 284 |
+
"model.language_model.layers.22.mlp.shared_expert.gate_proj.weight": "model-00009-of-00016.safetensors",
|
| 285 |
+
"model.language_model.layers.22.mlp.shared_expert.up_proj.weight": "model-00009-of-00016.safetensors",
|
| 286 |
+
"model.language_model.layers.22.mlp.shared_expert_gate.weight": "model-00009-of-00016.safetensors",
|
| 287 |
+
"model.language_model.layers.22.post_attention_layernorm.weight": "model-00009-of-00016.safetensors",
|
| 288 |
+
"model.language_model.layers.23.input_layernorm.weight": "model-00009-of-00016.safetensors",
|
| 289 |
+
"model.language_model.layers.23.mlp.experts.down_proj": "model-00010-of-00016.safetensors",
|
| 290 |
+
"model.language_model.layers.23.mlp.experts.gate_up_proj": "model-00010-of-00016.safetensors",
|
| 291 |
+
"model.language_model.layers.23.mlp.gate.weight": "model-00010-of-00016.safetensors",
|
| 292 |
+
"model.language_model.layers.23.mlp.shared_expert.down_proj.weight": "model-00010-of-00016.safetensors",
|
| 293 |
+
"model.language_model.layers.23.mlp.shared_expert.gate_proj.weight": "model-00010-of-00016.safetensors",
|
| 294 |
+
"model.language_model.layers.23.mlp.shared_expert.up_proj.weight": "model-00010-of-00016.safetensors",
|
| 295 |
+
"model.language_model.layers.23.mlp.shared_expert_gate.weight": "model-00010-of-00016.safetensors",
|
| 296 |
+
"model.language_model.layers.23.post_attention_layernorm.weight": "model-00010-of-00016.safetensors",
|
| 297 |
+
"model.language_model.layers.23.self_attn.k_norm.weight": "model-00010-of-00016.safetensors",
|
| 298 |
+
"model.language_model.layers.23.self_attn.k_proj.weight": "model-00010-of-00016.safetensors",
|
| 299 |
+
"model.language_model.layers.23.self_attn.o_proj.weight": "model-00010-of-00016.safetensors",
|
| 300 |
+
"model.language_model.layers.23.self_attn.q_norm.weight": "model-00010-of-00016.safetensors",
|
| 301 |
+
"model.language_model.layers.23.self_attn.q_proj.weight": "model-00010-of-00016.safetensors",
|
| 302 |
+
"model.language_model.layers.23.self_attn.v_proj.weight": "model-00010-of-00016.safetensors",
|
| 303 |
+
"model.language_model.layers.24.input_layernorm.weight": "model-00010-of-00016.safetensors",
|
| 304 |
+
"model.language_model.layers.24.linear_attn.A_log": "model-00010-of-00016.safetensors",
|
| 305 |
+
"model.language_model.layers.24.linear_attn.conv1d.weight": "model-00010-of-00016.safetensors",
|
| 306 |
+
"model.language_model.layers.24.linear_attn.dt_bias": "model-00010-of-00016.safetensors",
|
| 307 |
+
"model.language_model.layers.24.linear_attn.in_proj_a.weight": "model-00010-of-00016.safetensors",
|
| 308 |
+
"model.language_model.layers.24.linear_attn.in_proj_b.weight": "model-00010-of-00016.safetensors",
|
| 309 |
+
"model.language_model.layers.24.linear_attn.in_proj_qkv.weight": "model-00010-of-00016.safetensors",
|
| 310 |
+
"model.language_model.layers.24.linear_attn.in_proj_z.weight": "model-00010-of-00016.safetensors",
|
| 311 |
+
"model.language_model.layers.24.linear_attn.norm.weight": "model-00010-of-00016.safetensors",
|
| 312 |
+
"model.language_model.layers.24.linear_attn.out_proj.weight": "model-00010-of-00016.safetensors",
|
| 313 |
+
"model.language_model.layers.24.mlp.experts.down_proj": "model-00010-of-00016.safetensors",
|
| 314 |
+
"model.language_model.layers.24.mlp.experts.gate_up_proj": "model-00010-of-00016.safetensors",
|
| 315 |
+
"model.language_model.layers.24.mlp.gate.weight": "model-00010-of-00016.safetensors",
|
| 316 |
+
"model.language_model.layers.24.mlp.shared_expert.down_proj.weight": "model-00010-of-00016.safetensors",
|
| 317 |
+
"model.language_model.layers.24.mlp.shared_expert.gate_proj.weight": "model-00010-of-00016.safetensors",
|
| 318 |
+
"model.language_model.layers.24.mlp.shared_expert.up_proj.weight": "model-00010-of-00016.safetensors",
|
| 319 |
+
"model.language_model.layers.24.mlp.shared_expert_gate.weight": "model-00010-of-00016.safetensors",
|
| 320 |
+
"model.language_model.layers.24.post_attention_layernorm.weight": "model-00010-of-00016.safetensors",
|
| 321 |
+
"model.language_model.layers.25.input_layernorm.weight": "model-00010-of-00016.safetensors",
|
| 322 |
+
"model.language_model.layers.25.linear_attn.A_log": "model-00010-of-00016.safetensors",
|
| 323 |
+
"model.language_model.layers.25.linear_attn.conv1d.weight": "model-00010-of-00016.safetensors",
|
| 324 |
+
"model.language_model.layers.25.linear_attn.dt_bias": "model-00010-of-00016.safetensors",
|
| 325 |
+
"model.language_model.layers.25.linear_attn.in_proj_a.weight": "model-00010-of-00016.safetensors",
|
| 326 |
+
"model.language_model.layers.25.linear_attn.in_proj_b.weight": "model-00010-of-00016.safetensors",
|
| 327 |
+
"model.language_model.layers.25.linear_attn.in_proj_qkv.weight": "model-00010-of-00016.safetensors",
|
| 328 |
+
"model.language_model.layers.25.linear_attn.in_proj_z.weight": "model-00010-of-00016.safetensors",
|
| 329 |
+
"model.language_model.layers.25.linear_attn.norm.weight": "model-00010-of-00016.safetensors",
|
| 330 |
+
"model.language_model.layers.25.linear_attn.out_proj.weight": "model-00010-of-00016.safetensors",
|
| 331 |
+
"model.language_model.layers.25.mlp.experts.down_proj": "model-00010-of-00016.safetensors",
|
| 332 |
+
"model.language_model.layers.25.mlp.experts.gate_up_proj": "model-00011-of-00016.safetensors",
|
| 333 |
+
"model.language_model.layers.25.mlp.gate.weight": "model-00011-of-00016.safetensors",
|
| 334 |
+
"model.language_model.layers.25.mlp.shared_expert.down_proj.weight": "model-00011-of-00016.safetensors",
|
| 335 |
+
"model.language_model.layers.25.mlp.shared_expert.gate_proj.weight": "model-00011-of-00016.safetensors",
|
| 336 |
+
"model.language_model.layers.25.mlp.shared_expert.up_proj.weight": "model-00011-of-00016.safetensors",
|
| 337 |
+
"model.language_model.layers.25.mlp.shared_expert_gate.weight": "model-00011-of-00016.safetensors",
|
| 338 |
+
"model.language_model.layers.25.post_attention_layernorm.weight": "model-00011-of-00016.safetensors",
|
| 339 |
+
"model.language_model.layers.26.input_layernorm.weight": "model-00011-of-00016.safetensors",
|
| 340 |
+
"model.language_model.layers.26.linear_attn.A_log": "model-00011-of-00016.safetensors",
|
| 341 |
+
"model.language_model.layers.26.linear_attn.conv1d.weight": "model-00011-of-00016.safetensors",
|
| 342 |
+
"model.language_model.layers.26.linear_attn.dt_bias": "model-00011-of-00016.safetensors",
|
| 343 |
+
"model.language_model.layers.26.linear_attn.in_proj_a.weight": "model-00011-of-00016.safetensors",
|
| 344 |
+
"model.language_model.layers.26.linear_attn.in_proj_b.weight": "model-00011-of-00016.safetensors",
|
| 345 |
+
"model.language_model.layers.26.linear_attn.in_proj_qkv.weight": "model-00011-of-00016.safetensors",
|
| 346 |
+
"model.language_model.layers.26.linear_attn.in_proj_z.weight": "model-00011-of-00016.safetensors",
|
| 347 |
+
"model.language_model.layers.26.linear_attn.norm.weight": "model-00011-of-00016.safetensors",
|
| 348 |
+
"model.language_model.layers.26.linear_attn.out_proj.weight": "model-00011-of-00016.safetensors",
|
| 349 |
+
"model.language_model.layers.26.mlp.experts.down_proj": "model-00011-of-00016.safetensors",
|
| 350 |
+
"model.language_model.layers.26.mlp.experts.gate_up_proj": "model-00011-of-00016.safetensors",
|
| 351 |
+
"model.language_model.layers.26.mlp.gate.weight": "model-00011-of-00016.safetensors",
|
| 352 |
+
"model.language_model.layers.26.mlp.shared_expert.down_proj.weight": "model-00011-of-00016.safetensors",
|
| 353 |
+
"model.language_model.layers.26.mlp.shared_expert.gate_proj.weight": "model-00011-of-00016.safetensors",
|
| 354 |
+
"model.language_model.layers.26.mlp.shared_expert.up_proj.weight": "model-00011-of-00016.safetensors",
|
| 355 |
+
"model.language_model.layers.26.mlp.shared_expert_gate.weight": "model-00011-of-00016.safetensors",
|
| 356 |
+
"model.language_model.layers.26.post_attention_layernorm.weight": "model-00011-of-00016.safetensors",
|
| 357 |
+
"model.language_model.layers.27.input_layernorm.weight": "model-00011-of-00016.safetensors",
|
| 358 |
+
"model.language_model.layers.27.mlp.experts.down_proj": "model-00011-of-00016.safetensors",
|
| 359 |
+
"model.language_model.layers.27.mlp.experts.gate_up_proj": "model-00011-of-00016.safetensors",
|
| 360 |
+
"model.language_model.layers.27.mlp.gate.weight": "model-00011-of-00016.safetensors",
|
| 361 |
+
"model.language_model.layers.27.mlp.shared_expert.down_proj.weight": "model-00011-of-00016.safetensors",
|
| 362 |
+
"model.language_model.layers.27.mlp.shared_expert.gate_proj.weight": "model-00011-of-00016.safetensors",
|
| 363 |
+
"model.language_model.layers.27.mlp.shared_expert.up_proj.weight": "model-00011-of-00016.safetensors",
|
| 364 |
+
"model.language_model.layers.27.mlp.shared_expert_gate.weight": "model-00011-of-00016.safetensors",
|
| 365 |
+
"model.language_model.layers.27.post_attention_layernorm.weight": "model-00011-of-00016.safetensors",
|
| 366 |
+
"model.language_model.layers.27.self_attn.k_norm.weight": "model-00011-of-00016.safetensors",
|
| 367 |
+
"model.language_model.layers.27.self_attn.k_proj.weight": "model-00011-of-00016.safetensors",
|
| 368 |
+
"model.language_model.layers.27.self_attn.o_proj.weight": "model-00011-of-00016.safetensors",
|
| 369 |
+
"model.language_model.layers.27.self_attn.q_norm.weight": "model-00011-of-00016.safetensors",
|
| 370 |
+
"model.language_model.layers.27.self_attn.q_proj.weight": "model-00011-of-00016.safetensors",
|
| 371 |
+
"model.language_model.layers.27.self_attn.v_proj.weight": "model-00011-of-00016.safetensors",
|
| 372 |
+
"model.language_model.layers.28.input_layernorm.weight": "model-00011-of-00016.safetensors",
|
| 373 |
+
