Shubhra Pandit
commited on
Commit
·
91b1203
1
Parent(s):
2996fc0
Upload model files
Browse files- .gitattributes +2 -0
- chat_template.json +3 -0
- config.json +1267 -0
- generation_config.json +6 -0
- model-00001-of-00014.safetensors +3 -0
- model-00002-of-00014.safetensors +3 -0
- model-00003-of-00014.safetensors +3 -0
- model-00004-of-00014.safetensors +3 -0
- model-00005-of-00014.safetensors +3 -0
- model-00006-of-00014.safetensors +3 -0
- model-00007-of-00014.safetensors +3 -0
- model-00008-of-00014.safetensors +3 -0
- model-00009-of-00014.safetensors +3 -0
- model-00010-of-00014.safetensors +3 -0
- model-00011-of-00014.safetensors +3 -0
- model-00012-of-00014.safetensors +3 -0
- model-00013-of-00014.safetensors +3 -0
- model-00014-of-00014.safetensors +3 -0
- model.safetensors.index.json +0 -0
- preprocessor_config.json +27 -0
- processor_config.json +7 -0
- recipe.yaml +12 -0
- special_tokens_map.json +773 -0
- tokenizer.json +3 -0
- tokenizer.model +3 -0
- tokenizer_config.json +0 -0
.gitattributes
CHANGED
@@ -33,3 +33,5 @@ saved_model/**/* 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
|
|
|
|
|
|
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
|
37 |
+
|
chat_template.json
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"chat_template": "{%- set today = strftime_now(\"%Y-%m-%d\") %}\n{%- set default_system_message = \"You are Pixtral, a Large Language Model (LLM) created by Mistral AI, a French startup headquartered in Paris.\\nYou power an AI assistant called Le Chat.\\nYour knowledge base was last updated on 2023-10-01.\\nThe current date is \" + today + \"\\n\\nWhen you're not sure about some information, you say that you don't have the information and don't make up anything.\\nIf the user's question is not clear, ambiguous, or does not provide enough context for you to accurately answer the question, you do not try to answer it right away and you rather ask the user to clarify their request (e.g. \\\"What are some good restaurants around me?\\\" => \\\"Where are you?\\\" or \\\"When is the next flight to Tokyo\\\" => \\\"Where do you travel from?\\\").\\nYou are always very attentive to dates, in particular you try to resolve dates (e.g. \\\"today\\\" is \" + today + \") and when asked about information at specific dates, you discard information that is at another date.\\nYou follow these instructions in all languages, and always respond to the user in the language they use or request.\\nNext sections describe the capabilities that you have.\\n\\n# WEB BROWSING INSTRUCTIONS\\n\\nYou cannot perform any web search or access internet to open URLs, links etc. If it seems like the user is expecting you to do so, you clarify the situation and ask the user to copy paste the text directly in the chat.\\n\\n# MULTI-MODAL INSTRUCTIONS\\n\\nYou have the ability to read images, but you cannot generate images. You also cannot transcribe audio files or videos.\\nYou cannot read nor transcribe audio files or videos and you cannot read images.\" %}\n\n\n\n{{- bos_token }}\n{%- if messages[0]['role'] == 'system' %}\n {%- set system_message = messages[0]['content'] %}\n {%- set loop_messages = messages[1:] %}\n{%- else %}\n {%- set system_message = default_system_message %}\n {%- set loop_messages = messages %}\n{%- endif %}\n\n{{- '[SYSTEM_PROMPT] ' + system_message + '[/SYSTEM_PROMPT]' }}\n\n{%- for message in loop_messages %}\n {%- if message['role'] == 'user' %}\n {%- if message[\"content\"] is not string %}\n {{- '[INST]' }}\n {%- set ns = namespace(text_count=0, text_index=-1) %}\n {%- for chunk in message[\"content\"] %}\n {%- if chunk[\"type\"] == \"text\" %}\n {%- set ns.text_count = ns.text_count + 1 %}\n {%- set ns.text_index = loop.index0 %}\n {%- endif %}\n {%- endfor %}\n {#- When there is only one text chunk, put it at the end to match the behaviour of mistral-common #}\n {%- if ns.text_count == 1 and message[\"content\"]|length > 1 %}\n {%- set content = message[\"content\"][:ns.text_index] + message[\"content\"][ns.text_index + 1:] + message[\"content\"][ns.text_index: ns.text_index + 1] %}\n {%- else %}\n {%- set content = message[\"content\"] %}\n {%- endif %}\n {%- for chunk in content %}\n {%- if chunk[\"type\"] == \"text\" %}\n {{- \" \" + chunk[\"text\"]|trim }}\n {%- elif chunk[\"type\"] == \"image\" %}\n {{- \"[IMG]\" }}\n {%- else %}\n {{- raise_exception(\"Unrecognized content type!\") }}\n {%- endif %}\n {%- endfor %}\n {%- else %}\n {{- '[INST] ' }}\n {{- message[\"content\"] }}\n {%- endif %}\n {{- '[/INST]' }}\n {%- elif message['role'] == 'assistant' %}\n {{- ' ' + message['content'] + eos_token }}\n {%- else %}\n {{- raise_exception('Only user and assistant roles are supported, plus an optional initial system message!') }}\n {%- endif %}\n{%- endfor %}\n"
|
3 |
+
}
|
config.json
ADDED
@@ -0,0 +1,1267 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "/network/shubhra/Pixtral-Large-Instruct-2411-pr23/",
|
3 |
+
"architectures": [
|
4 |
+
"LlavaForConditionalGeneration"
|
5 |
+
],
|
6 |
+
"ignore_index": -100,
|
7 |
+
"image_seq_length": 1,
|
8 |
+
"image_token_index": 10,
|
9 |
+
"model_type": "llava",
|
10 |
+
"multimodal_projector_bias": false,
|
11 |
+
"projector_hidden_act": "gelu",
|
12 |
+
"quantization_config": {
|
13 |
+
"config_groups": {
|
14 |
+
"group_0": {
|
15 |
+
"input_activations": null,
|
16 |
+
"output_activations": null,
|
17 |
+
"targets": [
|
18 |
+
"Linear"
|
19 |
+
],
|
20 |
+
"weights": {
|
21 |
+
"actorder": "weight",
|
22 |
+
"block_structure": null,
|
23 |
+
"dynamic": false,
|
24 |
+
"group_size": 128,
|
25 |
+
"num_bits": 4,
|
26 |
+
"observer": "minmax",
|
27 |
+
"observer_kwargs": {},
|
28 |
+
"strategy": "group",
|
29 |
+
"symmetric": true,
|
30 |
+
"type": "int"
|
31 |
+
}
|
32 |
+
}
|
33 |
+
},
|
34 |
+
"format": "pack-quantized",
|
35 |
+
"global_compression_ratio": 1.5568737751955632,
|
36 |
+
"ignore": [
|
37 |
+
"vision_tower.transformer.layers.0.feed_forward.gate_proj",
|
38 |
+
"vision_tower.transformer.layers.0.feed_forward.up_proj",
|
39 |
+
"vision_tower.transformer.layers.0.feed_forward.down_proj",
|
40 |
+
"vision_tower.transformer.layers.0.attention.k_proj",
|
41 |
+
"vision_tower.transformer.layers.0.attention.v_proj",
|
42 |
+
"vision_tower.transformer.layers.0.attention.q_proj",
|
43 |
+
"vision_tower.transformer.layers.0.attention.o_proj",
|
44 |
+
"vision_tower.transformer.layers.1.feed_forward.gate_proj",
|
45 |
+
"vision_tower.transformer.layers.1.feed_forward.up_proj",
|
46 |
+
"vision_tower.transformer.layers.1.feed_forward.down_proj",
|
47 |
+
"vision_tower.transformer.layers.1.attention.k_proj",
|
48 |
+
"vision_tower.transformer.layers.1.attention.v_proj",
|
49 |
+
"vision_tower.transformer.layers.1.attention.q_proj",
|
50 |
+
"vision_tower.transformer.layers.1.attention.o_proj",
|
51 |
+
"vision_tower.transformer.layers.2.feed_forward.gate_proj",
|
52 |
+
"vision_tower.transformer.layers.2.feed_forward.up_proj",
|
53 |
+
"vision_tower.transformer.layers.2.feed_forward.down_proj",
|
54 |
+
"vision_tower.transformer.layers.2.attention.k_proj",
|
55 |
+
"vision_tower.transformer.layers.2.attention.v_proj",
|
56 |
+
"vision_tower.transformer.layers.2.attention.q_proj",
|
57 |
+
"vision_tower.transformer.layers.2.attention.o_proj",
|
58 |
+
"vision_tower.transformer.layers.3.feed_forward.gate_proj",
|
59 |
+
"vision_tower.transformer.layers.3.feed_forward.up_proj",
|
60 |
+
"vision_tower.transformer.layers.3.feed_forward.down_proj",
|
61 |
+
"vision_tower.transformer.layers.3.attention.k_proj",
|
62 |
+
"vision_tower.transformer.layers.3.attention.v_proj",
|
63 |
+
"vision_tower.transformer.layers.3.attention.q_proj",
|
64 |
+
"vision_tower.transformer.layers.3.attention.o_proj",
|
65 |
+
"vision_tower.transformer.layers.4.feed_forward.gate_proj",
|
66 |
+
"vision_tower.transformer.layers.4.feed_forward.up_proj",
|
67 |
+
"vision_tower.transformer.layers.4.feed_forward.down_proj",
|
68 |
+
"vision_tower.transformer.layers.4.attention.k_proj",
|
69 |
+
"vision_tower.transformer.layers.4.attention.v_proj",
|
70 |
+
"vision_tower.transformer.layers.4.attention.q_proj",
|
71 |
+
"vision_tower.transformer.layers.4.attention.o_proj",
|
72 |
+
"vision_tower.transformer.layers.5.feed_forward.gate_proj",
|
73 |
+
"vision_tower.transformer.layers.5.feed_forward.up_proj",
|
74 |
+
"vision_tower.transformer.layers.5.feed_forward.down_proj",
|
75 |
+
"vision_tower.transformer.layers.5.attention.k_proj",
|
76 |
+
"vision_tower.transformer.layers.5.attention.v_proj",
|
77 |
+
"vision_tower.transformer.layers.5.attention.q_proj",
|
78 |
+
"vision_tower.transformer.layers.5.attention.o_proj",
|
79 |
+
"vision_tower.transformer.layers.6.feed_forward.gate_proj",
|
80 |
+
"vision_tower.transformer.layers.6.feed_forward.up_proj",
|
81 |
+
"vision_tower.transformer.layers.6.feed_forward.down_proj",
|
82 |
+
"vision_tower.transformer.layers.6.attention.k_proj",
|
83 |
+
"vision_tower.transformer.layers.6.attention.v_proj",
|
84 |
+
"vision_tower.transformer.layers.6.attention.q_proj",
|
85 |
+
"vision_tower.transformer.layers.6.attention.o_proj",
|
86 |
+
"vision_tower.transformer.layers.7.feed_forward.gate_proj",
|
87 |
+
"vision_tower.transformer.layers.7.feed_forward.up_proj",
|
88 |
+
"vision_tower.transformer.layers.7.feed_forward.down_proj",
|
89 |
+
"vision_tower.transformer.layers.7.attention.k_proj",
|
90 |
+
"vision_tower.transformer.layers.7.attention.v_proj",
|
91 |
+
"vision_tower.transformer.layers.7.attention.q_proj",
|
92 |
+
"vision_tower.transformer.layers.7.attention.o_proj",
|
93 |
+
"vision_tower.transformer.layers.8.feed_forward.gate_proj",
|
94 |
+
"vision_tower.transformer.layers.8.feed_forward.up_proj",
|
95 |
+
"vision_tower.transformer.layers.8.feed_forward.down_proj",
|
96 |
+
"vision_tower.transformer.layers.8.attention.k_proj",
|
97 |
+
"vision_tower.transformer.layers.8.attention.v_proj",
|
98 |
+
"vision_tower.transformer.layers.8.attention.q_proj",
|
99 |
+
"vision_tower.transformer.layers.8.attention.o_proj",
|
100 |
+
"vision_tower.transformer.layers.9.feed_forward.gate_proj",
|
101 |
+
"vision_tower.transformer.layers.9.feed_forward.up_proj",
|
102 |
+
"vision_tower.transformer.layers.9.feed_forward.down_proj",
|
103 |
+
"vision_tower.transformer.layers.9.attention.k_proj",
|
104 |
+
"vision_tower.transformer.layers.9.attention.v_proj",
|
105 |
+
"vision_tower.transformer.layers.9.attention.q_proj",
|
106 |
+
"vision_tower.transformer.layers.9.attention.o_proj",
|
107 |
+
"vision_tower.transformer.layers.10.feed_forward.gate_proj",
|
108 |
+
"vision_tower.transformer.layers.10.feed_forward.up_proj",
|
109 |
+
"vision_tower.transformer.layers.10.feed_forward.down_proj",
|
110 |
+
"vision_tower.transformer.layers.10.attention.k_proj",
|
111 |
+
"vision_tower.transformer.layers.10.attention.v_proj",
|
112 |
+
"vision_tower.transformer.layers.10.attention.q_proj",
|
113 |
+
"vision_tower.transformer.layers.10.attention.o_proj",
|
114 |
+
"vision_tower.transformer.layers.11.feed_forward.gate_proj",
|
115 |
+
"vision_tower.transformer.layers.11.feed_forward.up_proj",
|
116 |
+
"vision_tower.transformer.layers.11.feed_forward.down_proj",
|
117 |
+
"vision_tower.transformer.layers.11.attention.k_proj",
|
118 |
+
"vision_tower.transformer.layers.11.attention.v_proj",
|
119 |
+
"vision_tower.transformer.layers.11.attention.q_proj",
|
120 |
+
"vision_tower.transformer.layers.11.attention.o_proj",
|
121 |
+
"vision_tower.transformer.layers.12.feed_forward.gate_proj",
|
122 |
+
"vision_tower.transformer.layers.12.feed_forward.up_proj",
|
123 |
+
"vision_tower.transformer.layers.12.feed_forward.down_proj",
|
124 |
+
"vision_tower.transformer.layers.12.attention.k_proj",
|
125 |
+
"vision_tower.transformer.layers.12.attention.v_proj",
|
126 |
+
"vision_tower.transformer.layers.12.attention.q_proj",
|
127 |
+
"vision_tower.transformer.layers.12.attention.o_proj",
|
128 |
+
"vision_tower.transformer.layers.13.feed_forward.gate_proj",
|
129 |
+
"vision_tower.transformer.layers.13.feed_forward.up_proj",
|
130 |
+
"vision_tower.transformer.layers.13.feed_forward.down_proj",
|
131 |
+
"vision_tower.transformer.layers.13.attention.k_proj",
|
132 |
+
"vision_tower.transformer.layers.13.attention.v_proj",
|
133 |
+
"vision_tower.transformer.layers.13.attention.q_proj",
|
134 |
+
"vision_tower.transformer.layers.13.attention.o_proj",
|
135 |
+
"vision_tower.transformer.layers.14.feed_forward.gate_proj",
|
136 |
+
"vision_tower.transformer.layers.14.feed_forward.up_proj",
|
137 |
+
"vision_tower.transformer.layers.14.feed_forward.down_proj",
|
138 |
+
"vision_tower.transformer.layers.14.attention.k_proj",
|
139 |
+
"vision_tower.transformer.layers.14.attention.v_proj",
|
140 |
+
"vision_tower.transformer.layers.14.attention.q_proj",
|
141 |
+
"vision_tower.transformer.layers.14.attention.o_proj",
|
142 |
+
"vision_tower.transformer.layers.15.feed_forward.gate_proj",
|
143 |
+
"vision_tower.transformer.layers.15.feed_forward.up_proj",
|
144 |
+
"vision_tower.transformer.layers.15.feed_forward.down_proj",
|
145 |
+
"vision_tower.transformer.layers.15.attention.k_proj",
|
146 |
+
"vision_tower.transformer.layers.15.attention.v_proj",
|
147 |
+
"vision_tower.transformer.layers.15.attention.q_proj",
|
148 |
+
"vision_tower.transformer.layers.15.attention.o_proj",
|
149 |
+
"vision_tower.transformer.layers.16.feed_forward.gate_proj",
|
150 |
+
