Dreano
commited on
Upload folder using huggingface_hub
Browse files- chat_template.json +1 -1
- model-00001-of-00002.safetensors +1 -1
- model-00002-of-00002.safetensors +1 -1
- preprocessor_config.json +2 -2
- quant_log.csv +196 -196
- tokenizer.json +2 -2
- tokenizer_config.json +3 -3
chat_template.json
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
{
|
| 2 |
-
"chat_template": "{% set image_placeholder = '<|vision_start|><|image_pad|><|vision_end|>' %}\n{% for message in messages %}\n {#--- Handle User Messages with Template and Examples ---#}\n {%- if message['role'] == 'user' and template -%}\n {% if loop.first and message['role'] != 'system' %}\n {{- '<|im_start|>system
|
| 3 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"chat_template": "{% set image_placeholder = '<|vision_start|><|image_pad|><|vision_end|>' %}\n{% for message in messages %}\n {#--- Handle User Messages with Template and Examples ---#}\n {%- if message['role'] == 'user' and template -%}\n {% if loop.first and message['role'] != 'system' %}\n {{- '<|im_start|>system\nYou are NuExtract, an information extraction tool created by NuMind.<|im_end|>' }}\n {% endif %}\n \n {{- '<|im_start|>' + message['role'] -}}\n \n {#--- Template Section ---#}\n {{ '\n# Template:' }}\n {{- '\n' + template + '\n' }}\n \n {#--- Examples Section (if provided) ---#}\n {% if examples -%}\n {{- '# Examples:' }}\n {% for example in examples %}\n {{- '## Input:\n' }}\n {#--- Handle image examples ---#}\n {% if example['input'] is mapping and example['input']['type'] == 'image' %}\n {{- image_placeholder | trim -}}\n {% elif example['input'] == '<image>' %}\n {{- image_placeholder | trim -}}\n {% else %}\n {{- example['input'] -}}\n {% endif %}\n {{- '\n## Output:\n' ~ example['output'] }}\n {% endfor %}\n {%- endif %}\n \n {#--- Context Section: Handle various content types ---#}\n {{- '# Context:\n' }}\n {%- if message['content'] is string -%}\n {#--- Simple string content ---#}\n {{- message['content'] | trim -}}\n {%- elif message['content'] is mapping and message['content']['type'] == 'image' -%}\n {#--- Single image document ---#}\n {{- image_placeholder | trim -}}\n {%- else -%}\n {#--- List of content items (mixed text/images) ---#}\n {#--- First, determine what the actual input content is (not ICL images) ---#}\n {%- set ns = namespace(has_text_input=false, text_content='') -%}\n \n {#--- Count content types and identify actual input document ---#}\n {%- for content in message['content'] -%}\n {%- if content is mapping and content.get('type') == 'text' -%}\n {%- if content.get('text') != '<image>' -%}\n {%- set ns.has_text_input = true -%}\n {%- set ns.text_content = content['text'] -%}\n {%- endif -%}\n {%- elif content is string -%}\n {%- if content != '<image>' -%}\n {%- set ns.has_text_input = true -%}\n {%- set ns.text_content = content -%}\n {%- endif -%}\n {%- endif -%}\n {%- endfor -%}\n \n {#--- Determine what to output based on actual input type ---#}\n {%- if ns.has_text_input -%}\n {#--- Main input is text, so output the text content ---#}\n {{- ns.text_content | trim -}}\n {%- else -%}\n {#--- Main input is image or <image> placeholder ---#}\n {%- set ns2 = namespace(found_image=false) -%}\n {%- for content in message['content'] -%}\n {%- if content is mapping and content.get('type') == 'image' and not ns2.found_image -%}\n {{- image_placeholder | trim -}}\n {%- set ns2.found_image = true -%}\n {%- elif content is mapping and content.get('type') == 'text' and content.get('text') == '<image>' and not ns2.found_image -%}\n {{- image_placeholder | trim -}}\n {%- set ns2.found_image = true -%}\n {%- elif content is string and content == '<image>' and not ns2.found_image -%}\n {{- image_placeholder | trim -}}\n {%- set ns2.found_image = true -%}\n {%- endif -%}\n {%- endfor -%}\n {%- endif -%}\n {%- endif -%}\n {{- '<|im_end|>\n'}}\n \n {#--- Handle All Other Messages (Assistant, System, etc.) ---#}\n {% else %}\n {% if loop.first and message['role'] != 'system' %}\n {{- '<|im_start|>system\nYou are a helpful assistant.