Hugging Face's logo

EZCon
/
GLM-OCR-mlx

Image-to-Text
Transformers
Safetensors
MLX
glm_ocr
Model card Files
xet
Community
1
hermeschen-ezcon commited on
Commit
81af169
·
verified ·
1 Parent(s): 224d8a0

Upload folder using huggingface_hub

Browse files
Files changed (10) hide show
  1. README.md +31 -0
  2. chat_template.jinja +140 -0
  3. config.json +69 -0
  4. generation_config.json +10 -0
  5. model.safetensors +3 -0
  6. model.safetensors.index.json +517 -0
  7. preprocessor_config.json +11 -0
  8. processor_config.json +30 -0
  9. tokenizer.json +0 -0
  10. tokenizer_config.json +49 -0
README.md ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: mit
3
+ language:
4
+ - zh
5
+ - en
6
+ - fr
7
+ - es
8
+ - ru
9
+ - de
10
+ - ja
11
+ - ko
12
+ pipeline_tag: image-to-text
13
+ library_name: transformers
14
+ base_model:
15
+ - zai-org/GLM-OCR
16
+ tags:
17
+ - mlx
18
+ ---
19
+
20
+ # EZCon/GLM-OCR-mlx
21
+ This model was converted to MLX format from [`unsloth/GLM-OCR`]() using mlx-vlm version **0.3.11**.
22
+ Refer to the [original model card](https://huggingface.co/unsloth/GLM-OCR) for more details on the model.
23
+ ## Use with mlx
24
+
25
+ ```bash
26
+ pip install -U mlx-vlm
27
+ ```
28
+
29
+ ```bash
30
+ python -m mlx_vlm.generate --model EZCon/GLM-OCR-mlx --max-tokens 100 --temperature 0.0 --prompt "Describe this image." --image <path_to_image>
31
+ ```
chat_template.jinja ADDED
@@ -0,0 +1,140 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [gMASK]<sop>
2
+ {%- if tools -%}
3
+ <|system|>
4
+ # Tools
5
+
6
+ You may call one or more functions to assist with the user query.
7
+
8
+ You are provided with function signatures within <tools></tools> XML tags:
9
+ <tools>
10
+ {% for tool in tools %}
11
+ {{ tool | tojson(ensure_ascii=False) }}
12
+ {% endfor %}
13
+ </tools>
14
+
15
+ For each function call, output the function name and arguments within the following XML format:
16
+ <tool_call>{function-name}
17
+ <arg_key>{arg-key-1}</arg_key>
18
+ <arg_value>{arg-value-1}</arg_value>
19
+ <arg_key>{arg-key-2}</arg_key>
20
+ <arg_value>{arg-value-2}</arg_value>
21
+ ...
22
+ </tool_call>{%- endif -%}
23
+ {%- macro visible_text(content) -%}
24
+ {%- if content is string -%}
25
+ {{- content }}
26
+ {%- elif content is iterable and content is not mapping -%}
27
+ {%- for item in content -%}
28
+ {%- if item is mapping and item.type == 'text' -%}
29
+ {{- item.text }}
30
+ {%- elif item is mapping and (item.type == 'image' or 'image' in item) -%}
31
+ <|begin_of_image|><|image|><|end_of_image|>
32
+ {%- elif item is mapping and (item.type == 'video' or 'video' in item) -%}
33
+ <|begin_of_video|><|video|><|end_of_video|>
34
+ {%- elif item is string -%}
35
+ {{- item }}
36
+ {%- endif -%}
37
+ {%- endfor -%}
38
+ {%- else -%}
39
+ {{- content }}
40
+ {%- endif -%}
41
+ {%- endmacro -%}
42
+ {%- set ns = namespace(last_user_index=-1) %}
43
+ {%- for m in messages %}
44
+ {%- if m.role == 'user' %}
45
+ {% set ns.last_user_index = loop.index0 -%}
46
+ {%- endif %}
47
+ {%- endfor %}
48
+ {% for m in messages %}
49
+ {%- if m.role == 'user' -%}<|user|>
50
+ {% if m.content is string %}
51
+ {{ m.content }}
52
+ {%- else %}
53
+ {%- for item in m.content %}
54
+ {% if item.type == 'video' or 'video' in item %}
55
+ <|begin_of_video|><|video|><|end_of_video|>{% elif item.type == 'image' or 'image' in item %}
56
+ <|begin_of_image|><|image|><|end_of_image|>{% elif item.type == 'text' %}
57
+ {{ item.text }}
58
+ {%- endif %}
59
+ {%- endfor %}
60
+ {%- endif %}
61
+ {{- '/nothink' if (enable_thinking is defined and not enable_thinking and not visible_text(m.content).endswith("/nothink")) else '' -}}
62
+ {%- elif m.role == 'assistant' -%}
63
+ <|assistant|>
64
+ {%- set reasoning_content = '' %}
65
+ {%- set content = visible_text(m.content) %}
66
+ {%- if m.reasoning_content is string %}
67
+ {%- set reasoning_content = m.reasoning_content %}
68
+ {%- else %}
69
+ {%- if '</think>' in content %}
70
+ {%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
71
+ {%- set content = content.split('</think>')[-1].lstrip('\n') %}
72
+ {%- endif %}
73
+ {%- endif %}
74
+ {%- if loop.index0 > ns.last_user_index and reasoning_content -%}
75
+ {{ '\n<think>' + reasoning_content.strip() + '</think>'}}
76
+ {%- else -%}
77
+ {{ '\n<think></think>' }}
78
+ {%- endif -%}
79
+ {%- if content.strip() -%}
80
+ {{ '\n' + content.strip() }}
81
+ {%- endif -%}
82
+ {% if m.tool_calls %}
83
+ {% for tc in m.tool_calls %}
84
+ {%- if tc.function %}
85
+ {%- set tc = tc.function %}
86
+ {%- endif %}
87
+ {{ '\n<tool_call>' + tc.name }}
88
+ {% set _args = tc.arguments %}
89
+ {% for k, v in _args.items() %}
90
+ <arg_key>{{ k }}</arg_key>
91
+ <arg_value>{{ v | tojson(ensure_ascii=False) if v is not string else v }}</arg_value>
92
+ {% endfor %}
93
+ </tool_call>{% endfor %}
94
+ {% endif %}
95
+ {%- elif m.role == 'tool' -%}
96
+ {%- if m.content is string -%}
97
+ {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
98
+ {{- '<|observation|>' }}
99
+ {%- endif %}
100
+ {{- '\n<tool_response>\n' }}
101
+ {{- m.content }}
102
+ {{- '\n</tool_response>' }}
103
+ {% elif m.content is iterable and m.content is not mapping %}
104
+ {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
105
+ {{- '<|observation|>' }}
106
+ {%- endif %}
107
+ {{- '\n<tool_response>\n' }}
108
+ {%- for tr in m.content -%}
109
+ {%- if tr is mapping and tr.type is defined -%}
110
+ {%- set t = tr.type | lower -%}
111
+ {%- if t == 'text' and tr.text is defined -%}
112
+ {{ tr.text }}
113
+ {%- elif t in ['image', 'image_url'] -%}
114
+ <|begin_of_image|><|image|><|end_of_image|>
115
+ {%- elif t in ['video', 'video_url'] -%}
116
+ <|begin_of_video|><|video|><|end_of_video|>
117
+ {%- else -%}
118
+ {{ tr | tojson(ensure_ascii=False) }}
119
+ {%- endif -%}
120
+ {%- else -%}
121
+ {{ tr.output if tr.output is defined else tr }}
122
+ {%- endif -%}
123
+ {%- endfor -%}
124
+ {{- '\n</tool_response>' }}
125
+ {%- else -%}
126
+ <|observation|>{% for tr in m.content %}
127
+
128
+ <tool_response>
129
+ {{ tr.output if tr.output is defined else tr }}
130