"model.language_model.layers.28.linear_attn.A_log": "model-00011-of-00016.safetensors",
|
| 374 |
+
"model.language_model.layers.28.linear_attn.conv1d.weight": "model-00011-of-00016.safetensors",
|
| 375 |
+
"model.language_model.layers.28.linear_attn.dt_bias": "model-00011-of-00016.safetensors",
|
| 376 |
+
"model.language_model.layers.28.linear_attn.in_proj_a.weight": "model-00011-of-00016.safetensors",
|
| 377 |
+
"model.language_model.layers.28.linear_attn.in_proj_b.weight": "model-00011-of-00016.safetensors",
|
| 378 |
+
"model.language_model.layers.28.linear_attn.in_proj_qkv.weight": "model-00011-of-00016.safetensors",
|
| 379 |
+
"model.language_model.layers.28.linear_attn.in_proj_z.weight": "model-00011-of-00016.safetensors",
|
| 380 |
+
"model.language_model.layers.28.linear_attn.norm.weight": "model-00011-of-00016.safetensors",
|
| 381 |
+
"model.language_model.layers.28.linear_attn.out_proj.weight": "model-00011-of-00016.safetensors",
|
| 382 |
+
"model.language_model.layers.28.mlp.experts.down_proj": "model-00012-of-00016.safetensors",
|
| 383 |
+
"model.language_model.layers.28.mlp.experts.gate_up_proj": "model-00012-of-00016.safetensors",
|
| 384 |
+
"model.language_model.layers.28.mlp.gate.weight": "model-00012-of-00016.safetensors",
|
| 385 |
+
"model.language_model.layers.28.mlp.shared_expert.down_proj.weight": "model-00012-of-00016.safetensors",
|
| 386 |
+
"model.language_model.layers.28.mlp.shared_expert.gate_proj.weight": "model-00012-of-00016.safetensors",
|
| 387 |
+
"model.language_model.layers.28.mlp.shared_expert.up_proj.weight": "model-00012-of-00016.safetensors",
|
| 388 |
+
"model.language_model.layers.28.mlp.shared_expert_gate.weight": "model-00012-of-00016.safetensors",
|
| 389 |
+
"model.language_model.layers.28.post_attention_layernorm.weight": "model-00012-of-00016.safetensors",
|
| 390 |
+
"model.language_model.layers.29.input_layernorm.weight": "model-00012-of-00016.safetensors",
|
| 391 |
+
"model.language_model.layers.29.linear_attn.A_log": "model-00012-of-00016.safetensors",
|
| 392 |
+
"model.language_model.layers.29.linear_attn.conv1d.weight": "model-00012-of-00016.safetensors",
|
| 393 |
+
"model.language_model.layers.29.linear_attn.dt_bias": "model-00012-of-00016.safetensors",
|
| 394 |
+
"model.language_model.layers.29.linear_attn.in_proj_a.weight": "model-00012-of-00016.safetensors",
|
| 395 |
+
"model.language_model.layers.29.linear_attn.in_proj_b.weight": "model-00012-of-00016.safetensors",
|
| 396 |
+
"model.language_model.layers.29.linear_attn.in_proj_qkv.weight": "model-00012-of-00016.safetensors",
|
| 397 |
+
"model.language_model.layers.29.linear_attn.in_proj_z.weight": "model-00012-of-00016.safetensors",
|
| 398 |
+
"model.language_model.layers.29.linear_attn.norm.weight": "model-00012-of-00016.safetensors",
|
| 399 |
+
"model.language_model.layers.29.linear_attn.out_proj.weight": "model-00012-of-00016.safetensors",
|
| 400 |
+
"model.language_model.layers.29.mlp.experts.down_proj": "model-00012-of-00016.safetensors",
|
| 401 |
+
"model.language_model.layers.29.mlp.experts.gate_up_proj": "model-00012-of-00016.safetensors",
|
| 402 |
+
"model.language_model.layers.29.mlp.gate.weight": "model-00012-of-00016.safetensors",
|
| 403 |
+
"model.language_model.layers.29.mlp.shared_expert.down_proj.weight": "model-00012-of-00016.safetensors",
|
| 404 |
+
"model.language_model.layers.29.mlp.shared_expert.gate_proj.weight": "model-00012-of-00016.safetensors",
|
| 405 |
+
"model.language_model.layers.29.mlp.shared_expert.up_proj.weight": "model-00012-of-00016.safetensors",
|
| 406 |
+
"model.language_model.layers.29.mlp.shared_expert_gate.weight": "model-00012-of-00016.safetensors",
|
| 407 |
+
"model.language_model.layers.29.post_attention_layernorm.weight": "model-00012-of-00016.safetensors",
|
| 408 |
+
"model.language_model.layers.3.input_layernorm.weight": "model-00002-of-00016.safetensors",
|
| 409 |
+
"model.language_model.layers.3.mlp.experts.down_proj": "model-00002-of-00016.safetensors",
|
| 410 |
+
"model.language_model.layers.3.mlp.experts.gate_up_proj": "model-00002-of-00016.safetensors",
|
| 411 |
+
"model.language_model.layers.3.mlp.gate.weight": "model-00002-of-00016.safetensors",
|
| 412 |
+
"model.language_model.layers.3.mlp.shared_expert.down_proj.weight": "model-00002-of-00016.safetensors",
|
| 413 |
+
"model.language_model.layers.3.mlp.shared_expert.gate_proj.weight": "model-00002-of-00016.safetensors",
|
| 414 |
+
"model.language_model.layers.3.mlp.shared_expert.up_proj.weight": "model-00002-of-00016.safetensors",
|
| 415 |
+
"model.language_model.layers.3.mlp.shared_expert_gate.weight": "model-00002-of-00016.safetensors",
|
| 416 |
+
"model.language_model.layers.3.post_attention_layernorm.weight": "model-00002-of-00016.safetensors",
|
| 417 |
+
"model.language_model.layers.3.self_attn.k_norm.weight": "model-00002-of-00016.safetensors",
|
| 418 |
+
"model.language_model.layers.3.self_attn.k_proj.weight": "model-00002-of-00016.safetensors",
|
| 419 |
+
"model.language_model.layers.3.self_attn.o_proj.weight": "model-00002-of-00016.safetensors",
|
| 420 |
+
"model.language_model.layers.3.self_attn.q_norm.weight": "model-00002-of-00016.safetensors",
|
| 421 |
+
"model.language_model.layers.3.self_attn.q_proj.weight": "model-00002-of-00016.safetensors",
|
| 422 |
+
"model.language_model.layers.3.self_attn.v_proj.weight": "model-00002-of-00016.safetensors",
|
| 423 |
+
"model.language_model.layers.30.input_layernorm.weight": "model-00012-of-00016.safetensors",
|
| 424 |
+
"model.language_model.layers.30.linear_attn.A_log": "model-00012-of-00016.safetensors",
|
| 425 |
+
"model.language_model.layers.30.linear_attn.conv1d.weight": "model-00012-of-00016.safetensors",
|
| 426 |
+
"model.language_model.layers.30.linear_attn.dt_bias": "model-00012-of-00016.safetensors",
|
| 427 |
+
"model.language_model.layers.30.linear_attn.in_proj_a.weight": "model-00012-of-00016.safetensors",
|
| 428 |
+
"model.language_model.layers.30.linear_attn.in_proj_b.weight": "model-00012-of-00016.safetensors",
|
| 429 |
+
"model.language_model.layers.30.linear_attn.in_proj_qkv.weight": "model-00012-of-00016.safetensors",
|
| 430 |
+
"model.language_model.layers.30.linear_attn.in_proj_z.weight": "model-00012-of-00016.safetensors",
|
| 431 |
+
"model.language_model.layers.30.linear_attn.norm.weight": "model-00012-of-00016.safetensors",
|
| 432 |
+
"model.language_model.layers.30.linear_attn.out_proj.weight": "model-00012-of-00016.safetensors",
|
| 433 |
+
"model.language_model.layers.30.mlp.experts.down_proj": "model-00012-of-00016.safetensors",
|
| 434 |
+
"model.language_model.layers.30.mlp.experts.gate_up_proj": "model-00012-of-00016.safetensors",
|
| 435 |
+
"model.language_model.layers.30.mlp.gate.weight": "model-00012-of-00016.safetensors",
|
| 436 |
+
"model.language_model.layers.30.mlp.shared_expert.down_proj.weight": "model-00012-of-00016.safetensors",
|
| 437 |
+
"model.language_model.layers.30.mlp.shared_expert.gate_proj.weight": "model-00012-of-00016.safetensors",
|
| 438 |
+
"model.language_model.layers.30.mlp.shared_expert.up_proj.weight": "model-00012-of-00016.safetensors",
|
| 439 |
+
"model.language_model.layers.30.mlp.shared_expert_gate.weight": "model-00012-of-00016.safetensors",
|
| 440 |
+
"model.language_model.layers.30.post_attention_layernorm.weight": "model-00012-of-00016.safetensors",
|
| 441 |
+
"model.language_model.layers.31.input_layernorm.weight": "model-00012-of-00016.safetensors",
|
| 442 |
+
"model.language_model.layers.31.mlp.experts.down_proj": "model-00013-of-00016.safetensors",
|
| 443 |
+
"model.language_model.layers.31.mlp.experts.gate_up_proj": "model-00013-of-00016.safetensors",
|
| 444 |
+
"model.language_model.layers.31.mlp.gate.weight": "model-00013-of-00016.safetensors",
|
| 445 |
+
"model.language_model.layers.31.mlp.shared_expert.down_proj.weight": "model-00013-of-00016.safetensors",
|
| 446 |
+
"model.language_model.layers.31.mlp.shared_expert.gate_proj.weight": "model-00013-of-00016.safetensors",
|
| 447 |
+
"model.language_model.layers.31.mlp.shared_expert.up_proj.weight": "model-00013-of-00016.safetensors",
|
| 448 |
+
"model.language_model.layers.31.mlp.shared_expert_gate.weight": "model-00013-of-00016.safetensors",
|
| 449 |
+
"model.language_model.layers.31.post_attention_layernorm.weight": "model-00013-of-00016.safetensors",
|
| 450 |
+
"model.language_model.layers.31.self_attn.k_norm.weight": "model-00013-of-00016.safetensors",
|
| 451 |
+
"model.language_model.layers.31.self_attn.k_proj.weight": "model-00013-of-00016.safetensors",
|
| 452 |
+
"model.language_model.layers.31.self_attn.o_proj.weight": "model-00013-of-00016.safetensors",
|
| 453 |
+
"model.language_model.layers.31.self_attn.q_norm.weight": "model-00013-of-00016.safetensors",
|
| 454 |
+
"model.language_model.layers.31.self_attn.q_proj.weight": "model-00013-of-00016.safetensors",
|
| 455 |
+
"model.language_model.layers.31.self_attn.v_proj.weight": "model-00013-of-00016.safetensors",
|
| 456 |
+
"model.language_model.layers.32.input_layernorm.weight": "model-00013-of-00016.safetensors",
|
| 457 |
+
"model.language_model.layers.32.linear_attn.A_log": "model-00013-of-00016.safetensors",
|
| 458 |
+
"model.language_model.layers.32.linear_attn.conv1d.weight": "model-00013-of-00016.safetensors",
|
| 459 |
+
"model.language_model.layers.32.linear_attn.dt_bias": "model-00013-of-00016.safetensors",
|
| 460 |
+
"model.language_model.layers.32.linear_attn.in_proj_a.weight": "model-00013-of-00016.safetensors",
|
| 461 |
+
"model.language_model.layers.32.linear_attn.in_proj_b.weight": "model-00013-of-00016.safetensors",
|
| 462 |
+
"model.language_model.layers.32.linear_attn.in_proj_qkv.weight": "model-00013-of-00016.safetensors",
