Hugging Face's logo

rooty2020
/
dinov3-vit7b16-spatialboost

Safetensors
dinov3_vit
Model card Files
xet
Community
rooty2020 commited on
Commit
5243fac
·
verified ·
1 Parent(s): ab52d11

Upload models

Browse files
Files changed (3) hide show
  1. config.json +28 -0
  2. model.safetensors +3 -0
  3. model.safetensors.index.json +934 -0
config.json ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "attention_dropout": 0.0,
3
+ "drop_path_rate": 0.0,
4
+ "hidden_act": "silu",
5
+ "hidden_size": 4096,
6
+ "image_size": 224,
7
+ "initializer_range": 0.02,
8
+ "intermediate_size": 8192,
9
+ "key_bias": false,
10
+ "layer_norm_eps": 1e-05,
11
+ "layerscale_value": 1.0,
12
+ "mlp_bias": true,
13
+ "model_type": "dinov3_vit",
14
+ "num_attention_heads": 32,
15
+ "num_channels": 3,
16
+ "num_hidden_layers": 40,
17
+ "num_register_tokens": 4,
18
+ "patch_size": 16,
19
+ "pos_embed_jitter": null,
20
+ "pos_embed_rescale": 2.0,
21
+ "pos_embed_shift": null,
22
+ "proj_bias": true,
23
+ "query_bias": false,
24
+ "rope_theta": 100.0,
25
+ "transformers_version": "4.49.0",
26
+ "use_gated_mlp": true,
27
+ "value_bias": false
28
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fdc9c4a188f2ba320efd0548d0ffa2523893107af0a81612886508eaed3cf218
3
+ size 37602314888
model.safetensors.index.json ADDED
@@ -0,0 +1,934 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 37602214048
4
+ },
5
+ "weight_map": {
6
+ "embeddings.cls_token": "vision_encoder.safetensors",
7
+ "embeddings.mask_token": "vision_encoder.safetensors",
8
+ "embeddings.patch_embeddings.bias": "vision_encoder.safetensors",
9
+ "embeddings.patch_embeddings.weight": "vision_encoder.safetensors",
10
+ "embeddings.register_tokens": "vision_encoder.safetensors",
11
+ "layer.0.attention.alpha_logit": "vision_encoder.safetensors",
12
+ "layer.0.attention.attn_base.k_proj.weight": "vision_encoder.safetensors",
13
+ "layer.0.attention.attn_base.o_proj.bias": "vision_encoder.safetensors",
14
+ "layer.0.attention.attn_base.o_proj.weight": "vision_encoder.safetensors",
15
+ "layer.0.attention.attn_base.q_proj.weight": "vision_encoder.safetensors",
16
+ "layer.0.attention.attn_base.v_proj.weight": "vision_encoder.safetensors",
17
+ "layer.0.attention.attn_plus.k_proj.weight": "vision_encoder.safetensors",
18
+ "layer.0.attention.attn_plus.o_proj.bias": "vision_encoder.safetensors",
19
+ "layer.0.attention.attn_plus.o_proj.weight": "vision_encoder.safetensors",
20
+ "layer.0.attention.attn_plus.q_proj.weight": "vision_encoder.safetensors",
21
+ "layer.0.attention.attn_plus.v_proj.weight": "vision_encoder.safetensors",
22
+ "layer.0.layer_scale1.lambda1": "vision_encoder.safetensors",
23
+ "layer.0.layer_scale2.lambda1": "vision_encoder.safetensors",
24
+ "layer.0.mlp.down_proj.bias": "vision_encoder.safetensors",
25
+ "layer.0.mlp.down_proj.weight": "vision_encoder.safetensors",
26
+ "layer.0.mlp.gate_proj.bias": "vision_encoder.safetensors",
27
+ "layer.0.mlp.gate_proj.weight": "vision_encoder.safetensors",
28
+ "layer.0.mlp.up_proj.bias": "vision_encoder.safetensors",
29
+ "layer.0.mlp.up_proj.weight": "vision_encoder.safetensors",
30
+ "layer.0.norm1.bias": "vision_encoder.safetensors",
31