+ </tool_response>{% endfor -%}
131
+ {% endif -%}
132
+ {%- elif m.role == 'system' -%}
133
+ <|system|>
134
+ {{ visible_text(m.content) }}
135
+ {%- endif -%}
136
+ {%- endfor -%}
137
+ {%- if add_generation_prompt -%}
138
+ <|assistant|>
139
+ {{'<think></think>\n' if (enable_thinking is defined and not enable_thinking) else ''}}
140
+ {%- endif -%}
config.json ADDED
@@ -0,0 +1,69 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "GlmOcrForConditionalGeneration"
4
+ ],
5
+ "eos_token_id": [
6
+ 59246,
7
+ 59253
8
+ ],
9
+ "image_end_token_id": 59257,
10
+ "image_start_token_id": 59256,
11
+ "image_token_id": 59280,
12
+ "model_type": "glm_ocr",
13
+ "text_config": {
14
+ "model_type": "glm_ocr_text",
15
+ "pad_token_id": 59246,
16
+ "vocab_size": 59392,
17
+ "eos_token_id": [
18
+ 59246,
19
+ 59253
20
+ ],
21
+ "attention_bias": false,
22
+ "attention_dropout": 0.0,
23
+ "head_dim": 128,
24
+ "hidden_act": "silu",
25
+ "hidden_size": 1536,
26
+ "initializer_range": 0.02,
27
+ "intermediate_size": 4608,
28
+ "max_position_embeddings": 131072,
29
+ "num_attention_heads": 16,
30
+ "num_hidden_layers": 16,
31
+ "num_nextn_predict_layers": 1,
32
+ "num_key_value_heads": 8,
33
+ "rms_norm_eps": 1e-05,
34
+ "dtype": "bfloat16",
35
+ "rope_parameters": {
36
+ "rope_type": "default",
37
+ "mrope_section": [
38
+ 16,
39
+ 24,
40
+ 24
41
+ ],
42
+ "partial_rotary_factor": 1.0,
43
+ "rope_theta": 10000
44
+ },
45
+ "tie_word_embeddings": false,
46
+ "use_cache": true
47
+ },
48
+ "transformers_version": "5.0.1dev0",
49
+ "video_end_token_id": 59259,
50
+ "video_start_token_id": 59258,
51
+ "video_token_id": 59281,
52
+ "vision_config": {
53
+ "model_type": "glm_ocr_vision",
54
+ "hidden_size": 1024,
55
+ "depth": 24,
56
+ "num_heads": 16,
57
+ "attention_bias": true,
58
+ "intermediate_size": 4096,
59
+ "hidden_act": "silu",
60
+ "hidden_dropout_prob": 0.0,
61
+ "initializer_range": 0.02,
62
+ "image_size": 336,
63
+ "patch_size": 14,
64
+ "out_hidden_size": 1536,
65
+ "rms_norm_eps": 1e-05,
66
+ "spatial_merge_size": 2,
67
+ "temporal_patch_size": 2
68
+ }
69
+ }
generation_config.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "do_sample": false,
4
+ "eos_token_id": [
5
+ 59246,
6
+ 59253
7
+ ],
8
+ "pad_token_id": 59246,
9
+ "transformers_version": "5.0.1dev0"
10
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3e93839427cf081600cd96651ba4d9edf6d6120f042555eb9fcfa36153905f74
3
+ size 2214872319
model.safetensors.index.json ADDED
@@ -0,0 +1,517 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 2214811648
4
+ },
5
+ "weight_map": {
6
+ "language_model.lm_head.weight": "model.safetensors",
7
+ "language_model.model.embed_tokens.weight": "model.safetensors",
8
+ "language_model.model.layers.0.input_layernorm.weight": "model.safetensors",
9
+ "language_model.model.layers.0.mlp.down_proj.weight": "model.safetensors",
10
+ "language_model.model.layers.0.mlp.gate_up_proj.weight": "model.safetensors",
11
+ "language_model.model.layers.0.post_attention_layernorm.weight": "model.safetensors",
12
+ "language_model.model.layers.0.post_mlp_layernorm.weight": "model.safetensors",
13
+ "language_model.model.layers.0.post_self_attn_layernorm.weight": "model.safetensors",
14
+ "language_model.model.layers.0.self_attn.k_proj.weight": "model.safetensors",
15
+ "language_model.model.layers.0.self_attn.o_proj.weight": "model.safetensors",
16
+ "language_model.model.layers.0.self_attn.q_proj.weight": "model.safetensors",
17
+ "language_model.model.layers.0.self_attn.v_proj.weight": "model.safetensors",
18
+ "language_model.model.layers.1.input_layernorm.weight": "model.safetensors",
19
+ "language_model.model.layers.1.mlp.down_proj.weight": "model.safetensors",
20
+ "language_model.model.layers.1.mlp.gate_up_proj.weight": "model.safetensors",
21
+ "language_model.model.layers.1.post_attention_layernorm.weight": "model.safetensors",
22
+ "language_model.model.layers.1.post_mlp_layernorm.weight": "model.safetensors",
23
+ "language_model.model.layers.1.post_self_attn_layernorm.weight": "model.safetensors",
24
+ "language_model.model.layers.1.self_attn.k_proj.weight": "model.safetensors",
25
+ "language_model.model.layers.1.self_attn.o_proj.weight": "model.safetensors",
26
+ "language_model.model.layers.1.self_attn.q_proj.weight": "model.safetensors",
27
+ "language_model.model.layers.1.self_attn.v_proj.weight": "model.safetensors",
28
+ "language_model.model.layers.10.input_layernorm.weight": "model.safetensors",
29
+ "language_model.model.layers.10.mlp.down_proj.weight": "model.safetensors",
30
+ "language_model.model.layers.10.mlp.gate_up_proj.weight": "model.safetensors",
31
+ "language_model.model.layers.10.post_attention_layernorm.weight": "model.safetensors",
32
+ "language_model.model.layers.10.post_mlp_layernorm.weight": "model.safetensors",
33
+ "language_model.model.layers.10.post_self_attn_layernorm.weight": "model.safetensors",
34
+ "language_model.model.layers.10.self_attn.k_proj.weight": "model.safetensors",
35
+ "language_model.model.layers.10.self_attn.o_proj.weight": "model.safetensors",
36
+ "language_model.model.layers.10.self_attn.q_proj.weight": "model.safetensors",
37
+ "language_model.model.layers.10.self_attn.v_proj.weight": "model.safetensors",
38
+ "language_model.model.layers.11.input_layernorm.weight": "model.safetensors",
39
+ "language_model.model.layers.11.mlp.down_proj.weight": "model.safetensors",
40
+ "language_model.model.layers.11.mlp.gate_up_proj.weight": "model.safetensors",
41
+ "language_model.model.layers.11.post_attention_layernorm.weight": "model.safetensors",
42
+ "language_model.model.layers.11.post_mlp_layernorm.weight": "model.safetensors",
43
+ "language_model.model.layers.11.post_self_attn_layernorm.weight": "model.safetensors",
44
+ "language_model.model.layers.11.self_attn.k_proj.weight": "model.safetensors",
45
+ "language_model.model.layers.11.self_attn.o_proj.weight": "model.safetensors",
46
+ "language_model.model.layers.11.self_attn.q_proj.weight": "model.safetensors",
47
+ "language_model.model.layers.11.self_attn.v_proj.weight": "model.safetensors",
48
+ "language_model.model.layers.12.input_layernorm.weight": "model.safetensors",