"vision_tower.transformer.layers.16.feed_forward.up_proj",
|
151 |
+
"vision_tower.transformer.layers.16.feed_forward.down_proj",
|
152 |
+
"vision_tower.transformer.layers.16.attention.k_proj",
|
153 |
+
"vision_tower.transformer.layers.16.attention.v_proj",
|
154 |
+
"vision_tower.transformer.layers.16.attention.q_proj",
|
155 |
+
"vision_tower.transformer.layers.16.attention.o_proj",
|
156 |
+
"vision_tower.transformer.layers.17.feed_forward.gate_proj",
|
157 |
+
"vision_tower.transformer.layers.17.feed_forward.up_proj",
|
158 |
+
"vision_tower.transformer.layers.17.feed_forward.down_proj",
|
159 |
+
"vision_tower.transformer.layers.17.attention.k_proj",
|
160 |
+
"vision_tower.transformer.layers.17.attention.v_proj",
|
161 |
+
"vision_tower.transformer.layers.17.attention.q_proj",
|
162 |
+
"vision_tower.transformer.layers.17.attention.o_proj",
|
163 |
+
"vision_tower.transformer.layers.18.feed_forward.gate_proj",
|
164 |
+
"vision_tower.transformer.layers.18.feed_forward.up_proj",
|
165 |
+
"vision_tower.transformer.layers.18.feed_forward.down_proj",
|
166 |
+
"vision_tower.transformer.layers.18.attention.k_proj",
|
167 |
+
"vision_tower.transformer.layers.18.attention.v_proj",
|
168 |
+
"vision_tower.transformer.layers.18.attention.q_proj",
|
169 |
+
"vision_tower.transformer.layers.18.attention.o_proj",
|
170 |
+
"vision_tower.transformer.layers.19.feed_forward.gate_proj",
|
171 |
+
"vision_tower.transformer.layers.19.feed_forward.up_proj",
|
172 |
+
"vision_tower.transformer.layers.19.feed_forward.down_proj",
|
173 |
+
"vision_tower.transformer.layers.19.attention.k_proj",
|
174 |
+
"vision_tower.transformer.layers.19.attention.v_proj",
|
175 |
+
"vision_tower.transformer.layers.19.attention.q_proj",
|
176 |
+
"vision_tower.transformer.layers.19.attention.o_proj",
|
177 |
+
"vision_tower.transformer.layers.20.feed_forward.gate_proj",
|
178 |
+
"vision_tower.transformer.layers.20.feed_forward.up_proj",
|
179 |
+
"vision_tower.transformer.layers.20.feed_forward.down_proj",
|
180 |
+
"vision_tower.transformer.layers.20.attention.k_proj",
|
181 |
+
"vision_tower.transformer.layers.20.attention.v_proj",
|
182 |
+
"vision_tower.transformer.layers.20.attention.q_proj",
|
183 |
+
"vision_tower.transformer.layers.20.attention.o_proj",
|
184 |
+
"vision_tower.transformer.layers.21.feed_forward.gate_proj",
|
185 |
+
"vision_tower.transformer.layers.21.feed_forward.up_proj",
|
186 |
+
"vision_tower.transformer.layers.21.feed_forward.down_proj",
|
187 |
+
"vision_tower.transformer.layers.21.attention.k_proj",
|
188 |
+
"vision_tower.transformer.layers.21.attention.v_proj",
|
189 |
+
"vision_tower.transformer.layers.21.attention.q_proj",
|
190 |
+
"vision_tower.transformer.layers.21.attention.o_proj",
|
191 |
+
"vision_tower.transformer.layers.22.feed_forward.gate_proj",
|
192 |
+
"vision_tower.transformer.layers.22.feed_forward.up_proj",
|
193 |
+
"vision_tower.transformer.layers.22.feed_forward.down_proj",
|
194 |
+
"vision_tower.transformer.layers.22.attention.k_proj",
|
195 |
+
"vision_tower.transformer.layers.22.attention.v_proj",
|
196 |
+
"vision_tower.transformer.layers.22.attention.q_proj",
|
197 |
+
"vision_tower.transformer.layers.22.attention.o_proj",
|
198 |
+
"vision_tower.transformer.layers.23.feed_forward.gate_proj",
|
199 |
+
"vision_tower.transformer.layers.23.feed_forward.up_proj",
|
200 |
+
"vision_tower.transformer.layers.23.feed_forward.down_proj",
|
201 |
+
"vision_tower.transformer.layers.23.attention.k_proj",
|
202 |
+
"vision_tower.transformer.layers.23.attention.v_proj",
|
203 |
+
"vision_tower.transformer.layers.23.attention.q_proj",
|
204 |
+
"vision_tower.transformer.layers.23.attention.o_proj",
|
205 |
+
"vision_tower.transformer.layers.24.feed_forward.gate_proj",
|
206 |
+
"vision_tower.transformer.layers.24.feed_forward.up_proj",
|
207 |
+
"vision_tower.transformer.layers.24.feed_forward.down_proj",
|
208 |
+
"vision_tower.transformer.layers.24.attention.k_proj",
|
209 |
+
"vision_tower.transformer.layers.24.attention.v_proj",
|
210 |
+
"vision_tower.transformer.layers.24.attention.q_proj",
|
211 |
+
"vision_tower.transformer.layers.24.attention.o_proj",
|
212 |
+
"vision_tower.transformer.layers.25.feed_forward.gate_proj",
|
213 |
+
"vision_tower.transformer.layers.25.feed_forward.up_proj",
|
214 |
+
"vision_tower.transformer.layers.25.feed_forward.down_proj",
|
215 |
+
"vision_tower.transformer.layers.25.attention.k_proj",
|
216 |
+
"vision_tower.transformer.layers.25.attention.v_proj",
|
217 |
+
"vision_tower.transformer.layers.25.attention.q_proj",
|
218 |
+
"vision_tower.transformer.layers.25.attention.o_proj",
|
219 |
+
"vision_tower.transformer.layers.26.feed_forward.gate_proj",
|
220 |
+
"vision_tower.transformer.layers.26.feed_forward.up_proj",
|
221 |
+
"vision_tower.transformer.layers.26.feed_forward.down_proj",
|
222 |
+
"vision_tower.transformer.layers.26.attention.k_proj",
|
223 |
+
"vision_tower.transformer.layers.26.attention.v_proj",
|
224 |
+
"vision_tower.transformer.layers.26.attention.q_proj",
|
225 |
+
"vision_tower.transformer.layers.26.attention.o_proj",
|
226 |
+
"vision_tower.transformer.layers.27.feed_forward.gate_proj",
|
227 |
+
"vision_tower.transformer.layers.27.feed_forward.up_proj",
|
228 |
+
"vision_tower.transformer.layers.27.feed_forward.down_proj",
|
229 |
+
"vision_tower.transformer.layers.27.attention.k_proj",
|
230 |
+
"vision_tower.transformer.layers.27.attention.v_proj",
|
231 |
+
"vision_tower.transformer.layers.27.attention.q_proj",
|
232 |
+
"vision_tower.transformer.layers.27.attention.o_proj",
|
233 |
+
"vision_tower.transformer.layers.28.feed_forward.gate_proj",
|
234 |
+
"vision_tower.transformer.layers.28.feed_forward.up_proj",
|
235 |
+
"vision_tower.transformer.layers.28.feed_forward.down_proj",
|
236 |
+
"vision_tower.transformer.layers.28.attention.k_proj",
|
237 |
+
"vision_tower.transformer.layers.28.attention.v_proj",
|
238 |
+
"vision_tower.transformer.layers.28.attention.q_proj",
|
239 |
+
"vision_tower.transformer.layers.28.attention.o_proj",
|
240 |
+
"vision_tower.transformer.layers.29.feed_forward.gate_proj",
|
241 |
+
"vision_tower.transformer.layers.29.feed_forward.up_proj",
|
242 |
+
"vision_tower.transformer.layers.29.feed_forward.down_proj",
|
243 |
+
"vision_tower.transformer.layers.29.attention.k_proj",
|
244 |
+
"vision_tower.transformer.layers.29.attention.v_proj",
|
245 |
+
"vision_tower.transformer.layers.29.attention.q_proj",
|
246 |
+
"vision_tower.transformer.layers.29.attention.o_proj",
|
247 |
+
"vision_tower.transformer.layers.30.feed_forward.gate_proj",
|
248 |
+
"vision_tower.transformer.layers.30.feed_forward.up_proj",
|
249 |
+
"vision_tower.transformer.layers.30.feed_forward.down_proj",
|
250 |
+
"vision_tower.transformer.layers.30.attention.k_proj",
|
251 |
+
"vision_tower.transformer.layers.30.attention.v_proj",
|
252 |
+
"vision_tower.transformer.layers.30.attention.q_proj",
|
253 |
+
"vision_tower.transformer.layers.30.attention.o_proj",
|
254 |
+
"vision_tower.transformer.layers.31.feed_forward.gate_proj",
|
255 |
+
"vision_tower.transformer.layers.31.feed_forward.up_proj",
|
256 |
+
"vision_tower.transformer.layers.31.feed_forward.down_proj",
|
257 |
+
"vision_tower.transformer.layers.31.attention.k_proj",
|
258 |
+
"vision_tower.transformer.layers.31.attention.v_proj",
|
259 |
+
"vision_tower.transformer.layers.31.attention.q_proj",
|
260 |
+
"vision_tower.transformer.layers.31.attention.o_proj",
|
261 |
+
"vision_tower.transformer.layers.32.feed_forward.gate_proj",
|
262 |
+
"vision_tower.transformer.layers.32.feed_forward.up_proj",
|
263 |
+
"vision_tower.transformer.layers.32.feed_forward.down_proj",
|
264 |
+
"vision_tower.transformer.layers.32.attention.k_proj",
|
265 |
+
"vision_tower.transformer.layers.32.attention.v_proj",
|
266 |
+
"vision_tower.transformer.layers.32.attention.q_proj",
|
267 |
+
"vision_tower.transformer.layers.32.attention.o_proj",
|
268 |
+
"vision_tower.transformer.layers.33.feed_forward.gate_proj",
|
269 |
+
"vision_tower.transformer.layers.33.feed_forward.up_proj",
|
270 |
+
"vision_tower.transformer.layers.33.feed_forward.down_proj",
|
271 |
+
"vision_tower.transformer.layers.33.attention.k_proj",
|
272 |
+
"vision_tower.transformer.layers.33.attention.v_proj",
|
273 |
+
"vision_tower.transformer.layers.33.attention.q_proj",
|
274 |
+
"vision_tower.transformer.layers.33.attention.o_proj",
|
275 |
+
"vision_tower.transformer.layers.34.feed_forward.gate_proj",
|
276 |
+
"vision_tower.transformer.layers.34.feed_forward.up_proj",
|
277 |
+
"vision_tower.transformer.layers.34.feed_forward.down_proj",
|
278 |
+
"vision_tower.transformer.layers.34.attention.k_proj",
|
279 |
+
"vision_tower.transformer.layers.34.attention.v_proj",
|
280 |
+
"vision_tower.transformer.layers.34.attention.q_proj",
|
281 |
+
"vision_tower.transformer.layers.34.attention.o_proj",
|
282 |
+
"vision_tower.transformer.layers.35.feed_forward.gate_proj",
|
283 |
+
"vision_tower.transformer.layers.35.feed_forward.up_proj",
|
284 |
+
"vision_tower.transformer.layers.35.feed_forward.down_proj",
|
285 |
+
"vision_tower.transformer.layers.35.attention.k_proj",
|
286 |
+
"vision_tower.transformer.layers.35.attention.v_proj",
|
287 |
+
"vision_tower.transformer.layers.35.attention.q_proj",
|
288 |
+
"vision_tower.transformer.layers.35.attention.o_proj",
|
289 |
+
"vision_tower.transformer.layers.36.feed_forward.gate_proj",
|
290 |
+
"vision_tower.transformer.layers.36.feed_forward.up_proj",
|
291 |
+
"vision_tower.transformer.layers.36.feed_forward.down_proj",
|
292 |
+
"vision_tower.transformer.layers.36.attention.k_proj",
|
293 |
+
"vision_tower.transformer.layers.36.attention.v_proj",
|
294 |
+
"vision_tower.transformer.layers.36.attention.q_proj",
|
295 |
+
"vision_tower.transformer.layers.36.attention.o_proj",
|
296 |
+
"vision_tower.transformer.layers.37.feed_forward.gate_proj",
|
297 |
+
"vision_tower.transformer.layers.37.feed_forward.up_proj",
|
298 |
+
"vision_tower.transformer.layers.37.feed_forward.down_proj",
|
299 |
+
"vision_tower.transformer.layers.37.attention.k_proj",
|
300 |
+
"vision_tower.transformer.layers.37.attention.v_proj",
|
301 |
+
"vision_tower.transformer.layers.37.attention.q_proj",
|
302 |
+
"vision_tower.transformer.layers.37.attention.o_proj",
|
303 |
+
"vision_tower.transformer.layers.38.feed_forward.gate_proj",
|
304 |
+
"vision_tower.transformer.layers.38.feed_forward.up_proj",
|
305 |
+
"vision_tower.transformer.layers.38.feed_forward.down_proj",
|
306 |
+
"vision_tower.transformer.layers.38.attention.k_proj",
|
307 |
+
"vision_tower.transformer.layers.38.attention.v_proj",
|
308 |
+
"vision_tower.transformer.layers.38.attention.q_proj",
|
309 |
+
"vision_tower.transformer.layers.38.attention.o_proj",
|
310 |
+
"vision_tower.transformer.layers.39.feed_forward.gate_proj",
|
311 |
+
"vision_tower.transformer.layers.39.feed_forward.up_proj",
|
312 |
+
"vision_tower.transformer.layers.39.feed_forward.down_proj",
|
313 |
+
"vision_tower.transformer.layers.39.attention.k_proj",
|
314 |
+
"vision_tower.transformer.layers.39.attention.v_proj",
|
315 |
+
"vision_tower.transformer.layers.39.attention.q_proj",
|
316 |
+
"vision_tower.transformer.layers.39.attention.o_proj",
|
317 |
+
"multi_modal_projector.linear_1",
|
318 |
+
"multi_modal_projector.linear_2",
|
319 |
+
"language_model.lm_head"
|
320 |
+
],
|
321 |
+
"kv_cache_scheme": null,
|
322 |
+
"quant_method": "compressed-tensors",
|
323 |
+
"quantization_status": "compressed",
|
324 |
+
"sparsity_config": {
|
325 |
+
"format": "dense",
|
326 |
+
"global_sparsity": 0.15425385709456318,
|
327 |
+
"ignore": [
|
328 |
+
"vision_tower.transformer.layers.0.feed_forward.gate_proj",
|
329 |
+
"vision_tower.transformer.layers.0.feed_forward.up_proj",
|
330 |
+
"vision_tower.transformer.layers.0.feed_forward.down_proj",
|
331 |
+
"vision_tower.transformer.layers.0.attention.k_proj",
|
332 |
+
"vision_tower.transformer.layers.0.attention.v_proj",
|
333 |
+
"vision_tower.transformer.layers.0.attention.q_proj",
|
334 |
+
"vision_tower.transformer.layers.0.attention.o_proj",
|
335 |
+
"vision_tower.transformer.layers.1.feed_forward.gate_proj",
|
336 |
+
"vision_tower.transformer.layers.1.feed_forward.up_proj",
|
337 |
+
"vision_tower.transformer.layers.1.feed_forward.down_proj",
|
338 |
+
"vision_tower.transformer.layers.1.attention.k_proj",
|
339 |
+
"vision_tower.transformer.layers.1.attention.v_proj",
|
340 |
+
"vision_tower.transformer.layers.1.attention.q_proj",
|
341 |
+
"vision_tower.transformer.layers.1.attention.o_proj",
|
342 |
+
"vision_tower.transformer.layers.2.feed_forward.gate_proj",
|
343 |
+
"vision_tower.transformer.layers.2.feed_forward.up_proj",
|
344 |
+
"vision_tower.transformer.layers.2.feed_forward.down_proj",
|
345 |
+
"vision_tower.transformer.layers.2.attention.k_proj",
|
346 |
+
"vision_tower.transformer.layers.2.attention.v_proj",
|
347 |
+
"vision_tower.transformer.layers.2.attention.q_proj",
|
348 |
+
"vision_tower.transformer.layers.2.attention.o_proj",
|
349 |
+
"vision_tower.transformer.layers.3.feed_forward.gate_proj",
|
350 |
+
"vision_tower.transformer.layers.3.feed_forward.up_proj",
|
351 |
+
"vision_tower.transformer.layers.3.feed_forward.down_proj",
|
352 |
+
"vision_tower.transformer.layers.3.attention.k_proj",
|
353 |
+
"vision_tower.transformer.layers.3.attention.v_proj",
|
354 |
+
"vision_tower.transformer.layers.3.attention.q_proj",