<|im_end|>' }}\n {% endif %}\n \n {{- '<|im_start|>' + message['role'] + '\n' }}\n \n {#--- Same content handling logic as above but without template/examples ---#}\n {%- if message['content'] is string -%}\n {{- message['content'] | trim }}\n {%- elif message['content'] is mapping and message['content']['type'] == 'image' -%}\n {{- image_placeholder | trim }}\n {%- else -%}\n {%- for content in message['content'] -%}\n {%- if content is string -%}\n {{- content | trim -}}\n {%- elif content is mapping and content.get('type') == 'text' and content.get('text') == '<image>' -%}\n {{- image_placeholder | trim }}\n {%- elif content is mapping and content.get('type') == 'text' -%}\n {{- content['text'] | trim -}}\n {%- elif content is mapping and content.get('type') == 'image' -%}\n {# Skip adding image placeholder - it's already in the text #}\n {%- endif -%}\n {%- endfor -%}\n {%- endif -%}\n {{- '<|im_end|>'}}\n {% endif %}\n{% endfor -%}\n{#--- Add Generation Prompt if Requested ---#}\n{%- if add_generation_prompt %}\n {{- '<|im_start|>assistant' }}\n{% endif -%}"
|
| 3 |
}
|
model-00001-of-00002.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 3984201128
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c031e6ade26c8bd2f3ebb988be752737479f30658c76e6738b90d7099dd7f4a5
|
| 3 |
size 3984201128
|
model-00002-of-00002.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 2944322576
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b377f9bec88da826022e7a0b469bd3189a45b4716b7587ac902aaed1a175ce68
|
| 3 |
size 2944322576
|
preprocessor_config.json
CHANGED
|
@@ -14,7 +14,7 @@
|
|
| 14 |
0.26130258,
|
| 15 |
0.27577711
|
| 16 |
],
|
| 17 |
-
"max_pixels":
|
| 18 |
"merge_size": 2,
|
| 19 |
"min_pixels": 200704,
|
| 20 |
"patch_size": 14,
|
|
@@ -22,7 +22,7 @@
|
|
| 22 |
"resample": 3,
|
| 23 |
"rescale_factor": 0.00392156862745098,
|
| 24 |
"size": {
|
| 25 |
-
"longest_edge":
|
| 26 |
"shortest_edge": 200704
|
| 27 |
},
|
| 28 |
"temporal_patch_size": 2
|
|
|
|
| 14 |
0.26130258,
|
| 15 |
0.27577711
|
| 16 |
],
|
| 17 |
+
"max_pixels": 23000000,
|
| 18 |
"merge_size": 2,
|
| 19 |
"min_pixels": 200704,
|
| 20 |
"patch_size": 14,
|
|
|
|
| 22 |
"resample": 3,
|
| 23 |
"rescale_factor": 0.00392156862745098,
|
| 24 |
"size": {
|
| 25 |
+
"longest_edge": 23000000,
|
| 26 |
"shortest_edge": 200704
|
| 27 |
},
|
| 28 |
"temporal_patch_size": 2
|
quant_log.csv
CHANGED
|
@@ -1,197 +1,197 @@
|
|
| 1 |
layer,module,loss,samples,damp,time
|
| 2 |
-
0,self_attn.k_proj,0.
|
| 3 |
-
0,self_attn.v_proj,0.
|
| 4 |
-
0,self_attn.q_proj,1.
|
| 5 |
-
0,self_attn.o_proj,0.
|
| 6 |
-
0,mlp.up_proj,1.
|
| 7 |
-
0,mlp.gate_proj,
|
| 8 |
-
0,mlp.down_proj,0.
|
| 9 |
-
1,self_attn.k_proj,0.
|
| 10 |
-
1,self_attn.v_proj,0.
|
| 11 |
-
1,self_attn.q_proj,0.
|
| 12 |
-
1,self_attn.o_proj,0.
|
| 13 |
-
1,mlp.up_proj,
|
| 14 |
-
1,mlp.gate_proj,
|
| 15 |
-
1,mlp.down_proj,
|
| 16 |
-
2,self_attn.k_proj,
|
| 17 |
-
2,self_attn.v_proj,0.
|
| 18 |
-
2,self_attn.q_proj,
|
| 19 |
-
2,self_attn.o_proj,0.
|
| 20 |
-
2,mlp.up_proj,
|
| 21 |
-
2,mlp.gate_proj,
|
| 22 |
-
2,mlp.down_proj,
|
| 23 |
-
3,self_attn.k_proj,
|
| 24 |
-
3,self_attn.v_proj,0.
|
| 25 |
-
3,self_attn.q_proj,
|
| 26 |
-
3,self_attn.o_proj,0.
|
| 27 |
-
3,mlp.up_proj,
|
| 28 |
-
3,mlp.gate_proj,
|
| 29 |
-
3,mlp.down_proj,0.
|
| 30 |
-
4,self_attn.k_proj,
|
| 31 |
-
4,self_attn.v_proj,0.
|
| 32 |
-
4,self_attn.q_proj,
|
| 33 |
-
4,self_attn.o_proj,0.
|
| 34 |
-
4,mlp.up_proj,
|
| 35 |
-
4,mlp.gate_proj,
|
| 36 |
-
4,mlp.down_proj,
|
| 37 |
-
5,self_attn.k_proj,
|
| 38 |
-
5,self_attn.v_proj,0.
|
| 39 |
-
5,self_attn.q_proj,
|
| 40 |
-
5,self_attn.o_proj,0.
|
| 41 |
-
5,mlp.up_proj,
|
| 42 |
-
5,mlp.gate_proj,
|
| 43 |
-
5,mlp.down_proj,
|
| 44 |
-
6,self_attn.k_proj,
|
| 45 |
-
6,self_attn.v_proj,0.
|
| 46 |
-
6,self_attn.q_proj,
|
| 47 |
-
6,self_attn.o_proj,0.
|
| 48 |
-
6,mlp.up_proj,
|
| 49 |
-
6,mlp.gate_proj,
|
| 50 |
-
6,mlp.down_proj,
|
| 51 |
-
7,self_attn.k_proj,1.
|
| 52 |
-
7,self_attn.v_proj,
|
| 53 |
-
7,self_attn.q_proj,
|
| 54 |
-
7,self_attn.o_proj,
|
| 55 |
-
7,mlp.up_proj,
|
| 56 |
-
7,mlp.gate_proj,
|
| 57 |
-
7,mlp.down_proj,
|
| 58 |
-
8,self_attn.k_proj,
|
| 59 |
-
8,self_attn.v_proj,
|
| 60 |
-
8,self_attn.q_proj,
|
| 61 |
-
8,self_attn.o_proj,
|
| 62 |
-
8,mlp.up_proj,
|
| 63 |
-
8,mlp.gate_proj,
|
| 64 |
-
8,mlp.down_proj,
|
| 65 |
-
9,self_attn.k_proj,
|
| 66 |
-
9,self_attn.v_proj,
|
| 67 |
-
9,self_attn.q_proj,
|
| 68 |
-
9,self_attn.o_proj,1.