49
+ "language_model.model.layers.12.mlp.down_proj.weight": "model.safetensors",
50
+ "language_model.model.layers.12.mlp.gate_up_proj.weight": "model.safetensors",
51
+ "language_model.model.layers.12.post_attention_layernorm.weight": "model.safetensors",
52
+ "language_model.model.layers.12.post_mlp_layernorm.weight": "model.safetensors",
53
+ "language_model.model.layers.12.post_self_attn_layernorm.weight": "model.safetensors",
54
+ "language_model.model.layers.12.self_attn.k_proj.weight": "model.safetensors",
55
+ "language_model.model.layers.12.self_attn.o_proj.weight": "model.safetensors",
56
+ "language_model.model.layers.12.self_attn.q_proj.weight": "model.safetensors",
57
+ "language_model.model.layers.12.self_attn.v_proj.weight": "model.safetensors",
58
+ "language_model.model.layers.13.input_layernorm.weight": "model.safetensors",
59
+ "language_model.model.layers.13.mlp.down_proj.weight": "model.safetensors",
60
+ "language_model.model.layers.13.mlp.gate_up_proj.weight": "model.safetensors",
61
+ "language_model.model.layers.13.post_attention_layernorm.weight": "model.safetensors",
62
+ "language_model.model.layers.13.post_mlp_layernorm.weight": "model.safetensors",
63
+ "language_model.model.layers.13.post_self_attn_layernorm.weight": "model.safetensors",
64
+ "language_model.model.layers.13.self_attn.k_proj.weight": "model.safetensors",
65
+ "language_model.model.layers.13.self_attn.o_proj.weight": "model.safetensors",
66
+ "language_model.model.layers.13.self_attn.q_proj.weight": "model.safetensors",
67
+ "language_model.model.layers.13.self_attn.v_proj.weight": "model.safetensors",
68
+ "language_model.model.layers.14.input_layernorm.weight": "model.safetensors",
69
+ "language_model.model.layers.14.mlp.down_proj.weight": "model.safetensors",
70
+ "language_model.model.layers.14.mlp.gate_up_proj.weight": "model.safetensors",
71
+ "language_model.model.layers.14.post_attention_layernorm.weight": "model.safetensors",
72
+ "language_model.model.layers.14.post_mlp_layernorm.weight": "model.safetensors",
73
+ "language_model.model.layers.14.post_self_attn_layernorm.weight": "model.safetensors",
74
+ "language_model.model.layers.14.self_attn.k_proj.weight": "model.safetensors",
75
+ "language_model.model.layers.14.self_attn.o_proj.weight": "model.safetensors",
76
+ "language_model.model.layers.14.self_attn.q_proj.weight": "model.safetensors",
77
+ "language_model.model.layers.14.self_attn.v_proj.weight": "model.safetensors",
78
+ "language_model.model.layers.15.input_layernorm.weight": "model.safetensors",
79
+ "language_model.model.layers.15.mlp.down_proj.weight": "model.safetensors",
80
+ "language_model.model.layers.15.mlp.gate_up_proj.weight": "model.safetensors",
81
+ "language_model.model.layers.15.post_attention_layernorm.weight": "model.safetensors",
82
+ "language_model.model.layers.15.post_mlp_layernorm.weight": "model.safetensors",
83
+ "language_model.model.layers.15.post_self_attn_layernorm.weight": "model.safetensors",
84
+ "language_model.model.layers.15.self_attn.k_proj.weight": "model.safetensors",
85
+ "language_model.model.layers.15.self_attn.o_proj.weight": "model.safetensors",
86
+ "language_model.model.layers.15.self_attn.q_proj.weight": "model.safetensors",
87
+ "language_model.model.layers.15.self_attn.v_proj.weight": "model.safetensors",
88
+ "language_model.model.layers.2.input_layernorm.weight": "model.safetensors",
89
+ "language_model.model.layers.2.mlp.down_proj.weight": "model.safetensors",
90
+ "language_model.model.layers.2.mlp.gate_up_proj.weight": "model.safetensors",
91
+ "language_model.model.layers.2.post_attention_layernorm.weight": "model.safetensors",
92
+ "language_model.model.layers.2.post_mlp_layernorm.weight": "model.safetensors",
93
+ "language_model.model.layers.2.post_self_attn_layernorm.weight": "model.safetensors",
94
+ "language_model.model.layers.2.self_attn.k_proj.weight": "model.safetensors",
95
+ "language_model.model.layers.2.self_attn.o_proj.weight": "model.safetensors",
96
+ "language_model.model.layers.2.self_attn.q_proj.weight": "model.safetensors",
97
+ "language_model.model.layers.2.self_attn.v_proj.weight": "model.safetensors",
98
+ "language_model.model.layers.3.input_layernorm.weight": "model.safetensors",
99
+ "language_model.model.layers.3.mlp.down_proj.weight": "model.safetensors",
100
+ "language_model.model.layers.3.mlp.gate_up_proj.weight": "model.safetensors",
101
+ "language_model.model.layers.3.post_attention_layernorm.weight": "model.safetensors",
102
+ "language_model.model.layers.3.post_mlp_layernorm.weight": "model.safetensors",
103
+ "language_model.model.layers.3.post_self_attn_layernorm.weight": "model.safetensors",
104
+ "language_model.model.layers.3.self_attn.k_proj.weight": "model.safetensors",
105
+ "language_model.model.layers.3.self_attn.o_proj.weight": "model.safetensors",
106
+ "language_model.model.layers.3.self_attn.q_proj.weight": "model.safetensors",
107
+ "language_model.model.layers.3.self_attn.v_proj.weight": "model.safetensors",
108
+ "language_model.model.layers.4.input_layernorm.weight": "model.safetensors",
109
+ "language_model.model.layers.4.mlp.down_proj.weight": "model.safetensors",
110
+ "language_model.model.layers.4.mlp.gate_up_proj.weight": "model.safetensors",
111
+ "language_model.model.layers.4.post_attention_layernorm.weight": "model.safetensors",
112
+ "language_model.model.layers.4.post_mlp_layernorm.weight": "model.safetensors",
113
+ "language_model.model.layers.4.post_self_attn_layernorm.weight": "model.safetensors",
114
+ "language_model.model.layers.4.self_attn.k_proj.weight": "model.safetensors",
115
+ "language_model.model.layers.4.self_attn.o_proj.weight": "model.safetensors",
116
+ "language_model.model.layers.4.self_attn.q_proj.weight": "model.safetensors",
117
+ "language_model.model.layers.4.self_attn.v_proj.weight": "model.safetensors",
118
+ "language_model.model.layers.5.input_layernorm.weight": "model.safetensors",
119
+ "language_model.model.layers.5.mlp.down_proj.weight": "model.safetensors",
120
+ "language_model.model.layers.5.mlp.gate_up_proj.weight": "model.safetensors",