|
| 463 |
+
"model.language_model.layers.32.linear_attn.in_proj_z.weight": "model-00013-of-00016.safetensors",
|
| 464 |
+
"model.language_model.layers.32.linear_attn.norm.weight": "model-00013-of-00016.safetensors",
|
| 465 |
+
"model.language_model.layers.32.linear_attn.out_proj.weight": "model-00013-of-00016.safetensors",
|
| 466 |
+
"model.language_model.layers.32.mlp.experts.down_proj": "model-00013-of-00016.safetensors",
|
| 467 |
+
"model.language_model.layers.32.mlp.experts.gate_up_proj": "model-00013-of-00016.safetensors",
|
| 468 |
+
"model.language_model.layers.32.mlp.gate.weight": "model-00013-of-00016.safetensors",
|
| 469 |
+
"model.language_model.layers.32.mlp.shared_expert.down_proj.weight": "model-00013-of-00016.safetensors",
|
| 470 |
+
"model.language_model.layers.32.mlp.shared_expert.gate_proj.weight": "model-00013-of-00016.safetensors",
|
| 471 |
+
"model.language_model.layers.32.mlp.shared_expert.up_proj.weight": "model-00013-of-00016.safetensors",
|
| 472 |
+
"model.language_model.layers.32.mlp.shared_expert_gate.weight": "model-00013-of-00016.safetensors",
|
| 473 |
+
"model.language_model.layers.32.post_attention_layernorm.weight": "model-00013-of-00016.safetensors",
|
| 474 |
+
"model.language_model.layers.33.input_layernorm.weight": "model-00013-of-00016.safetensors",
|
| 475 |
+
"model.language_model.layers.33.linear_attn.A_log": "model-00013-of-00016.safetensors",
|
| 476 |
+
"model.language_model.layers.33.linear_attn.conv1d.weight": "model-00013-of-00016.safetensors",
|
| 477 |
+
"model.language_model.layers.33.linear_attn.dt_bias": "model-00013-of-00016.safetensors",
|
| 478 |
+
"model.language_model.layers.33.linear_attn.in_proj_a.weight": "model-00013-of-00016.safetensors",
|
| 479 |
+
"model.language_model.layers.33.linear_attn.in_proj_b.weight": "model-00013-of-00016.safetensors",
|
| 480 |
+
"model.language_model.layers.33.linear_attn.in_proj_qkv.weight": "model-00013-of-00016.safetensors",
|
| 481 |
+
"model.language_model.layers.33.linear_attn.in_proj_z.weight": "model-00013-of-00016.safetensors",
|
| 482 |
+
"model.language_model.layers.33.linear_attn.norm.weight": "model-00013-of-00016.safetensors",
|
| 483 |
+
"model.language_model.layers.33.linear_attn.out_proj.weight": "model-00013-of-00016.safetensors",
|
| 484 |
+
"model.language_model.layers.33.mlp.experts.down_proj": "model-00013-of-00016.safetensors",
|
| 485 |
+
"model.language_model.layers.33.mlp.experts.gate_up_proj": "model-00014-of-00016.safetensors",
|
| 486 |
+
"model.language_model.layers.33.mlp.gate.weight": "model-00014-of-00016.safetensors",
|
| 487 |
+
"model.language_model.layers.33.mlp.shared_expert.down_proj.weight": "model-00014-of-00016.safetensors",
|
| 488 |
+
"model.language_model.layers.33.mlp.shared_expert.gate_proj.weight": "model-00014-of-00016.safetensors",
|
| 489 |
+
"model.language_model.layers.33.mlp.shared_expert.up_proj.weight": "model-00014-of-00016.safetensors",
|
| 490 |
+
"model.language_model.layers.33.mlp.shared_expert_gate.weight": "model-00014-of-00016.safetensors",
|
| 491 |
+
"model.language_model.layers.33.post_attention_layernorm.weight": "model-00014-of-00016.safetensors",
|
| 492 |
+
"model.language_model.layers.34.input_layernorm.weight": "model-00014-of-00016.safetensors",
|
| 493 |
+
"model.language_model.layers.34.linear_attn.A_log": "model-00014-of-00016.safetensors",
|
| 494 |
+
"model.language_model.layers.34.linear_attn.conv1d.weight": "model-00014-of-00016.safetensors",
|
| 495 |
+
"model.language_model.layers.34.linear_attn.dt_bias": "model-00014-of-00016.safetensors",
|
| 496 |
+
"model.language_model.layers.34.linear_attn.in_proj_a.weight": "model-00014-of-00016.safetensors",
|
| 497 |
+
"model.language_model.layers.34.linear_attn.in_proj_b.weight": "model-00014-of-00016.safetensors",
|
| 498 |
+
"model.language_model.layers.34.linear_attn.in_proj_qkv.weight": "model-00014-of-00016.safetensors",
|
| 499 |
+
"model.language_model.layers.34.linear_attn.in_proj_z.weight": "model-00014-of-00016.safetensors",
|
| 500 |
+
"model.language_model.layers.34.linear_attn.norm.weight": "model-00014-of-00016.safetensors",
|
| 501 |
+
"model.language_model.layers.34.linear_attn.out_proj.weight": "model-00014-of-00016.safetensors",
|
| 502 |
+
"model.language_model.layers.34.mlp.experts.down_proj": "model-00014-of-00016.safetensors",
|
| 503 |
+
"model.language_model.layers.34.mlp.experts.gate_up_proj": "model-00014-of-00016.safetensors",
|
| 504 |
+
"model.language_model.layers.34.mlp.gate.weight": "model-00014-of-00016.safetensors",
|
| 505 |
+
"model.language_model.layers.34.mlp.shared_expert.down_proj.weight": "model-00014-of-00016.safetensors",
|
| 506 |
+
"model.language_model.layers.34.mlp.shared_expert.gate_proj.weight": "model-00014-of-00016.safetensors",
|
| 507 |
+
"model.language_model.layers.34.mlp.shared_expert.up_proj.weight": "model-00014-of-00016.safetensors",
|
| 508 |
+
"model.language_model.layers.34.mlp.shared_expert_gate.weight": "model-00014-of-00016.safetensors",
|
| 509 |
+
"model.language_model.layers.34.post_attention_layernorm.weight": "model-00014-of-00016.safetensors",
|
| 510 |
+
"model.language_model.layers.35.input_layernorm.weight": "model-00014-of-00016.safetensors",
|
| 511 |
+
"model.language_model.layers.35.mlp.experts.down_proj": "model-00014-of-00016.safetensors",
|
| 512 |
+
"model.language_model.layers.35.mlp.experts.gate_up_proj": "model-00014-of-00016.safetensors",
|
| 513 |
+
"model.language_model.layers.35.mlp.gate.weight": "model-00014-of-00016.safetensors",
|
| 514 |
+
"model.language_model.layers.35.mlp.shared_expert.down_proj.weight": "model-00014-of-00016.safetensors",
|
| 515 |
+
"model.language_model.layers.35.mlp.shared_expert.gate_proj.weight": "model-00014-of-00016.safetensors",
|
| 516 |
+
"model.language_model.layers.35.mlp.shared_expert.up_proj.weight": "model-00014-of-00016.safetensors",
|
| 517 |
+
"model.language_model.layers.35.mlp.shared_expert_gate.weight": "model-00014-of-00016.safetensors",
|
| 518 |
+
"model.language_model.layers.35.post_attention_layernorm.weight": "model-00014-of-00016.safetensors",
|
| 519 |
+
"model.language_model.layers.35.self_attn.k_norm.weight": "model-00014-of-00016.safetensors",
|
| 520 |
+
"model.language_model.layers.35.self_attn.k_proj.weight": "model-00014-of-00016.safetensors",
|
| 521 |
+
"model.language_model.layers.35.self_attn.o_proj.weight": "model-00014-of-00016.safetensors",
|
| 522 |
+
"model.language_model.layers.35.self_attn.q_norm.weight": "model-00014-of-00016.safetensors",
|
| 523 |
+
"model.language_model.layers.35.self_attn.q_proj.weight": "model-00014-of-00016.safetensors",
|
| 524 |
+
"model.language_model.layers.35.self_attn.v_proj.weight": "model-00014-of-00016.safetensors",
|
| 525 |
+
"model.language_model.layers.36.input_layernorm.weight": "model-00014-of-00016.safetensors",
|
| 526 |
+
"model.language_model.layers.36.linear_attn.A_log": "model-00014-of-00016.safetensors",
|
| 527 |
+
"model.language_model.layers.36.linear_attn.conv1d.weight": "model-00014-of-00016.safetensors",
|
| 528 |
+
"model.language_model.layers.36.linear_attn.dt_bias": "model-00014-of-00016.safetensors",
|
| 529 |
+
"model.language_model.layers.36.linear_attn.in_proj_a.weight": "model-00014-of-00016.safetensors",
|
| 530 |
+
"model.language_model.layers.36.linear_attn.in_proj_b.weight": "model-00014-of-00016.safetensors",
|
| 531 |
+
"model.language_model.layers.36.linear_attn.in_proj_qkv.weight": "model-00014-of-00016.safetensors",
|
| 532 |
+
"model.language_model.layers.36.linear_attn.in_proj_z.weight": "model-00014-of-00016.safetensors",
|
| 533 |
+
"model.language_model.layers.36.linear_attn.norm.weight": "model-00014-of-00016.safetensors",
|
| 534 |
+
"model.language_model.layers.36.linear_attn.out_proj.weight": "model-00014-of-00016.safetensors",
|
| 535 |
+
"model.language_model.layers.36.mlp.experts.down_proj": "model-00015-of-00016.safetensors",
|
| 536 |
+
"model.language_model.layers.36.mlp.experts.gate_up_proj": "model-00015-of-00016.safetensors",
|
| 537 |
+
"model.language_model.layers.36.mlp.gate.weight": "model-00015-of-00016.safetensors",
|
| 538 |
+
"model.language_model.layers.36.mlp.shared_expert.down_proj.weight": "model-00015-of-00016.safetensors",
|
| 539 |
+
"model.language_model.layers.36.mlp.shared_expert.gate_proj.weight": "model-00015-of-00016.safetensors",
|
| 540 |
+
"model.language_model.layers.36.mlp.shared_expert.up_proj.weight": "model-00015-of-00016.safetensors",
|
| 541 |
+
"model.language_model.layers.36.mlp.shared_expert_gate.weight": "model-00015-of-00016.safetensors",
|
| 542 |
+
"model.language_model.layers.36.post_attention_layernorm.weight": "model-00015-of-00016.safetensors",
|
| 543 |
+
"model.language_model.layers.37.input_layernorm.weight": "model-00015-of-00016.safetensors",
|
| 544 |
+
"model.language_model.layers.37.linear_attn.A_log": "model-00015-of-00016.safetensors",
|
| 545 |
+
"model.language_model.layers.37.linear_attn.conv1d.weight": "model-00015-of-00016.safetensors",
|
| 546 |
+
"model.language_model.layers.37.linear_attn.dt_bias": "model-00015-of-00016.safetensors",
|
| 547 |
+
"model.language_model.layers.37.linear_attn.in_proj_a.weight": "model-00015-of-00016.safetensors",
|
| 548 |
+
"model.language_model.layers.37.linear_attn.in_proj_b.weight": "model-00015-of-00016.safetensors",
|
| 549 |
+
"model.language_model.layers.37.linear_attn.in_proj_qkv.weight": "model-00015-of-00016.safetensors",
|
| 550 |
+
"model.language_model.layers.37.linear_attn.in_proj_z.weight": "model-00015-of-00016.safetensors",
|
| 551 |
+
"model.language_model.layers.37.linear_attn.norm.weight": "model-00015-of-00016.safetensors",
|
| 552 |
+