+ "layer.0.norm1.weight": "vision_encoder.safetensors",
32
+ "layer.0.norm2.bias": "vision_encoder.safetensors",
33
+ "layer.0.norm2.weight": "vision_encoder.safetensors",
34
+ "layer.1.attention.alpha_logit": "vision_encoder.safetensors",
35
+ "layer.1.attention.attn_base.k_proj.weight": "vision_encoder.safetensors",
36
+ "layer.1.attention.attn_base.o_proj.bias": "vision_encoder.safetensors",
37
+ "layer.1.attention.attn_base.o_proj.weight": "vision_encoder.safetensors",
38
+ "layer.1.attention.attn_base.q_proj.weight": "vision_encoder.safetensors",
39
+ "layer.1.attention.attn_base.v_proj.weight": "vision_encoder.safetensors",
40
+ "layer.1.attention.attn_plus.k_proj.weight": "vision_encoder.safetensors",
41
+ "layer.1.attention.attn_plus.o_proj.bias": "vision_encoder.safetensors",
42
+ "layer.1.attention.attn_plus.o_proj.weight": "vision_encoder.safetensors",
43
+ "layer.1.attention.attn_plus.q_proj.weight": "vision_encoder.safetensors",
44
+ "layer.1.attention.attn_plus.v_proj.weight": "vision_encoder.safetensors",
45
+ "layer.1.layer_scale1.lambda1": "vision_encoder.safetensors",
46
+ "layer.1.layer_scale2.lambda1": "vision_encoder.safetensors",
47
+ "layer.1.mlp.down_proj.bias": "vision_encoder.safetensors",
48
+ "layer.1.mlp.down_proj.weight": "vision_encoder.safetensors",
49
+ "layer.1.mlp.gate_proj.bias": "vision_encoder.safetensors",
50
+ "layer.1.mlp.gate_proj.weight": "vision_encoder.safetensors",
51
+ "layer.1.mlp.up_proj.bias": "vision_encoder.safetensors",
52
+ "layer.1.mlp.up_proj.weight": "vision_encoder.safetensors",
53
+ "layer.1.norm1.bias": "vision_encoder.safetensors",
54
+ "layer.1.norm1.weight": "vision_encoder.safetensors",
55
+ "layer.1.norm2.bias": "vision_encoder.safetensors",
56
+ "layer.1.norm2.weight": "vision_encoder.safetensors",
57
+ "layer.10.attention.alpha_logit": "vision_encoder.safetensors",
58
+ "layer.10.attention.attn_base.k_proj.weight": "vision_encoder.safetensors",
59
+ "layer.10.attention.attn_base.o_proj.bias": "vision_encoder.safetensors",
60
+ "layer.10.attention.attn_base.o_proj.weight": "vision_encoder.safetensors",
61
+ "layer.10.attention.attn_base.q_proj.weight": "vision_encoder.safetensors",
62
+ "layer.10.attention.attn_base.v_proj.weight": "vision_encoder.safetensors",
63
+ "layer.10.attention.attn_plus.k_proj.weight": "vision_encoder.safetensors",
64
+ "layer.10.attention.attn_plus.o_proj.bias": "vision_encoder.safetensors",
65
+ "layer.10.attention.attn_plus.o_proj.weight": "vision_encoder.safetensors",
66
+ "layer.10.attention.attn_plus.q_proj.weight": "vision_encoder.safetensors",
67
+ "layer.10.attention.attn_plus.v_proj.weight": "vision_encoder.safetensors",
68
+ "layer.10.layer_scale1.lambda1": "vision_encoder.safetensors",
69
+ "layer.10.layer_scale2.lambda1": "vision_encoder.safetensors",
70
+ "layer.10.mlp.down_proj.bias": "vision_encoder.safetensors",
71
+ "layer.10.mlp.down_proj.weight": "vision_encoder.safetensors",
72
+ "layer.10.mlp.gate_proj.bias": "vision_encoder.safetensors",
73
+ "layer.10.mlp.gate_proj.weight": "vision_encoder.safetensors",
74
+ "layer.10.mlp.up_proj.bias": "vision_encoder.safetensors",
75
+ "layer.10.mlp.up_proj.weight": "vision_encoder.safetensors",