|
355 |
+
"vision_tower.transformer.layers.3.attention.o_proj",
|
356 |
+
"vision_tower.transformer.layers.4.feed_forward.gate_proj",
|
357 |
+
"vision_tower.transformer.layers.4.feed_forward.up_proj",
|
358 |
+
"vision_tower.transformer.layers.4.feed_forward.down_proj",
|
359 |
+
"vision_tower.transformer.layers.4.attention.k_proj",
|
360 |
+
"vision_tower.transformer.layers.4.attention.v_proj",
|
361 |
+
"vision_tower.transformer.layers.4.attention.q_proj",
|
362 |
+
"vision_tower.transformer.layers.4.attention.o_proj",
|
363 |
+
"vision_tower.transformer.layers.5.feed_forward.gate_proj",
|
364 |
+
"vision_tower.transformer.layers.5.feed_forward.up_proj",
|
365 |
+
"vision_tower.transformer.layers.5.feed_forward.down_proj",
|
366 |
+
"vision_tower.transformer.layers.5.attention.k_proj",
|
367 |
+
"vision_tower.transformer.layers.5.attention.v_proj",
|
368 |
+
"vision_tower.transformer.layers.5.attention.q_proj",
|
369 |
+
"vision_tower.transformer.layers.5.attention.o_proj",
|
370 |
+
"vision_tower.transformer.layers.6.feed_forward.gate_proj",
|
371 |
+
"vision_tower.transformer.layers.6.feed_forward.up_proj",
|
372 |
+
"vision_tower.transformer.layers.6.feed_forward.down_proj",
|
373 |
+
"vision_tower.transformer.layers.6.attention.k_proj",
|
374 |
+
"vision_tower.transformer.layers.6.attention.v_proj",
|
375 |
+
"vision_tower.transformer.layers.6.attention.q_proj",
|
376 |
+
"vision_tower.transformer.layers.6.attention.o_proj",
|
377 |
+
"vision_tower.transformer.layers.7.feed_forward.gate_proj",
|
378 |
+
"vision_tower.transformer.layers.7.feed_forward.up_proj",
|
379 |
+
"vision_tower.transformer.layers.7.feed_forward.down_proj",
|
380 |
+
"vision_tower.transformer.layers.7.attention.k_proj",
|
381 |
+
"vision_tower.transformer.layers.7.attention.v_proj",
|
382 |
+
"vision_tower.transformer.layers.7.attention.q_proj",
|
383 |
+
"vision_tower.transformer.layers.7.attention.o_proj",
|
384 |
+
"vision_tower.transformer.layers.8.feed_forward.gate_proj",
|
385 |
+
"vision_tower.transformer.layers.8.feed_forward.up_proj",
|
386 |
+
"vision_tower.transformer.layers.8.feed_forward.down_proj",
|
387 |
+
"vision_tower.transformer.layers.8.attention.k_proj",
|
388 |
+
"vision_tower.transformer.layers.8.attention.v_proj",
|
389 |
+
"vision_tower.transformer.layers.8.attention.q_proj",
|
390 |
+
"vision_tower.transformer.layers.8.attention.o_proj",
|
391 |
+
"vision_tower.transformer.layers.9.feed_forward.gate_proj",
|
392 |
+
"vision_tower.transformer.layers.9.feed_forward.up_proj",
|
393 |
+
"vision_tower.transformer.layers.9.feed_forward.down_proj",
|
394 |
+
"vision_tower.transformer.layers.9.attention.k_proj",
|
395 |
+
"vision_tower.transformer.layers.9.attention.v_proj",
|
396 |
+
"vision_tower.transformer.layers.9.attention.q_proj",
|
397 |
+
"vision_tower.transformer.layers.9.attention.o_proj",
|
398 |
+
"vision_tower.transformer.layers.10.feed_forward.gate_proj",
|
399 |
+
"vision_tower.transformer.layers.10.feed_forward.up_proj",
|
400 |
+
"vision_tower.transformer.layers.10.feed_forward.down_proj",
|
401 |
+
"vision_tower.transformer.layers.10.attention.k_proj",
|
402 |
+
"vision_tower.transformer.layers.10.attention.v_proj",
|
403 |
+
"vision_tower.transformer.layers.10.attention.q_proj",
|
404 |
+
"vision_tower.transformer.layers.10.attention.o_proj",
|
405 |
+
"vision_tower.transformer.layers.11.feed_forward.gate_proj",
|
406 |
+
"vision_tower.transformer.layers.11.feed_forward.up_proj",
|
407 |
+
"vision_tower.transformer.layers.11.feed_forward.down_proj",
|
408 |
+
"vision_tower.transformer.layers.11.attention.k_proj",
|
409 |
+
"vision_tower.transformer.layers.11.attention.v_proj",
|
410 |
+
"vision_tower.transformer.layers.11.attention.q_proj",
|
411 |
+
"vision_tower.transformer.layers.11.attention.o_proj",
|
412 |
+
"vision_tower.transformer.layers.12.feed_forward.gate_proj",
|
413 |
+
"vision_tower.transformer.layers.12.feed_forward.up_proj",
|
414 |
+
"vision_tower.transformer.layers.12.feed_forward.down_proj",
|
415 |
+
"vision_tower.transformer.layers.12.attention.k_proj",
|
416 |
+
"vision_tower.transformer.layers.12.attention.v_proj",
|
417 |
+
"vision_tower.transformer.layers.12.attention.q_proj",
|
418 |
+
"vision_tower.transformer.layers.12.attention.o_proj",
|
419 |
+
"vision_tower.transformer.layers.13.feed_forward.gate_proj",
|
420 |
+
"vision_tower.transformer.layers.13.feed_forward.up_proj",
|
421 |
+
"vision_tower.transformer.layers.13.feed_forward.down_proj",
|
422 |
+
"vision_tower.transformer.layers.13.attention.k_proj",
|
423 |
+
"vision_tower.transformer.layers.13.attention.v_proj",
|
424 |
+
"vision_tower.transformer.layers.13.attention.q_proj",
|
425 |
+
"vision_tower.transformer.layers.13.attention.o_proj",
|
426 |
+
"vision_tower.transformer.layers.14.feed_forward.gate_proj",
|
427 |
+
"vision_tower.transformer.layers.14.feed_forward.up_proj",
|
428 |
+
"vision_tower.transformer.layers.14.feed_forward.down_proj",
|
429 |
+
"vision_tower.transformer.layers.14.attention.k_proj",
|
430 |
+
"vision_tower.transformer.layers.14.attention.v_proj",
|
431 |
+
"vision_tower.transformer.layers.14.attention.q_proj",
|
432 |
+
"vision_tower.transformer.layers.14.attention.o_proj",
|
433 |
+
"vision_tower.transformer.layers.15.feed_forward.gate_proj",
|
434 |
+
"vision_tower.transformer.layers.15.feed_forward.up_proj",
|
435 |
+
"vision_tower.transformer.layers.15.feed_forward.down_proj",
|
436 |
+
"vision_tower.transformer.layers.15.attention.k_proj",
|
437 |
+
"vision_tower.transformer.layers.15.attention.v_proj",
|
438 |
+
"vision_tower.transformer.layers.15.attention.q_proj",
|
439 |
+
"vision_tower.transformer.layers.15.attention.o_proj",
|
440 |
+
"vision_tower.transformer.layers.16.feed_forward.gate_proj",
|
441 |
+
"vision_tower.transformer.layers.16.feed_forward.up_proj",
|
442 |
+
"vision_tower.transformer.layers.16.feed_forward.down_proj",
|
443 |
+
"vision_tower.transformer.layers.16.attention.k_proj",
|
444 |
+
"vision_tower.transformer.layers.16.attention.v_proj",
|
445 |
+
"vision_tower.transformer.layers.16.attention.q_proj",
|
446 |
+
"vision_tower.transformer.layers.16.attention.o_proj",
|
447 |
+
"vision_tower.transformer.layers.17.feed_forward.gate_proj",
|
448 |
+
"vision_tower.transformer.layers.17.feed_forward.up_proj",
|
449 |
+
"vision_tower.transformer.layers.17.feed_forward.down_proj",
|
450 |
+
"vision_tower.transformer.layers.17.attention.k_proj",
|
451 |
+
"vision_tower.transformer.layers.17.attention.v_proj",
|
452 |
+
"vision_tower.transformer.layers.17.attention.q_proj",
|
453 |
+
"vision_tower.transformer.layers.17.attention.o_proj",
|
454 |
+
"vision_tower.transformer.layers.18.feed_forward.gate_proj",
|
455 |
+
"vision_tower.transformer.layers.18.feed_forward.up_proj",
|
456 |
+
"vision_tower.transformer.layers.18.feed_forward.down_proj",
|
457 |
+
"vision_tower.transformer.layers.18.attention.k_proj",
|
458 |
+
"vision_tower.transformer.layers.18.attention.v_proj",
|
459 |
+
"vision_tower.transformer.layers.18.attention.q_proj",
|
460 |
+
"vision_tower.transformer.layers.18.attention.o_proj",
|
461 |
+
"vision_tower.transformer.layers.19.feed_forward.gate_proj",
|
462 |
+
"vision_tower.transformer.layers.19.feed_forward.up_proj",
|
463 |
+
"vision_tower.transformer.layers.19.feed_forward.down_proj",
|
464 |
+
"vision_tower.transformer.layers.19.attention.k_proj",
|
465 |
+
"vision_tower.transformer.layers.19.attention.v_proj",
|
466 |
+
"vision_tower.transformer.layers.19.attention.q_proj",
|
467 |
+
"vision_tower.transformer.layers.19.attention.o_proj",
|
468 |
+
"vision_tower.transformer.layers.20.feed_forward.gate_proj",
|
469 |
+
"vision_tower.transformer.layers.20.feed_forward.up_proj",
|
470 |
+
"vision_tower.transformer.layers.20.feed_forward.down_proj",
|
471 |
+
"vision_tower.transformer.layers.20.attention.k_proj",
|
472 |
+
"vision_tower.transformer.layers.20.attention.v_proj",
|
473 |
+
"vision_tower.transformer.layers.20.attention.q_proj",
|
474 |
+
"vision_tower.transformer.layers.20.attention.o_proj",
|
475 |
+
"vision_tower.transformer.layers.21.feed_forward.gate_proj",
|
476 |
+
"vision_tower.transformer.layers.21.feed_forward.up_proj",
|
477 |
+
"vision_tower.transformer.layers.21.feed_forward.down_proj",
|
478 |
+
"vision_tower.transformer.layers.21.attention.k_proj",
|
479 |
+
"vision_tower.transformer.layers.21.attention.v_proj",
|
480 |
+
"vision_tower.transformer.layers.21.attention.q_proj",
|
481 |
+
"vision_tower.transformer.layers.21.attention.o_proj",
|
482 |
+
"vision_tower.transformer.layers.22.feed_forward.gate_proj",
|
483 |
+
"vision_tower.transformer.layers.22.feed_forward.up_proj",
|
484 |
+
"vision_tower.transformer.layers.22.feed_forward.down_proj",
|
485 |
+
"vision_tower.transformer.layers.22.attention.k_proj",
|
486 |
+
"vision_tower.transformer.layers.22.attention.v_proj",
|
487 |
+
"vision_tower.transformer.layers.22.attention.q_proj",
|
488 |
+
"vision_tower.transformer.layers.22.attention.o_proj",
|
489 |
+
"vision_tower.transformer.layers.23.feed_forward.gate_proj",
|
490 |
+
"vision_tower.transformer.layers.23.feed_forward.up_proj",
|
491 |
+
"vision_tower.transformer.layers.23.feed_forward.down_proj",
|
492 |
+
"vision_tower.transformer.layers.23.attention.k_proj",
|
493 |
+
"vision_tower.transformer.layers.23.attention.v_proj",
|
494 |
+
"vision_tower.transformer.layers.23.attention.q_proj",
|
495 |
+
"vision_tower.transformer.layers.23.attention.o_proj",
|
496 |
+
"vision_tower.transformer.layers.24.feed_forward.gate_proj",
|
497 |
+
"vision_tower.transformer.layers.24.feed_forward.up_proj",
|
498 |
+
"vision_tower.transformer.layers.24.feed_forward.down_proj",
|
499 |
+
"vision_tower.transformer.layers.24.attention.k_proj",
|
500 |
+
"vision_tower.transformer.layers.24.attention.v_proj",
|
501 |
+
"vision_tower.transformer.layers.24.attention.q_proj",
|
502 |
+
"vision_tower.transformer.layers.24.attention.o_proj",
|
503 |
+
"vision_tower.transformer.layers.25.feed_forward.gate_proj",
|
504 |
+
"vision_tower.transformer.layers.25.feed_forward.up_proj",
|
505 |
+
"vision_tower.transformer.layers.25.feed_forward.down_proj",
|
506 |
+
"vision_tower.transformer.layers.25.attention.k_proj",
|
507 |
+
"vision_tower.transformer.layers.25.attention.v_proj",
|
508 |
+
"vision_tower.transformer.layers.25.attention.q_proj",
|
509 |
+
"vision_tower.transformer.layers.25.attention.o_proj",
|
510 |
+
"vision_tower.transformer.layers.26.feed_forward.gate_proj",
|
511 |
+
"vision_tower.transformer.layers.26.feed_forward.up_proj",
|
512 |
+
"vision_tower.transformer.layers.26.feed_forward.down_proj",
|
513 |
+
"vision_tower.transformer.layers.26.attention.k_proj",
|
514 |
+
"vision_tower.transformer.layers.26.attention.v_proj",
|
515 |
+
"vision_tower.transformer.layers.26.attention.q_proj",
|
516 |
+
"vision_tower.transformer.layers.26.attention.o_proj",
|
517 |
+
"vision_tower.transformer.layers.27.feed_forward.gate_proj",
|
518 |
+
"vision_tower.transformer.layers.27.feed_forward.up_proj",
|
519 |
+
"vision_tower.transformer.layers.27.feed_forward.down_proj",
|
520 |
+
"vision_tower.transformer.layers.27.attention.k_proj",
|
521 |
+
"vision_tower.transformer.layers.27.attention.v_proj",
|
522 |
+
"vision_tower.transformer.layers.27.attention.q_proj",
|
523 |
+
"vision_tower.transformer.layers.27.attention.o_proj",
|
524 |
+
"vision_tower.transformer.layers.28.feed_forward.gate_proj",
|
525 |
+
"vision_tower.transformer.layers.28.feed_forward.up_proj",
|
526 |
+
"vision_tower.transformer.layers.28.feed_forward.down_proj",
|
527 |
+
"vision_tower.transformer.layers.28.attention.k_proj",
|
528 |
+
"vision_tower.transformer.layers.28.attention.v_proj",
|
529 |
+
"vision_tower.transformer.layers.28.attention.q_proj",
|
530 |
+
"vision_tower.transformer.layers.28.attention.o_proj",
|
531 |
+
"vision_tower.transformer.layers.29.feed_forward.gate_proj",
|
532 |
+
"vision_tower.transformer.layers.29.feed_forward.up_proj",
|
533 |
+
"vision_tower.transformer.layers.29.feed_forward.down_proj",
|
534 |
+
"vision_tower.transformer.layers.29.attention.k_proj",
|
535 |
+
"vision_tower.transformer.layers.29.attention.v_proj",
|
536 |
+
"vision_tower.transformer.layers.29.attention.q_proj",
|
537 |
+
"vision_tower.transformer.layers.29.attention.o_proj",
|
538 |
+
"vision_tower.transformer.layers.30.feed_forward.gate_proj",
|
539 |
+
"vision_tower.transformer.layers.30.feed_forward.up_proj",
|
540 |
+
"vision_tower.transformer.layers.30.feed_forward.down_proj",
|
541 |
+
"vision_tower.transformer.layers.30.attention.k_proj",
|
542 |
+
"vision_tower.transformer.layers.30.attention.v_proj",
|
543 |
+
"vision_tower.transformer.layers.30.attention.q_proj",
|
544 |
+
"vision_tower.transformer.layers.30.attention.o_proj",
|
545 |
+
"vision_tower.transformer.layers.31.feed_forward.gate_proj",
|
546 |
+
"vision_tower.transformer.layers.31.feed_forward.up_proj",
|
547 |
+
"vision_tower.transformer.layers.31.feed_forward.down_proj",
|
548 |
+
"vision_tower.transformer.layers.31.attention.k_proj",
|
549 |
+
"vision_tower.transformer.layers.31.attention.v_proj",
|
550 |
+
"vision_tower.transformer.layers.31.attention.q_proj",
|
551 |
+
"vision_tower.transformer.layers.31.attention.o_proj",
|
552 |
+
"vision_tower.transformer.layers.32.feed_forward.gate_proj",
|
553 |
+
"vision_tower.transformer.layers.32.feed_forward.up_proj",
|
554 |
+
"vision_tower.transformer.layers.32.feed_forward.down_proj",