"model.language_model.layers.37.linear_attn.out_proj.weight": "model-00015-of-00016.safetensors",
|
| 553 |
+
"model.language_model.layers.37.mlp.experts.down_proj": "model-00015-of-00016.safetensors",
|
| 554 |
+
"model.language_model.layers.37.mlp.experts.gate_up_proj": "model-00015-of-00016.safetensors",
|
| 555 |
+
"model.language_model.layers.37.mlp.gate.weight": "model-00015-of-00016.safetensors",
|
| 556 |
+
"model.language_model.layers.37.mlp.shared_expert.down_proj.weight": "model-00015-of-00016.safetensors",
|
| 557 |
+
"model.language_model.layers.37.mlp.shared_expert.gate_proj.weight": "model-00015-of-00016.safetensors",
|
| 558 |
+
"model.language_model.layers.37.mlp.shared_expert.up_proj.weight": "model-00015-of-00016.safetensors",
|
| 559 |
+
"model.language_model.layers.37.mlp.shared_expert_gate.weight": "model-00015-of-00016.safetensors",
|
| 560 |
+
"model.language_model.layers.37.post_attention_layernorm.weight": "model-00015-of-00016.safetensors",
|
| 561 |
+
"model.language_model.layers.38.input_layernorm.weight": "model-00015-of-00016.safetensors",
|
| 562 |
+
"model.language_model.layers.38.linear_attn.A_log": "model-00015-of-00016.safetensors",
|
| 563 |
+
"model.language_model.layers.38.linear_attn.conv1d.weight": "model-00015-of-00016.safetensors",
|
| 564 |
+
"model.language_model.layers.38.linear_attn.dt_bias": "model-00015-of-00016.safetensors",
|
| 565 |
+
"model.language_model.layers.38.linear_attn.in_proj_a.weight": "model-00015-of-00016.safetensors",
|
| 566 |
+
"model.language_model.layers.38.linear_attn.in_proj_b.weight": "model-00015-of-00016.safetensors",
|
| 567 |
+
"model.language_model.layers.38.linear_attn.in_proj_qkv.weight": "model-00015-of-00016.safetensors",
|
| 568 |
+
"model.language_model.layers.38.linear_attn.in_proj_z.weight": "model-00015-of-00016.safetensors",
|
| 569 |
+
"model.language_model.layers.38.linear_attn.norm.weight": "model-00015-of-00016.safetensors",
|
| 570 |
+
"model.language_model.layers.38.linear_attn.out_proj.weight": "model-00015-of-00016.safetensors",
|
| 571 |
+
"model.language_model.layers.38.mlp.experts.down_proj": "model-00015-of-00016.safetensors",
|
| 572 |
+
"model.language_model.layers.38.mlp.experts.gate_up_proj": "model-00015-of-00016.safetensors",
|
| 573 |
+
"model.language_model.layers.38.mlp.gate.weight": "model-00015-of-00016.safetensors",
|
| 574 |
+
"model.language_model.layers.38.mlp.shared_expert.down_proj.weight": "model-00015-of-00016.safetensors",
|
| 575 |
+
"model.language_model.layers.38.mlp.shared_expert.gate_proj.weight": "model-00015-of-00016.safetensors",
|
| 576 |
+
"model.language_model.layers.38.mlp.shared_expert.up_proj.weight": "model-00015-of-00016.safetensors",
|
| 577 |
+
"model.language_model.layers.38.mlp.shared_expert_gate.weight": "model-00015-of-00016.safetensors",
|
| 578 |
+
"model.language_model.layers.38.post_attention_layernorm.weight": "model-00015-of-00016.safetensors",
|
| 579 |
+
"model.language_model.layers.39.input_layernorm.weight": "model-00015-of-00016.safetensors",
|
| 580 |
+
"model.language_model.layers.39.mlp.experts.down_proj": "model-00016-of-00016.safetensors",
|
| 581 |
+
"model.language_model.layers.39.mlp.experts.gate_up_proj": "model-00016-of-00016.safetensors",
|
| 582 |
+
"model.language_model.layers.39.mlp.gate.weight": "model-00016-of-00016.safetensors",
|
| 583 |
+
"model.language_model.layers.39.mlp.shared_expert.down_proj.weight": "model-00016-of-00016.safetensors",
|
| 584 |
+
"model.language_model.layers.39.mlp.shared_expert.gate_proj.weight": "model-00016-of-00016.safetensors",
|
| 585 |
+
"model.language_model.layers.39.mlp.shared_expert.up_proj.weight": "model-00016-of-00016.safetensors",
|
| 586 |
+
"model.language_model.layers.39.mlp.shared_expert_gate.weight": "model-00016-of-00016.safetensors",
|
| 587 |
+
"model.language_model.layers.39.post_attention_layernorm.weight": "model-00016-of-00016.safetensors",
|
| 588 |
+
"model.language_model.layers.39.self_attn.k_norm.weight": "model-00016-of-00016.safetensors",
|
| 589 |
+
"model.language_model.layers.39.self_attn.k_proj.weight": "model-00016-of-00016.safetensors",
|
| 590 |
+
"model.language_model.layers.39.self_attn.o_proj.weight": "model-00016-of-00016.safetensors",
|
| 591 |
+
"model.language_model.layers.39.self_attn.q_norm.weight": "model-00016-of-00016.safetensors",
|
| 592 |
+
"model.language_model.layers.39.self_attn.q_proj.weight": "model-00016-of-00016.safetensors",
|
| 593 |
+
"model.language_model.layers.39.self_attn.v_proj.weight": "model-00016-of-00016.safetensors",
|
| 594 |
+
"model.language_model.layers.4.input_layernorm.weight": "model-00002-of-00016.safetensors",
|
| 595 |
+
"model.language_model.layers.4.linear_attn.A_log": "model-00002-of-00016.safetensors",
|
| 596 |
+
"model.language_model.layers.4.linear_attn.conv1d.weight": "model-00002-of-00016.safetensors",
|
| 597 |
+
"model.language_model.layers.4.linear_attn.dt_bias": "model-00002-of-00016.safetensors",
|
| 598 |
+
"model.language_model.layers.4.linear_attn.in_proj_a.weight": "model-00002-of-00016.safetensors",
|
| 599 |
+
"model.language_model.layers.4.linear_attn.in_proj_b.weight": "model-00002-of-00016.safetensors",
|
| 600 |
+
"model.language_model.layers.4.linear_attn.in_proj_qkv.weight": "model-00002-of-00016.safetensors",
|
| 601 |
+
"model.language_model.layers.4.linear_attn.in_proj_z.weight": "model-00002-of-00016.safetensors",
|
| 602 |
+
"model.language_model.layers.4.linear_attn.norm.weight": "model-00002-of-00016.safetensors",
|
| 603 |
+
"model.language_model.layers.4.linear_attn.out_proj.weight": "model-00002-of-00016.safetensors",
|
| 604 |
+
"model.language_model.layers.4.mlp.experts.down_proj": "model-00003-of-00016.safetensors",
|
| 605 |
+
"model.language_model.layers.4.mlp.experts.gate_up_proj": "model-00003-of-00016.safetensors",
|
| 606 |
+
"model.language_model.layers.4.mlp.gate.weight": "model-00003-of-00016.safetensors",
|
| 607 |
+
"model.language_model.layers.4.mlp.shared_expert.down_proj.weight": "model-00003-of-00016.safetensors",
|
| 608 |
+
"model.language_model.layers.4.mlp.shared_expert.gate_proj.weight": "model-00003-of-00016.safetensors",
|
| 609 |
+
"model.language_model.layers.4.mlp.shared_expert.up_proj.weight": "model-00003-of-00016.safetensors",
|
| 610 |
+
"model.language_model.layers.4.mlp.shared_expert_gate.weight": "model-00003-of-00016.safetensors",
|
| 611 |
+
"model.language_model.layers.4.post_attention_layernorm.weight": "model-00003-of-00016.safetensors",
|
| 612 |
+
"model.language_model.layers.5.input_layernorm.weight": "model-00003-of-00016.safetensors",
|
| 613 |
+
"model.language_model.layers.5.linear_attn.A_log": "model-00003-of-00016.safetensors",
|
| 614 |
+
"model.language_model.layers.5.linear_attn.conv1d.weight": "model-00003-of-00016.safetensors",
|
| 615 |
+
"model.language_model.layers.5.linear_attn.dt_bias": "model-00003-of-00016.safetensors",
|
| 616 |
+
"model.language_model.layers.5.linear_attn.in_proj_a.weight": "model-00003-of-00016.safetensors",
|
| 617 |
+
"model.language_model.layers.5.linear_attn.in_proj_b.weight": "model-00003-of-00016.safetensors",
|
| 618 |
+
"model.language_model.layers.5.linear_attn.in_proj_qkv.weight": "model-00003-of-00016.safetensors",
|
| 619 |
+
"model.language_model.layers.5.linear_attn.in_proj_z.weight": "model-00003-of-00016.safetensors",
|
| 620 |
+
"model.language_model.layers.5.linear_attn.norm.weight": "model-00003-of-00016.safetensors",
|
| 621 |
+
"model.language_model.layers.5.linear_attn.out_proj.weight": "model-00003-of-00016.safetensors",
|
| 622 |
+
"model.language_model.layers.5.mlp.experts.down_proj": "model-00003-of-00016.safetensors",
|
| 623 |
+
"model.language_model.layers.5.mlp.experts.gate_up_proj": "model-00003-of-00016.safetensors",
|
| 624 |
+
"model.language_model.layers.5.mlp.gate.weight": "model-00003-of-00016.safetensors",
|
| 625 |
+
"model.language_model.layers.5.mlp.shared_expert.down_proj.weight": "model-00003-of-00016.safetensors",
|
| 626 |
+
"model.language_model.layers.5.mlp.shared_expert.gate_proj.weight": "model-00003-of-00016.safetensors",
|
| 627 |
+
"model.language_model.layers.5.mlp.shared_expert.up_proj.weight": "model-00003-of-00016.safetensors",
|
| 628 |
+
"model.language_model.layers.5.mlp.shared_expert_gate.weight": "model-00003-of-00016.safetensors",
|
| 629 |
+
"model.language_model.layers.5.post_attention_layernorm.weight": "model-00003-of-00016.safetensors",
|
| 630 |
+
"model.language_model.layers.6.input_layernorm.weight": "model-00003-of-00016.safetensors",
|
| 631 |
+
"model.language_model.layers.6.linear_attn.A_log": "model-00003-of-00016.safetensors",
|
| 632 |
+
"model.language_model.layers.6.linear_attn.conv1d.weight": "model-00003-of-00016.safetensors",
|
| 633 |
+
"model.language_model.layers.6.linear_attn.dt_bias": "model-00003-of-00016.safetensors",
|
| 634 |
+
"model.language_model.layers.6.linear_attn.in_proj_a.weight": "model-00003-of-00016.safetensors",
|
| 635 |
+
"model.language_model.layers.6.linear_attn.in_proj_b.weight": "model-00003-of-00016.safetensors",
|
| 636 |
+
"model.language_model.layers.6.linear_attn.in_proj_qkv.weight": "model-00003-of-00016.safetensors",
|
| 637 |
+
"model.language_model.layers.6.linear_attn.in_proj_z.weight": "model-00003-of-00016.safetensors",
|
| 638 |
+
"model.language_model.layers.6.linear_attn.norm.weight": "model-00003-of-00016.safetensors",
|
| 639 |
+
"model.language_model.layers.6.linear_attn.out_proj.weight": "model-00003-of-00016.safetensors",
|
| 640 |
+
"model.language_model.layers.6.mlp.experts.down_proj": "model-00003-of-00016.safetensors",
|
| 641 |
+
"model.language_model.layers.6.mlp.experts.gate_up_proj": "model-00003-of-00016.safetensors",
|
| 642 |
+