|
| 69 |
-
9,mlp.up_proj,
|
| 70 |
-
9,mlp.gate_proj,
|
| 71 |
-
9,mlp.down_proj,
|
| 72 |
-
10,self_attn.k_proj,
|
| 73 |
-
10,self_attn.v_proj,0.
|
| 74 |
-
10,self_attn.q_proj,
|
| 75 |
-
10,self_attn.o_proj,
|
| 76 |
-
10,mlp.up_proj,
|
| 77 |
-
10,mlp.gate_proj,
|
| 78 |
-
10,mlp.down_proj,
|
| 79 |
-
11,self_attn.k_proj,
|
| 80 |
-
11,self_attn.v_proj,0.
|
| 81 |
-
11,self_attn.q_proj,
|
| 82 |
-
11,self_attn.o_proj,
|
| 83 |
-
11,mlp.up_proj,
|
| 84 |
-
11,mlp.gate_proj,
|
| 85 |
-
11,mlp.down_proj,
|
| 86 |
-
12,self_attn.k_proj,
|
| 87 |
-
12,self_attn.v_proj,
|
| 88 |
-
12,self_attn.q_proj,
|
| 89 |
-
12,self_attn.o_proj,
|
| 90 |
-
12,mlp.up_proj,
|
| 91 |
-
12,mlp.gate_proj,
|
| 92 |
-
12,mlp.down_proj,
|
| 93 |
-
13,self_attn.k_proj,
|
| 94 |
-
13,self_attn.v_proj,
|
| 95 |
-
13,self_attn.q_proj,
|
| 96 |
-
13,self_attn.o_proj,
|
| 97 |
-
13,mlp.up_proj,
|
| 98 |
-
13,mlp.gate_proj,
|
| 99 |
-
13,mlp.down_proj,
|
| 100 |
-
14,self_attn.k_proj,
|
| 101 |
-
14,self_attn.v_proj,
|
| 102 |
-
14,self_attn.q_proj,
|
| 103 |
-
14,self_attn.o_proj,
|
| 104 |
-
14,mlp.up_proj,
|
| 105 |
-
14,mlp.gate_proj,
|
| 106 |
-
14,mlp.down_proj,
|
| 107 |
-
15,self_attn.k_proj,
|
| 108 |
-
15,self_attn.v_proj,
|
| 109 |
-
15,self_attn.q_proj,
|
| 110 |
-
15,self_attn.o_proj,1.
|
| 111 |
-
15,mlp.up_proj,
|
| 112 |
-
15,mlp.gate_proj,
|
| 113 |
-
15,mlp.down_proj,
|
| 114 |
-
16,self_attn.k_proj,
|
| 115 |
-
16,self_attn.v_proj,
|
| 116 |
-
16,self_attn.q_proj,
|
| 117 |
-
16,self_attn.o_proj,
|
| 118 |
-
16,mlp.up_proj,
|
| 119 |
-
16,mlp.gate_proj,
|
| 120 |
-
16,mlp.down_proj,
|
| 121 |
-
17,self_attn.k_proj,
|
| 122 |
-
17,self_attn.v_proj,
|
| 123 |
-
17,self_attn.q_proj,
|
| 124 |
-
17,self_attn.o_proj,
|
| 125 |
-
17,mlp.up_proj,
|
| 126 |
-
17,mlp.gate_proj,
|
| 127 |
-
17,mlp.down_proj,
|
| 128 |
-
18,self_attn.k_proj,
|
| 129 |
-
18,self_attn.v_proj,1.
|
| 130 |
-
18,self_attn.q_proj,
|
| 131 |
-
18,self_attn.o_proj,
|
| 132 |
-
18,mlp.up_proj,
|
| 133 |
-
18,mlp.gate_proj,
|
| 134 |
-
18,mlp.down_proj,
|
| 135 |
-
19,self_attn.k_proj,
|
| 136 |
-
19,self_attn.v_proj,
|
| 137 |
-
19,self_attn.q_proj,
|
| 138 |
-
19,self_attn.o_proj,
|
| 139 |
-
19,mlp.up_proj,
|
| 140 |
-
19,mlp.gate_proj,
|
| 141 |
-
19,mlp.down_proj,
|
| 142 |
-
20,self_attn.k_proj,
|
| 143 |
-
20,self_attn.v_proj,
|
| 144 |
-
20,self_attn.q_proj,
|
| 145 |
-
20,self_attn.o_proj,
|
| 146 |
-
20,mlp.up_proj,
|
| 147 |
-
20,mlp.gate_proj,
|
| 148 |
-
20,mlp.down_proj,
|
| 149 |
-
21,self_attn.k_proj,
|
| 150 |
-
21,self_attn.v_proj,
|
| 151 |
-
21,self_attn.q_proj,
|
| 152 |
-
21,self_attn.o_proj,
|
| 153 |
-
21,mlp.up_proj,
|
| 154 |
-
21,mlp.gate_proj,
|
| 155 |
-
21,mlp.down_proj,
|
| 156 |
-
22,self_attn.k_proj,
|
| 157 |
-
22,self_attn.v_proj,
|
| 158 |
-
22,self_attn.q_proj,
|
| 159 |
-
22,self_attn.o_proj,
|
| 160 |
-
22,mlp.up_proj,
|
| 161 |
-
22,mlp.gate_proj,
|
| 162 |
-
22,mlp.down_proj,
|
| 163 |
-
23,self_attn.k_proj,
|
| 164 |
-
23,self_attn.v_proj,
|
| 165 |
-
23,self_attn.q_proj,
|
| 166 |
-
23,self_attn.o_proj,
|
| 167 |
-
23,mlp.up_proj,
|
| 168 |
-
23,mlp.gate_proj,
|
| 169 |
-
23,mlp.down_proj,
|
| 170 |
-
24,self_attn.k_proj,
|
| 171 |
-
24,self_attn.v_proj,
|
| 172 |
-
24,self_attn.q_proj,
|
| 173 |
-
24,self_attn.o_proj,
|
| 174 |
-
24,mlp.up_proj,
|
| 175 |
-
24,mlp.gate_proj,
|
| 176 |
-
24,mlp.down_proj,
|
| 177 |
-
25,self_attn.k_proj,
|
| 178 |
-
25,self_attn.v_proj,
|
| 179 |
-
25,self_attn.q_proj,
|
| 180 |
-
25,self_attn.o_proj,
|
| 181 |
-
25,mlp.up_proj,
|
| 182 |