|
555 |
+
"vision_tower.transformer.layers.32.attention.k_proj",
|
556 |
+
"vision_tower.transformer.layers.32.attention.v_proj",
|
557 |
+
"vision_tower.transformer.layers.32.attention.q_proj",
|
558 |
+
"vision_tower.transformer.layers.32.attention.o_proj",
|
559 |
+
"vision_tower.transformer.layers.33.feed_forward.gate_proj",
|
560 |
+
"vision_tower.transformer.layers.33.feed_forward.up_proj",
|
561 |
+
"vision_tower.transformer.layers.33.feed_forward.down_proj",
|
562 |
+
"vision_tower.transformer.layers.33.attention.k_proj",
|
563 |
+
"vision_tower.transformer.layers.33.attention.v_proj",
|
564 |
+
"vision_tower.transformer.layers.33.attention.q_proj",
|
565 |
+
"vision_tower.transformer.layers.33.attention.o_proj",
|
566 |
+
"vision_tower.transformer.layers.34.feed_forward.gate_proj",
|
567 |
+
"vision_tower.transformer.layers.34.feed_forward.up_proj",
|
568 |
+
"vision_tower.transformer.layers.34.feed_forward.down_proj",
|
569 |
+
"vision_tower.transformer.layers.34.attention.k_proj",
|
570 |
+
"vision_tower.transformer.layers.34.attention.v_proj",
|
571 |
+
"vision_tower.transformer.layers.34.attention.q_proj",
|
572 |
+
"vision_tower.transformer.layers.34.attention.o_proj",
|
573 |
+
"vision_tower.transformer.layers.35.feed_forward.gate_proj",
|
574 |
+
"vision_tower.transformer.layers.35.feed_forward.up_proj",
|
575 |
+
"vision_tower.transformer.layers.35.feed_forward.down_proj",
|
576 |
+
"vision_tower.transformer.layers.35.attention.k_proj",
|
577 |
+
"vision_tower.transformer.layers.35.attention.v_proj",
|
578 |
+
"vision_tower.transformer.layers.35.attention.q_proj",
|
579 |
+
"vision_tower.transformer.layers.35.attention.o_proj",
|
580 |
+
"vision_tower.transformer.layers.36.feed_forward.gate_proj",
|
581 |
+
"vision_tower.transformer.layers.36.feed_forward.up_proj",
|
582 |
+
"vision_tower.transformer.layers.36.feed_forward.down_proj",
|
583 |
+
"vision_tower.transformer.layers.36.attention.k_proj",
|
584 |
+
"vision_tower.transformer.layers.36.attention.v_proj",
|
585 |
+
"vision_tower.transformer.layers.36.attention.q_proj",
|
586 |
+
"vision_tower.transformer.layers.36.attention.o_proj",
|
587 |
+
"vision_tower.transformer.layers.37.feed_forward.gate_proj",
|
588 |
+
"vision_tower.transformer.layers.37.feed_forward.up_proj",
|
589 |
+
"vision_tower.transformer.layers.37.feed_forward.down_proj",
|
590 |
+
"vision_tower.transformer.layers.37.attention.k_proj",
|
591 |
+
"vision_tower.transformer.layers.37.attention.v_proj",
|
592 |
+
"vision_tower.transformer.layers.37.attention.q_proj",
|
593 |
+
"vision_tower.transformer.layers.37.attention.o_proj",
|
594 |
+
"vision_tower.transformer.layers.38.feed_forward.gate_proj",
|
595 |
+
"vision_tower.transformer.layers.38.feed_forward.up_proj",
|
596 |
+
"vision_tower.transformer.layers.38.feed_forward.down_proj",
|
597 |
+
"vision_tower.transformer.layers.38.attention.k_proj",
|
598 |
+
"vision_tower.transformer.layers.38.attention.v_proj",
|
599 |
+
"vision_tower.transformer.layers.38.attention.q_proj",
|
600 |
+
"vision_tower.transformer.layers.38.attention.o_proj",
|
601 |
+
"vision_tower.transformer.layers.39.feed_forward.gate_proj",
|
602 |
+
"vision_tower.transformer.layers.39.feed_forward.up_proj",
|
603 |
+
"vision_tower.transformer.layers.39.feed_forward.down_proj",
|
604 |
+
"vision_tower.transformer.layers.39.attention.k_proj",
|
605 |
+
"vision_tower.transformer.layers.39.attention.v_proj",
|
606 |
+
"vision_tower.transformer.layers.39.attention.q_proj",
|
607 |
+
"vision_tower.transformer.layers.39.attention.o_proj",
|
608 |
+
"multi_modal_projector.linear_1",
|
609 |
+
"multi_modal_projector.linear_2",
|
610 |
+
"language_model.model.layers.0.self_attn.v_proj",
|
611 |
+
"language_model.model.layers.0.self_attn.o_proj",
|
612 |
+
"language_model.model.layers.0.mlp.gate_proj",
|
613 |
+
"language_model.model.layers.0.mlp.up_proj",
|
614 |
+
"language_model.model.layers.0.mlp.down_proj",
|
615 |
+
"language_model.model.layers.1.self_attn.q_proj",
|
616 |
+
"language_model.model.layers.1.self_attn.k_proj",
|
617 |
+
"language_model.model.layers.1.self_attn.v_proj",
|
618 |
+
"language_model.model.layers.1.self_attn.o_proj",
|
619 |
+
"language_model.model.layers.1.mlp.gate_proj",
|
620 |
+
"language_model.model.layers.1.mlp.up_proj",
|
621 |
+
"language_model.model.layers.1.mlp.down_proj",
|
622 |
+
"language_model.model.layers.2.self_attn.q_proj",
|
623 |
+
"language_model.model.layers.2.self_attn.k_proj",
|
624 |
+
"language_model.model.layers.2.self_attn.v_proj",
|
625 |
+
"language_model.model.layers.2.self_attn.o_proj",
|
626 |
+
"language_model.model.layers.2.mlp.gate_proj",
|
627 |
+
"language_model.model.layers.2.mlp.up_proj",
|
628 |
+
"language_model.model.layers.2.mlp.down_proj",
|
629 |
+
"language_model.model.layers.3.self_attn.q_proj",
|
630 |
+
"language_model.model.layers.3.self_attn.k_proj",
|
631 |
+
"language_model.model.layers.3.self_attn.v_proj",
|
632 |
+
"language_model.model.layers.3.self_attn.o_proj",
|
633 |
+
"language_model.model.layers.3.mlp.gate_proj",
|
634 |
+
"language_model.model.layers.3.mlp.up_proj",
|
635 |
+
"language_model.model.layers.3.mlp.down_proj",
|
636 |
+
"language_model.model.layers.4.self_attn.q_proj",
|
637 |
+
"language_model.model.layers.4.self_attn.k_proj",
|
638 |
+
"language_model.model.layers.4.self_attn.v_proj",
|
639 |
+
"language_model.model.layers.4.self_attn.o_proj",
|
640 |
+
"language_model.model.layers.4.mlp.gate_proj",
|
641 |
+
"language_model.model.layers.4.mlp.up_proj",
|
642 |
+
"language_model.model.layers.4.mlp.down_proj",
|
643 |
+
"language_model.model.layers.5.self_attn.q_proj",
|
644 |
+
"language_model.model.layers.5.self_attn.k_proj",
|
645 |
+
"language_model.model.layers.5.self_attn.v_proj",
|
646 |
+
"language_model.model.layers.5.self_attn.o_proj",
|
647 |
+
"language_model.model.layers.5.mlp.gate_proj",
|
648 |
+
"language_model.model.layers.5.mlp.up_proj",
|
649 |
+
"language_model.model.layers.5.mlp.down_proj",
|
650 |
+
"language_model.model.layers.6.self_attn.q_proj",
|
651 |
+
"language_model.model.layers.6.self_attn.k_proj",
|
652 |
+
"language_model.model.layers.6.self_attn.v_proj",
|
653 |
+
"language_model.model.layers.6.self_attn.o_proj",
|
654 |
+
"language_model.model.layers.6.mlp.gate_proj",
|
655 |
+
"language_model.model.layers.6.mlp.up_proj",
|
656 |
+
"language_model.model.layers.6.mlp.down_proj",
|
657 |
+
"language_model.model.layers.7.self_attn.q_proj",
|
658 |
+
"language_model.model.layers.7.self_attn.k_proj",
|
659 |
+
"language_model.model.layers.7.self_attn.v_proj",
|
660 |
+
"language_model.model.layers.7.self_attn.o_proj",
|
661 |
+
"language_model.model.layers.7.mlp.gate_proj",
|
662 |
+
"language_model.model.layers.7.mlp.up_proj",
|
663 |
+
"language_model.model.layers.7.mlp.down_proj",
|
664 |
+
"language_model.model.layers.8.self_attn.q_proj",
|
665 |
+
"language_model.model.layers.8.self_attn.k_proj",
|
666 |
+
"language_model.model.layers.8.self_attn.v_proj",
|
667 |
+
"language_model.model.layers.8.self_attn.o_proj",
|
668 |
+
"language_model.model.layers.8.mlp.gate_proj",
|
669 |
+
"language_model.model.layers.8.mlp.up_proj",
|
670 |
+
"language_model.model.layers.8.mlp.down_proj",
|
671 |
+
"language_model.model.layers.9.self_attn.q_proj",
|
672 |
+
"language_model.model.layers.9.self_attn.k_proj",
|
673 |
+
"language_model.model.layers.9.self_attn.v_proj",
|
674 |
+
"language_model.model.layers.9.self_attn.o_proj",
|
675 |
+
"language_model.model.layers.9.mlp.gate_proj",
|
676 |
+
"language_model.model.layers.9.mlp.up_proj",
|
677 |
+
"language_model.model.layers.9.mlp.down_proj",
|
678 |
+
"language_model.model.layers.10.self_attn.q_proj",
|
679 |
+
"language_model.model.layers.10.self_attn.k_proj",
|
680 |
+
"language_model.model.layers.10.self_attn.v_proj",
|
681 |
+
"language_model.model.layers.10.self_attn.o_proj",
|
682 |
+
"language_model.model.layers.10.mlp.gate_proj",
|
683 |
+
"language_model.model.layers.10.mlp.up_proj",
|
684 |
+
"language_model.model.layers.10.mlp.down_proj",
|
685 |
+
"language_model.model.layers.11.self_attn.q_proj",
|
686 |
+
"language_model.model.layers.11.self_attn.k_proj",
|
687 |
+
"language_model.model.layers.11.self_attn.v_proj",
|
688 |
+
"language_model.model.layers.11.self_attn.o_proj",
|
689 |
+
"language_model.model.layers.11.mlp.gate_proj",
|
690 |
+
"language_model.model.layers.11.mlp.up_proj",
|
691 |
+
"language_model.model.layers.11.mlp.down_proj",
|
692 |
+
"language_model.model.layers.12.self_attn.q_proj",
|
693 |
+
"language_model.model.layers.12.self_attn.k_proj",
|
694 |
+
"language_model.model.layers.12.self_attn.v_proj",
|
695 |
+
"language_model.model.layers.12.self_attn.o_proj",
|
696 |
+
"language_model.model.layers.12.mlp.gate_proj",
|
697 |
+
"language_model.model.layers.12.mlp.up_proj",
|
698 |
+
"language_model.model.layers.12.mlp.down_proj",
|
699 |
+
"language_model.model.layers.13.self_attn.q_proj",
|
700 |
+
"language_model.model.layers.13.self_attn.k_proj",
|
701 |
+
"language_model.model.layers.13.self_attn.v_proj",
|
702 |
+
"language_model.model.layers.13.self_attn.o_proj",
|
703 |
+
"language_model.model.layers.13.mlp.gate_proj",
|
704 |
+
"language_model.model.layers.13.mlp.up_proj",
|
705 |
+
"language_model.model.layers.13.mlp.down_proj",
|
706 |
+
"language_model.model.layers.14.self_attn.q_proj",
|
707 |
+
"language_model.model.layers.14.self_attn.k_proj",
|
708 |
+
"language_model.model.layers.14.self_attn.v_proj",
|
709 |
+
"language_model.model.layers.14.self_attn.o_proj",
|
710 |
+
"language_model.model.layers.14.mlp.gate_proj",
|
711 |
+
"language_model.model.layers.14.mlp.up_proj",
|
712 |
+
"language_model.model.layers.14.mlp.down_proj",
|
713 |
+
"language_model.model.layers.15.self_attn.q_proj",
|
714 |
+
"language_model.model.layers.15.self_attn.k_proj",
|
715 |
+
"language_model.model.layers.15.self_attn.v_proj",
|
716 |
+
"language_model.model.layers.15.self_attn.o_proj",
|
717 |
+
"language_model.model.layers.15.mlp.gate_proj",
|
718 |
+
"language_model.model.layers.15.mlp.up_proj",
|
719 |
+
"language_model.model.layers.15.mlp.down_proj",
|
720 |
+
"language_model.model.layers.16.self_attn.q_proj",
|
721 |
+
"language_model.model.layers.16.self_attn.k_proj",
|
722 |
+
"language_model.model.layers.16.self_attn.v_proj",
|
723 |
+
"language_model.model.layers.16.self_attn.o_proj",
|
724 |
+
"language_model.model.layers.16.mlp.gate_proj",
|
725 |
+
"language_model.model.layers.16.mlp.up_proj",
|
726 |
+
"language_model.model.layers.16.mlp.down_proj",
|
727 |
+
"language_model.model.layers.17.self_attn.q_proj",
|
728 |
+
"language_model.model.layers.17.self_attn.k_proj",
|
729 |
+
"language_model.model.layers.17.self_attn.v_proj",
|
730 |
+
"language_model.model.layers.17.self_attn.o_proj",
|
731 |
+
"language_model.model.layers.17.mlp.gate_proj",
|
732 |
+
"language_model.model.layers.17.mlp.up_proj",
|
733 |
+
"language_model.model.layers.17.mlp.down_proj",
|
734 |
+
"language_model.model.layers.18.self_attn.q_proj",
|
735 |
+
"language_model.model.layers.18.self_attn.k_proj",
|
736 |
+
"language_model.model.layers.18.self_attn.v_proj",
|
737 |
+
"language_model.model.layers.18.self_attn.o_proj",
|
738 |
+
"language_model.model.layers.18.mlp.gate_proj",
|
739 |
+
"language_model.model.layers.18.mlp.up_proj",
|
740 |
+
"language_model.model.layers.18.mlp.down_proj",
|
741 |
+
"language_model.model.layers.19.self_attn.q_proj",
|
742 |
+
"language_model.model.layers.19.self_attn.k_proj",
|
743 |
+
"language_model.model.layers.19.self_attn.v_proj",
|
744 |
+
"language_model.model.layers.19.self_attn.o_proj",
|
745 |
+
"language_model.model.layers.19.mlp.gate_proj",
|
746 |
+
"language_model.model.layers.19.mlp.up_proj",
|
747 |
+
"language_model.model.layers.19.mlp.down_proj",
|
748 |
+
"language_model.model.layers.20.self_attn.q_proj",
|
749 |
+
"language_model.model.layers.20.self_attn.k_proj",
|
750 |
+
"language_model.model.layers.20.self_attn.v_proj",
|
751 |
+
"language_model.model.layers.20.self_attn.o_proj",
|
752 |
+
"language_model.model.layers.20.mlp.gate_proj",
|
753 |
+
"language_model.model.layers.20.mlp.up_proj",
|
754 |
+
"language_model.model.layers.20.mlp.down_proj",
|
755 |
+
"language_model.model.layers.21.self_attn.q_proj",
|
756 |
+
"language_model.model.layers.21.self_attn.k_proj",
|
757 |
+
"language_model.model.layers.21.self_attn.v_proj",
|
758 |
+
"language_model.model.layers.21.self_attn.o_proj",
|
759 |
+
"language_model.model.layers.21.mlp.gate_proj",
|
760 |
+
"language_model.model.layers.21.mlp.up_proj",
|
761 |
+
"language_model.model.layers.21.mlp.down_proj",
|
762 |
+
"language_model.model.layers.22.self_attn.q_proj",
|
763 |
+
"language_model.model.layers.22.self_attn.k_proj",
|
764 |
+
"language_model.model.layers.22.self_attn.v_proj",
|
765 |
+
"language_model.model.layers.22.self_attn.o_proj",
|
766 |
+
"language_model.model.layers.22.mlp.gate_proj",
|
767 |
+
"language_model.model.layers.22.mlp.up_proj",
|
768 |
+
"language_model.model.layers.22.mlp.down_proj",
|
769 |
+
"language_model.model.layers.23.self_attn.q_proj",
|
770 |
+
"language_model.model.layers.23.self_attn.k_proj",
|
771 |
+
"language_model.model.layers.23.self_attn.v_proj",
|
772 |
+
"language_model.model.layers.23.self_attn.o_proj",
|
773 |
+
"language_model.model.layers.23.mlp.gate_proj",
|
774 |
+
"language_model.model.layers.23.mlp.up_proj",