121
+ "language_model.model.layers.5.post_attention_layernorm.weight": "model.safetensors",
122
+ "language_model.model.layers.5.post_mlp_layernorm.weight": "model.safetensors",
123
+ "language_model.model.layers.5.post_self_attn_layernorm.weight": "model.safetensors",
124
+ "language_model.model.layers.5.self_attn.k_proj.weight": "model.safetensors",
125
+ "language_model.model.layers.5.self_attn.o_proj.weight": "model.safetensors",
126
+ "language_model.model.layers.5.self_attn.q_proj.weight": "model.safetensors",
127
+ "language_model.model.layers.5.self_attn.v_proj.weight": "model.safetensors",
128
+ "language_model.model.layers.6.input_layernorm.weight": "model.safetensors",
129
+ "language_model.model.layers.6.mlp.down_proj.weight": "model.safetensors",
130
+ "language_model.model.layers.6.mlp.gate_up_proj.weight": "model.safetensors",
131
+ "language_model.model.layers.6.post_attention_layernorm.weight": "model.safetensors",
132
+ "language_model.model.layers.6.post_mlp_layernorm.weight": "model.safetensors",
133
+ "language_model.model.layers.6.post_self_attn_layernorm.weight": "model.safetensors",
134
+ "language_model.model.layers.6.self_attn.k_proj.weight": "model.safetensors",
135
+ "language_model.model.layers.6.self_attn.o_proj.weight": "model.safetensors",
136
+ "language_model.model.layers.6.self_attn.q_proj.weight": "model.safetensors",
137
+ "language_model.model.layers.6.self_attn.v_proj.weight": "model.safetensors",
138
+ "language_model.model.layers.7.input_layernorm.weight": "model.safetensors",
139
+ "language_model.model.layers.7.mlp.down_proj.weight": "model.safetensors",
140
+ "language_model.model.layers.7.mlp.gate_up_proj.weight": "model.safetensors",
141
+ "language_model.model.layers.7.post_attention_layernorm.weight": "model.safetensors",
142
+ "language_model.model.layers.7.post_mlp_layernorm.weight": "model.safetensors",
143
+ "language_model.model.layers.7.post_self_attn_layernorm.weight": "model.safetensors",
144
+ "language_model.model.layers.7.self_attn.k_proj.weight": "model.safetensors",
145
+ "language_model.model.layers.7.self_attn.o_proj.weight": "model.safetensors",
146
+ "language_model.model.layers.7.self_attn.q_proj.weight": "model.safetensors",
147
+ "language_model.model.layers.7.self_attn.v_proj.weight": "model.safetensors",
148
+ "language_model.model.layers.8.input_layernorm.weight": "model.safetensors",
149
+ "language_model.model.layers.8.mlp.down_proj.weight": "model.safetensors",
150
+ "language_model.model.layers.8.mlp.gate_up_proj.weight": "model.safetensors",
151
+ "language_model.model.layers.8.post_attention_layernorm.weight": "model.safetensors",
152
+ "language_model.model.layers.8.post_mlp_layernorm.weight": "model.safetensors",
153
+ "language_model.model.layers.8.post_self_attn_layernorm.weight": "model.safetensors",
154
+ "language_model.model.layers.8.self_attn.k_proj.weight": "model.safetensors",
155
+ "language_model.model.layers.8.self_attn.o_proj.weight": "model.safetensors",
156
+ "language_model.model.layers.8.self_attn.q_proj.weight": "model.safetensors",
157
+ "language_model.model.layers.8.self_attn.v_proj.weight": "model.safetensors",
158
+ "language_model.model.layers.9.input_layernorm.weight": "model.safetensors",
159
+ "language_model.model.layers.9.mlp.down_proj.weight": "model.safetensors",
160
+ "language_model.model.layers.9.mlp.gate_up_proj.weight": "model.safetensors",
161
+ "language_model.model.layers.9.post_attention_layernorm.weight": "model.safetensors",
162
+ "language_model.model.layers.9.post_mlp_layernorm.weight": "model.safetensors",
163
+ "language_model.model.layers.9.post_self_attn_layernorm.weight": "model.safetensors",
164
+ "language_model.model.layers.9.self_attn.k_proj.weight": "model.safetensors",
165
+ "language_model.model.layers.9.self_attn.o_proj.weight": "model.safetensors",
166
+ "language_model.model.layers.9.self_attn.q_proj.weight": "model.safetensors",
167
+ "language_model.model.layers.9.self_attn.v_proj.weight": "model.safetensors",
168
+ "language_model.model.norm.weight": "model.safetensors",
169
+ "vision_tower.blocks.0.attn.k_norm.weight": "model.safetensors",
170
+ "vision_tower.blocks.0.attn.proj.bias": "model.safetensors",
171
+ "vision_tower.blocks.0.attn.proj.weight": "model.safetensors",
172
+ "vision_tower.blocks.0.attn.q_norm.weight": "model.safetensors",
173
+ "vision_tower.blocks.0.attn.qkv.bias": "model.safetensors",
174
+ "vision_tower.blocks.0.attn.qkv.weight": "model.safetensors",
175
+ "vision_tower.blocks.0.mlp.down_proj.bias": "model.safetensors",
176
+ "vision_tower.blocks.0.mlp.down_proj.weight": "model.safetensors",
177
+ "vision_tower.blocks.0.mlp.gate_proj.bias": "model.safetensors",
178
+ "vision_tower.blocks.0.mlp.gate_proj.weight": "model.safetensors",
179
+ "vision_tower.blocks.0.mlp.up_proj.bias": "model.safetensors",
180
+ "vision_tower.blocks.0.mlp.up_proj.weight": "model.safetensors",
181
+ "vision_tower.blocks.0.norm1.weight": "model.safetensors",
182
+ "vision_tower.blocks.0.norm2.weight": "model.safetensors",
183
+ "vision_tower.blocks.1.attn.k_norm.weight": "model.safetensors",
184
+ "vision_tower.blocks.1.attn.proj.bias": "model.safetensors",
185
+ "vision_tower.blocks.1.attn.proj.weight": "model.safetensors",
186
+ "vision_tower.blocks.1.attn.q_norm.weight": "model.safetensors",
187
+ "vision_tower.blocks.1.attn.qkv.bias": "model.safetensors",
188
+ "vision_tower.blocks.1.attn.qkv.weight": "model.safetensors",
189
+ "vision_tower.blocks.1.mlp.down_proj.bias": "model.safetensors",
190
+ "vision_tower.blocks.1.mlp.down_proj.weight": "model.safetensors",
191
+ "vision_tower.blocks.1.mlp.gate_proj.bias": "model.safetensors",
192
+ "vision_tower.blocks.1.mlp.gate_proj.weight": "model.safetensors",
193
+ "vision_tower.blocks.1.mlp.up_proj.bias": "model.safetensors",
194
+ "vision_tower.blocks.1.mlp.up_proj.weight": "model.safetensors",
195
+ "vision_tower.blocks.1.norm1.weight": "model.safetensors",
196
+ "vision_tower.blocks.1.norm2.weight": "model.safetensors",
197
+ "vision_tower.blocks.10.attn.k_norm.weight": "model.safetensors",
198