-
25,mlp.gate_proj,
|
| 183 |
-
25,mlp.down_proj,
|
| 184 |
-
26,self_attn.k_proj,
|
| 185 |
-
26,self_attn.v_proj,
|
| 186 |
-
26,self_attn.q_proj,
|
| 187 |
-
26,self_attn.o_proj,
|
| 188 |
-
26,mlp.up_proj,
|
| 189 |
-
26,mlp.gate_proj,
|
| 190 |
-
26,mlp.down_proj,0.
|
| 191 |
-
27,self_attn.k_proj,
|
| 192 |
-
27,self_attn.v_proj,
|
| 193 |
-
27,self_attn.q_proj,
|
| 194 |
-
27,self_attn.o_proj,
|
| 195 |
-
27,mlp.up_proj,
|
| 196 |
-
27,mlp.gate_proj,
|
| 197 |
-
27,mlp.down_proj,
|
|
|
|
| 1 |
layer,module,loss,samples,damp,time
|
| 2 |
+
0,self_attn.k_proj,0.37272908,0.01000,0.963
|
| 3 |
+
0,self_attn.v_proj,0.06249057,0.01000,0.568
|
| 4 |
+
0,self_attn.q_proj,1.71414546,0.01000,0.582
|
| 5 |
+
0,self_attn.o_proj,0.27218719,0.01000,0.579
|
| 6 |
+
0,mlp.up_proj,1.97148761,0.01000,0.602
|
| 7 |
+
0,mlp.gate_proj,5.45231162,0.01000,0.596
|
| 8 |
+
0,mlp.down_proj,0.61297246,0.01000,3.775
|
| 9 |
+
1,self_attn.k_proj,0.27201448,0.01000,0.597
|
| 10 |
+
1,self_attn.v_proj,0.06198664,0.01000,0.593
|
| 11 |
+
1,self_attn.q_proj,0.98380148,0.01000,0.580
|
| 12 |
+
1,self_attn.o_proj,0.07505655,0.01000,0.641
|
| 13 |
+
1,mlp.up_proj,29.94800979,0.01000,0.609
|
| 14 |
+
1,mlp.gate_proj,46.14816320,0.01000,0.605
|
| 15 |
+
1,mlp.down_proj,1.18726570,0.01000,3.571
|
| 16 |
+
2,self_attn.k_proj,1.10431638,0.01000,0.573
|
| 17 |
+
2,self_attn.v_proj,0.17621659,0.01000,0.580
|
| 18 |
+
2,self_attn.q_proj,3.89487222,0.01000,0.582
|
| 19 |
+
2,self_attn.o_proj,0.19335656,0.01000,0.599
|
| 20 |
+
2,mlp.up_proj,39.82002804,0.01000,0.599
|
| 21 |
+
2,mlp.gate_proj,65.63606855,0.01000,0.596
|
| 22 |
+
2,mlp.down_proj,2.19458135,0.01000,3.704
|
| 23 |
+
3,self_attn.k_proj,1.41276497,0.01000,0.583
|
| 24 |
+
3,self_attn.v_proj,0.32184546,0.01000,0.565
|
| 25 |
+
3,self_attn.q_proj,5.03698810,0.01000,0.568
|
| 26 |
+
3,self_attn.o_proj,0.52726446,0.01000,0.599
|
| 27 |
+
3,mlp.up_proj,114.33669945,0.01000,0.638
|
| 28 |
+
3,mlp.gate_proj,148.85794254,0.01000,0.621
|
| 29 |
+
3,mlp.down_proj,0.01009648,0.01250,4.438
|
| 30 |
+
4,self_attn.k_proj,2.52166661,0.01000,0.639
|
| 31 |
+
4,self_attn.v_proj,0.67622127,0.01000,0.620
|
| 32 |
+
4,self_attn.q_proj,10.74245164,0.01000,0.600
|
| 33 |
+
4,self_attn.o_proj,0.42747642,0.01000,0.602
|
| 34 |
+
4,mlp.up_proj,110.46822172,0.01000,0.661
|
| 35 |
+
4,mlp.gate_proj,161.16622643,0.01000,0.668
|
| 36 |
+
4,mlp.down_proj,2.72548815,0.01000,3.644
|
| 37 |
+
5,self_attn.k_proj,2.36542153,0.01000,0.596
|
| 38 |
+
5,self_attn.v_proj,0.82987875,0.01000,0.580
|
| 39 |
+
5,self_attn.q_proj,10.98670529,0.01000,0.583
|
| 40 |
+
5,self_attn.o_proj,0.48163516,0.01000,0.578
|
| 41 |
+
5,mlp.up_proj,169.33053275,0.01000,0.606
|
| 42 |
+
5,mlp.gate_proj,208.03480575,0.01000,0.632
|
| 43 |
+
5,mlp.down_proj,2.63673970,0.01000,3.785
|
| 44 |
+
6,self_attn.k_proj,1.51797450,0.01000,0.567
|
| 45 |
+
6,self_attn.v_proj,0.70319183,0.01000,0.574
|
| 46 |
+
6,self_attn.q_proj,7.39946994,0.01000,0.575
|
| 47 |
+
6,self_attn.o_proj,0.58445273,0.01000,0.597
|
| 48 |
+
6,mlp.up_proj,40.56810539,0.01000,0.589
|
| 49 |
+
6,mlp.gate_proj,57.70911391,0.01000,0.594
|
| 50 |
+
6,mlp.down_proj,4.15842980,0.01000,3.864
|
| 51 |
+
7,self_attn.k_proj,1.65347762,0.01000,0.605
|
| 52 |
+
7,self_attn.v_proj,1.26919441,0.01000,0.621
|
| 53 |
+
7,self_attn.q_proj,8.91557116,0.01000,0.617
|
| 54 |
+
7,self_attn.o_proj,1.26652386,0.01000,0.634
|
| 55 |
+
7,mlp.up_proj,37.50803346,0.01000,0.644
|
| 56 |
+
7,mlp.gate_proj,41.65068270,0.01000,0.663
|
| 57 |
+