|
775 |
+
"language_model.model.layers.23.mlp.down_proj",
|
776 |
+
"language_model.model.layers.24.self_attn.q_proj",
|
777 |
+
"language_model.model.layers.24.self_attn.k_proj",
|
778 |
+
"language_model.model.layers.24.self_attn.v_proj",
|
779 |
+
"language_model.model.layers.24.self_attn.o_proj",
|
780 |
+
"language_model.model.layers.24.mlp.gate_proj",
|
781 |
+
"language_model.model.layers.24.mlp.up_proj",
|
782 |
+
"language_model.model.layers.24.mlp.down_proj",
|
783 |
+
"language_model.model.layers.25.self_attn.q_proj",
|
784 |
+
"language_model.model.layers.25.self_attn.k_proj",
|
785 |
+
"language_model.model.layers.25.self_attn.v_proj",
|
786 |
+
"language_model.model.layers.25.self_attn.o_proj",
|
787 |
+
"language_model.model.layers.25.mlp.gate_proj",
|
788 |
+
"language_model.model.layers.25.mlp.up_proj",
|
789 |
+
"language_model.model.layers.25.mlp.down_proj",
|
790 |
+
"language_model.model.layers.26.self_attn.q_proj",
|
791 |
+
"language_model.model.layers.26.self_attn.k_proj",
|
792 |
+
"language_model.model.layers.26.self_attn.v_proj",
|
793 |
+
"language_model.model.layers.26.self_attn.o_proj",
|
794 |
+
"language_model.model.layers.26.mlp.gate_proj",
|
795 |
+
"language_model.model.layers.26.mlp.up_proj",
|
796 |
+
"language_model.model.layers.26.mlp.down_proj",
|
797 |
+
"language_model.model.layers.27.self_attn.q_proj",
|
798 |
+
"language_model.model.layers.27.self_attn.k_proj",
|
799 |
+
"language_model.model.layers.27.self_attn.v_proj",
|
800 |
+
"language_model.model.layers.27.self_attn.o_proj",
|
801 |
+
"language_model.model.layers.27.mlp.gate_proj",
|
802 |
+
"language_model.model.layers.27.mlp.up_proj",
|
803 |
+
"language_model.model.layers.27.mlp.down_proj",
|
804 |
+
"language_model.model.layers.28.self_attn.q_proj",
|
805 |
+
"language_model.model.layers.28.self_attn.k_proj",
|
806 |
+
"language_model.model.layers.28.self_attn.v_proj",
|
807 |
+
"language_model.model.layers.28.self_attn.o_proj",
|
808 |
+
"language_model.model.layers.28.mlp.gate_proj",
|
809 |
+
"language_model.model.layers.28.mlp.up_proj",
|
810 |
+
"language_model.model.layers.28.mlp.down_proj",
|
811 |
+
"language_model.model.layers.29.self_attn.q_proj",
|
812 |
+
"language_model.model.layers.29.self_attn.k_proj",
|
813 |
+
"language_model.model.layers.29.self_attn.v_proj",
|
814 |
+
"language_model.model.layers.29.self_attn.o_proj",
|
815 |
+
"language_model.model.layers.29.mlp.gate_proj",
|
816 |
+
"language_model.model.layers.29.mlp.up_proj",
|
817 |
+
"language_model.model.layers.29.mlp.down_proj",
|
818 |
+
"language_model.model.layers.30.self_attn.q_proj",
|
819 |
+
"language_model.model.layers.30.self_attn.k_proj",
|
820 |
+
"language_model.model.layers.30.self_attn.v_proj",
|
821 |
+
"language_model.model.layers.30.self_attn.o_proj",
|
822 |
+
"language_model.model.layers.30.mlp.gate_proj",
|
823 |
+
"language_model.model.layers.30.mlp.up_proj",
|
824 |
+
"language_model.model.layers.30.mlp.down_proj",
|
825 |
+
"language_model.model.layers.31.self_attn.q_proj",
|
826 |
+
"language_model.model.layers.31.self_attn.k_proj",
|
827 |
+
"language_model.model.layers.31.self_attn.v_proj",
|
828 |
+
"language_model.model.layers.31.self_attn.o_proj",
|
829 |
+
"language_model.model.layers.31.mlp.gate_proj",
|
830 |
+
"language_model.model.layers.31.mlp.up_proj",
|
831 |
+
"language_model.model.layers.31.mlp.down_proj",
|
832 |
+
"language_model.model.layers.32.self_attn.q_proj",
|
833 |
+
"language_model.model.layers.32.self_attn.k_proj",
|
834 |
+
"language_model.model.layers.32.self_attn.v_proj",
|
835 |
+
"language_model.model.layers.32.self_attn.o_proj",
|
836 |
+
"language_model.model.layers.32.mlp.gate_proj",
|
837 |
+
"language_model.model.layers.32.mlp.up_proj",
|
838 |
+
"language_model.model.layers.32.mlp.down_proj",
|
839 |
+
"language_model.model.layers.33.self_attn.q_proj",
|
840 |
+
"language_model.model.layers.33.self_attn.k_proj",
|
841 |
+
"language_model.model.layers.33.self_attn.v_proj",
|
842 |
+
"language_model.model.layers.33.self_attn.o_proj",
|
843 |
+
"language_model.model.layers.33.mlp.gate_proj",
|
844 |
+
"language_model.model.layers.33.mlp.up_proj",
|
845 |
+
"language_model.model.layers.33.mlp.down_proj",
|
846 |
+
"language_model.model.layers.34.self_attn.q_proj",
|
847 |
+
"language_model.model.layers.34.self_attn.k_proj",
|
848 |
+
"language_model.model.layers.34.self_attn.v_proj",
|
849 |
+
"language_model.model.layers.34.self_attn.o_proj",
|
850 |
+
"language_model.model.layers.34.mlp.gate_proj",
|
851 |
+
"language_model.model.layers.34.mlp.up_proj",
|
852 |
+
"language_model.model.layers.34.mlp.down_proj",
|
853 |
+
"language_model.model.layers.35.self_attn.q_proj",
|
854 |
+
"language_model.model.layers.35.self_attn.k_proj",
|
855 |
+
"language_model.model.layers.35.self_attn.v_proj",
|
856 |
+
"language_model.model.layers.35.self_attn.o_proj",
|
857 |
+
"language_model.model.layers.35.mlp.gate_proj",
|
858 |
+
"language_model.model.layers.35.mlp.up_proj",
|
859 |
+
"language_model.model.layers.35.mlp.down_proj",
|
860 |
+
"language_model.model.layers.36.self_attn.q_proj",
|
861 |
+
"language_model.model.layers.36.self_attn.k_proj",
|
862 |
+
"language_model.model.layers.36.self_attn.v_proj",
|
863 |
+
"language_model.model.layers.36.self_attn.o_proj",
|
864 |
+
"language_model.model.layers.36.mlp.gate_proj",
|
865 |
+
"language_model.model.layers.36.mlp.up_proj",
|
866 |
+
"language_model.model.layers.36.mlp.down_proj",
|
867 |
+
"language_model.model.layers.37.self_attn.q_proj",
|
868 |
+
"language_model.model.layers.37.self_attn.k_proj",
|
869 |
+
"language_model.model.layers.37.self_attn.v_proj",
|
870 |
+
"language_model.model.layers.37.self_attn.o_proj",
|
871 |
+
"language_model.model.layers.37.mlp.gate_proj",
|
872 |
+
"language_model.model.layers.37.mlp.up_proj",
|
873 |
+
"language_model.model.layers.37.mlp.down_proj",
|
874 |
+
"language_model.model.layers.38.self_attn.q_proj",
|
875 |
+
"language_model.model.layers.38.self_attn.k_proj",
|
876 |
+
"language_model.model.layers.38.self_attn.v_proj",
|
877 |
+
"language_model.model.layers.38.self_attn.o_proj",
|
878 |
+
"language_model.model.layers.38.mlp.gate_proj",
|
879 |
+
"language_model.model.layers.38.mlp.up_proj",
|
880 |
+
"language_model.model.layers.38.mlp.down_proj",
|
881 |
+
"language_model.model.layers.39.self_attn.q_proj",
|
882 |
+
"language_model.model.layers.39.self_attn.k_proj",
|
883 |
+
"language_model.model.layers.39.self_attn.v_proj",
|
884 |
+
"language_model.model.layers.39.self_attn.o_proj",
|
885 |
+
"language_model.model.layers.39.mlp.gate_proj",
|
886 |
+
"language_model.model.layers.39.mlp.up_proj",
|
887 |
+
"language_model.model.layers.39.mlp.down_proj",
|
888 |
+
"language_model.model.layers.40.self_attn.q_proj",
|
889 |
+
"language_model.model.layers.40.self_attn.k_proj",
|
890 |
+
"language_model.model.layers.40.self_attn.v_proj",
|
891 |
+
"language_model.model.layers.40.self_attn.o_proj",
|
892 |
+
"language_model.model.layers.40.mlp.gate_proj",
|
893 |
+
"language_model.model.layers.40.mlp.up_proj",
|
894 |
+
"language_model.model.layers.40.mlp.down_proj",
|
895 |
+
"language_model.model.layers.41.self_attn.q_proj",
|
896 |
+
"language_model.model.layers.41.self_attn.k_proj",
|
897 |
+
"language_model.model.layers.41.self_attn.v_proj",
|
898 |
+
"language_model.model.layers.41.self_attn.o_proj",
|
899 |
+
"language_model.model.layers.41.mlp.gate_proj",
|
900 |
+
"language_model.model.layers.41.mlp.up_proj",
|
901 |
+
"language_model.model.layers.41.mlp.down_proj",
|
902 |
+
"language_model.model.layers.42.self_attn.q_proj",
|
903 |
+
"language_model.model.layers.42.self_attn.k_proj",
|
904 |
+
"language_model.model.layers.42.self_attn.v_proj",
|
905 |
+
"language_model.model.layers.42.self_attn.o_proj",
|
906 |
+
"language_model.model.layers.42.mlp.gate_proj",
|
907 |
+
"language_model.model.layers.42.mlp.up_proj",
|
908 |
+
"language_model.model.layers.42.mlp.down_proj",
|
909 |
+
"language_model.model.layers.43.self_attn.q_proj",
|
910 |
+
"language_model.model.layers.43.self_attn.k_proj",
|
911 |
+
"language_model.model.layers.43.self_attn.v_proj",
|
912 |
+
"language_model.model.layers.43.self_attn.o_proj",
|
913 |
+
"language_model.model.layers.43.mlp.gate_proj",
|
914 |
+
"language_model.model.layers.43.mlp.up_proj",
|
915 |
+
"language_model.model.layers.43.mlp.down_proj",
|
916 |
+
"language_model.model.layers.44.self_attn.q_proj",
|
917 |
+
"language_model.model.layers.44.self_attn.k_proj",
|
918 |
+
"language_model.model.layers.44.self_attn.v_proj",
|
919 |
+
"language_model.model.layers.44.self_attn.o_proj",
|
920 |
+
"language_model.model.layers.44.mlp.gate_proj",
|
921 |
+
"language_model.model.layers.44.mlp.up_proj",
|
922 |
+
"language_model.model.layers.44.mlp.down_proj",
|
923 |
+
"language_model.model.layers.45.self_attn.q_proj",
|
924 |
+
"language_model.model.layers.45.self_attn.k_proj",
|
925 |
+
"language_model.model.layers.45.self_attn.v_proj",
|
926 |
+
"language_model.model.layers.45.self_attn.o_proj",
|
927 |
+
"language_model.model.layers.45.mlp.gate_proj",
|
928 |
+
"language_model.model.layers.45.mlp.up_proj",
|
929 |
+
"language_model.model.layers.45.mlp.down_proj",
|
930 |
+
"language_model.model.layers.46.self_attn.q_proj",
|
931 |
+
"language_model.model.layers.46.self_attn.k_proj",
|
932 |
+
"language_model.model.layers.46.self_attn.v_proj",
|
933 |
+
"language_model.model.layers.46.self_attn.o_proj",
|
934 |
+
"language_model.model.layers.46.mlp.gate_proj",
|
935 |
+
"language_model.model.layers.46.mlp.up_proj",
|
936 |
+
"language_model.model.layers.46.mlp.down_proj",
|
937 |
+
"language_model.model.layers.47.self_attn.q_proj",
|
938 |
+
"language_model.model.layers.47.self_attn.k_proj",
|
939 |
+
"language_model.model.layers.47.self_attn.v_proj",
|
940 |
+
"language_model.model.layers.47.self_attn.o_proj",
|
941 |
+
"language_model.model.layers.47.mlp.gate_proj",
|
942 |
+
"language_model.model.layers.47.mlp.up_proj",
|
943 |
+
"language_model.model.layers.47.mlp.down_proj",
|
944 |
+
"language_model.model.layers.48.self_attn.q_proj",
|
945 |
+
"language_model.model.layers.48.self_attn.k_proj",
|
946 |
+
"language_model.model.layers.48.self_attn.v_proj",
|
947 |
+
"language_model.model.layers.48.self_attn.o_proj",
|
948 |
+
"language_model.model.layers.48.mlp.gate_proj",
|
949 |
+
"language_model.model.layers.48.mlp.up_proj",
|
950 |
+
"language_model.model.layers.48.mlp.down_proj",
|
951 |
+
"language_model.model.layers.49.self_attn.q_proj",
|
952 |
+
"language_model.model.layers.49.self_attn.k_proj",
|
953 |
+
"language_model.model.layers.49.self_attn.v_proj",
|
954 |
+
"language_model.model.layers.49.self_attn.o_proj",
|
955 |
+
"language_model.model.layers.49.mlp.gate_proj",
|
956 |
+
"language_model.model.layers.49.mlp.up_proj",
|
957 |
+
"language_model.model.layers.49.mlp.down_proj",
|
958 |
+
"language_model.model.layers.50.self_attn.q_proj",
|
959 |
+
"language_model.model.layers.50.self_attn.k_proj",
|
960 |
+
"language_model.model.layers.50.self_attn.v_proj",
|
961 |
+
"language_model.model.layers.50.self_attn.o_proj",
|
962 |
+
"language_model.model.layers.50.mlp.gate_proj",
|
963 |
+
"language_model.model.layers.50.mlp.up_proj",
|
964 |
+
"language_model.model.layers.50.mlp.down_proj",
|
965 |
+
"language_model.model.layers.51.self_attn.q_proj",
|
966 |
+
"language_model.model.layers.51.self_attn.k_proj",
|
967 |
+
"language_model.model.layers.51.self_attn.v_proj",
|
968 |
+
"language_model.model.layers.51.self_attn.o_proj",
|
969 |
+
"language_model.model.layers.51.mlp.gate_proj",
|
970 |
+
"language_model.model.layers.51.mlp.up_proj",
|
971 |
+
"language_model.model.layers.51.mlp.down_proj",
|
972 |
+
"language_model.model.layers.52.self_attn.q_proj",
|
973 |
+
"language_model.model.layers.52.self_attn.k_proj",
|
974 |
+
"language_model.model.layers.52.self_attn.v_proj",
|
975 |
+
"language_model.model.layers.52.self_attn.o_proj",
|
976 |
+
"language_model.model.layers.52.mlp.gate_proj",
|
977 |
+
"language_model.model.layers.52.mlp.up_proj",
|
978 |
+
"language_model.model.layers.52.mlp.down_proj",
|
979 |
+
"language_model.model.layers.53.self_attn.q_proj",
|
980 |
+
"language_model.model.layers.53.self_attn.k_proj",
|
981 |
+
"language_model.model.layers.53.self_attn.v_proj",
|
982 |
+
"language_model.model.layers.53.self_attn.o_proj",
|
983 |
+
"language_model.model.layers.53.mlp.gate_proj",
|
984 |
+
"language_model.model.layers.53.mlp.up_proj",
|
985 |
+
"language_model.model.layers.53.mlp.down_proj",
|
986 |
+
"language_model.model.layers.54.self_attn.q_proj",
|
987 |
+
"language_model.model.layers.54.self_attn.k_proj",
|
988 |
+
"language_model.model.layers.54.self_attn.v_proj",
|
989 |
+
"language_model.model.layers.54.self_attn.o_proj",
|
990 |
+
"language_model.model.layers.54.mlp.gate_proj",
|
991 |
+
"language_model.model.layers.54.mlp.up_proj",
|
992 |
+
"language_model.model.layers.54.mlp.down_proj",
|
993 |
+
"language_model.model.layers.55.self_attn.q_proj",
|
994 |
+
"language_model.model.layers.55.self_attn.k_proj",
|
995 |
+
"language_model.model.layers.55.self_attn.v_proj",
|
996 |
+
"language_model.model.layers.55.self_attn.o_proj",
|
997 |
+
"language_model.model.layers.55.mlp.gate_proj",
|
998 |
+
"language_model.model.layers.55.mlp.up_proj",
|
999 |
+
"language_model.model.layers.55.mlp.down_proj",
|
1000 |
+