+ "vision_tower.blocks.10.attn.proj.bias": "model.safetensors",
199
+ "vision_tower.blocks.10.attn.proj.weight": "model.safetensors",
200
+ "vision_tower.blocks.10.attn.q_norm.weight": "model.safetensors",
201
+ "vision_tower.blocks.10.attn.qkv.bias": "model.safetensors",
202
+ "vision_tower.blocks.10.attn.qkv.weight": "model.safetensors",
203
+ "vision_tower.blocks.10.mlp.down_proj.bias": "model.safetensors",
204
+ "vision_tower.blocks.10.mlp.down_proj.weight": "model.safetensors",
205
+ "vision_tower.blocks.10.mlp.gate_proj.bias": "model.safetensors",
206
+ "vision_tower.blocks.10.mlp.gate_proj.weight": "model.safetensors",
207
+ "vision_tower.blocks.10.mlp.up_proj.bias": "model.safetensors",
208
+ "vision_tower.blocks.10.mlp.up_proj.weight": "model.safetensors",
209
+ "vision_tower.blocks.10.norm1.weight": "model.safetensors",
210
+ "vision_tower.blocks.10.norm2.weight": "model.safetensors",
211
+ "vision_tower.blocks.11.attn.k_norm.weight": "model.safetensors",
212
+ "vision_tower.blocks.11.attn.proj.bias": "model.safetensors",
213
+ "vision_tower.blocks.11.attn.proj.weight": "model.safetensors",
214
+ "vision_tower.blocks.11.attn.q_norm.weight": "model.safetensors",
215
+ "vision_tower.blocks.11.attn.qkv.bias": "model.safetensors",
216
+ "vision_tower.blocks.11.attn.qkv.weight": "model.safetensors",
217
+ "vision_tower.blocks.11.mlp.down_proj.bias": "model.safetensors",
218
+ "vision_tower.blocks.11.mlp.down_proj.weight": "model.safetensors",
219
+ "vision_tower.blocks.11.mlp.gate_proj.bias": "model.safetensors",
220
+ "vision_tower.blocks.11.mlp.gate_proj.weight": "model.safetensors",
221
+ "vision_tower.blocks.11.mlp.up_proj.bias": "model.safetensors",
222
+ "vision_tower.blocks.11.mlp.up_proj.weight": "model.safetensors",
223
+ "vision_tower.blocks.11.norm1.weight": "model.safetensors",
224
+ "vision_tower.blocks.11.norm2.weight": "model.safetensors",
225
+ "vision_tower.blocks.12.attn.k_norm.weight": "model.safetensors",
226
+ "vision_tower.blocks.12.attn.proj.bias": "model.safetensors",
227
+ "vision_tower.blocks.12.attn.proj.weight": "model.safetensors",
228
+ "vision_tower.blocks.12.attn.q_norm.weight": "model.safetensors",
229
+ "vision_tower.blocks.12.attn.qkv.bias": "model.safetensors",
230
+ "vision_tower.blocks.12.attn.qkv.weight": "model.safetensors",
231
+ "vision_tower.blocks.12.mlp.down_proj.bias": "model.safetensors",
232
+ "vision_tower.blocks.12.mlp.down_proj.weight": "model.safetensors",
233
+ "vision_tower.blocks.12.mlp.gate_proj.bias": "model.safetensors",
234
+ "vision_tower.blocks.12.mlp.gate_proj.weight": "model.safetensors",
235
+ "vision_tower.blocks.12.mlp.up_proj.bias": "model.safetensors",
236
+ "vision_tower.blocks.12.mlp.up_proj.weight": "model.safetensors",
237
+ "vision_tower.blocks.12.norm1.weight": "model.safetensors",
238
+ "vision_tower.blocks.12.norm2.weight": "model.safetensors",
239
+ "vision_tower.blocks.13.attn.k_norm.weight": "model.safetensors",
240
+ "vision_tower.blocks.13.attn.proj.bias": "model.safetensors",
241
+ "vision_tower.blocks.13.attn.proj.weight": "model.safetensors",
242
+ "vision_tower.blocks.13.attn.q_norm.weight": "model.safetensors",
243
+ "vision_tower.blocks.13.attn.qkv.bias": "model.safetensors",
244
+ "vision_tower.blocks.13.attn.qkv.weight": "model.safetensors",
245
+ "vision_tower.blocks.13.mlp.down_proj.bias": "model.safetensors",
246
+ "vision_tower.blocks.13.mlp.down_proj.weight": "model.safetensors",
247
+ "vision_tower.blocks.13.mlp.gate_proj.bias": "model.safetensors",
248
+ "vision_tower.blocks.13.mlp.gate_proj.weight": "model.safetensors",
249
+ "vision_tower.blocks.13.mlp.up_proj.bias": "model.safetensors",
250
+ "vision_tower.blocks.13.mlp.up_proj.weight": "model.safetensors",
251
+ "vision_tower.blocks.13.norm1.weight": "model.safetensors",
252
+ "vision_tower.blocks.13.norm2.weight": "model.safetensors",
253
+ "vision_tower.blocks.14.attn.k_norm.weight": "model.safetensors",
254
+ "vision_tower.blocks.14.attn.proj.bias": "model.safetensors",
255
+ "vision_tower.blocks.14.attn.proj.weight": "model.safetensors",
256
+ "vision_tower.blocks.14.attn.q_norm.weight": "model.safetensors",
257
+ "vision_tower.blocks.14.attn.qkv.bias": "model.safetensors",
258
+ "vision_tower.blocks.14.attn.qkv.weight": "model.safetensors",
259
+ "vision_tower.blocks.14.mlp.down_proj.bias": "model.safetensors",
260
+ "vision_tower.blocks.14.mlp.down_proj.weight": "model.safetensors",
261
+ "vision_tower.blocks.14.mlp.gate_proj.bias": "model.safetensors",
262
+ "vision_tower.blocks.14.mlp.gate_proj.weight": "model.safetensors",
263
+ "vision_tower.blocks.14.mlp.up_proj.bias": "model.safetensors",
264
+ "vision_tower.blocks.14.mlp.up_proj.weight": "model.safetensors",
265
+ "vision_tower.blocks.14.norm1.weight": "model.safetensors",
266
+ "vision_tower.blocks.14.norm2.weight": "model.safetensors",
267
+ "vision_tower.blocks.15.attn.k_norm.weight": "model.safetensors",
268
+ "vision_tower.blocks.15.attn.proj.bias": "model.safetensors",
269
+ "vision_tower.blocks.15.attn.proj.weight": "model.safetensors",
270
+ "vision_tower.blocks.15.attn.q_norm.weight": "model.safetensors",
271
+ "vision_tower.blocks.15.attn.qkv.bias": "model.safetensors",
272
+ "vision_tower.blocks.15.attn.qkv.weight": "model.safetensors",
273
+ "vision_tower.blocks.15.mlp.down_proj.bias": "model.safetensors",
274
+ "vision_tower.blocks.15.mlp.down_proj.weight": "model.safetensors",
275
+ "vision_tower.blocks.15.mlp.gate_proj.bias": "model.safetensors",
276
+ "vision_tower.blocks.15.mlp.gate_proj.weight": "model.safetensors",
277
+ "vision_tower.blocks.15.mlp.up_proj.bias": "model.safetensors",
278
+ "vision_tower.blocks.15.mlp.up_proj.weight": "model.safetensors",
279
+ "vision_tower.blocks.15.norm1.weight": "model.safetensors",
280
+ "vision_tower.blocks.15.norm2.weight": "model.safetensors",
281
+ "vision_tower.blocks.16.attn.k_norm.weight": "model.safetensors",
282
+ "vision_tower.blocks.16.attn.proj.bias": "model.safetensors",
283
+ "vision_tower.blocks.16.attn.proj.weight": "model.safetensors",
284
+ "vision_tower.blocks.16.attn.q_norm.weight": "model.safetensors",
285
+ "vision_tower.blocks.16.attn.qkv.bias": "model.safetensors",