76
+ "layer.10.norm1.bias": "vision_encoder.safetensors",
77
+ "layer.10.norm1.weight": "vision_encoder.safetensors",
78
+ "layer.10.norm2.bias": "vision_encoder.safetensors",
79
+ "layer.10.norm2.weight": "vision_encoder.safetensors",
80
+ "layer.11.attention.alpha_logit": "vision_encoder.safetensors",
81
+ "layer.11.attention.attn_base.k_proj.weight": "vision_encoder.safetensors",
82
+ "layer.11.attention.attn_base.o_proj.bias": "vision_encoder.safetensors",
83
+ "layer.11.attention.attn_base.o_proj.weight": "vision_encoder.safetensors",
84
+ "layer.11.attention.attn_base.q_proj.weight": "vision_encoder.safetensors",
85
+ "layer.11.attention.attn_base.v_proj.weight": "vision_encoder.safetensors",
86
+ "layer.11.attention.attn_plus.k_proj.weight": "vision_encoder.safetensors",
87
+ "layer.11.attention.attn_plus.o_proj.bias": "vision_encoder.safetensors",
88
+ "layer.11.attention.attn_plus.o_proj.weight": "vision_encoder.safetensors",
89
+ "layer.11.attention.attn_plus.q_proj.weight": "vision_encoder.safetensors",
90
+ "layer.11.attention.attn_plus.v_proj.weight": "vision_encoder.safetensors",
91
+ "layer.11.layer_scale1.lambda1": "vision_encoder.safetensors",
92
+ "layer.11.layer_scale2.lambda1": "vision_encoder.safetensors",
93
+ "layer.11.mlp.down_proj.bias": "vision_encoder.safetensors",
94
+ "layer.11.mlp.down_proj.weight": "vision_encoder.safetensors",
95
+ "layer.11.mlp.gate_proj.bias": "vision_encoder.safetensors",
96
+ "layer.11.mlp.gate_proj.weight": "vision_encoder.safetensors",
97
+ "layer.11.mlp.up_proj.bias": "vision_encoder.safetensors",
98
+ "layer.11.mlp.up_proj.weight": "vision_encoder.safetensors",
99
+ "layer.11.norm1.bias": "vision_encoder.safetensors",
100
+ "layer.11.norm1.weight": "vision_encoder.safetensors",
101
+ "layer.11.norm2.bias": "vision_encoder.safetensors",
102
+ "layer.11.norm2.weight": "vision_encoder.safetensors",
103
+ "layer.12.attention.alpha_logit": "vision_encoder.safetensors",
104
+ "layer.12.attention.attn_base.k_proj.weight": "vision_encoder.safetensors",
105
+ "layer.12.attention.attn_base.o_proj.bias": "vision_encoder.safetensors",
106
+ "layer.12.attention.attn_base.o_proj.weight": "vision_encoder.safetensors",
107
+ "layer.12.attention.attn_base.q_proj.weight": "vision_encoder.safetensors",
108
+ "layer.12.attention.attn_base.v_proj.weight": "vision_encoder.safetensors",
109
+ "layer.12.attention.attn_plus.k_proj.weight": "vision_encoder.safetensors",
110
+ "layer.12.attention.attn_plus.o_proj.bias": "vision_encoder.safetensors",
111
+ "layer.12.attention.attn_plus.o_proj.weight": "vision_encoder.safetensors",
112
+ "layer.12.attention.attn_plus.q_proj.weight": "vision_encoder.safetensors",
113
+ "layer.12.attention.attn_plus.v_proj.weight": "vision_encoder.safetensors",
114
+ "layer.12.layer_scale1.lambda1": "vision_encoder.safetensors",
115
+ "layer.12.layer_scale2.lambda1": "vision_encoder.safetensors",
116
+ "layer.12.mlp.down_proj.bias": "vision_encoder.safetensors",
117
+ "layer.12.mlp.down_proj.weight": "vision_encoder.safetensors",
118
+ "layer.12.mlp.gate_proj.bias": "vision_encoder.safetensors",
119
+ "layer.12.mlp.gate_proj.weight": "vision_encoder.safetensors",