"language_model.model.layers.56.self_attn.q_proj",
|
1001 |
+
"language_model.model.layers.56.self_attn.k_proj",
|
1002 |
+
"language_model.model.layers.56.self_attn.v_proj",
|
1003 |
+
"language_model.model.layers.56.self_attn.o_proj",
|
1004 |
+
"language_model.model.layers.56.mlp.gate_proj",
|
1005 |
+
"language_model.model.layers.56.mlp.up_proj",
|
1006 |
+
"language_model.model.layers.56.mlp.down_proj",
|
1007 |
+
"language_model.model.layers.57.self_attn.q_proj",
|
1008 |
+
"language_model.model.layers.57.self_attn.k_proj",
|
1009 |
+
"language_model.model.layers.57.self_attn.v_proj",
|
1010 |
+
"language_model.model.layers.57.self_attn.o_proj",
|
1011 |
+
"language_model.model.layers.57.mlp.gate_proj",
|
1012 |
+
"language_model.model.layers.57.mlp.up_proj",
|
1013 |
+
"language_model.model.layers.57.mlp.down_proj",
|
1014 |
+
"language_model.model.layers.58.self_attn.q_proj",
|
1015 |
+
"language_model.model.layers.58.self_attn.k_proj",
|
1016 |
+
"language_model.model.layers.58.self_attn.v_proj",
|
1017 |
+
"language_model.model.layers.58.self_attn.o_proj",
|
1018 |
+
"language_model.model.layers.58.mlp.gate_proj",
|
1019 |
+
"language_model.model.layers.58.mlp.up_proj",
|
1020 |
+
"language_model.model.layers.58.mlp.down_proj",
|
1021 |
+
"language_model.model.layers.59.self_attn.q_proj",
|
1022 |
+
"language_model.model.layers.59.self_attn.k_proj",
|
1023 |
+
"language_model.model.layers.59.self_attn.v_proj",
|
1024 |
+
"language_model.model.layers.59.self_attn.o_proj",
|
1025 |
+
"language_model.model.layers.59.mlp.gate_proj",
|
1026 |
+
"language_model.model.layers.59.mlp.up_proj",
|
1027 |
+
"language_model.model.layers.59.mlp.down_proj",
|
1028 |
+
"language_model.model.layers.60.self_attn.q_proj",
|
1029 |
+
"language_model.model.layers.60.self_attn.k_proj",
|
1030 |
+
"language_model.model.layers.60.self_attn.v_proj",
|
1031 |
+
"language_model.model.layers.60.self_attn.o_proj",
|
1032 |
+
"language_model.model.layers.60.mlp.gate_proj",
|
1033 |
+
"language_model.model.layers.60.mlp.up_proj",
|
1034 |
+
"language_model.model.layers.60.mlp.down_proj",
|
1035 |
+
"language_model.model.layers.61.self_attn.q_proj",
|
1036 |
+
"language_model.model.layers.61.self_attn.k_proj",
|
1037 |
+
"language_model.model.layers.61.self_attn.v_proj",
|
1038 |
+
"language_model.model.layers.61.self_attn.o_proj",
|
1039 |
+
"language_model.model.layers.61.mlp.gate_proj",
|
1040 |
+
"language_model.model.layers.61.mlp.up_proj",
|
1041 |
+
"language_model.model.layers.61.mlp.down_proj",
|
1042 |
+
"language_model.model.layers.62.self_attn.q_proj",
|
1043 |
+
"language_model.model.layers.62.self_attn.k_proj",
|
1044 |
+
"language_model.model.layers.62.self_attn.v_proj",
|
1045 |
+
"language_model.model.layers.62.self_attn.o_proj",
|
1046 |
+
"language_model.model.layers.62.mlp.gate_proj",
|
1047 |
+
"language_model.model.layers.62.mlp.up_proj",
|
1048 |
+
"language_model.model.layers.62.mlp.down_proj",
|
1049 |
+
"language_model.model.layers.63.self_attn.q_proj",
|
1050 |
+
"language_model.model.layers.63.self_attn.k_proj",
|
1051 |
+
"language_model.model.layers.63.self_attn.v_proj",
|
1052 |
+
"language_model.model.layers.63.self_attn.o_proj",
|
1053 |
+
"language_model.model.layers.63.mlp.gate_proj",
|
1054 |
+
"language_model.model.layers.63.mlp.up_proj",
|
1055 |
+
"language_model.model.layers.63.mlp.down_proj",
|
1056 |
+
"language_model.model.layers.64.self_attn.q_proj",
|
1057 |
+
"language_model.model.layers.64.self_attn.k_proj",
|
1058 |
+
"language_model.model.layers.64.self_attn.v_proj",
|
1059 |
+
"language_model.model.layers.64.self_attn.o_proj",
|
1060 |
+
"language_model.model.layers.64.mlp.gate_proj",
|
1061 |
+
"language_model.model.layers.64.mlp.up_proj",
|
1062 |
+
"language_model.model.layers.64.mlp.down_proj",
|
1063 |
+
"language_model.model.layers.65.self_attn.q_proj",
|
1064 |
+
"language_model.model.layers.65.self_attn.k_proj",
|
1065 |
+
"language_model.model.layers.65.self_attn.v_proj",
|
1066 |
+
"language_model.model.layers.65.self_attn.o_proj",
|
1067 |
+
"language_model.model.layers.65.mlp.gate_proj",
|
1068 |
+
"language_model.model.layers.65.mlp.up_proj",
|
1069 |
+
"language_model.model.layers.65.mlp.down_proj",
|
1070 |
+
"language_model.model.layers.66.self_attn.q_proj",
|
1071 |
+
"language_model.model.layers.66.self_attn.k_proj",
|
1072 |
+
"language_model.model.layers.66.self_attn.v_proj",
|
1073 |
+
"language_model.model.layers.66.self_attn.o_proj",
|
1074 |
+
"language_model.model.layers.66.mlp.gate_proj",
|
1075 |
+
"language_model.model.layers.66.mlp.up_proj",
|
1076 |
+
"language_model.model.layers.66.mlp.down_proj",
|
1077 |
+
"language_model.model.layers.67.self_attn.q_proj",
|
1078 |
+
"language_model.model.layers.67.self_attn.k_proj",
|
1079 |
+
"language_model.model.layers.67.self_attn.v_proj",
|
1080 |
+
"language_model.model.layers.67.self_attn.o_proj",
|
1081 |
+
"language_model.model.layers.67.mlp.gate_proj",
|
1082 |
+
"language_model.model.layers.67.mlp.up_proj",
|
1083 |
+
"language_model.model.layers.67.mlp.down_proj",
|
1084 |
+
"language_model.model.layers.68.self_attn.q_proj",
|
1085 |
+
"language_model.model.layers.68.self_attn.k_proj",
|
1086 |
+
"language_model.model.layers.68.self_attn.v_proj",
|
1087 |
+
"language_model.model.layers.68.self_attn.o_proj",
|
1088 |
+
"language_model.model.layers.68.mlp.gate_proj",
|
1089 |
+
"language_model.model.layers.68.mlp.up_proj",
|
1090 |
+
"language_model.model.layers.68.mlp.down_proj",
|
1091 |
+
"language_model.model.layers.69.self_attn.q_proj",
|
1092 |
+
"language_model.model.layers.69.self_attn.k_proj",
|
1093 |
+
"language_model.model.layers.69.self_attn.v_proj",
|
1094 |
+
"language_model.model.layers.69.self_attn.o_proj",
|
1095 |
+
"language_model.model.layers.69.mlp.gate_proj",
|
1096 |
+
"language_model.model.layers.69.mlp.up_proj",
|
1097 |
+
"language_model.model.layers.69.mlp.down_proj",
|
1098 |
+
"language_model.model.layers.70.self_attn.q_proj",
|
1099 |
+
"language_model.model.layers.70.self_attn.k_proj",
|
1100 |
+
"language_model.model.layers.70.self_attn.v_proj",
|
1101 |
+
"language_model.model.layers.70.self_attn.o_proj",
|
1102 |
+
"language_model.model.layers.70.mlp.gate_proj",
|
1103 |
+
"language_model.model.layers.70.mlp.up_proj",
|
1104 |
+
"language_model.model.layers.70.mlp.down_proj",
|
1105 |
+
"language_model.model.layers.71.self_attn.q_proj",
|
1106 |
+
"language_model.model.layers.71.self_attn.k_proj",
|
1107 |
+
"language_model.model.layers.71.self_attn.v_proj",
|
1108 |
+
"language_model.model.layers.71.self_attn.o_proj",
|
1109 |
+
"language_model.model.layers.71.mlp.gate_proj",
|
1110 |
+
"language_model.model.layers.71.mlp.up_proj",
|
1111 |
+
"language_model.model.layers.71.mlp.down_proj",
|
1112 |
+
"language_model.model.layers.72.self_attn.q_proj",
|
1113 |
+
"language_model.model.layers.72.self_attn.k_proj",
|
1114 |
+
"language_model.model.layers.72.self_attn.v_proj",
|
1115 |
+
"language_model.model.layers.72.self_attn.o_proj",
|
1116 |
+
"language_model.model.layers.72.mlp.gate_proj",
|
1117 |
+
"language_model.model.layers.72.mlp.up_proj",
|
1118 |
+
"language_model.model.layers.72.mlp.down_proj",
|
1119 |
+
"language_model.model.layers.73.self_attn.q_proj",
|
1120 |
+
"language_model.model.layers.73.self_attn.k_proj",
|
1121 |
+
"language_model.model.layers.73.self_attn.v_proj",
|
1122 |
+
"language_model.model.layers.73.self_attn.o_proj",
|
1123 |
+
"language_model.model.layers.73.mlp.gate_proj",
|
1124 |
+
"language_model.model.layers.73.mlp.up_proj",
|
1125 |
+
"language_model.model.layers.73.mlp.down_proj",
|
1126 |
+
"language_model.model.layers.74.self_attn.q_proj",
|
1127 |
+
"language_model.model.layers.74.self_attn.k_proj",
|
1128 |
+
"language_model.model.layers.74.self_attn.v_proj",
|
1129 |
+
"language_model.model.layers.74.self_attn.o_proj",
|
1130 |
+
"language_model.model.layers.74.mlp.gate_proj",
|
1131 |
+
"language_model.model.layers.74.mlp.up_proj",
|
1132 |
+
"language_model.model.layers.74.mlp.down_proj",
|
1133 |
+
"language_model.model.layers.75.self_attn.q_proj",
|
1134 |
+
"language_model.model.layers.75.self_attn.k_proj",
|
1135 |
+
"language_model.model.layers.75.self_attn.v_proj",
|
1136 |
+
"language_model.model.layers.75.self_attn.o_proj",
|
1137 |
+
"language_model.model.layers.75.mlp.gate_proj",
|
1138 |
+
"language_model.model.layers.75.mlp.up_proj",
|
1139 |
+
"language_model.model.layers.75.mlp.down_proj",
|
1140 |
+
"language_model.model.layers.76.self_attn.q_proj",
|
1141 |
+
"language_model.model.layers.76.self_attn.k_proj",
|
1142 |
+
"language_model.model.layers.76.self_attn.v_proj",
|
1143 |
+
"language_model.model.layers.76.self_attn.o_proj",
|
1144 |
+
"language_model.model.layers.76.mlp.gate_proj",
|
1145 |
+
"language_model.model.layers.76.mlp.up_proj",
|
1146 |
+
"language_model.model.layers.76.mlp.down_proj",
|
1147 |
+
"language_model.model.layers.77.self_attn.q_proj",
|
1148 |
+
"language_model.model.layers.77.self_attn.k_proj",
|
1149 |
+
"language_model.model.layers.77.self_attn.v_proj",
|
1150 |
+
"language_model.model.layers.77.self_attn.o_proj",
|
1151 |
+
"language_model.model.layers.77.mlp.gate_proj",
|
1152 |
+
"language_model.model.layers.77.mlp.up_proj",
|
1153 |
+
"language_model.model.layers.77.mlp.down_proj",
|
1154 |
+
"language_model.model.layers.78.self_attn.q_proj",
|
1155 |
+
"language_model.model.layers.78.self_attn.k_proj",
|
1156 |
+
"language_model.model.layers.78.self_attn.v_proj",
|
1157 |
+
"language_model.model.layers.78.self_attn.o_proj",
|
1158 |
+
"language_model.model.layers.78.mlp.gate_proj",
|
1159 |
+
"language_model.model.layers.78.mlp.up_proj",
|
1160 |
+
"language_model.model.layers.78.mlp.down_proj",
|
1161 |
+
"language_model.model.layers.79.self_attn.q_proj",
|
1162 |
+
"language_model.model.layers.79.self_attn.k_proj",
|
1163 |
+
"language_model.model.layers.79.self_attn.v_proj",
|
1164 |
+
"language_model.model.layers.79.self_attn.o_proj",
|
1165 |
+
"language_model.model.layers.79.mlp.gate_proj",
|
1166 |
+
"language_model.model.layers.79.mlp.up_proj",
|
1167 |
+
"language_model.model.layers.79.mlp.down_proj",
|
1168 |
+
"language_model.model.layers.80.self_attn.q_proj",
|
1169 |
+
"language_model.model.layers.80.self_attn.k_proj",
|
1170 |
+
"language_model.model.layers.80.self_attn.v_proj",
|
1171 |
+
"language_model.model.layers.80.self_attn.o_proj",
|
1172 |
+
"language_model.model.layers.80.mlp.gate_proj",
|
1173 |
+
"language_model.model.layers.80.mlp.up_proj",
|
1174 |
+
"language_model.model.layers.80.mlp.down_proj",
|
1175 |
+
"language_model.model.layers.81.self_attn.q_proj",
|
1176 |
+
"language_model.model.layers.81.self_attn.k_proj",
|
1177 |
+
"language_model.model.layers.81.self_attn.v_proj",
|
1178 |
+
"language_model.model.layers.81.self_attn.o_proj",
|
1179 |
+
"language_model.model.layers.81.mlp.gate_proj",
|
1180 |
+
"language_model.model.layers.81.mlp.up_proj",
|
1181 |
+
"language_model.model.layers.81.mlp.down_proj",
|
1182 |
+
"language_model.model.layers.82.self_attn.q_proj",
|
1183 |
+
"language_model.model.layers.82.self_attn.k_proj",
|
1184 |
+
"language_model.model.layers.82.self_attn.v_proj",
|
1185 |
+
"language_model.model.layers.82.self_attn.o_proj",
|
1186 |
+
"language_model.model.layers.82.mlp.gate_proj",
|
1187 |
+
"language_model.model.layers.82.mlp.up_proj",
|
1188 |
+
"language_model.model.layers.82.mlp.down_proj",
|
1189 |
+
"language_model.model.layers.83.self_attn.q_proj",
|
1190 |
+
"language_model.model.layers.83.self_attn.k_proj",
|
1191 |
+
"language_model.model.layers.83.self_attn.v_proj",
|
1192 |
+
"language_model.model.layers.83.self_attn.o_proj",
|
1193 |
+
"language_model.model.layers.83.mlp.gate_proj",
|
1194 |
+
"language_model.model.layers.83.mlp.up_proj",
|
1195 |
+
"language_model.model.layers.83.mlp.down_proj",
|
1196 |
+
"language_model.model.layers.84.self_attn.q_proj",
|
1197 |
+
"language_model.model.layers.84.self_attn.k_proj",
|
1198 |
+
"language_model.model.layers.84.self_attn.v_proj",
|
1199 |
+
"language_model.model.layers.84.self_attn.o_proj",
|
1200 |
+
"language_model.model.layers.84.mlp.gate_proj",
|
1201 |
+
"language_model.model.layers.84.mlp.up_proj",
|
1202 |
+
"language_model.model.layers.84.mlp.down_proj",
|
1203 |
+
"language_model.model.layers.85.self_attn.q_proj",
|
1204 |
+
"language_model.model.layers.85.self_attn.k_proj",
|
1205 |
+
"language_model.model.layers.85.self_attn.v_proj",
|
1206 |
+
"language_model.model.layers.85.self_attn.o_proj",
|
1207 |
+
"language_model.model.layers.85.mlp.gate_proj",
|
1208 |
+
"language_model.model.layers.85.mlp.up_proj",
|
1209 |
+
"language_model.model.layers.85.mlp.down_proj",
|
1210 |
+
"language_model.model.layers.86.self_attn.q_proj",
|
1211 |
+
"language_model.model.layers.86.self_attn.k_proj",
|
1212 |
+
"language_model.model.layers.86.self_attn.v_proj",
|
1213 |
+
"language_model.model.layers.86.self_attn.o_proj",
|
1214 |
+
"language_model.model.layers.86.mlp.gate_proj",
|
1215 |
+
"language_model.model.layers.86.mlp.up_proj",
|
1216 |
+
"language_model.model.layers.86.mlp.down_proj",
|
1217 |
+
"language_model.model.layers.87.self_attn.q_proj",
|
1218 |
+
"language_model.model.layers.87.self_attn.k_proj",
|
1219 |
+
"language_model.model.layers.87.self_attn.v_proj",
|
1220 |
+
"language_model.model.layers.87.self_attn.o_proj",
|
1221 |
+
"language_model.model.layers.87.mlp.gate_proj",
|