"model.language_model.layers.6.mlp.gate.weight": "model-00003-of-00016.safetensors",
|
| 643 |
+
"model.language_model.layers.6.mlp.shared_expert.down_proj.weight": "model-00003-of-00016.safetensors",
|
| 644 |
+
"model.language_model.layers.6.mlp.shared_expert.gate_proj.weight": "model-00003-of-00016.safetensors",
|
| 645 |
+
"model.language_model.layers.6.mlp.shared_expert.up_proj.weight": "model-00003-of-00016.safetensors",
|
| 646 |
+
"model.language_model.layers.6.mlp.shared_expert_gate.weight": "model-00003-of-00016.safetensors",
|
| 647 |
+
"model.language_model.layers.6.post_attention_layernorm.weight": "model-00003-of-00016.safetensors",
|
| 648 |
+
"model.language_model.layers.7.input_layernorm.weight": "model-00003-of-00016.safetensors",
|
| 649 |
+
"model.language_model.layers.7.mlp.experts.down_proj": "model-00004-of-00016.safetensors",
|
| 650 |
+
"model.language_model.layers.7.mlp.experts.gate_up_proj": "model-00004-of-00016.safetensors",
|
| 651 |
+
"model.language_model.layers.7.mlp.gate.weight": "model-00004-of-00016.safetensors",
|
| 652 |
+
"model.language_model.layers.7.mlp.shared_expert.down_proj.weight": "model-00004-of-00016.safetensors",
|
| 653 |
+
"model.language_model.layers.7.mlp.shared_expert.gate_proj.weight": "model-00004-of-00016.safetensors",
|
| 654 |
+
"model.language_model.layers.7.mlp.shared_expert.up_proj.weight": "model-00004-of-00016.safetensors",
|
| 655 |
+
"model.language_model.layers.7.mlp.shared_expert_gate.weight": "model-00004-of-00016.safetensors",
|
| 656 |
+
"model.language_model.layers.7.post_attention_layernorm.weight": "model-00004-of-00016.safetensors",
|
| 657 |
+
"model.language_model.layers.7.self_attn.k_norm.weight": "model-00004-of-00016.safetensors",
|
| 658 |
+
"model.language_model.layers.7.self_attn.k_proj.weight": "model-00004-of-00016.safetensors",
|
| 659 |
+
"model.language_model.layers.7.self_attn.o_proj.weight": "model-00004-of-00016.safetensors",
|
| 660 |
+
"model.language_model.layers.7.self_attn.q_norm.weight": "model-00004-of-00016.safetensors",
|
| 661 |
+
"model.language_model.layers.7.self_attn.q_proj.weight": "model-00004-of-00016.safetensors",
|
| 662 |
+
"model.language_model.layers.7.self_attn.v_proj.weight": "model-00004-of-00016.safetensors",
|
| 663 |
+
"model.language_model.layers.8.input_layernorm.weight": "model-00004-of-00016.safetensors",
|
| 664 |
+
"model.language_model.layers.8.linear_attn.A_log": "model-00004-of-00016.safetensors",
|
| 665 |
+
"model.language_model.layers.8.linear_attn.conv1d.weight": "model-00004-of-00016.safetensors",
|
| 666 |
+
"model.language_model.layers.8.linear_attn.dt_bias": "model-00004-of-00016.safetensors",
|
| 667 |
+
"model.language_model.layers.8.linear_attn.in_proj_a.weight": "model-00004-of-00016.safetensors",
|
| 668 |
+
"model.language_model.layers.8.linear_attn.in_proj_b.weight": "model-00004-of-00016.safetensors",
|
| 669 |
+
"model.language_model.layers.8.linear_attn.in_proj_qkv.weight": "model-00004-of-00016.safetensors",
|
| 670 |
+
"model.language_model.layers.8.linear_attn.in_proj_z.weight": "model-00004-of-00016.safetensors",
|
| 671 |
+
"model.language_model.layers.8.linear_attn.norm.weight": "model-00004-of-00016.safetensors",
|
| 672 |
+
"model.language_model.layers.8.linear_attn.out_proj.weight": "model-00004-of-00016.safetensors",
|
| 673 |
+
"model.language_model.layers.8.mlp.experts.down_proj": "model-00004-of-00016.safetensors",
|
| 674 |
+
"model.language_model.layers.8.mlp.experts.gate_up_proj": "model-00004-of-00016.safetensors",
|
| 675 |
+
"model.language_model.layers.8.mlp.gate.weight": "model-00004-of-00016.safetensors",
|
| 676 |
+
"model.language_model.layers.8.mlp.shared_expert.down_proj.weight": "model-00004-of-00016.safetensors",
|
| 677 |
+
"model.language_model.layers.8.mlp.shared_expert.gate_proj.weight": "model-00004-of-00016.safetensors",
|
| 678 |
+
"model.language_model.layers.8.mlp.shared_expert.up_proj.weight": "model-00004-of-00016.safetensors",
|
| 679 |
+
"model.language_model.layers.8.mlp.shared_expert_gate.weight": "model-00004-of-00016.safetensors",
|
| 680 |
+
"model.language_model.layers.8.post_attention_layernorm.weight": "model-00004-of-00016.safetensors",
|
| 681 |
+
"model.language_model.layers.9.input_layernorm.weight": "model-00004-of-00016.safetensors",
|
| 682 |
+
"model.language_model.layers.9.linear_attn.A_log": "model-00004-of-00016.safetensors",
|
| 683 |
+
"model.language_model.layers.9.linear_attn.conv1d.weight": "model-00004-of-00016.safetensors",
|
| 684 |
+
"model.language_model.layers.9.linear_attn.dt_bias": "model-00004-of-00016.safetensors",
|
| 685 |
+
"model.language_model.layers.9.linear_attn.in_proj_a.weight": "model-00004-of-00016.safetensors",
|
| 686 |
+
"model.language_model.layers.9.linear_attn.in_proj_b.weight": "model-00004-of-00016.safetensors",
|
| 687 |
+
"model.language_model.layers.9.linear_attn.in_proj_qkv.weight": "model-00004-of-00016.safetensors",
|
| 688 |
+
"model.language_model.layers.9.linear_attn.in_proj_z.weight": "model-00004-of-00016.safetensors",
|
| 689 |
+
"model.language_model.layers.9.linear_attn.norm.weight": "model-00004-of-00016.safetensors",
|
| 690 |
+
"model.language_model.layers.9.linear_attn.out_proj.weight": "model-00004-of-00016.safetensors",
|
| 691 |
+
"model.language_model.layers.9.mlp.experts.down_proj": "model-00004-of-00016.safetensors",
|
| 692 |
+
"model.language_model.layers.9.mlp.experts.gate_up_proj": "model-00005-of-00016.safetensors",
|
| 693 |
+
"model.language_model.layers.9.mlp.gate.weight": "model-00005-of-00016.safetensors",
|
| 694 |
+
"model.language_model.layers.9.mlp.shared_expert.down_proj.weight": "model-00005-of-00016.safetensors",
|
| 695 |
+
"model.language_model.layers.9.mlp.shared_expert.gate_proj.weight": "model-00005-of-00016.safetensors",
|
| 696 |
+
"model.language_model.layers.9.mlp.shared_expert.up_proj.weight": "model-00005-of-00016.safetensors",
|
| 697 |
+
"model.language_model.layers.9.mlp.shared_expert_gate.weight": "model-00005-of-00016.safetensors",
|
| 698 |
+
"model.language_model.layers.9.post_attention_layernorm.weight": "model-00005-of-00016.safetensors",
|
| 699 |
+
"model.language_model.norm.weight": "model-00016-of-00016.safetensors",
|
| 700 |
+
"model.visual.blocks.15.attn.proj.weight": "model-visual-restored.safetensors",
|
| 701 |
+
"model.visual.blocks.11.attn.qkv.bias": "model-visual-restored.safetensors",
|
| 702 |
+
"model.visual.blocks.18.mlp.linear_fc2.bias": "model-visual-restored.safetensors",
|
| 703 |
+
"model.visual.blocks.11.norm2.bias": "model-visual-restored.safetensors",
|
| 704 |
+
"model.visual.blocks.0.mlp.linear_fc2.weight": "model-visual-restored.safetensors",
|
| 705 |
+
"model.visual.blocks.26.mlp.linear_fc1.weight": "model-visual-restored.safetensors",
|
| 706 |
+
"model.visual.blocks.23.mlp.linear_fc2.weight": "model-visual-restored.safetensors",
|
| 707 |
+
"model.visual.blocks.5.norm1.weight": "model-visual-restored.safetensors",
|
| 708 |
+
"model.visual.blocks.22.norm1.bias": "model-visual-restored.safetensors",
|
| 709 |
+
"model.visual.blocks.4.norm2.bias": "model-visual-restored.safetensors",
|
| 710 |
+
"model.visual.blocks.15.mlp.linear_fc2.weight": "model-visual-restored.safetensors",
|
| 711 |
+
"model.visual.blocks.1.norm2.bias": "model-visual-restored.safetensors",
|
| 712 |
+
"model.visual.blocks.5.mlp.linear_fc2.bias": "model-visual-restored.safetensors",
|
| 713 |
+
"model.visual.blocks.12.mlp.linear_fc2.weight": "model-visual-restored.safetensors",
|
| 714 |
+
"model.visual.blocks.15.norm2.bias": "model-visual-restored.safetensors",
|
| 715 |
+
"model.visual.blocks.26.attn.qkv.bias": "model-visual-restored.safetensors",
|
| 716 |
+
"model.visual.blocks.17.mlp.linear_fc2.weight": "model-visual-restored.safetensors",
|
| 717 |
+
"model.visual.blocks.5.attn.qkv.bias": "model-visual-restored.safetensors",
|
| 718 |
+
"model.visual.blocks.21.norm1.bias": "model-visual-restored.safetensors",
|
| 719 |
+
"model.visual.blocks.24.norm2.bias": "model-visual-restored.safetensors",
|
| 720 |
+
"model.visual.blocks.22.attn.proj.bias": "model-visual-restored.safetensors",
|
| 721 |
+
"model.visual.blocks.26.norm2.weight": "model-visual-restored.safetensors",
|
| 722 |
+
"model.visual.blocks.11.mlp.linear_fc1.weight": "model-visual-restored.safetensors",
|
| 723 |
+
"model.visual.blocks.6.norm1.weight": "model-visual-restored.safetensors",
|
| 724 |
+
"model.visual.blocks.22.attn.qkv.bias": "model-visual-restored.safetensors",
|
| 725 |
+
"model.visual.blocks.14.norm1.bias": "model-visual-restored.safetensors",
|
| 726 |
+
"model.visual.blocks.14.norm2.bias": "model-visual-restored.safetensors",
|
| 727 |
+
"model.visual.blocks.19.attn.qkv.bias": "model-visual-restored.safetensors",
|
| 728 |
+
"model.visual.blocks.22.norm2.bias": "model-visual-restored.safetensors",
|
| 729 |
+
"model.visual.blocks.25.attn.qkv.bias": "model-visual-restored.safetensors",
|
| 730 |
+
"model.visual.blocks.12.norm1.weight": "model-visual-restored.safetensors",
|
| 731 |
+
"model.visual.blocks.21.attn.proj.weight": "model-visual-restored.safetensors",
|
| 732 |
+
"model.visual.blocks.10.mlp.linear_fc2.weight": "model-visual-restored.safetensors",
|
| 733 |
+
"model.visual.blocks.19.norm1.weight": "model-visual-restored.safetensors",
|
| 734 |
+
"model.visual.blocks.22.mlp.linear_fc1.weight": "model-visual-restored.safetensors",
|
| 735 |
+
"model.visual.blocks.3.norm1.weight": "model-visual-restored.safetensors",
|
| 736 |
+
"model.visual.blocks.14.attn.proj.bias": "model-visual-restored.safetensors",
|
| 737 |
+
"model.visual.blocks.20.mlp.linear_fc2.bias": "model-visual-restored.safetensors",
|
| 738 |
+
"model.visual.blocks.19.mlp.linear_fc1.bias": "model-visual-restored.safetensors",