7,mlp.down_proj,5.51070028,0.01000,3.738
|
| 58 |
+
8,self_attn.k_proj,3.05815341,0.01000,0.599
|
| 59 |
+
8,self_attn.v_proj,1.05918791,0.01000,0.605
|
| 60 |
+
8,self_attn.q_proj,13.26367949,0.01000,0.607
|
| 61 |
+
8,self_attn.o_proj,1.30098944,0.01000,0.605
|
| 62 |
+
8,mlp.up_proj,39.01938902,0.01000,0.591
|
| 63 |
+
8,mlp.gate_proj,40.47557959,0.01000,0.582
|
| 64 |
+
8,mlp.down_proj,5.03287961,0.01000,3.769
|
| 65 |
+
9,self_attn.k_proj,2.22347360,0.01000,0.609
|
| 66 |
+
9,self_attn.v_proj,1.48193391,0.01000,0.608
|
| 67 |
+
9,self_attn.q_proj,11.94389274,0.01000,0.585
|
| 68 |
+
9,self_attn.o_proj,1.67733308,0.01000,0.602
|
| 69 |
+
9,mlp.up_proj,72.89947883,0.01000,0.633
|
| 70 |
+
9,mlp.gate_proj,116.10071928,0.01000,0.614
|
| 71 |
+
9,mlp.down_proj,4.55470750,0.01000,3.589
|
| 72 |
+
10,self_attn.k_proj,2.02291953,0.01000,0.587
|
| 73 |
+
10,self_attn.v_proj,0.91743158,0.01000,0.572
|
| 74 |
+
10,self_attn.q_proj,10.23927115,0.01000,0.586
|
| 75 |
+
10,self_attn.o_proj,1.26303708,0.01000,0.628
|
| 76 |
+
10,mlp.up_proj,38.33327576,0.01000,0.631
|
| 77 |
+
10,mlp.gate_proj,42.22324041,0.01000,0.631
|
| 78 |
+
10,mlp.down_proj,4.19643478,0.01000,3.880
|
| 79 |
+
11,self_attn.k_proj,2.49720556,0.01000,0.604
|
| 80 |
+
11,self_attn.v_proj,0.84860099,0.01000,0.604
|
| 81 |
+
11,self_attn.q_proj,10.59318743,0.01000,0.614
|
| 82 |
+
11,self_attn.o_proj,1.44926018,0.01000,0.613
|
| 83 |
+
11,mlp.up_proj,34.05972397,0.01000,0.600
|
| 84 |
+
11,mlp.gate_proj,34.98881461,0.01000,0.621
|
| 85 |
+
11,mlp.down_proj,3.78504640,0.01000,3.701
|
| 86 |
+
12,self_attn.k_proj,2.51501925,0.01000,0.643
|
| 87 |
+
12,self_attn.v_proj,1.06571939,0.01000,0.612
|
| 88 |
+
12,self_attn.q_proj,11.37260368,0.01000,0.595
|
| 89 |
+
12,self_attn.o_proj,1.42345318,0.01000,0.574
|
| 90 |
+
12,mlp.up_proj,33.02424516,0.01000,0.609
|
| 91 |
+
12,mlp.gate_proj,32.15129480,0.01000,0.631
|
| 92 |
+
12,mlp.down_proj,4.32310599,0.01000,3.762
|
| 93 |
+
13,self_attn.k_proj,2.17874396,0.01000,0.600
|
| 94 |
+
13,self_attn.v_proj,1.34991701,0.01000,0.585
|
| 95 |
+
13,self_attn.q_proj,11.63673653,0.01000,0.595
|
| 96 |
+
13,self_attn.o_proj,2.12498794,0.01000,0.577
|
| 97 |
+
13,mlp.up_proj,31.13961415,0.01000,0.606
|
| 98 |
+
13,mlp.gate_proj,32.60532906,0.01000,0.599
|
| 99 |
+
13,mlp.down_proj,3.95873230,0.01000,3.753
|
| 100 |
+
14,self_attn.k_proj,2.99430476,0.01000,0.563
|
| 101 |
+
14,self_attn.v_proj,1.20193668,0.01000,0.557
|
| 102 |
+
14,self_attn.q_proj,15.81873781,0.01000,0.563
|
| 103 |
+
14,self_attn.o_proj,2.06551350,0.01000,0.634
|
| 104 |
+
14,mlp.up_proj,33.95725983,0.01000,0.578
|
| 105 |
+
14,mlp.gate_proj,33.69845680,0.01000,0.593
|
| 106 |
+
14,mlp.down_proj,4.71709626,0.01000,3.694
|
| 107 |
+
15,self_attn.k_proj,2.77116690,0.01000,0.609
|
| 108 |
+
15,self_attn.v_proj,1.10220412,0.01000,0.591
|
| 109 |
+
15,self_attn.q_proj,12.72785654,0.01000,0.599
|
| 110 |
+
15,self_attn.o_proj,1.87682963,0.01000,0.609
|
| 111 |
+
15,mlp.up_proj,31.97431222,0.01000,0.659
|
| 112 |
+
15,mlp.gate_proj,30.82866852,0.01000,0.627
|
| 113 |
+
15,mlp.down_proj,4.68322897,0.01000,3.653
|
| 114 |
+
16,self_attn.k_proj,2.65178245,0.01000,0.597
|
| 115 |
+
16,self_attn.v_proj,1.33414724,0.01000,0.595
|
| 116 |
+
16,self_attn.q_proj,13.40043761,0.01000,0.587
|
| 117 |
+
16,self_attn.o_proj,2.56995311,0.01000,0.605
|
| 118 |
+
16,mlp.up_proj,34.49688110,0.01000,0.593
|
| 119 |
+
16,mlp.gate_proj,33.07141986,0.01000,0.594
|
| 120 |
+
16,mlp.down_proj,4.67668955,0.01000,3.801
|
| 121 |
+
17,self_attn.k_proj,2.56790221,0.01000,0.573