1222 |
+
"language_model.model.layers.87.mlp.up_proj",
|
1223 |
+
"language_model.model.layers.87.mlp.down_proj",
|
1224 |
+
"language_model.lm_head"
|
1225 |
+
],
|
1226 |
+
"registry_requires_subclass": false,
|
1227 |
+
"sparsity_structure": "unstructured",
|
1228 |
+
"targets": [
|
1229 |
+
"language_model.model.layers.0.self_attn.q_proj",
|
1230 |
+
"language_model.model.layers.0.self_attn.k_proj"
|
1231 |
+
]
|
1232 |
+
}
|
1233 |
+
},
|
1234 |
+
"text_config": {
|
1235 |
+
"hidden_size": 12288,
|
1236 |
+
"intermediate_size": 28672,
|
1237 |
+
"is_composition": true,
|
1238 |
+
"max_position_embeddings": 131072,
|
1239 |
+
"model_type": "mistral",
|
1240 |
+
"norm_eps": 1e-05,
|
1241 |
+
"num_attention_heads": 96,
|
1242 |
+
"num_hidden_layers": 88,
|
1243 |
+
"num_key_value_heads": 8,
|
1244 |
+
"rms_norm_eps": 1e-05,
|
1245 |
+
"rope_theta": 1000000000.0,
|
1246 |
+
"sliding_window": null,
|
1247 |
+
"vocab_size": 32768
|
1248 |
+
},
|
1249 |
+
"torch_dtype": "bfloat16",
|
1250 |
+
"transformers_version": "4.48.3",
|
1251 |
+
"vision_config": {
|
1252 |
+
"head_dim": 88,
|
1253 |
+
"hidden_act": "silu",
|
1254 |
+
"hidden_size": 1408,
|
1255 |
+
"image_break_token_id": 14,
|
1256 |
+
"image_end_token_id": 15,
|
1257 |
+
"image_size": 1024,
|
1258 |
+
"image_token_id": 10,
|
1259 |
+
"intermediate_size": 6144,
|
1260 |
+
"model_type": "pixtral",
|
1261 |
+
"num_hidden_layers": 40,
|
1262 |
+
"patch_size": 16,
|
1263 |
+
"rope_theta": 10000.0
|
1264 |
+
},
|
1265 |
+
"vision_feature_layer": -1,
|
1266 |
+
"vision_feature_select_strategy": "full"
|
1267 |
+
}
|
generation_config.json
ADDED
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_from_model_config": true,
|
3 |
+
"bos_token_id": 1,
|
4 |
+
"eos_token_id": 2,
|
5 |
+
"transformers_version": "4.48.3"
|
6 |
+
}
|
model-00001-of-00014.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:76231892d853fe67709fa8fd9a360b4b391c5ea15fd081ac00b2996a153db2af
|
3 |
+
size 4919005520
|
model-00002-of-00014.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:be0ea06c677fc12892653e75cc641398571d917b135d737e3ccc321d54e4eec4
|
3 |
+
size 4996174888
|
model-00003-of-00014.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:1b49e034399811932bfb058524b25ad55554213b5a69cd8c1d639ba3e7bf6d88
|
3 |
+
size 4996175016
|
model-00004-of-00014.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ddc7fe1ac41ca6378a4bca1de8f2594cfb38f822c99c5ed36dac72128a6b069c
|
3 |
+
size 4996175048
|
model-00005-of-00014.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e98e430c51272fc5b2df6d3992679c66c29320cb95ce0280dc813e994c4b813b
|
3 |
+
size 4996175048
|
model-00006-of-00014.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:74263fd992d581d373a64a3d7a68a8924a2ce41d4cfc23bbc07f73777412a54a
|
3 |
+
size 4996175048
|
model-00007-of-00014.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a3ef6e584beb471deba8eea4e8026d8624f4ca92e4aa0a097ae19a9973d52a65
|
3 |
+
size 4996175048
|
model-00008-of-00014.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:9e60e6ff7c71a13980790a8da1bff67a0072cdcca803af3f3b3e2e261493cd5b
|
3 |
+
size 4996175048
|
model-00009-of-00014.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:cd03b144c487ec5d048ee56962f936be5e6923df6640d61f20bda5bbcca744a7
|
3 |
+
size 4996175048
|
model-00010-of-00014.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e9a0def27fad589a78d451098e1f4beb1555864a847645b0fd8ca9f76be31f6e
|
3 |
+
size 4996175048
|
model-00011-of-00014.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f4493a1a7f373728a7c0cd87f7d8ae08577eafbdc64801b72a0a1c2db946f7f9
|
3 |
+
size 4996175048
|
model-00012-of-00014.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:efb4813ae0945a8389fe36eeddc57bcc9124a174b7d37c088b740981f91804e1
|
3 |
+
size 4996175048
|
model-00013-of-00014.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:370aca4404ba1205a0d3ce4e4afa98f253cbcfe33740511db32cea1e787d14a9
|
3 |
+
size 4996175048
|
model-00014-of-00014.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:0198b24bd9b4851880fdd74224c9849e104dac7d5aaf3ac06d44fe2ff7fc7806
|
3 |
+
size 2596191424
|
model.safetensors.index.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
preprocessor_config.json
ADDED
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"do_convert_rgb": true,
|
3 |
+
"do_normalize": true,
|
4 |
+
"do_rescale": true,
|
5 |
+
"do_resize": true,
|
6 |
+
"image_mean": [
|
7 |
+
0.48145466,
|
8 |
+
0.4578275,
|
9 |
+
0.40821073
|
10 |
+
],
|
11 |
+
"image_processor_type": "PixtralImageProcessor",
|
12 |
+
"image_std": [
|
13 |
+
0.26862954,
|
14 |
+
0.26130258,
|
15 |
+
0.27577711
|
16 |
+
],
|
17 |
+
"patch_size": {
|
18 |
+
"height": 16,
|
19 |
+
"width": 16
|
20 |
+
},
|
21 |
+
"processor_class": "PixtralProcessor",
|
22 |
+
"resample": 3,
|
23 |
+
"rescale_factor": 0.00392156862745098,
|
24 |
+
"size": {
|
25 |
+
"longest_edge": 1024
|
26 |
+
}
|
27 |
+
}
|
processor_config.json
ADDED
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"image_break_token": "[IMG_BREAK]",
|
3 |
+
"image_end_token": "[IMG_END]",
|
4 |
+
"image_token": "[IMG]",
|
5 |
+
"patch_size": 16,
|
6 |
+
"processor_class": "PixtralProcessor"
|
7 |
+
}
|
recipe.yaml
ADDED
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
DEFAULT_stage:
|
2 |
+
DEFAULT_modifiers:
|
3 |
+
GPTQModifier:
|
4 |
+
sequential_targets: [MistralDecoderLayer]
|
5 |
+
dampening_frac: 0.01
|
6 |
+
config_groups:
|
7 |
+
config_group:
|
8 |
+
targets: [Linear]
|
9 |
+
weights: {num_bits: 4, type: int, symmetric: true, group_size: 128, strategy: group,
|
10 |
+
dynamic: false, actorder: weight, observer: minmax}
|
11 |
+
targets: Linear
|
12 |
+
ignore: ['re:.*lm_head', 're:vision_tower.*', 're:multi_modal_projector.*']
|
special_tokens_map.json
ADDED
@@ -0,0 +1,773 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"additional_special_tokens": [
|
3 |
+
"<s>",
|
4 |
+
"</s>",
|
5 |
+
"[INST]",
|
6 |
+
"[/INST]",
|
7 |
+
"[TOOL_CALLS]",
|
8 |
+
"[AVAILABLE_TOOLS]",
|
9 |
+
"[/AVAILABLE_TOOLS]",
|
10 |
+
"[TOOL_RESULTS]",
|
11 |
+
"[/TOOL_RESULTS]",
|
12 |
+
"[IMG]",
|
13 |
+
"[PREFIX]",
|
14 |
+
"[MIDDLE]",
|
15 |
+
"[SUFFIX]",
|
16 |
+
"[IMG_BREAK]",
|
17 |
+
"[IMG_END]",
|
18 |
+
"[SYSTEM_PROMPT]",
|
19 |
+
"[/SYSTEM_PROMPT]",
|
20 |
+
"[TOOL_CONTENT]",
|
21 |
+
"[control_17]",
|
22 |
+
"[control_18]",
|
23 |
+
"[control_19]",
|
24 |
+
"[control_20]",
|
25 |
+
"[control_21]",
|
26 |
+
"[control_22]",
|
27 |
+
"[control_23]",
|
28 |
+
"[control_24]",
|
29 |
+
"[control_25]",
|
30 |
+
"[control_26]",
|
31 |
+
"[control_27]",
|
32 |
+
"[control_28]",
|
33 |
+
"[control_29]",
|
34 |
+
"[control_30]",
|
35 |
+
"[control_31]",
|
36 |
+
"[control_32]",
|
37 |
+
"[control_33]",
|
38 |
+
"[control_34]",
|
39 |
+
"[control_35]",
|
40 |
+
"[control_36]",
|
41 |
+
"[control_37]",
|
42 |
+
"[control_38]",
|
43 |
+
"[control_39]",
|
44 |
+
"[control_40]",
|
45 |
+
"[control_41]",
|
46 |
+
"[control_42]",
|
47 |
+
"[control_43]",
|
48 |
+
"[control_44]",
|
49 |
+
"[control_45]",
|
50 |
+
"[control_46]",
|
51 |
+
"[control_47]",
|
52 |
+
"[control_48]",
|
53 |
+
"[control_49]",
|
54 |
+
"[control_50]",
|
55 |
+
"[control_51]",
|
56 |
+
"[control_52]",
|
57 |
+
"[control_53]",
|
58 |
+
"[control_54]",
|
59 |
+
"[control_55]",
|
60 |
+
"[control_56]",
|
61 |
+
"[control_57]",
|
62 |
+
"[control_58]",
|
63 |
+
"[control_59]",
|
64 |
+
"[control_60]",
|
65 |
+
"[control_61]",
|
66 |
+
"[control_62]",
|
67 |
+
"[control_63]",
|
68 |
+
"[control_64]",
|
69 |
+
"[control_65]",
|
70 |
+
"[control_66]",
|
71 |
+
"[control_67]",
|
72 |
+
"[control_68]",
|
73 |
+
"[control_69]",
|
74 |
+
"[control_70]",
|
75 |
+
"[control_71]",
|
76 |
+
"[control_72]",
|
77 |
+
"[control_73]",
|
78 |
+
"[control_74]",
|
79 |
+
"[control_75]",
|
80 |
+
"[control_76]",
|
81 |
+
"[control_77]",
|
82 |
+
"[control_78]",
|
83 |
+
"[control_79]",
|
84 |
+
"[control_80]",
|
85 |
+
"[control_81]",
|
86 |
+
"[control_82]",
|
87 |
+
"[control_83]",
|
88 |
+
"[control_84]",
|
89 |
+
"[control_85]",
|
90 |
+
"[control_86]",
|
91 |
+
"[control_87]",
|
92 |
+
"[control_88]",
|
93 |
+
"[control_89]",
|
94 |
+
"[control_90]",
|
95 |
+
"[control_91]",
|
96 |
+
"[control_92]",
|
97 |
+
"[control_93]",
|
98 |
+
"[control_94]",
|
99 |
+
"[control_95]",
|
100 |
+
"[control_96]",
|
101 |
+
"[control_97]",
|
102 |
+
"[control_98]",
|
103 |
+
"[control_99]",
|
104 |
+
"[control_100]",
|
105 |
+
"[control_101]",
|
106 |
+
"[control_102]",
|
107 |
+
"[control_103]",
|
108 |
+
"[control_104]",
|
109 |
+
"[control_105]",
|
110 |
+
"[control_106]",
|
111 |
+
"[control_107]",
|
112 |
+
"[control_108]",
|
113 |
+
"[control_109]",
|
114 |
+
"[control_110]",
|
115 |
+
"[control_111]",
|
116 |
+
"[control_112]",
|
117 |
+
"[control_113]",
|
118 |
+
"[control_114]",
|
119 |
+
"[control_115]",
|
120 |
+
"[control_116]",
|
121 |
+
"[control_117]",
|
122 |
+
"[control_118]",
|
123 |
+
"[control_119]",
|
124 |
+
"[control_120]",
|
125 |
+
"[control_121]",
|
126 |
+
"[control_122]",
|
127 |
+
"[control_123]",
|
128 |
+
"[control_124]",
|
129 |
+
"[control_125]",
|
130 |
+
"[control_126]",
|
131 |
+
"[control_127]",
|
132 |
+
"[control_128]",
|
133 |
+
"[control_129]",
|
134 |
+
"[control_130]",
|
135 |
+
"[control_131]",
|
136 |
+
"[control_132]",
|
137 |
+
"[control_133]",
|
138 |
+
"[control_134]",
|
139 |
+
"[control_135]",
|
140 |
+
"[control_136]",
|
141 |
+
"[control_137]",
|
142 |
+
"[control_138]",
|
143 |
+
"[control_139]",
|
144 |
+
"[control_140]",
|
145 |
+
"[control_141]",
|
146 |
+
"[control_142]",
|
147 |
+
"[control_143]",
|
148 |
+
"[control_144]",
|
149 |
+
"[control_145]",
|
150 |
+
"[control_146]",
|
151 |
+
"[control_147]",
|
152 |
+
"[control_148]",
|
153 |
+
"[control_149]",
|
154 |
+
"[control_150]",
|
155 |
+
"[control_151]",
|
156 |
+
"[control_152]",
|
157 |
+
"[control_153]",
|
158 |
+
"[control_154]",
|
159 |
+
"[control_155]",
|
160 |
+
"[control_156]",
|
161 |
+
"[control_157]",
|
162 |
+
"[control_158]",
|
163 |
+
"[control_159]",
|
164 |
+
"[control_160]",
|
165 |
+
"[control_161]",
|
166 |
+
"[control_162]",
|
167 |
+
"[control_163]",
|
168 |
+
"[control_164]",
|
169 |
+
"[control_165]",
|
170 |
+
"[control_166]",
|
171 |
+
"[control_167]",
|
172 |
+
"[control_168]",
|
173 |
+
"[control_169]",
|
174 |
+
"[control_170]",
|
175 |
+
"[control_171]",
|
176 |
+
"[control_172]",
|
177 |
+
"[control_173]",
|
178 |
+
"[control_174]",
|
179 |
+
"[control_175]",
|
180 |
+
"[control_176]",
|
181 |
+
"[control_177]",
|
182 |
+
"[control_178]",
|
183 |
+
"[control_179]",
|
184 |
+
"[control_180]",
|
185 |
+
"[control_181]",
|
186 |
+
"[control_182]",
|
187 |
+
"[control_183]",
|
188 |
+
"[control_184]",
|
189 |
+
"[control_185]",
|
190 |
+
"[control_186]",
|
191 |
+
"[control_187]",
|
192 |
+
"[control_188]",
|
193 |
+
"[control_189]",
|
194 |
+
"[control_190]",
|
195 |
+
"[control_191]",
|
196 |
+
"[control_192]",
|
197 |
+
"[control_193]",
|
198 |
+
"[control_194]",
|
199 |
+
"[control_195]",
|
200 |
+
"[control_196]",
|
201 |
+
"[control_197]",
|
202 |
+
"[control_198]",
|
203 |
+
"[control_199]",
|
204 |
+
"[control_200]",
|
205 |
+
"[control_201]",
|
206 |
+
"[control_202]",
|
207 |
+
"[control_203]",
|
208 |
+
"[control_204]",
|
209 |
+
"[control_205]",
|
210 |
+
"[control_206]",
|
211 |
+
"[control_207]",
|
212 |
+
"[control_208]",
|
213 |
+
"[control_209]",
|
214 |
+
"[control_210]",
|
215 |
+
"[control_211]",
|
216 |
+
"[control_212]",
|
217 |
+
"[control_213]",
|
218 |
+
"[control_214]",
|
219 |
+
"[control_215]",
|
220 |
+
"[control_216]",
|
221 |
+
"[control_217]",
|
222 |
+
"[control_218]",
|
223 |
+
"[control_219]",
|
224 |
+
"[control_220]",
|
225 |
+
"[control_221]",
|
226 |
+
"[control_222]",
|
227 |
+
"[control_223]",
|
228 |
+
"[control_224]",
|
229 |
+
"[control_225]",
|
230 |
+
"[control_226]",
|
231 |
+
"[control_227]",
|
232 |
+
"[control_228]",
|
233 |
+
"[control_229]",
|
234 |
+
"[control_230]",
|
235 |
+
"[control_231]",
|
236 |
+
"[control_232]",
|
237 |
+
"[control_233]",
|
238 |
+
"[control_234]",
|
239 |
+
"[control_235]",
|
240 |
+
"[control_236]",
|
241 |
+
"[control_237]",
|
242 |
+
"[control_238]",
|
243 |
+
"[control_239]",
|
244 |
+
"[control_240]",
|
245 |
+
"[control_241]",
|
246 |
+
"[control_242]",
|
247 |
+
"[control_243]",
|
248 |
+
"[control_244]",
|
249 |
+
"[control_245]",
|
250 |
+
"[control_246]",
|
251 |
+
"[control_247]",
|
252 |
+
"[control_248]",
|
253 |
+
"[control_249]",
|
254 |
+
"[control_250]",
|
255 |
+
"[control_251]",
|
256 |
+
"[control_252]",
|
257 |
+
"[control_253]",
|
258 |
+
"[control_254]",
|
259 |
+
"[control_255]",
|
260 |
+
"[control_256]",
|
261 |
+
"[control_257]",
|
262 |
+
"[control_258]",
|
263 |
+
"[control_259]",
|
264 |
+
"[control_260]",
|
265 |
+
"[control_261]",
|
266 |
+
"[control_262]",
|
267 |
+
"[control_263]",
|
268 |
+
"[control_264]",
|
269 |
+
"[control_265]",
|
270 |
+
"[control_266]",
|
271 |
+
"[control_267]",
|
272 |
+
"[control_268]",
|
273 |
+
"[control_269]",
|
274 |
+
"[control_270]",
|
275 |
+
"[control_271]",
|
276 |
+
"[control_272]",
|
277 |
+
"[control_273]",
|
278 |
+
"[control_274]",
|
279 |
+
"[control_275]",
|
280 |
+
"[control_276]",
|
281 |
+
"[control_277]",
|
282 |
+
"[control_278]",
|
283 |
+
"[control_279]",
|
284 |
+
"[control_280]",
|
285 |
+
"[control_281]",
|
286 |
+
"[control_282]",
|
287 |
+
"[control_283]",
|
288 |
+
"[control_284]",
|
289 |
+
"[control_285]",
|
290 |