|
| 739 |
+
"model.visual.blocks.0.norm2.weight": "model-visual-restored.safetensors",
|
| 740 |
+
"model.visual.blocks.22.mlp.linear_fc2.bias": "model-visual-restored.safetensors",
|
| 741 |
+
"model.visual.blocks.8.norm2.bias": "model-visual-restored.safetensors",
|
| 742 |
+
"model.visual.blocks.19.attn.qkv.weight": "model-visual-restored.safetensors",
|
| 743 |
+
"model.visual.blocks.2.mlp.linear_fc1.bias": "model-visual-restored.safetensors",
|
| 744 |
+
"model.visual.blocks.1.attn.proj.weight": "model-visual-restored.safetensors",
|
| 745 |
+
"model.visual.blocks.7.attn.qkv.bias": "model-visual-restored.safetensors",
|
| 746 |
+
"model.visual.blocks.15.norm1.bias": "model-visual-restored.safetensors",
|
| 747 |
+
"model.visual.blocks.0.mlp.linear_fc2.bias": "model-visual-restored.safetensors",
|
| 748 |
+
"model.visual.blocks.23.mlp.linear_fc2.bias": "model-visual-restored.safetensors",
|
| 749 |
+
"model.visual.blocks.23.norm1.weight": "model-visual-restored.safetensors",
|
| 750 |
+
"model.visual.blocks.21.attn.qkv.bias": "model-visual-restored.safetensors",
|
| 751 |
+
"model.visual.blocks.14.norm2.weight": "model-visual-restored.safetensors",
|
| 752 |
+
"model.visual.blocks.11.mlp.linear_fc2.bias": "model-visual-restored.safetensors",
|
| 753 |
+
"model.visual.blocks.9.attn.qkv.weight": "model-visual-restored.safetensors",
|
| 754 |
+
"model.visual.blocks.17.mlp.linear_fc2.bias": "model-visual-restored.safetensors",
|
| 755 |
+
"model.visual.blocks.20.attn.qkv.weight": "model-visual-restored.safetensors",
|
| 756 |
+
"model.visual.blocks.1.mlp.linear_fc2.bias": "model-visual-restored.safetensors",
|
| 757 |
+
"model.visual.blocks.9.mlp.linear_fc1.weight": "model-visual-restored.safetensors",
|
| 758 |
+
"model.visual.blocks.2.norm2.weight": "model-visual-restored.safetensors",
|
| 759 |
+
"model.visual.blocks.6.attn.proj.bias": "model-visual-restored.safetensors",
|
| 760 |
+
"model.visual.blocks.5.mlp.linear_fc1.bias": "model-visual-restored.safetensors",
|
| 761 |
+
"model.visual.blocks.26.norm1.bias": "model-visual-restored.safetensors",
|
| 762 |
+
"model.visual.blocks.15.norm2.weight": "model-visual-restored.safetensors",
|
| 763 |
+
"model.visual.blocks.12.mlp.linear_fc1.bias": "model-visual-restored.safetensors",
|
| 764 |
+
"model.visual.merger.linear_fc1.weight": "model-visual-restored.safetensors",
|
| 765 |
+
"model.visual.blocks.14.attn.qkv.weight": "model-visual-restored.safetensors",
|
| 766 |
+
"model.visual.blocks.11.mlp.linear_fc1.bias": "model-visual-restored.safetensors",
|
| 767 |
+
"model.visual.blocks.6.mlp.linear_fc2.weight": "model-visual-restored.safetensors",
|
| 768 |
+
"model.visual.blocks.22.norm2.weight": "model-visual-restored.safetensors",
|
| 769 |
+
"model.visual.blocks.19.attn.proj.weight": "model-visual-restored.safetensors",
|
| 770 |
+
"model.visual.blocks.12.norm1.bias": "model-visual-restored.safetensors",
|
| 771 |
+
"model.visual.blocks.25.attn.proj.weight": "model-visual-restored.safetensors",
|
| 772 |
+
"model.visual.blocks.1.norm2.weight": "model-visual-restored.safetensors",
|
| 773 |
+
"model.visual.blocks.26.attn.qkv.weight": "model-visual-restored.safetensors",
|
| 774 |
+
"model.visual.blocks.11.mlp.linear_fc2.weight": "model-visual-restored.safetensors",
|
| 775 |
+
"model.visual.blocks.20.mlp.linear_fc1.weight": "model-visual-restored.safetensors",
|
| 776 |
+
"model.visual.blocks.8.norm1.bias": "model-visual-restored.safetensors",
|
| 777 |
+
"model.visual.blocks.7.attn.proj.bias": "model-visual-restored.safetensors",
|
| 778 |
+
"model.visual.blocks.8.attn.qkv.bias": "model-visual-restored.safetensors",
|
| 779 |
+
"model.visual.blocks.17.mlp.linear_fc1.weight": "model-visual-restored.safetensors",
|
| 780 |
+
"model.visual.blocks.7.norm1.bias": "model-visual-restored.safetensors",
|
| 781 |
+
"model.visual.blocks.14.norm1.weight": "model-visual-restored.safetensors",
|
| 782 |
+
"model.visual.blocks.18.norm2.bias": "model-visual-restored.safetensors",
|
| 783 |
+
"model.visual.blocks.15.mlp.linear_fc1.bias": "model-visual-restored.safetensors",
|
| 784 |
+
"model.visual.blocks.17.norm1.weight": "model-visual-restored.safetensors",
|
| 785 |
+
"model.visual.blocks.19.norm2.bias": "model-visual-restored.safetensors",
|
| 786 |
+
"model.visual.blocks.25.attn.qkv.weight": "model-visual-restored.safetensors",
|
| 787 |
+
"model.visual.blocks.9.norm2.weight": "model-visual-restored.safetensors",
|
| 788 |
+
"model.visual.blocks.13.norm1.bias": "model-visual-restored.safetensors",
|
| 789 |
+
"model.visual.blocks.7.mlp.linear_fc2.weight": "model-visual-restored.safetensors",
|
| 790 |
+
"model.visual.blocks.20.attn.proj.weight": "model-visual-restored.safetensors",
|
| 791 |
+
"model.visual.blocks.1.mlp.linear_fc1.bias": "model-visual-restored.safetensors",
|
| 792 |
+
"model.visual.blocks.2.mlp.linear_fc2.weight": "model-visual-restored.safetensors",
|
| 793 |
+
"model.visual.blocks.24.mlp.linear_fc2.bias": "model-visual-restored.safetensors",
|
| 794 |
+
"model.visual.blocks.3.attn.proj.bias": "model-visual-restored.safetensors",
|
| 795 |
+
"model.visual.blocks.0.norm1.weight": "model-visual-restored.safetensors",
|
| 796 |
+
"model.visual.blocks.9.mlp.linear_fc1.bias": "model-visual-restored.safetensors",
|
| 797 |
+
"model.visual.blocks.5.norm1.bias": "model-visual-restored.safetensors",
|
| 798 |
+
"model.visual.blocks.20.attn.proj.bias": "model-visual-restored.safetensors",
|
| 799 |
+
"model.visual.merger.linear_fc2.bias": "model-visual-restored.safetensors",
|
| 800 |
+
"model.visual.blocks.15.norm1.weight": "model-visual-restored.safetensors",
|
| 801 |
+
"model.visual.blocks.16.norm2.weight": "model-visual-restored.safetensors",
|
| 802 |
+
"model.visual.blocks.4.attn.proj.weight": "model-visual-restored.safetensors",
|
| 803 |
+
"model.visual.blocks.7.norm2.bias": "model-visual-restored.safetensors",
|
| 804 |
+
"model.visual.blocks.24.norm1.weight": "model-visual-restored.safetensors",
|
| 805 |
+
"model.visual.blocks.17.norm2.weight": "model-visual-restored.safetensors",
|
| 806 |
+
"model.visual.blocks.3.mlp.linear_fc1.bias": "model-visual-restored.safetensors",
|
| 807 |
+
"model.visual.blocks.8.mlp.linear_fc2.weight": "model-visual-restored.safetensors",
|
| 808 |
+
"model.visual.blocks.0.mlp.linear_fc1.bias": "model-visual-restored.safetensors",
|
| 809 |
+
"model.visual.blocks.6.norm2.bias": "model-visual-restored.safetensors",
|
| 810 |
+
"model.visual.blocks.13.attn.proj.bias": "model-visual-restored.safetensors",
|
| 811 |
+
"model.visual.blocks.2.attn.qkv.weight": "model-visual-restored.safetensors",
|
| 812 |
+
"model.visual.blocks.8.mlp.linear_fc1.bias": "model-visual-restored.safetensors",
|
| 813 |
+
"model.visual.blocks.25.mlp.linear_fc1.weight": "model-visual-restored.safetensors",
|
| 814 |
+
"model.visual.blocks.21.attn.qkv.weight": "model-visual-restored.safetensors",
|
| 815 |
+
"model.visual.blocks.20.norm2.weight": "model-visual-restored.safetensors",
|
| 816 |
+
"model.visual.blocks.6.attn.proj.weight": "model-visual-restored.safetensors",
|
| 817 |
+
"model.visual.blocks.25.mlp.linear_fc1.bias": "model-visual-restored.safetensors",
|
| 818 |
+
"model.visual.blocks.16.attn.qkv.bias": "model-visual-restored.safetensors",
|
| 819 |
+
"model.visual.blocks.22.norm1.weight": "model-visual-restored.safetensors",
|
| 820 |
+
"model.visual.blocks.17.attn.qkv.bias": "model-visual-restored.safetensors",
|
| 821 |
+
"model.visual.blocks.4.norm1.weight": "model-visual-restored.safetensors",
|
| 822 |
+
"model.visual.blocks.2.attn.proj.bias": "model-visual-restored.safetensors",
|
| 823 |
+
"model.visual.blocks.5.norm2.weight": "model-visual-restored.safetensors",
|
| 824 |
+
"model.visual.blocks.18.mlp.linear_fc1.weight": "model-visual-restored.safetensors",
|
| 825 |
+
"model.visual.blocks.18.mlp.linear_fc2.weight": "model-visual-restored.safetensors",
|
| 826 |
+
"model.visual.blocks.8.norm1.weight": "model-visual-restored.safetensors",
|
| 827 |
+
"model.visual.blocks.6.attn.qkv.bias": "model-visual-restored.safetensors",
|
| 828 |
+
"model.visual.blocks.24.mlp.linear_fc1.weight": "model-visual-restored.safetensors",
|
| 829 |
+
"model.visual.blocks.3.norm2.bias": "model-visual-restored.safetensors",
|
| 830 |
+
"model.visual.blocks.9.norm2.bias": "model-visual-restored.safetensors",
|
| 831 |
+
"model.visual.blocks.16.mlp.linear_fc1.bias": "model-visual-restored.safetensors",
|
| 832 |
+
"model.visual.blocks.14.mlp.linear_fc1.bias": "model-visual-restored.safetensors",
|
| 833 |
+
"model.visual.blocks.25.norm2.weight": "model-visual-restored.safetensors",
|
| 834 |
+
"model.visual.blocks.15.attn.qkv.weight": "model-visual-restored.safetensors",
|
| 835 |
+
"model.visual.blocks.12.attn.proj.bias": "model-visual-restored.safetensors",
|
| 836 |
+
"model.visual.blocks.21.mlp.linear_fc2.bias": "model-visual-restored.safetensors",
|
| 837 |
+
"model.visual.blocks.16.mlp.linear_fc2.bias": "model-visual-restored.safetensors",
|
| 838 |
+
"model.visual.blocks.3.mlp.linear_fc1.weight": "model-visual-restored.safetensors",
|
| 839 |
+
"model.visual.blocks.10.attn.proj.bias": "model-visual-restored.safetensors",
|
| 840 |
+
"model.visual.blocks.21.mlp.linear_fc2.weight": "model-visual-restored.safetensors",
|
| 841 |
+
"model.visual.blocks.13.mlp.linear_fc1.weight": "model-visual-restored.safetensors",
|
| 842 |
+
"model.visual.blocks.13.mlp.linear_fc1.bias": "model-visual-restored.safetensors",
|
| 843 |
+
"model.visual.blocks.19.norm2.weight": "model-visual-restored.safetensors",
|
| 844 |
+
"model.visual.blocks.3.norm1.bias": "model-visual-restored.safetensors",
|
| 845 |
+
"model.visual.blocks.8.attn.proj.weight": "model-visual-restored.safetensors",