286
+ "vision_tower.blocks.16.attn.qkv.weight": "model.safetensors",
287
+ "vision_tower.blocks.16.mlp.down_proj.bias": "model.safetensors",
288
+ "vision_tower.blocks.16.mlp.down_proj.weight": "model.safetensors",
289
+ "vision_tower.blocks.16.mlp.gate_proj.bias": "model.safetensors",
290
+ "vision_tower.blocks.16.mlp.gate_proj.weight": "model.safetensors",
291
+ "vision_tower.blocks.16.mlp.up_proj.bias": "model.safetensors",
292
+ "vision_tower.blocks.16.mlp.up_proj.weight": "model.safetensors",
293
+ "vision_tower.blocks.16.norm1.weight": "model.safetensors",
294
+ "vision_tower.blocks.16.norm2.weight": "model.safetensors",
295
+ "vision_tower.blocks.17.attn.k_norm.weight": "model.safetensors",
296
+ "vision_tower.blocks.17.attn.proj.bias": "model.safetensors",
297
+ "vision_tower.blocks.17.attn.proj.weight": "model.safetensors",
298
+ "vision_tower.blocks.17.attn.q_norm.weight": "model.safetensors",
299
+ "vision_tower.blocks.17.attn.qkv.bias": "model.safetensors",
300
+ "vision_tower.blocks.17.attn.qkv.weight": "model.safetensors",
301
+ "vision_tower.blocks.17.mlp.down_proj.bias": "model.safetensors",
302
+ "vision_tower.blocks.17.mlp.down_proj.weight": "model.safetensors",
303
+ "vision_tower.blocks.17.mlp.gate_proj.bias": "model.safetensors",
304
+ "vision_tower.blocks.17.mlp.gate_proj.weight": "model.safetensors",
305
+ "vision_tower.blocks.17.mlp.up_proj.bias": "model.safetensors",
306
+ "vision_tower.blocks.17.mlp.up_proj.weight": "model.safetensors",
307
+ "vision_tower.blocks.17.norm1.weight": "model.safetensors",
308
+ "vision_tower.blocks.17.norm2.weight": "model.safetensors",
309
+ "vision_tower.blocks.18.attn.k_norm.weight": "model.safetensors",
310
+ "vision_tower.blocks.18.attn.proj.bias": "model.safetensors",
311
+ "vision_tower.blocks.18.attn.proj.weight": "model.safetensors",
312
+ "vision_tower.blocks.18.attn.q_norm.weight": "model.safetensors",
313
+ "vision_tower.blocks.18.attn.qkv.bias": "model.safetensors",
314
+ "vision_tower.blocks.18.attn.qkv.weight": "model.safetensors",
315
+ "vision_tower.blocks.18.mlp.down_proj.bias": "model.safetensors",
316
+ "vision_tower.blocks.18.mlp.down_proj.weight": "model.safetensors",
317
+ "vision_tower.blocks.18.mlp.gate_proj.bias": "model.safetensors",
318
+ "vision_tower.blocks.18.mlp.gate_proj.weight": "model.safetensors",
319
+ "vision_tower.blocks.18.mlp.up_proj.bias": "model.safetensors",
320
+ "vision_tower.blocks.18.mlp.up_proj.weight": "model.safetensors",
321
+ "vision_tower.blocks.18.norm1.weight": "model.safetensors",
322
+ "vision_tower.blocks.18.norm2.weight": "model.safetensors",
323
+ "vision_tower.blocks.19.attn.k_norm.weight": "model.safetensors",
324
+ "vision_tower.blocks.19.attn.proj.bias": "model.safetensors",
325
+ "vision_tower.blocks.19.attn.proj.weight": "model.safetensors",
326
+ "vision_tower.blocks.19.attn.q_norm.weight": "model.safetensors",
327
+ "vision_tower.blocks.19.attn.qkv.bias": "model.safetensors",
328
+ "vision_tower.blocks.19.attn.qkv.weight": "model.safetensors",
329
+ "vision_tower.blocks.19.mlp.down_proj.bias": "model.safetensors",
330
+ "vision_tower.blocks.19.mlp.down_proj.weight": "model.safetensors",
331
+ "vision_tower.blocks.19.mlp.gate_proj.bias": "model.safetensors",
332
+ "vision_tower.blocks.19.mlp.gate_proj.weight": "model.safetensors",
333
+ "vision_tower.blocks.19.mlp.up_proj.bias": "model.safetensors",
334
+ "vision_tower.blocks.19.mlp.up_proj.weight": "model.safetensors",
335
+ "vision_tower.blocks.19.norm1.weight": "model.safetensors",
336
+ "vision_tower.blocks.19.norm2.weight": "model.safetensors",
337
+ "vision_tower.blocks.2.attn.k_norm.weight": "model.safetensors",
338
+ "vision_tower.blocks.2.attn.proj.bias": "model.safetensors",
339
+ "vision_tower.blocks.2.attn.proj.weight": "model.safetensors",
340
+ "vision_tower.blocks.2.attn.q_norm.weight": "model.safetensors",
341
+ "vision_tower.blocks.2.attn.qkv.bias": "model.safetensors",
342
+ "vision_tower.blocks.2.attn.qkv.weight": "model.safetensors",
343
+ "vision_tower.blocks.2.mlp.down_proj.bias": "model.safetensors",
344
+ "vision_tower.blocks.2.mlp.down_proj.weight": "model.safetensors",
345
+ "vision_tower.blocks.2.mlp.gate_proj.bias": "model.safetensors",
346
+ "vision_tower.blocks.2.mlp.gate_proj.weight": "model.safetensors",
347
+ "vision_tower.blocks.2.mlp.up_proj.bias": "model.safetensors",
348
+ "vision_tower.blocks.2.mlp.up_proj.weight": "model.safetensors",
349
+ "vision_tower.blocks.2.norm1.weight": "model.safetensors",
350
+ "vision_tower.blocks.2.norm2.weight": "model.safetensors",
351
+ "vision_tower.blocks.20.attn.k_norm.weight": "model.safetensors",
352
+ "vision_tower.blocks.20.attn.proj.bias": "model.safetensors",
353
+ "vision_tower.blocks.20.attn.proj.weight": "model.safetensors",
354
+ "vision_tower.blocks.20.attn.q_norm.weight": "model.safetensors",
355
+ "vision_tower.blocks.20.attn.qkv.bias": "model.safetensors",
356
+ "vision_tower.blocks.20.attn.qkv.weight": "model.safetensors",
357
+ "vision_tower.blocks.20.mlp.down_proj.bias": "model.safetensors",
358
+ "vision_tower.blocks.20.mlp.down_proj.weight": "model.safetensors",
359
+ "vision_tower.blocks.20.mlp.gate_proj.bias": "model.safetensors",
360
+ "vision_tower.blocks.20.mlp.gate_proj.weight": "model.safetensors",
361
+ "vision_tower.blocks.20.mlp.up_proj.bias": "model.safetensors",
362
+ "vision_tower.blocks.20.mlp.up_proj.weight": "model.safetensors",
363
+ "vision_tower.blocks.20.norm1.weight": "model.safetensors",
364
+ "vision_tower.blocks.20.norm2.weight": "model.safetensors",
365
+ "vision_tower.blocks.21.attn.k_norm.weight": "model.safetensors",
366
+ "vision_tower.blocks.21.attn.proj.bias": "model.safetensors",
367
+ "vision_tower.blocks.21.attn.proj.weight": "model.safetensors",
368
+ "vision_tower.blocks.21.attn.q_norm.weight": "model.safetensors",
369
+ "vision_tower.blocks.21.attn.qkv.bias": "model.safetensors",
370
+ "vision_tower.blocks.21.attn.qkv.weight": "model.safetensors",
371
+ "vision_tower.blocks.21.mlp.down_proj.bias": "model.safetensors",
372
+ "vision_tower.blocks.21.mlp.down_proj.weight": "model.safetensors",
373