|
| 122 |
+
17,self_attn.v_proj,1.63155973,0.01000,0.569
|
| 123 |
+
17,self_attn.q_proj,14.37411106,0.01000,0.567
|
| 124 |
+
17,self_attn.o_proj,2.11580409,0.01000,0.613
|
| 125 |
+
17,mlp.up_proj,40.19091352,0.01000,0.624
|
| 126 |
+
17,mlp.gate_proj,37.72127967,0.01000,0.617
|
| 127 |
+
17,mlp.down_proj,6.41667661,0.01000,3.904
|
| 128 |
+
18,self_attn.k_proj,2.01086315,0.01000,0.627
|
| 129 |
+
18,self_attn.v_proj,1.82605520,0.01000,0.622
|
| 130 |
+
18,self_attn.q_proj,12.26445795,0.01000,0.628
|
| 131 |
+
18,self_attn.o_proj,2.60375627,0.01000,0.626
|
| 132 |
+
18,mlp.up_proj,42.72995571,0.01000,0.637
|
| 133 |
+
18,mlp.gate_proj,39.51592978,0.01000,0.633
|
| 134 |
+
18,mlp.down_proj,7.30472941,0.01000,3.770
|
| 135 |
+
19,self_attn.k_proj,2.03734621,0.01000,0.609
|
| 136 |
+
19,self_attn.v_proj,2.17895159,0.01000,0.608
|
| 137 |
+
19,self_attn.q_proj,13.97130506,0.01000,0.602
|
| 138 |
+
19,self_attn.o_proj,2.98630336,0.01000,0.589
|
| 139 |
+
19,mlp.up_proj,45.40066239,0.01000,0.608
|
| 140 |
+
19,mlp.gate_proj,43.89151394,0.01000,0.612
|
| 141 |
+
19,mlp.down_proj,7.82231198,0.01000,3.688
|
| 142 |
+
20,self_attn.k_proj,2.03618901,0.01000,0.600
|
| 143 |
+
20,self_attn.v_proj,2.32651697,0.01000,0.609
|
| 144 |
+
20,self_attn.q_proj,12.49319835,0.01000,0.644
|
| 145 |
+
20,self_attn.o_proj,2.02054115,0.01000,0.607
|
| 146 |
+
20,mlp.up_proj,53.92699021,0.01000,0.606
|
| 147 |
+
20,mlp.gate_proj,51.74172017,0.01000,0.599
|
| 148 |
+
20,mlp.down_proj,14.09048887,0.01000,3.809
|
| 149 |
+
21,self_attn.k_proj,2.22723000,0.01000,0.632
|
| 150 |
+
21,self_attn.v_proj,3.41219732,0.01000,0.620
|
| 151 |
+
21,self_attn.q_proj,14.44116497,0.01000,0.618
|
| 152 |
+
21,self_attn.o_proj,4.64260670,0.01000,0.616
|
| 153 |
+
21,mlp.up_proj,64.68185143,0.01000,0.632
|
| 154 |
+
21,mlp.gate_proj,65.52687134,0.01000,0.628
|
| 155 |
+
21,mlp.down_proj,17.97917598,0.01000,3.809
|
| 156 |
+
22,self_attn.k_proj,2.89653644,0.01000,0.588
|
| 157 |
+
22,self_attn.v_proj,5.82066236,0.01000,0.582
|
| 158 |
+
22,self_attn.q_proj,18.60508524,0.01000,0.594
|
| 159 |
+
22,self_attn.o_proj,5.02134183,0.01000,0.634
|
| 160 |
+
22,mlp.up_proj,90.02371180,0.01000,0.611
|
| 161 |
+
22,mlp.gate_proj,91.14663930,0.01000,0.615
|
| 162 |
+
22,mlp.down_proj,34.05776069,0.01000,3.731
|
| 163 |
+
23,self_attn.k_proj,4.18426553,0.01000,0.582
|
| 164 |
+
23,self_attn.v_proj,10.33130930,0.01000,0.578
|
| 165 |
+
23,self_attn.q_proj,25.64362199,0.01000,0.571
|
| 166 |
+
23,self_attn.o_proj,13.37783952,0.01000,0.588
|
| 167 |
+
23,mlp.up_proj,125.68233908,0.01000,0.604
|
| 168 |
+
23,mlp.gate_proj,128.76723017,0.01000,0.603
|
| 169 |
+
23,mlp.down_proj,43.81595416,0.01000,3.752
|
| 170 |
+
24,self_attn.k_proj,3.36082683,0.01000,0.582
|
| 171 |
+
24,self_attn.v_proj,10.30109492,0.01000,0.590
|
| 172 |
+
24,self_attn.q_proj,23.33862463,0.01000,0.593
|
| 173 |
+
24,self_attn.o_proj,5.05974270,0.01000,0.585
|
| 174 |
+
24,mlp.up_proj,139.16557624,0.01000,0.603
|
| 175 |
+
24,mlp.gate_proj,132.34724480,0.01000,0.586
|
| 176 |
+
24,mlp.down_proj,59.95090011,0.01000,3.698
|
| 177 |
+
25,self_attn.k_proj,4.10498364,0.01000,0.576
|
| 178 |
+
25,self_attn.v_proj,17.18356073,0.01000,0.562
|
| 179 |
+
25,self_attn.q_proj,27.32380730,0.01000,0.571
|
| 180 |
+
25,self_attn.o_proj,9.62589656,0.01000,0.578
|
| 181 |
+
25,mlp.up_proj,171.75121912,0.01000,0.598
|
| 182 |
+
25,mlp.gate_proj,154.04677341,0.01000,0.583
|
| 183 |
+
25,mlp.down_proj,83.74196908,0.01000,3.576
|
| 184 |
+
26,self_attn.k_proj,5.83665500,0.01000,0.589