|
| 846 |
+
"model.visual.blocks.22.attn.qkv.weight": "model-visual-restored.safetensors",
|
| 847 |
+
"model.visual.blocks.5.norm2.bias": "model-visual-restored.safetensors",
|
| 848 |
+
"model.visual.blocks.11.norm1.bias": "model-visual-restored.safetensors",
|
| 849 |
+
"model.visual.blocks.16.norm1.weight": "model-visual-restored.safetensors",
|
| 850 |
+
"model.visual.blocks.23.attn.proj.weight": "model-visual-restored.safetensors",
|
| 851 |
+
"model.visual.blocks.21.norm1.weight": "model-visual-restored.safetensors",
|
| 852 |
+
"model.visual.blocks.22.attn.proj.weight": "model-visual-restored.safetensors",
|
| 853 |
+
"model.visual.blocks.14.mlp.linear_fc2.weight": "model-visual-restored.safetensors",
|
| 854 |
+
"model.visual.blocks.17.attn.proj.weight": "model-visual-restored.safetensors",
|
| 855 |
+
"model.visual.blocks.0.norm2.bias": "model-visual-restored.safetensors",
|
| 856 |
+
"model.visual.blocks.19.norm1.bias": "model-visual-restored.safetensors",
|
| 857 |
+
"model.visual.blocks.10.attn.qkv.weight": "model-visual-restored.safetensors",
|
| 858 |
+
"model.visual.blocks.18.attn.proj.bias": "model-visual-restored.safetensors",
|
| 859 |
+
"model.visual.blocks.5.mlp.linear_fc1.weight": "model-visual-restored.safetensors",
|
| 860 |
+
"model.visual.blocks.12.attn.proj.weight": "model-visual-restored.safetensors",
|
| 861 |
+
"model.visual.blocks.5.attn.proj.bias": "model-visual-restored.safetensors",
|
| 862 |
+
"model.visual.blocks.3.mlp.linear_fc2.weight": "model-visual-restored.safetensors",
|
| 863 |
+
"model.visual.blocks.14.attn.proj.weight": "model-visual-restored.safetensors",
|
| 864 |
+
"model.visual.blocks.17.norm1.bias": "model-visual-restored.safetensors",
|
| 865 |
+
"model.visual.blocks.24.mlp.linear_fc1.bias": "model-visual-restored.safetensors",
|
| 866 |
+
"model.visual.blocks.12.norm2.weight": "model-visual-restored.safetensors",
|
| 867 |
+
"model.visual.blocks.0.mlp.linear_fc1.weight": "model-visual-restored.safetensors",
|
| 868 |
+
"model.visual.blocks.11.norm1.weight": "model-visual-restored.safetensors",
|
| 869 |
+
"model.visual.blocks.4.mlp.linear_fc1.bias": "model-visual-restored.safetensors",
|
| 870 |
+
"model.visual.blocks.24.norm2.weight": "model-visual-restored.safetensors",
|
| 871 |
+
"model.visual.blocks.13.attn.qkv.bias": "model-visual-restored.safetensors",
|
| 872 |
+
"model.visual.merger.linear_fc2.weight": "model-visual-restored.safetensors",
|
| 873 |
+
"model.visual.blocks.1.mlp.linear_fc2.weight": "model-visual-restored.safetensors",
|
| 874 |
+
"model.visual.blocks.26.mlp.linear_fc2.bias": "model-visual-restored.safetensors",
|
| 875 |
+
"model.visual.blocks.3.attn.qkv.weight": "model-visual-restored.safetensors",
|
| 876 |
+
"model.visual.blocks.2.norm2.bias": "model-visual-restored.safetensors",
|
| 877 |
+
"model.visual.blocks.0.attn.proj.weight": "model-visual-restored.safetensors",
|
| 878 |
+
"model.visual.blocks.26.mlp.linear_fc1.bias": "model-visual-restored.safetensors",
|
| 879 |
+
"model.visual.blocks.25.mlp.linear_fc2.bias": "model-visual-restored.safetensors",
|
| 880 |
+
"model.visual.blocks.15.mlp.linear_fc2.bias": "model-visual-restored.safetensors",
|
| 881 |
+
"model.visual.blocks.7.mlp.linear_fc2.bias": "model-visual-restored.safetensors",
|
| 882 |
+
"model.visual.blocks.4.mlp.linear_fc2.bias": "model-visual-restored.safetensors",
|
| 883 |
+
"model.visual.blocks.5.attn.proj.weight": "model-visual-restored.safetensors",
|
| 884 |
+
"model.visual.blocks.6.mlp.linear_fc1.weight": "model-visual-restored.safetensors",
|
| 885 |
+
"model.visual.patch_embed.proj.weight": "model-visual-restored.safetensors",
|
| 886 |
+
"model.visual.blocks.11.attn.qkv.weight": "model-visual-restored.safetensors",
|
| 887 |
+
"model.visual.blocks.7.attn.proj.weight": "model-visual-restored.safetensors",
|
| 888 |
+
"model.visual.blocks.10.norm2.bias": "model-visual-restored.safetensors",
|
| 889 |
+
"model.visual.blocks.17.attn.proj.bias": "model-visual-restored.safetensors",
|
| 890 |
+
"model.visual.blocks.9.attn.proj.bias": "model-visual-restored.safetensors",
|
| 891 |
+
"model.visual.merger.norm.weight": "model-visual-restored.safetensors",
|
| 892 |
+
"model.visual.blocks.24.attn.proj.weight": "model-visual-restored.safetensors",
|
| 893 |
+
"model.visual.blocks.12.norm2.bias": "model-visual-restored.safetensors",
|
| 894 |
+
"model.visual.blocks.10.attn.qkv.bias": "model-visual-restored.safetensors",
|
| 895 |
+
"model.visual.blocks.9.norm1.bias": "model-visual-restored.safetensors",
|
| 896 |
+
"model.visual.blocks.15.attn.qkv.bias": "model-visual-restored.safetensors",
|
| 897 |
+
"model.visual.blocks.10.mlp.linear_fc2.bias": "model-visual-restored.safetensors",
|
| 898 |
+
"model.visual.blocks.19.attn.proj.bias": "model-visual-restored.safetensors",
|
| 899 |
+
"model.visual.blocks.4.attn.qkv.bias": "model-visual-restored.safetensors",
|
| 900 |
+
"model.visual.blocks.24.attn.qkv.bias": "model-visual-restored.safetensors",
|
| 901 |
+
"model.visual.blocks.4.norm1.bias": "model-visual-restored.safetensors",
|
| 902 |
+
"model.visual.blocks.2.mlp.linear_fc2.bias": "model-visual-restored.safetensors",
|
| 903 |
+
"model.visual.blocks.6.norm1.bias": "model-visual-restored.safetensors",
|
| 904 |
+
"model.visual.blocks.24.attn.qkv.weight": "model-visual-restored.safetensors",
|
| 905 |
+
"model.visual.blocks.4.norm2.weight": "model-visual-restored.safetensors",
|
| 906 |
+
"model.visual.blocks.8.attn.qkv.weight": "model-visual-restored.safetensors",
|
| 907 |
+
"model.visual.blocks.7.norm2.weight": "model-visual-restored.safetensors",
|
| 908 |
+
"model.visual.blocks.26.norm1.weight": "model-visual-restored.safetensors",
|
| 909 |
+
"model.visual.blocks.1.attn.proj.bias": "model-visual-restored.safetensors",
|
| 910 |
+
"model.visual.blocks.11.attn.proj.bias": "model-visual-restored.safetensors",
|
| 911 |
+
"model.visual.blocks.6.norm2.weight": "model-visual-restored.safetensors",
|
| 912 |
+
"model.visual.blocks.10.mlp.linear_fc1.bias": "model-visual-restored.safetensors",
|
| 913 |
+
"model.visual.blocks.6.attn.qkv.weight": "model-visual-restored.safetensors",
|
| 914 |
+
"model.visual.blocks.26.attn.proj.weight": "model-visual-restored.safetensors",
|
| 915 |
+
"model.visual.blocks.16.attn.qkv.weight": "model-visual-restored.safetensors",
|
| 916 |
+
"model.visual.blocks.2.attn.proj.weight": "model-visual-restored.safetensors",
|
| 917 |
+
"model.visual.blocks.13.mlp.linear_fc2.weight": "model-visual-restored.safetensors",
|
| 918 |
+
"model.visual.blocks.23.norm2.bias": "model-visual-restored.safetensors",
|
| 919 |
+
"model.visual.blocks.7.attn.qkv.weight": "model-visual-restored.safetensors",
|
| 920 |
+
"model.visual.blocks.11.attn.proj.weight": "model-visual-restored.safetensors",
|
| 921 |
+
"model.visual.blocks.16.norm1.bias": "model-visual-restored.safetensors",
|
| 922 |
+
"model.visual.blocks.21.mlp.linear_fc1.bias": "model-visual-restored.safetensors",
|
| 923 |
+
"model.visual.blocks.17.norm2.bias": "model-visual-restored.safetensors",
|
| 924 |
+
"model.visual.blocks.18.attn.qkv.weight": "model-visual-restored.safetensors",
|
| 925 |
+
"model.visual.blocks.23.mlp.linear_fc1.bias": "model-visual-restored.safetensors",
|
| 926 |
+
"model.visual.blocks.20.norm1.weight": "model-visual-restored.safetensors",
|
| 927 |
+
"model.visual.blocks.17.mlp.linear_fc1.bias": "model-visual-restored.safetensors",
|
| 928 |
+
"model.visual.blocks.20.attn.qkv.bias": "model-visual-restored.safetensors",
|
| 929 |
+
"model.visual.blocks.23.attn.qkv.weight": "model-visual-restored.safetensors",
|
| 930 |
+
"model.visual.blocks.1.norm1.bias": "model-visual-restored.safetensors",
|
| 931 |
+
"model.visual.blocks.0.attn.qkv.bias": "model-visual-restored.safetensors",
|
| 932 |
+
"model.visual.blocks.24.norm1.bias": "model-visual-restored.safetensors",
|
| 933 |
+
"model.visual.blocks.2.norm1.weight": "model-visual-restored.safetensors",
|
| 934 |
+
"model.visual.blocks.24.attn.proj.bias": "model-visual-restored.safetensors",
|
| 935 |
+
"model.visual.blocks.1.norm1.weight": "model-visual-restored.safetensors",
|
| 936 |
+
"model.visual.blocks.0.attn.qkv.weight": "model-visual-restored.safetensors",
|
| 937 |
+
"model.visual.blocks.12.mlp.linear_fc1.weight": "model-visual-restored.safetensors",
|
| 938 |
+
"model.visual.blocks.23.norm1.bias": "model-visual-restored.safetensors",
|
| 939 |
+
"model.visual.blocks.12.mlp.linear_fc2.bias": "model-visual-restored.safetensors",
|
| 940 |
+
"model.visual.blocks.4.mlp.linear_fc2.weight": "model-visual-restored.safetensors",
|
| 941 |
+
"model.visual.blocks.13.attn.qkv.weight": "model-visual-restored.safetensors",
|
| 942 |
+
"model.visual.blocks.26.mlp.linear_fc2.weight": "model-visual-restored.safetensors",
|
| 943 |
+
"model.visual.blocks.23.norm2.weight": "model-visual-restored.safetensors",
|
| 944 |
+
"model.visual.blocks.20.mlp.linear_fc2.weight": "model-visual-restored.safetensors",
|
| 945 |
+
"model.visual.blocks.6.mlp.linear_fc2.bias": "model-visual-restored.safetensors",
|
| 946 |
+
"model.visual.blocks.21.attn.proj.bias": "model-visual-restored.safetensors",
|
| 947 |
+
"model.visual.blocks.8.mlp.linear_fc1.weight": "model-visual-restored.safetensors",
|
| 948 |
+
"model.visual.blocks.12.attn.qkv.bias": "model-visual-restored.safetensors",
|
| 949 |
+
"model.visual.blocks.13.mlp.linear_fc2.bias": "model-visual-restored.safetensors",
|
| 950 |
+
"model.visual.blocks.25.mlp.linear_fc2.weight": "model-visual-restored.safetensors",
|
| 951 |
+
"model.visual.pos_embed.weight": "model-visual-restored.safetensors",
|
| 952 |
+
"model.visual.blocks.25.attn.proj.bias": "model-visual-restored.safetensors",
|
| 953 |