+ "vision_tower.blocks.21.mlp.gate_proj.bias": "model.safetensors",
374
+ "vision_tower.blocks.21.mlp.gate_proj.weight": "model.safetensors",
375
+ "vision_tower.blocks.21.mlp.up_proj.bias": "model.safetensors",
376
+ "vision_tower.blocks.21.mlp.up_proj.weight": "model.safetensors",
377
+ "vision_tower.blocks.21.norm1.weight": "model.safetensors",
378
+ "vision_tower.blocks.21.norm2.weight": "model.safetensors",
379
+ "vision_tower.blocks.22.attn.k_norm.weight": "model.safetensors",
380
+ "vision_tower.blocks.22.attn.proj.bias": "model.safetensors",
381
+ "vision_tower.blocks.22.attn.proj.weight": "model.safetensors",
382
+ "vision_tower.blocks.22.attn.q_norm.weight": "model.safetensors",
383
+ "vision_tower.blocks.22.attn.qkv.bias": "model.safetensors",
384
+ "vision_tower.blocks.22.attn.qkv.weight": "model.safetensors",
385
+ "vision_tower.blocks.22.mlp.down_proj.bias": "model.safetensors",
386
+ "vision_tower.blocks.22.mlp.down_proj.weight": "model.safetensors",
387
+ "vision_tower.blocks.22.mlp.gate_proj.bias": "model.safetensors",
388
+ "vision_tower.blocks.22.mlp.gate_proj.weight": "model.safetensors",
389
+ "vision_tower.blocks.22.mlp.up_proj.bias": "model.safetensors",
390
+ "vision_tower.blocks.22.mlp.up_proj.weight": "model.safetensors",
391
+ "vision_tower.blocks.22.norm1.weight": "model.safetensors",
392
+ "vision_tower.blocks.22.norm2.weight": "model.safetensors",
393
+ "vision_tower.blocks.23.attn.k_norm.weight": "model.safetensors",
394
+ "vision_tower.blocks.23.attn.proj.bias": "model.safetensors",
395
+ "vision_tower.blocks.23.attn.proj.weight": "model.safetensors",
396
+ "vision_tower.blocks.23.attn.q_norm.weight": "model.safetensors",
397
+ "vision_tower.blocks.23.attn.qkv.bias": "model.safetensors",
398
+ "vision_tower.blocks.23.attn.qkv.weight": "model.safetensors",
399
+ "vision_tower.blocks.23.mlp.down_proj.bias": "model.safetensors",
400
+ "vision_tower.blocks.23.mlp.down_proj.weight": "model.safetensors",
401
+ "vision_tower.blocks.23.mlp.gate_proj.bias": "model.safetensors",
402
+ "vision_tower.blocks.23.mlp.gate_proj.weight": "model.safetensors",
403
+ "vision_tower.blocks.23.mlp.up_proj.bias": "model.safetensors",
404
+ "vision_tower.blocks.23.mlp.up_proj.weight": "model.safetensors",
405
+ "vision_tower.blocks.23.norm1.weight": "model.safetensors",
406
+ "vision_tower.blocks.23.norm2.weight": "model.safetensors",
407
+ "vision_tower.blocks.3.attn.k_norm.weight": "model.safetensors",
408
+ "vision_tower.blocks.3.attn.proj.bias": "model.safetensors",
409
+ "vision_tower.blocks.3.attn.proj.weight": "model.safetensors",
410
+ "vision_tower.blocks.3.attn.q_norm.weight": "model.safetensors",
411
+ "vision_tower.blocks.3.attn.qkv.bias": "model.safetensors",
412
+ "vision_tower.blocks.3.attn.qkv.weight": "model.safetensors",
413
+ "vision_tower.blocks.3.mlp.down_proj.bias": "model.safetensors",
414
+ "vision_tower.blocks.3.mlp.down_proj.weight": "model.safetensors",
415
+ "vision_tower.blocks.3.mlp.gate_proj.bias": "model.safetensors",
416
+ "vision_tower.blocks.3.mlp.gate_proj.weight": "model.safetensors",
417
+ "vision_tower.blocks.3.mlp.up_proj.bias": "model.safetensors",
418
+ "vision_tower.blocks.3.mlp.up_proj.weight": "model.safetensors",
419
+ "vision_tower.blocks.3.norm1.weight": "model.safetensors",
420
+ "vision_tower.blocks.3.norm2.weight": "model.safetensors",
421
+ "vision_tower.blocks.4.attn.k_norm.weight": "model.safetensors",
422
+ "vision_tower.blocks.4.attn.proj.bias": "model.safetensors",
423
+ "vision_tower.blocks.4.attn.proj.weight": "model.safetensors",
424
+ "vision_tower.blocks.4.attn.q_norm.weight": "model.safetensors",
425
+ "vision_tower.blocks.4.attn.qkv.bias": "model.safetensors",
426
+ "vision_tower.blocks.4.attn.qkv.weight": "model.safetensors",
427
+ "vision_tower.blocks.4.mlp.down_proj.bias": "model.safetensors",
428
+ "vision_tower.blocks.4.mlp.down_proj.weight": "model.safetensors",
429
+ "vision_tower.blocks.4.mlp.gate_proj.bias": "model.safetensors",
430
+ "vision_tower.blocks.4.mlp.gate_proj.weight": "model.safetensors",
431
+ "vision_tower.blocks.4.mlp.up_proj.bias": "model.safetensors",
432
+ "vision_tower.blocks.4.mlp.up_proj.weight": "model.safetensors",
433
+ "vision_tower.blocks.4.norm1.weight": "model.safetensors",
434
+ "vision_tower.blocks.4.norm2.weight": "model.safetensors",
435
+ "vision_tower.blocks.5.attn.k_norm.weight": "model.safetensors",
436
+ "vision_tower.blocks.5.attn.proj.bias": "model.safetensors",
437
+ "vision_tower.blocks.5.attn.proj.weight": "model.safetensors",
438
+ "vision_tower.blocks.5.attn.q_norm.weight": "model.safetensors",
439
+ "vision_tower.blocks.5.attn.qkv.bias": "model.safetensors",
440
+ "vision_tower.blocks.5.attn.qkv.weight": "model.safetensors",
441
+ "vision_tower.blocks.5.mlp.down_proj.bias": "model.safetensors",
442
+ "vision_tower.blocks.5.mlp.down_proj.weight": "model.safetensors",
443
+ "vision_tower.blocks.5.mlp.gate_proj.bias": "model.safetensors",
444
+ "vision_tower.blocks.5.mlp.gate_proj.weight": "model.safetensors",
445
+ "vision_tower.blocks.5.mlp.up_proj.bias": "model.safetensors",
446
+ "vision_tower.blocks.5.mlp.up_proj.weight": "model.safetensors",
447
+ "vision_tower.blocks.5.norm1.weight": "model.safetensors",
448
+ "vision_tower.blocks.5.norm2.weight": "model.safetensors",
449
+ "vision_tower.blocks.6.attn.k_norm.weight": "model.safetensors",
450
+ "vision_tower.blocks.6.attn.proj.bias": "model.safetensors",
451
+ "vision_tower.blocks.6.attn.proj.weight": "model.safetensors",
452
+ "vision_tower.blocks.6.attn.q_norm.weight": "model.safetensors",
453
+ "vision_tower.blocks.6.attn.qkv.bias": "model.safetensors",
454
+ "vision_tower.blocks.6.attn.qkv.weight": "model.safetensors",
455
+ "vision_tower.blocks.6.mlp.down_proj.bias": "model.safetensors",
456
+ "vision_tower.blocks.6.mlp.down_proj.weight": "model.safetensors",
457
+ "vision_tower.blocks.6.mlp.gate_proj.bias": "model.safetensors",
458
+ "vision_tower.blocks.6.mlp.gate_proj.weight": "model.safetensors",
459
+ "vision_tower.blocks.6.mlp.up_proj.bias": "model.safetensors",
460
+ "vision_tower.blocks.6.mlp.up_proj.weight": "model.safetensors",
461
+ "vision_tower.blocks.6.norm1.weight": "model.safetensors",