+
"[control_286]",
|
291 |
+
"[control_287]",
|
292 |
+
"[control_288]",
|
293 |
+
"[control_289]",
|
294 |
+
"[control_290]",
|
295 |
+
"[control_291]",
|
296 |
+
"[control_292]",
|
297 |
+
"[control_293]",
|
298 |
+
"[control_294]",
|
299 |
+
"[control_295]",
|
300 |
+
"[control_296]",
|
301 |
+
"[control_297]",
|
302 |
+
"[control_298]",
|
303 |
+
"[control_299]",
|
304 |
+
"[control_300]",
|
305 |
+
"[control_301]",
|
306 |
+
"[control_302]",
|
307 |
+
"[control_303]",
|
308 |
+
"[control_304]",
|
309 |
+
"[control_305]",
|
310 |
+
"[control_306]",
|
311 |
+
"[control_307]",
|
312 |
+
"[control_308]",
|
313 |
+
"[control_309]",
|
314 |
+
"[control_310]",
|
315 |
+
"[control_311]",
|
316 |
+
"[control_312]",
|
317 |
+
"[control_313]",
|
318 |
+
"[control_314]",
|
319 |
+
"[control_315]",
|
320 |
+
"[control_316]",
|
321 |
+
"[control_317]",
|
322 |
+
"[control_318]",
|
323 |
+
"[control_319]",
|
324 |
+
"[control_320]",
|
325 |
+
"[control_321]",
|
326 |
+
"[control_322]",
|
327 |
+
"[control_323]",
|
328 |
+
"[control_324]",
|
329 |
+
"[control_325]",
|
330 |
+
"[control_326]",
|
331 |
+
"[control_327]",
|
332 |
+
"[control_328]",
|
333 |
+
"[control_329]",
|
334 |
+
"[control_330]",
|
335 |
+
"[control_331]",
|
336 |
+
"[control_332]",
|
337 |
+
"[control_333]",
|
338 |
+
"[control_334]",
|
339 |
+
"[control_335]",
|
340 |
+
"[control_336]",
|
341 |
+
"[control_337]",
|
342 |
+
"[control_338]",
|
343 |
+
"[control_339]",
|
344 |
+
"[control_340]",
|
345 |
+
"[control_341]",
|
346 |
+
"[control_342]",
|
347 |
+
"[control_343]",
|
348 |
+
"[control_344]",
|
349 |
+
"[control_345]",
|
350 |
+
"[control_346]",
|
351 |
+
"[control_347]",
|
352 |
+
"[control_348]",
|
353 |
+
"[control_349]",
|
354 |
+
"[control_350]",
|
355 |
+
"[control_351]",
|
356 |
+
"[control_352]",
|
357 |
+
"[control_353]",
|
358 |
+
"[control_354]",
|
359 |
+
"[control_355]",
|
360 |
+
"[control_356]",
|
361 |
+
"[control_357]",
|
362 |
+
"[control_358]",
|
363 |
+
"[control_359]",
|
364 |
+
"[control_360]",
|
365 |
+
"[control_361]",
|
366 |
+
"[control_362]",
|
367 |
+
"[control_363]",
|
368 |
+
"[control_364]",
|
369 |
+
"[control_365]",
|
370 |
+
"[control_366]",
|
371 |
+
"[control_367]",
|
372 |
+
"[control_368]",
|
373 |
+
"[control_369]",
|
374 |
+
"[control_370]",
|
375 |
+
"[control_371]",
|
376 |
+
"[control_372]",
|
377 |
+
"[control_373]",
|
378 |
+
"[control_374]",
|
379 |
+
"[control_375]",
|
380 |
+
"[control_376]",
|
381 |
+
"[control_377]",
|
382 |
+
"[control_378]",
|
383 |
+
"[control_379]",
|
384 |
+
"[control_380]",
|
385 |
+
"[control_381]",
|
386 |
+
"[control_382]",
|
387 |
+
"[control_383]",
|
388 |
+
"[control_384]",
|
389 |
+
"[control_385]",
|
390 |
+
"[control_386]",
|
391 |
+
"[control_387]",
|
392 |
+
"[control_388]",
|
393 |
+
"[control_389]",
|
394 |
+
"[control_390]",
|
395 |
+
"[control_391]",
|
396 |
+
"[control_392]",
|
397 |
+
"[control_393]",
|
398 |
+
"[control_394]",
|
399 |
+
"[control_395]",
|
400 |
+
"[control_396]",
|
401 |
+
"[control_397]",
|
402 |
+
"[control_398]",
|
403 |
+
"[control_399]",
|
404 |
+
"[control_400]",
|
405 |
+
"[control_401]",
|
406 |
+
"[control_402]",
|
407 |
+
"[control_403]",
|
408 |
+
"[control_404]",
|
409 |
+
"[control_405]",
|
410 |
+
"[control_406]",
|
411 |
+
"[control_407]",
|
412 |
+
"[control_408]",
|
413 |
+
"[control_409]",
|
414 |
+
"[control_410]",
|
415 |
+
"[control_411]",
|
416 |
+
"[control_412]",
|
417 |
+
"[control_413]",
|
418 |
+
"[control_414]",
|
419 |
+
"[control_415]",
|
420 |
+
"[control_416]",
|
421 |
+
"[control_417]",
|
422 |
+
"[control_418]",
|
423 |
+
"[control_419]",
|
424 |
+
"[control_420]",
|
425 |
+
"[control_421]",
|
426 |
+
"[control_422]",
|
427 |
+
"[control_423]",
|
428 |
+
"[control_424]",
|
429 |
+
"[control_425]",
|
430 |
+
"[control_426]",
|
431 |
+
"[control_427]",
|
432 |
+
"[control_428]",
|
433 |
+
"[control_429]",
|
434 |
+
"[control_430]",
|
435 |
+
"[control_431]",
|
436 |
+
"[control_432]",
|
437 |
+
"[control_433]",
|
438 |
+
"[control_434]",
|
439 |
+
"[control_435]",
|
440 |
+
"[control_436]",
|
441 |
+
"[control_437]",
|
442 |
+
"[control_438]",
|
443 |
+
"[control_439]",
|
444 |
+
"[control_440]",
|
445 |
+
"[control_441]",
|
446 |
+
"[control_442]",
|
447 |
+
"[control_443]",
|
448 |
+
"[control_444]",
|
449 |
+
"[control_445]",
|
450 |
+
"[control_446]",
|
451 |
+
"[control_447]",
|
452 |
+
"[control_448]",
|
453 |
+
"[control_449]",
|
454 |
+
"[control_450]",
|
455 |
+
"[control_451]",
|
456 |
+
"[control_452]",
|
457 |
+
"[control_453]",
|
458 |
+
"[control_454]",
|
459 |
+
"[control_455]",
|
460 |
+
"[control_456]",
|
461 |
+
"[control_457]",
|
462 |
+
"[control_458]",
|
463 |
+
"[control_459]",
|
464 |
+
"[control_460]",
|
465 |
+
"[control_461]",
|
466 |
+
"[control_462]",
|
467 |
+
"[control_463]",
|
468 |
+
"[control_464]",
|
469 |
+
"[control_465]",
|
470 |
+
"[control_466]",
|
471 |
+
"[control_467]",
|
472 |
+
"[control_468]",
|
473 |
+
"[control_469]",
|
474 |
+
"[control_470]",
|
475 |
+
"[control_471]",
|
476 |
+
"[control_472]",
|
477 |
+
"[control_473]",
|
478 |
+
"[control_474]",
|
479 |
+
"[control_475]",
|
480 |
+
"[control_476]",
|
481 |
+
"[control_477]",
|
482 |
+
"[control_478]",
|
483 |
+
"[control_479]",
|
484 |
+
"[control_480]",
|
485 |
+
"[control_481]",
|
486 |
+
"[control_482]",
|
487 |
+
"[control_483]",
|
488 |
+
"[control_484]",
|
489 |
+
"[control_485]",
|
490 |
+
"[control_486]",
|
491 |
+
"[control_487]",
|
492 |
+
"[control_488]",
|
493 |
+
"[control_489]",
|
494 |
+
"[control_490]",
|
495 |
+
"[control_491]",
|
496 |
+
"[control_492]",
|
497 |
+
"[control_493]",
|
498 |
+
"[control_494]",
|
499 |
+
"[control_495]",
|
500 |
+
"[control_496]",
|
501 |
+
"[control_497]",
|
502 |
+
"[control_498]",
|
503 |
+
"[control_499]",
|
504 |
+
"[control_500]",
|
505 |
+
"[control_501]",
|
506 |
+
"[control_502]",
|
507 |
+
"[control_503]",
|
508 |
+
"[control_504]",
|
509 |
+
"[control_505]",
|
510 |
+
"[control_506]",
|
511 |
+
"[control_507]",
|
512 |
+
"[control_508]",
|
513 |
+
"[control_509]",
|
514 |
+
"[control_510]",
|
515 |
+
"[control_511]",
|
516 |
+
"[control_512]",
|
517 |
+
"[control_513]",
|
518 |
+
"[control_514]",
|
519 |
+
"[control_515]",
|
520 |
+
"[control_516]",
|
521 |
+
"[control_517]",
|
522 |
+
"[control_518]",
|
523 |
+
"[control_519]",
|
524 |
+
"[control_520]",
|
525 |
+
"[control_521]",
|
526 |
+
"[control_522]",
|
527 |
+
"[control_523]",
|
528 |
+
"[control_524]",
|
529 |
+
"[control_525]",
|
530 |
+
"[control_526]",
|
531 |
+
"[control_527]",
|
532 |
+
"[control_528]",
|
533 |
+
"[control_529]",
|
534 |
+
"[control_530]",
|
535 |
+
"[control_531]",
|
536 |
+
"[control_532]",
|
537 |
+
"[control_533]",
|
538 |
+
"[control_534]",
|
539 |
+
"[control_535]",
|
540 |
+
"[control_536]",
|
541 |
+
"[control_537]",
|
542 |
+
"[control_538]",
|
543 |
+
"[control_539]",
|
544 |
+
"[control_540]",
|
545 |
+
"[control_541]",
|
546 |
+
"[control_542]",
|
547 |
+
"[control_543]",
|
548 |
+
"[control_544]",
|
549 |
+
"[control_545]",
|
550 |
+
"[control_546]",
|
551 |
+
"[control_547]",
|
552 |
+
"[control_548]",
|
553 |
+
"[control_549]",
|
554 |
+
"[control_550]",
|
555 |
+
"[control_551]",
|
556 |
+
"[control_552]",
|
557 |
+
"[control_553]",
|
558 |
+
"[control_554]",
|
559 |
+
"[control_555]",
|
560 |
+
"[control_556]",
|
561 |
+
"[control_557]",
|
562 |
+
"[control_558]",
|
563 |
+
"[control_559]",
|
564 |
+
"[control_560]",
|
565 |
+
"[control_561]",
|
566 |
+
"[control_562]",
|
567 |
+
"[control_563]",
|
568 |
+
"[control_564]",
|
569 |
+
"[control_565]",
|
570 |
+
"[control_566]",
|
571 |
+
"[control_567]",
|
572 |
+
"[control_568]",
|
573 |
+
"[control_569]",
|
574 |
+
"[control_570]",
|
575 |
+
"[control_571]",
|
576 |
+
"[control_572]",
|
577 |
+
"[control_573]",
|
578 |
+
"[control_574]",
|
579 |
+
"[control_575]",
|
580 |
+
"[control_576]",
|
581 |
+
"[control_577]",
|
582 |
+
"[control_578]",
|
583 |
+
"[control_579]",
|
584 |
+
"[control_580]",
|
585 |
+
"[control_581]",
|
586 |
+
"[control_582]",
|
587 |
+
"[control_583]",
|
588 |
+
"[control_584]",
|
589 |
+
"[control_585]",
|
590 |
+
"[control_586]",
|
591 |
+
"[control_587]",
|
592 |
+
"[control_588]",
|
593 |
+
"[control_589]",
|
594 |
+
"[control_590]",
|
595 |
+
"[control_591]",
|
596 |
+
"[control_592]",
|
597 |
+
"[control_593]",
|
598 |
+
"[control_594]",
|
599 |
+
"[control_595]",
|
600 |
+
"[control_596]",
|
601 |
+
"[control_597]",
|
602 |
+
"[control_598]",
|
603 |
+
"[control_599]",
|
604 |
+
"[control_600]",
|
605 |
+
"[control_601]",
|
606 |
+
"[control_602]",
|
607 |
+
"[control_603]",
|
608 |
+
"[control_604]",
|
609 |
+
"[control_605]",
|
610 |
+
"[control_606]",
|
611 |
+
"[control_607]",
|
612 |
+
"[control_608]",
|
613 |
+
"[control_609]",
|
614 |
+
"[control_610]",
|
615 |
+
"[control_611]",
|
616 |
+
"[control_612]",
|
617 |
+
"[control_613]",
|
618 |
+
"[control_614]",
|
619 |
+
"[control_615]",
|
620 |
+
"[control_616]",
|
621 |
+
"[control_617]",
|
622 |
+
"[control_618]",
|
623 |
+
"[control_619]",
|
624 |
+
"[control_620]",
|
625 |
+
"[control_621]",
|
626 |
+
"[control_622]",
|
627 |
+
"[control_623]",
|
628 |
+
"[control_624]",
|
629 |
+
"[control_625]",
|
630 |
+
"[control_626]",
|
631 |
+
"[control_627]",
|
632 |
+
"[control_628]",
|
633 |
+
"[control_629]",
|
634 |
+
"[control_630]",
|
635 |
+
"[control_631]",
|
636 |
+
"[control_632]",
|
637 |
+
"[control_633]",
|
638 |
+
"[control_634]",
|
639 |
+
"[control_635]",
|
640 |
+
"[control_636]",
|
641 |
+
"[control_637]",
|
642 |
+
"[control_638]",
|
643 |
+
"[control_639]",
|
644 |
+
"[control_640]",
|
645 |
+
"[control_641]",
|
646 |
+
"[control_642]",
|
647 |
+
"[control_643]",
|
648 |
+
"[control_644]",
|
649 |
+
"[control_645]",
|
650 |
+
"[control_646]",
|
651 |
+
"[control_647]",
|
652 |
+
"[control_648]",
|
653 |
+
"[control_649]",
|
654 |
+
"[control_650]",
|
655 |
+
"[control_651]",
|
656 |
+
"[control_652]",
|
657 |
+
"[control_653]",
|
658 |
+
"[control_654]",
|
659 |
+
"[control_655]",
|
660 |
+
"[control_656]",
|
661 |
+
"[control_657]",
|
662 |
+
"[control_658]",
|
663 |
+
"[control_659]",
|
664 |
+
"[control_660]",
|
665 |
+
"[control_661]",
|
666 |
+
"[control_662]",
|
667 |
+
"[control_663]",
|
668 |
+
"[control_664]",
|
669 |
+
"[control_665]",
|
670 |
+
"[control_666]",
|
671 |
+
"[control_667]",
|
672 |
+
"[control_668]",
|
673 |
+
"[control_669]",
|
674 |
+
"[control_670]",
|
675 |
+
"[control_671]",
|
676 |
+
"[control_672]",
|
677 |
+
"[control_673]",
|
678 |
+
"[control_674]",
|
679 |
+
"[control_675]",
|
680 |
+
"[control_676]",
|
681 |
+
"[control_677]",
|
682 |
+
"[control_678]",
|
683 |
+
"[control_679]",
|
684 |
+
"[control_680]",
|
685 |
+
"[control_681]",
|
686 |
+
"[control_682]",
|
687 |
+
"[control_683]",
|
688 |
+
"[control_684]",
|
689 |
+
"[control_685]",
|
690 |
+
"[control_686]",
|
691 |
+
"[control_687]",
|
692 |
+
"[control_688]",
|
693 |
+
"[control_689]",
|
694 |
+
"[control_690]",
|
695 |
+
"[control_691]",
|
696 |
+
"[control_692]",
|
697 |
+
"[control_693]",
|
698 |
+
"[control_694]",
|
699 |
+
"[control_695]",
|
700 |
+
"[control_696]",
|
701 |
+
"[control_697]",
|
702 |
+
"[control_698]",
|
703 |
+
"[control_699]",
|
704 |
+
"[control_700]",
|
705 |
+
"[control_701]",
|
706 |
+
"[control_702]",
|
707 |
+
"[control_703]",
|
708 |
+
"[control_704]",
|
709 |
+
"[control_705]",
|
710 |
+
"[control_706]",
|
711 |
+
"[control_707]",
|
712 |
+
"[control_708]",
|
713 |
+
"[control_709]",
|
714 |
+
"[control_710]",
|
715 |
+
"[control_711]",
|
716 |
+
"[control_712]",
|
717 |
+
"[control_713]",
|
718 |
+
"[control_714]",
|
719 |
+
"[control_715]",
|
720 |
+
"[control_716]",
|
721 |
+
"[control_717]",
|
722 |
+
"[control_718]",
|
723 |
+
"[control_719]",
|
724 |
+
"[control_720]",
|
725 |
+
"[control_721]",
|
726 |
+
"[control_722]",
|
727 |
+
"[control_723]",
|
728 |
+
"[control_724]",
|
729 |
+
"[control_725]",
|
730 |
+
"[control_726]",
|
731 |
+
"[control_727]",
|
732 |
+
"[control_728]",
|
733 |
+
"[control_729]",
|
734 |
+
"[control_730]",
|
735 |
+
"[control_731]",
|
736 |
+
"[control_732]",
|
737 |
+
"[control_733]",
|
738 |
+
"[control_734]",
|
739 |
+
"[control_735]",
|
740 |
+
"[control_736]",
|
741 |
+
"[control_737]",
|
742 |
+
"[control_738]",
|
743 |
+
"[control_739]",
|
744 |
+
"[control_740]",
|
745 |
+
"[control_741]",
|
746 |
+
"[control_742]",
|
747 |
+
"[control_743]",
|
748 |
+
"[control_744]",
|
749 |
+
"[control_745]",
|
750 |
+
"[control_746]"
|
751 |
+
],
|
752 |
+
"bos_token": {
|
753 |
+
"content": "<s>",
|
754 |
+
"lstrip": false,
|
755 |
+
"normalized": false,
|
756 |
+
"rstrip": false,
|
757 |
+
"single_word": false
|
758 |
+
},
|
759 |
+
"eos_token": {
|
760 |
+
"content": "</s>",
|
761 |
+
"lstrip": false,
|
762 |
+
"normalized": false,
|
763 |
+
"rstrip": false,
|
764 |
+
"single_word": false
|
765 |
+
},
|
766 |
+
"unk_token": {
|
767 |
+
"content": "<unk>",
|
768 |
+
"lstrip": false,
|
769 |
+
"normalized": false,
|
770 |
+
"rstrip": false,
|
771 |
+
"single_word": false
|
772 |
+
}
|
773 |
+
}
|
tokenizer.json
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:93346fbb7e676ad3c177ae3ed72700dd93ffeb380d9b083b58227cb0ea4a75da
|
3 |
+
size 3672259
|
tokenizer.model
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:1b968b8dc352f42192367337c78ccc61e1eaddc6d641a579372d4f20694beb7a
|
3 |
+
size 587562
|
tokenizer_config.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|