+
"model.visual.blocks.26.attn.proj.bias": "model-visual-restored.safetensors",
|
| 954 |
+
"model.visual.merger.linear_fc1.bias": "model-visual-restored.safetensors",
|
| 955 |
+
"model.visual.blocks.1.mlp.linear_fc1.weight": "model-visual-restored.safetensors",
|
| 956 |
+
"model.visual.blocks.9.norm1.weight": "model-visual-restored.safetensors",
|
| 957 |
+
"model.visual.blocks.12.attn.qkv.weight": "model-visual-restored.safetensors",
|
| 958 |
+
"model.visual.blocks.18.norm1.bias": "model-visual-restored.safetensors",
|
| 959 |
+
"model.visual.merger.norm.bias": "model-visual-restored.safetensors",
|
| 960 |
+
"model.visual.blocks.19.mlp.linear_fc1.weight": "model-visual-restored.safetensors",
|
| 961 |
+
"model.visual.blocks.3.attn.qkv.bias": "model-visual-restored.safetensors",
|
| 962 |
+
"model.visual.blocks.3.attn.proj.weight": "model-visual-restored.safetensors",
|
| 963 |
+
"model.visual.blocks.3.mlp.linear_fc2.bias": "model-visual-restored.safetensors",
|
| 964 |
+
"model.visual.blocks.4.mlp.linear_fc1.weight": "model-visual-restored.safetensors",
|
| 965 |
+
"model.visual.blocks.20.norm2.bias": "model-visual-restored.safetensors",
|
| 966 |
+
"model.visual.blocks.18.mlp.linear_fc1.bias": "model-visual-restored.safetensors",
|
| 967 |
+
"model.visual.blocks.19.mlp.linear_fc2.weight": "model-visual-restored.safetensors",
|
| 968 |
+
"model.visual.blocks.9.mlp.linear_fc2.weight": "model-visual-restored.safetensors",
|
| 969 |
+
"model.visual.blocks.4.attn.proj.bias": "model-visual-restored.safetensors",
|
| 970 |
+
"model.visual.blocks.10.attn.proj.weight": "model-visual-restored.safetensors",
|
| 971 |
+
"model.visual.blocks.25.norm2.bias": "model-visual-restored.safetensors",
|
| 972 |
+
"model.visual.blocks.10.norm1.weight": "model-visual-restored.safetensors",
|
| 973 |
+
"model.visual.blocks.7.mlp.linear_fc1.bias": "model-visual-restored.safetensors",
|
| 974 |
+
"model.visual.blocks.4.attn.qkv.weight": "model-visual-restored.safetensors",
|
| 975 |
+
"model.visual.blocks.3.norm2.weight": "model-visual-restored.safetensors",
|
| 976 |
+
"model.visual.blocks.2.attn.qkv.bias": "model-visual-restored.safetensors",
|
| 977 |
+
"model.visual.blocks.6.mlp.linear_fc1.bias": "model-visual-restored.safetensors",
|
| 978 |
+
"model.visual.blocks.9.mlp.linear_fc2.bias": "model-visual-restored.safetensors",
|
| 979 |
+
"model.visual.blocks.2.norm1.bias": "model-visual-restored.safetensors",
|
| 980 |
+
"model.visual.blocks.10.norm1.bias": "model-visual-restored.safetensors",
|
| 981 |
+
"model.visual.blocks.16.attn.proj.weight": "model-visual-restored.safetensors",
|
| 982 |
+
"model.visual.blocks.8.attn.proj.bias": "model-visual-restored.safetensors",
|
| 983 |
+
"model.visual.blocks.9.attn.proj.weight": "model-visual-restored.safetensors",
|
| 984 |
+
"model.visual.blocks.18.norm1.weight": "model-visual-restored.safetensors",
|
| 985 |
+
"model.visual.blocks.8.norm2.weight": "model-visual-restored.safetensors",
|
| 986 |
+
"model.visual.blocks.21.norm2.weight": "model-visual-restored.safetensors",
|
| 987 |
+
"model.visual.blocks.1.attn.qkv.weight": "model-visual-restored.safetensors",
|
| 988 |
+
"model.visual.blocks.21.norm2.bias": "model-visual-restored.safetensors",
|
| 989 |
+
"model.visual.blocks.0.norm1.bias": "model-visual-restored.safetensors",
|
| 990 |
+
"model.visual.blocks.20.norm1.bias": "model-visual-restored.safetensors",
|
| 991 |
+
"model.visual.blocks.5.attn.qkv.weight": "model-visual-restored.safetensors",
|
| 992 |
+
"model.visual.blocks.5.mlp.linear_fc2.weight": "model-visual-restored.safetensors",
|
| 993 |
+
"model.visual.blocks.18.attn.qkv.bias": "model-visual-restored.safetensors",
|
| 994 |
+
"model.visual.blocks.13.attn.proj.weight": "model-visual-restored.safetensors",
|
| 995 |
+
"model.visual.blocks.20.mlp.linear_fc1.bias": "model-visual-restored.safetensors",
|
| 996 |
+
"model.visual.blocks.24.mlp.linear_fc2.weight": "model-visual-restored.safetensors",
|
| 997 |
+
"model.visual.blocks.23.attn.proj.bias": "model-visual-restored.safetensors",
|
| 998 |
+
"model.visual.blocks.7.mlp.linear_fc1.weight": "model-visual-restored.safetensors",
|
| 999 |
+
"model.visual.blocks.13.norm1.weight": "model-visual-restored.safetensors",
|
| 1000 |
+
"model.visual.blocks.9.attn.qkv.bias": "model-visual-restored.safetensors",
|
| 1001 |
+
"model.visual.blocks.7.norm1.weight": "model-visual-restored.safetensors",
|
| 1002 |
+
"model.visual.blocks.17.attn.qkv.weight": "model-visual-restored.safetensors",
|
| 1003 |
+
"model.visual.blocks.10.norm2.weight": "model-visual-restored.safetensors",
|
| 1004 |
+
"model.visual.blocks.16.mlp.linear_fc2.weight": "model-visual-restored.safetensors",
|
| 1005 |
+
"model.visual.blocks.16.mlp.linear_fc1.weight": "model-visual-restored.safetensors",
|
| 1006 |
+
"model.visual.blocks.1.attn.qkv.bias": "model-visual-restored.safetensors",
|
| 1007 |
+
"model.visual.blocks.22.mlp.linear_fc1.bias": "model-visual-restored.safetensors",
|
| 1008 |
+
"model.visual.blocks.16.attn.proj.bias": "model-visual-restored.safetensors",
|
| 1009 |
+
"model.visual.blocks.22.mlp.linear_fc2.weight": "model-visual-restored.safetensors",
|
| 1010 |
+
"model.visual.blocks.10.mlp.linear_fc1.weight": "model-visual-restored.safetensors",
|
| 1011 |
+
"model.visual.blocks.25.norm1.weight": "model-visual-restored.safetensors",
|
| 1012 |
+
"model.visual.blocks.15.mlp.linear_fc1.weight": "model-visual-restored.safetensors",
|
| 1013 |
+
"model.visual.blocks.8.mlp.linear_fc2.bias": "model-visual-restored.safetensors",
|
| 1014 |
+
"model.visual.blocks.14.attn.qkv.bias": "model-visual-restored.safetensors",
|
| 1015 |
+
"model.visual.blocks.11.norm2.weight": "model-visual-restored.safetensors",
|
| 1016 |
+
"model.visual.blocks.25.norm1.bias": "model-visual-restored.safetensors",
|
| 1017 |
+
"model.visual.blocks.13.norm2.weight": "model-visual-restored.safetensors",
|
| 1018 |
+
"model.visual.blocks.21.mlp.linear_fc1.weight": "model-visual-restored.safetensors",
|
| 1019 |
+
"model.visual.blocks.2.mlp.linear_fc1.weight": "model-visual-restored.safetensors",
|
| 1020 |
+
"model.visual.blocks.19.mlp.linear_fc2.bias": "model-visual-restored.safetensors",
|
| 1021 |
+
"model.visual.patch_embed.proj.bias": "model-visual-restored.safetensors",
|
| 1022 |
+
"model.visual.blocks.26.norm2.bias": "model-visual-restored.safetensors",
|
| 1023 |
+
"model.visual.blocks.18.norm2.weight": "model-visual-restored.safetensors",
|
| 1024 |
+
"model.visual.blocks.14.mlp.linear_fc2.bias": "model-visual-restored.safetensors",
|
| 1025 |
+
"model.visual.blocks.13.norm2.bias": "model-visual-restored.safetensors",
|
| 1026 |
+
"model.visual.blocks.18.attn.proj.weight": "model-visual-restored.safetensors",
|
| 1027 |
+
"model.visual.blocks.15.attn.proj.bias": "model-visual-restored.safetensors",
|
| 1028 |
+
"model.visual.blocks.23.mlp.linear_fc1.weight": "model-visual-restored.safetensors",
|
| 1029 |
+
"model.visual.blocks.14.mlp.linear_fc1.weight": "model-visual-restored.safetensors",
|
| 1030 |
+
"model.visual.blocks.0.attn.proj.bias": "model-visual-restored.safetensors",
|
| 1031 |
+
"model.visual.blocks.16.norm2.bias": "model-visual-restored.safetensors",
|
| 1032 |
+
"model.visual.blocks.23.attn.qkv.bias": "model-visual-restored.safetensors"
|
| 1033 |
+
}
|
| 1034 |
+
}
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_convert_rgb": true,
|
| 3 |
+
"do_normalize": true,
|
| 4 |
+
"do_rescale": true,
|
| 5 |
+
"do_resize": true,
|
| 6 |
+
"image_mean": [
|
| 7 |
+
0.5,
|
| 8 |
+
0.5,
|
| 9 |
+
0.5
|
| 10 |
+
],
|
| 11 |
+
"image_processor_type": "Qwen2VLImageProcessor",
|
| 12 |
+
"image_std": [
|
| 13 |
+
0.5,
|
| 14 |
+
0.5,
|
| 15 |
+
0.5
|
| 16 |
+
],
|
| 17 |
+
"merge_size": 2,
|
| 18 |
+
"patch_size": 16,
|
| 19 |
+
"resample": 3,
|
| 20 |
+
"rescale_factor": 0.00392156862745098,
|
| 21 |
+
"size": {
|
| 22 |
+
"longest_edge": 16777216,
|
| 23 |
+
"shortest_edge": 65536
|
| 24 |
+
},
|
| 25 |
+
"temporal_patch_size": 2
|
| 26 |
+
}
|
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 |
+
}
|
tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:00df80e45f6b4537e96d223831a3bc141c6c0c9eef3e2d400f9b612316f46048
|
| 3 |
+
size 19989323
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,37 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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 |
+
"extra_special_tokens": [
|
| 12 |
+
"<think>",
|
| 13 |
+
"</think>"
|
| 14 |
+
],
|
| 15 |
+
"image_token": "<|image_pad|>",
|
| 16 |
+
"is_local": true,
|
| 17 |
+
"local_files_only": true,
|
| 18 |
+
"model_max_length": 262144,
|
| 19 |
+
"model_specific_special_tokens": {
|
| 20 |
+
"audio_bos_token": "<|audio_start|>",
|
| 21 |
+
"audio_eos_token": "<|audio_end|>",
|
| 22 |
+
"audio_token": "<|audio_pad|>",
|
| 23 |
+
"image_token": "<|image_pad|>",
|
| 24 |
+
"video_token": "<|video_pad|>",
|
| 25 |
+
"vision_bos_token": "<|vision_start|>",
|
| 26 |
+
"vision_eos_token": "<|vision_end|>"
|
| 27 |
+
},
|
| 28 |
+
"pad_token": "<|endoftext|>",
|
| 29 |
+
"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+",
|
| 30 |
+
"processor_class": "Qwen3VLProcessor",
|
| 31 |
+
"split_special_tokens": false,
|
| 32 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 33 |
+
"unk_token": null,
|
| 34 |
+
"video_token": "<|video_pad|>",
|
| 35 |
+
"vision_bos_token": "<|vision_start|>",
|
| 36 |
+
"vision_eos_token": "<|vision_end|>"
|
| 37 |
+
}
|
video_preprocessor_config.json
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"size": {
|
| 3 |
+
"longest_edge": 25165824,
|
| 4 |
+
"shortest_edge": 4096
|
| 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 |
+
"video_processor_type": "Qwen3VLVideoProcessor"
|
| 21 |
+
}
|