462
+ "vision_tower.blocks.6.norm2.weight": "model.safetensors",
463
+ "vision_tower.blocks.7.attn.k_norm.weight": "model.safetensors",
464
+ "vision_tower.blocks.7.attn.proj.bias": "model.safetensors",
465
+ "vision_tower.blocks.7.attn.proj.weight": "model.safetensors",
466
+ "vision_tower.blocks.7.attn.q_norm.weight": "model.safetensors",
467
+ "vision_tower.blocks.7.attn.qkv.bias": "model.safetensors",
468
+ "vision_tower.blocks.7.attn.qkv.weight": "model.safetensors",
469
+ "vision_tower.blocks.7.mlp.down_proj.bias": "model.safetensors",
470
+ "vision_tower.blocks.7.mlp.down_proj.weight": "model.safetensors",
471
+ "vision_tower.blocks.7.mlp.gate_proj.bias": "model.safetensors",
472
+ "vision_tower.blocks.7.mlp.gate_proj.weight": "model.safetensors",
473
+ "vision_tower.blocks.7.mlp.up_proj.bias": "model.safetensors",
474
+ "vision_tower.blocks.7.mlp.up_proj.weight": "model.safetensors",
475
+ "vision_tower.blocks.7.norm1.weight": "model.safetensors",
476
+ "vision_tower.blocks.7.norm2.weight": "model.safetensors",
477
+ "vision_tower.blocks.8.attn.k_norm.weight": "model.safetensors",
478
+ "vision_tower.blocks.8.attn.proj.bias": "model.safetensors",
479
+ "vision_tower.blocks.8.attn.proj.weight": "model.safetensors",
480
+ "vision_tower.blocks.8.attn.q_norm.weight": "model.safetensors",
481
+ "vision_tower.blocks.8.attn.qkv.bias": "model.safetensors",
482
+ "vision_tower.blocks.8.attn.qkv.weight": "model.safetensors",
483
+ "vision_tower.blocks.8.mlp.down_proj.bias": "model.safetensors",
484
+ "vision_tower.blocks.8.mlp.down_proj.weight": "model.safetensors",
485
+ "vision_tower.blocks.8.mlp.gate_proj.bias": "model.safetensors",
486
+ "vision_tower.blocks.8.mlp.gate_proj.weight": "model.safetensors",
487
+ "vision_tower.blocks.8.mlp.up_proj.bias": "model.safetensors",
488
+ "vision_tower.blocks.8.mlp.up_proj.weight": "model.safetensors",
489
+ "vision_tower.blocks.8.norm1.weight": "model.safetensors",
490
+ "vision_tower.blocks.8.norm2.weight": "model.safetensors",
491
+ "vision_tower.blocks.9.attn.k_norm.weight": "model.safetensors",
492
+ "vision_tower.blocks.9.attn.proj.bias": "model.safetensors",
493
+ "vision_tower.blocks.9.attn.proj.weight": "model.safetensors",
494
+ "vision_tower.blocks.9.attn.q_norm.weight": "model.safetensors",
495
+ "vision_tower.blocks.9.attn.qkv.bias": "model.safetensors",
496
+ "vision_tower.blocks.9.attn.qkv.weight": "model.safetensors",
497
+ "vision_tower.blocks.9.mlp.down_proj.bias": "model.safetensors",
498
+ "vision_tower.blocks.9.mlp.down_proj.weight": "model.safetensors",
499
+ "vision_tower.blocks.9.mlp.gate_proj.bias": "model.safetensors",
500
+ "vision_tower.blocks.9.mlp.gate_proj.weight": "model.safetensors",
501
+ "vision_tower.blocks.9.mlp.up_proj.bias": "model.safetensors",
502
+ "vision_tower.blocks.9.mlp.up_proj.weight": "model.safetensors",
503
+ "vision_tower.blocks.9.norm1.weight": "model.safetensors",
504
+ "vision_tower.blocks.9.norm2.weight": "model.safetensors",
505
+ "vision_tower.downsample.bias": "model.safetensors",
506
+ "vision_tower.downsample.weight": "model.safetensors",
507
+ "vision_tower.merger.down_proj.weight": "model.safetensors",
508
+ "vision_tower.merger.gate_proj.weight": "model.safetensors",
509
+ "vision_tower.merger.post_projection_norm.bias": "model.safetensors",
510
+ "vision_tower.merger.post_projection_norm.weight": "model.safetensors",
511
+ "vision_tower.merger.proj.weight": "model.safetensors",
512
+ "vision_tower.merger.up_proj.weight": "model.safetensors",
513
+ "vision_tower.patch_embed.proj.bias": "model.safetensors",
514
+ "vision_tower.patch_embed.proj.weight": "model.safetensors",
515
+ "vision_tower.post_layernorm.weight": "model.safetensors"
516
+ }
517
+ }
preprocessor_config.json ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "size": {"shortest_edge": 12544, "longest_edge": 9633792},
3
+ "do_rescale": true,
4
+ "patch_size": 14,
5
+ "temporal_patch_size": 2,
6
+ "merge_size": 2,
7
+ "image_mean": [0.48145466, 0.4578275, 0.40821073],
8
+ "image_std": [0.26862954, 0.26130258, 0.27577711],
9
+ "image_processor_type": "Glm46VImageProcessor",
10
+ "processor_class": "Glm46VProcessor"
11
+ }
processor_config.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "image_processor": {
3
+ "data_format": "channels_first",
4
+ "do_convert_rgb": true,
5
+ "do_normalize": true,
6
+ "do_rescale": true,
7
+ "do_resize": true,
8
+ "image_mean": [
9
+ 0.48145466,
10
+ 0.4578275,
11
+ 0.40821073
12
+ ],
13
+ "image_processor_type": "Glm46VImageProcessorFast",
14
+ "image_std": [
15
+ 0.26862954,
16
+ 0.26130258,
17
+ 0.27577711
18
+ ],
19
+ "merge_size": 2,
20
+ "patch_size": 14,
21
+ "resample": 3,
22
+ "rescale_factor": 0.00392156862745098,
23
+ "size": {
24
+ "longest_edge": 9633792,
25
+ "shortest_edge": 12544
26
+ },
27
+ "temporal_patch_size": 2
28
+ },
29
+ "processor_class": "GlmOcrProcessor"
30
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,49 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "backend": "tokenizers",
3
+ "clean_up_tokenization_spaces": false,
4
+ "eos_token": "<|endoftext|>",
5
+ "extra_special_tokens": [
6
+ "<|endoftext|>",
7
+ "[MASK]",
8
+ "[gMASK]",
9
+ "[sMASK]",
10
+ "<sop>",
11
+ "<eop>",
12
+ "<|system|>",
13
+ "<|user|>",
14
+ "<|assistant|>",
15
+ "<|observation|>",
16
+ "<|begin_of_image|>",
17
+ "<|end_of_image|>",
18
+ "<|begin_of_video|>",
19
+ "<|end_of_video|>",
20
+ "<|begin_of_audio|>",
21
+ "<|end_of_audio|>",
22
+ "<|begin_of_transcription|>",
23
+ "<|end_of_transcription|>",
24
+ "<|code_prefix|>",
25
+ "<|code_middle|>",
26
+ "<|code_suffix|>",
27
+ "<think>",
28
+ "</think>",
29
+ "<tool_call>",
30
+ "</tool_call>",
31
+ "<tool_response>",
32
+ "</tool_response>",
33
+ "<arg_key>",
34
+ "</arg_key>",
35
+ "<arg_value>",
36
+ "</arg_value>",
37
+ "/nothink",
38
+ "<|begin_of_box|>",
39
+ "<|end_of_box|>",
40
+ "<|image|>",
41
+ "<|video|>"
42
+ ],
43
+ "is_local": true,
44
+ "model_max_length": 655380,
45
+ "pad_token": "<|endoftext|>",
46
+ "padding_side": "left",
47
+ "processor_class": "GlmOcrProcessor",
48
+ "tokenizer_class": "TokenizersBackend"
49
+ }