120
+ "layer.12.mlp.up_proj.bias": "vision_encoder.safetensors",
121
+ "layer.12.mlp.up_proj.weight": "vision_encoder.safetensors",
122
+ "layer.12.norm1.bias": "vision_encoder.safetensors",
123
+ "layer.12.norm1.weight": "vision_encoder.safetensors",
124
+ "layer.12.norm2.bias": "vision_encoder.safetensors",
125
+ "layer.12.norm2.weight": "vision_encoder.safetensors",
126
+ "layer.13.attention.alpha_logit": "vision_encoder.safetensors",
127
+ "layer.13.attention.attn_base.k_proj.weight": "vision_encoder.safetensors",
128
+ "layer.13.attention.attn_base.o_proj.bias": "vision_encoder.safetensors",
129
+ "layer.13.attention.attn_base.o_proj.weight": "vision_encoder.safetensors",
130
+ "layer.13.attention.attn_base.q_proj.weight": "vision_encoder.safetensors",
131
+ "layer.13.attention.attn_base.v_proj.weight": "vision_encoder.safetensors",
132
+ "layer.13.attention.attn_plus.k_proj.weight": "vision_encoder.safetensors",
133
+ "layer.13.attention.attn_plus.o_proj.bias": "vision_encoder.safetensors",
134
+ "layer.13.attention.attn_plus.o_proj.weight": "vision_encoder.safetensors",
135
+ "layer.13.attention.attn_plus.q_proj.weight": "vision_encoder.safetensors",
136
+ "layer.13.attention.attn_plus.v_proj.weight": "vision_encoder.safetensors",
137
+ "layer.13.layer_scale1.lambda1": "vision_encoder.safetensors",
138
+ "layer.13.layer_scale2.lambda1": "vision_encoder.safetensors",
139
+ "layer.13.mlp.down_proj.bias": "vision_encoder.safetensors",
140
+ "layer.13.mlp.down_proj.weight": "vision_encoder.safetensors",
141
+ "layer.13.mlp.gate_proj.bias": "vision_encoder.safetensors",
142
+ "layer.13.mlp.gate_proj.weight": "vision_encoder.safetensors",
143
+ "layer.13.mlp.up_proj.bias": "vision_encoder.safetensors",
144
+ "layer.13.mlp.up_proj.weight": "vision_encoder.safetensors",
145
+ "layer.13.norm1.bias": "vision_encoder.safetensors",
146
+ "layer.13.norm1.weight": "vision_encoder.safetensors",
147
+ "layer.13.norm2.bias": "vision_encoder.safetensors",
148
+ "layer.13.norm2.weight": "vision_encoder.safetensors",
149
+ "layer.14.attention.alpha_logit": "vision_encoder.safetensors",
150
+ "layer.14.attention.attn_base.k_proj.weight": "vision_encoder.safetensors",
151
+ "layer.14.attention.attn_base.o_proj.bias": "vision_encoder.safetensors",
152
+ "layer.14.attention.attn_base.o_proj.weight": "vision_encoder.safetensors",
153
+ "layer.14.attention.attn_base.q_proj.weight": "vision_encoder.safetensors",
154
+ "layer.14.attention.attn_base.v_proj.weight": "vision_encoder.safetensors",
155
+ "layer.14.attention.attn_plus.k_proj.weight": "vision_encoder.safetensors",
156
+ "layer.14.attention.attn_plus.o_proj.bias": "vision_encoder.safetensors",
157
+ "layer.14.attention.attn_plus.o_proj.weight": "vision_encoder.safetensors",
158
+ "layer.14.attention.attn_plus.q_proj.weight": "vision_encoder.safetensors",
159
+ "layer.14.attention.attn_plus.v_proj.weight": "vision_encoder.safetensors",
160
+ "layer.14.layer_scale1.lambda1": "vision_encoder.safetensors",
161
+ "layer.14.layer_scale2.lambda1": "vision_encoder.safetensors",
162
+ "layer.14.mlp.down_proj.bias": "vision_encoder.safetensors",
163
+ "layer.14.mlp.down_proj.weight": "vision_encoder.safetensors",