|
| 185 |
+
26,self_attn.v_proj,42.81872257,0.01000,0.590
|
| 186 |
+
26,self_attn.q_proj,46.47709586,0.01000,0.595
|
| 187 |
+
26,self_attn.o_proj,23.30883757,0.01000,0.583
|
| 188 |
+
26,mlp.up_proj,170.47679616,0.01000,0.612
|
| 189 |
+
26,mlp.gate_proj,150.27404706,0.01000,0.595
|
| 190 |
+
26,mlp.down_proj,0.00017362,0.01250,4.201
|
| 191 |
+
27,self_attn.k_proj,6.51690253,0.01000,0.592
|
| 192 |
+
27,self_attn.v_proj,61.79766743,0.01000,0.568
|
| 193 |
+
27,self_attn.q_proj,64.76963792,0.01000,0.587
|
| 194 |
+
27,self_attn.o_proj,43.01269404,0.01000,0.577
|
| 195 |
+
27,mlp.up_proj,221.63391986,0.01000,0.607
|
| 196 |
+
27,mlp.gate_proj,214.62520319,0.01000,0.615
|
| 197 |
+
27,mlp.down_proj,415.36760403,0.01000,3.916
|
tokenizer.json
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ba0c439f7be467bf47d12a7e6f9adc6116201056fc60c67f431c679b7c16afc8
|
| 3 |
+
size 11422064
|
tokenizer_config.json
CHANGED
|
@@ -1,4 +1,5 @@
|
|
| 1 |
{
|
|
|
|
| 2 |
"add_bos_token": false,
|
| 3 |
"add_prefix_space": false,
|
| 4 |
"added_tokens_decoder": {
|
|
@@ -195,16 +196,15 @@
|
|
| 195 |
"<|video_pad|>"
|
| 196 |
],
|
| 197 |
"bos_token": null,
|
| 198 |
-
"chat_template": "{% set image_placeholder = '<|vision_start|><|image_pad|><|vision_end|>' %}\n{% for message in messages %}\n {#--- Handle User Messages with Template and Examples ---#}\n {%- if message['role'] == 'user' and template -%}\n {% if loop.first and message['role'] != 'system' %}\n {{- '<|im_start|>system
|
| 199 |
"clean_up_tokenization_spaces": false,
|
| 200 |
"eos_token": "<|im_end|>",
|
| 201 |
"errors": "replace",
|
| 202 |
"extra_special_tokens": {},
|
| 203 |
-
"max_pixels":
|
| 204 |
"min_pixels": 200704,
|
| 205 |
"model_max_length": 131072,
|
| 206 |
"pad_token": "<|endoftext|>",
|
| 207 |
-
"padding_side": "left",
|
| 208 |
"processor_class": "Qwen2_5_VLProcessor",
|
| 209 |
"split_special_tokens": false,
|
| 210 |
"tokenizer_class": "Qwen2Tokenizer",
|
|
|
|
| 1 |
{
|
| 2 |
+
"_commit_hash": null,
|
| 3 |
"add_bos_token": false,
|
| 4 |
"add_prefix_space": false,
|
| 5 |
"added_tokens_decoder": {
|
|
|
|
| 196 |
"<|video_pad|>"
|
| 197 |
],
|
| 198 |
"bos_token": null,
|
| 199 |
+
"chat_template": "{% set image_placeholder = '<|vision_start|><|image_pad|><|vision_end|>' %}\n{% for message in messages %}\n {#--- Handle User Messages with Template and Examples ---#}\n {%- if message['role'] == 'user' and template -%}\n {% if loop.first and message['role'] != 'system' %}\n {{- '<|im_start|>system\nYou are NuExtract, an information extraction tool created by NuMind.<|im_end|>' }}\n {% endif %}\n \n {{- '<|im_start|>' + message['role'] -}}\n \n {#--- Template Section ---#}\n {{ '\n# Template:' }}\n {{- '\n' + template + '\n' }}\n \n {#--- Examples Section (if provided) ---#}\n {% if examples -%}\n {{- '# Examples:' }}\n {% for example in examples %}\n {{- '## Input:\n' }}\n {#--- Handle image examples ---#}\n {% if example['input'] is mapping and example['input']['type'] == 'image' %}\n {{- image_placeholder | trim -}}\n {% elif example['input'] == '<image>' %}\n {{- image_placeholder | trim -}}\n {% else %}\n {{- example['input'] -}}\n {% endif %}\n {{- '\n## Output:\n' ~ example['output'] }}\n {% endfor %}\n {%- endif %}\n \n {#--- Context Section: Handle various content types ---#}\n {{- '# Context:\n' }}\n {%- if message['content'] is string -%}\n {#--- Simple string content ---#}\n {{- message['content'] | trim -}}\n {%- elif message['content'] is mapping and