164
+ "layer.14.mlp.gate_proj.bias": "vision_encoder.safetensors",
165
+ "layer.14.mlp.gate_proj.weight": "vision_encoder.safetensors",
166
+ "layer.14.mlp.up_proj.bias": "vision_encoder.safetensors",
167
+ "layer.14.mlp.up_proj.weight": "vision_encoder.safetensors",
168
+ "layer.14.norm1.bias": "vision_encoder.safetensors",
169
+ "layer.14.norm1.weight": "vision_encoder.safetensors",
170
+ "layer.14.norm2.bias": "vision_encoder.safetensors",
171
+ "layer.14.norm2.weight": "vision_encoder.safetensors",
172
+ "layer.15.attention.alpha_logit": "vision_encoder.safetensors",
173
+ "layer.15.attention.attn_base.k_proj.weight": "vision_encoder.safetensors",
174
+ "layer.15.attention.attn_base.o_proj.bias": "vision_encoder.safetensors",
175
+ "layer.15.attention.attn_base.o_proj.weight": "vision_encoder.safetensors",
176
+ "layer.15.attention.attn_base.q_proj.weight": "vision_encoder.safetensors",
177
+ "layer.15.attention.attn_base.v_proj.weight": "vision_encoder.safetensors",
178
+ "layer.15.attention.attn_plus.k_proj.weight": "vision_encoder.safetensors",
179
+ "layer.15.attention.attn_plus.o_proj.bias": "vision_encoder.safetensors",
180
+ "layer.15.attention.attn_plus.o_proj.weight": "vision_encoder.safetensors",
181
+ "layer.15.attention.attn_plus.q_proj.weight": "vision_encoder.safetensors",
182
+ "layer.15.attention.attn_plus.v_proj.weight": "vision_encoder.safetensors",
183
+ "layer.15.layer_scale1.lambda1": "vision_encoder.safetensors",
184
+ "layer.15.layer_scale2.lambda1": "vision_encoder.safetensors",
185
+ "layer.15.mlp.down_proj.bias": "vision_encoder.safetensors",
186
+ "layer.15.mlp.down_proj.weight": "vision_encoder.safetensors",
187
+ "layer.15.mlp.gate_proj.bias": "vision_encoder.safetensors",
188
+ "layer.15.mlp.gate_proj.weight": "vision_encoder.safetensors",
189
+ "layer.15.mlp.up_proj.bias": "vision_encoder.safetensors",
190
+ "layer.15.mlp.up_proj.weight": "vision_encoder.safetensors",
191
+ "layer.15.norm1.bias": "vision_encoder.safetensors",
192
+ "layer.15.norm1.weight": "vision_encoder.safetensors",
193
+ "layer.15.norm2.bias": "vision_encoder.safetensors",
194
+ "layer.15.norm2.weight": "vision_encoder.safetensors",
195
+ "layer.16.attention.alpha_logit": "vision_encoder.safetensors",
196
+ "layer.16.attention.attn_base.k_proj.weight": "vision_encoder.safetensors",
197
+ "layer.16.attention.attn_base.o_proj.bias": "vision_encoder.safetensors",
198
+ "layer.16.attention.attn_base.o_proj.weight": "vision_encoder.safetensors",
199
+ "layer.16.attention.attn_base.q_proj.weight": "vision_encoder.safetensors",
200
+ "layer.16.attention.attn_base.v_proj.weight": "vision_encoder.safetensors",
201
+ "layer.16.attention.attn_plus.k_proj.weight": "vision_encoder.safetensors",
202
+ "layer.16.attention.attn_plus.o_proj.bias": "vision_encoder.safetensors",
203
+ "layer.16.attention.attn_plus.o_proj.weight": "vision_encoder.safetensors",
204
+ "layer.16.attention.attn_plus.q_proj.weight": "vision_encoder.safetensors",
205
+ "layer.16.attention.attn_plus.v_proj.weight": "vision_encoder.safetensors",
206
+ "layer.16.layer_scale1.lambda1": "vision_encoder.safetensors",
207
+ "layer.16.layer_scale2.lambda1": "vision_encoder.safetensors",