message['content']['type'] == 'image' -%}\n {#--- Single image document ---#}\n {{- image_placeholder | trim -}}\n {%- else -%}\n {#--- List of content items (mixed text/images) ---#}\n {#--- First, determine what the actual input content is (not ICL images) ---#}\n {%- set ns = namespace(has_text_input=false, text_content='') -%}\n \n {#--- Count content types and identify actual input document ---#}\n {%- for content in message['content'] -%}\n {%- if content is mapping and content.get('type') == 'text' -%}\n {%- if content.get('text') != '<image>' -%}\n {%- set ns.has_text_input = true -%}\n {%- set ns.text_content = content['text'] -%}\n {%- endif -%}\n {%- elif content is string -%}\n {%- if content != '<image>' -%}\n {%- set ns.has_text_input = true -%}\n {%- set ns.text_content = content -%}\n {%- endif -%}\n {%- endif -%}\n {%- endfor -%}\n \n {#--- Determine what to output based on actual input type ---#}\n {%- if ns.has_text_input -%}\n {#--- Main input is text, so output the text content ---#}\n {{- ns.text_content | trim -}}\n {%- else -%}\n {#--- Main input is image or <image> placeholder ---#}\n {%- set ns2 = namespace(found_image=false) -%}\n {%- for content in message['content'] -%}\n {%- if content is mapping and content.get('type') == 'image' and not ns2.found_image -%}\n {{- image_placeholder | trim -}}\n {%- set ns2.found_image = true -%}\n {%- elif content is mapping and content.get('type') == 'text' and content.get('text') == '<image>' and not ns2.found_image -%}\n {{- image_placeholder | trim -}}\n {%- set ns2.found_image = true -%}\n {%- elif content is string and content == '<image>' and not ns2.found_image -%}\n {{- image_placeholder | trim -}}\n {%- set ns2.found_image = true -%}\n {%- endif -%}\n {%- endfor -%}\n {%- endif -%}\n {%- endif -%}\n {{- '<|im_end|>\n'}}\n \n {#--- Handle All Other Messages (Assistant, System, etc.) ---#}\n {% else %}\n {% if loop.first and message['role'] != 'system' %}\n {{- '<|im_start|>system\nYou are a helpful assistant.<|im_end|>' }}\n {% endif %}\n \n {{- '<|im_start|>' + message['role'] + '\n' }}\n \n {#--- Same content handling logic as above but without template/examples ---#}\n {%- if message['content'] is string -%}\n {{- message['content'] | trim }}\n {%- elif message['content'] is mapping and message['content']['type'] == 'image' -%}\n {{- image_placeholder | trim }}\n {%- else -%}\n {%- for content in message['content'] -%}\n {%- if content is string -%}\n {{- content | trim -}}\n {%- elif content is mapping and content.get('type') == 'text' and content.get('text') == '<image>' -%}\n {{- image_placeholder | trim }}\n {%- elif content is mapping and content.get('type') == 'text' -%}\n {{- content['text'] | trim -}}\n {%- elif content is mapping and content.get('type') == 'image' -%}\n {# Skip adding image placeholder - it's already in the text #}\n {%- endif -%}\n {%- endfor -%}\n {%- endif -%}\n {{- '<|im_end|>'}}\n {% endif %}\n{% endfor -%}\n{#--- Add Generation Prompt if Requested ---#}\n{%- if add_generation_prompt %}\n {{- '<|im_start|>assistant' }}\n{% endif -%}",
|
| 200 |
"clean_up_tokenization_spaces": false,
|
| 201 |
"eos_token": "<|im_end|>",
|
| 202 |
"errors": "replace",
|
| 203 |
"extra_special_tokens": {},
|
| 204 |
+
"max_pixels": 23000000,
|
| 205 |
"min_pixels": 200704,
|
| 206 |
"model_max_length": 131072,
|
| 207 |
"pad_token": "<|endoftext|>",
|
|
|
|
| 208 |
"processor_class": "Qwen2_5_VLProcessor",
|
| 209 |
"split_special_tokens": false,
|
| 210 |
"tokenizer_class": "Qwen2Tokenizer",
|