208
+ "layer.16.mlp.down_proj.bias": "vision_encoder.safetensors",
209
+ "layer.16.mlp.down_proj.weight": "vision_encoder.safetensors",
210
+ "layer.16.mlp.gate_proj.bias": "vision_encoder.safetensors",
211
+ "layer.16.mlp.gate_proj.weight": "vision_encoder.safetensors",
212
+ "layer.16.mlp.up_proj.bias": "vision_encoder.safetensors",
213
+ "layer.16.mlp.up_proj.weight": "vision_encoder.safetensors",
214
+ "layer.16.norm1.bias": "vision_encoder.safetensors",
215
+ "layer.16.norm1.weight": "vision_encoder.safetensors",
216
+ "layer.16.norm2.bias": "vision_encoder.safetensors",
217
+ "layer.16.norm2.weight": "vision_encoder.safetensors",
218
+ "layer.17.attention.alpha_logit": "vision_encoder.safetensors",
219
+ "layer.17.attention.attn_base.k_proj.weight": "vision_encoder.safetensors",
220
+ "layer.17.attention.attn_base.o_proj.bias": "vision_encoder.safetensors",
221
+ "layer.17.attention.attn_base.o_proj.weight": "vision_encoder.safetensors",
222
+ "layer.17.attention.attn_base.q_proj.weight": "vision_encoder.safetensors",
223
+ "layer.17.attention.attn_base.v_proj.weight": "vision_encoder.safetensors",
224
+ "layer.17.attention.attn_plus.k_proj.weight": "vision_encoder.safetensors",
225
+ "layer.17.attention.attn_plus.o_proj.bias": "vision_encoder.safetensors",
226
+ "layer.17.attention.attn_plus.o_proj.weight": "vision_encoder.safetensors",
227
+ "layer.17.attention.attn_plus.q_proj.weight": "vision_encoder.safetensors",
228
+ "layer.17.attention.attn_plus.v_proj.weight": "vision_encoder.safetensors",
229
+ "layer.17.layer_scale1.lambda1": "vision_encoder.safetensors",
230
+ "layer.17.layer_scale2.lambda1": "vision_encoder.safetensors",
231
+ "layer.17.mlp.down_proj.bias": "vision_encoder.safetensors",
232
+ "layer.17.mlp.down_proj.weight": "vision_encoder.safetensors",
233
+ "layer.17.mlp.gate_proj.bias": "vision_encoder.safetensors",
234
+ "layer.17.mlp.gate_proj.weight": "vision_encoder.safetensors",
235
+ "layer.17.mlp.up_proj.bias": "vision_encoder.safetensors",
236
+ "layer.17.mlp.up_proj.weight": "vision_encoder.safetensors",
237
+ "layer.17.norm1.bias": "vision_encoder.safetensors",
238
+ "layer.17.norm1.weight": "vision_encoder.safetensors",
239
+ "layer.17.norm2.bias": "vision_encoder.safetensors",
240
+ "layer.17.norm2.weight": "vision_encoder.safetensors",
241
+ "layer.18.attention.alpha_logit": "vision_encoder.safetensors",
242
+ "layer.18.attention.attn_base.k_proj.weight": "vision_encoder.safetensors",
243
+ "layer.18.attention.attn_base.o_proj.bias": "vision_encoder.safetensors",
244
+ "layer.18.attention.attn_base.o_proj.weight": "vision_encoder.safetensors",
245
+ "layer.18.attention.attn_base.q_proj.weight": "vision_encoder.safetensors",
246
+ "layer.18.attention.attn_base.v_proj.weight": "vision_encoder.safetensors",
247
+ "layer.18.attention.attn_plus.k_proj.weight": "vision_encoder.safetensors",
248
+ "layer.18.attention.attn_plus.o_proj.bias": "vision_encoder.safetensors",
249
+ "layer.18.attention.attn_plus.o_proj.weight": "vision_encoder.safetensors",
250
+ "layer.18.attention.attn_plus.q_proj.weight": "vision_encoder.safetensors",
251
+ "layer.18.attention.attn_plus.v_proj.weight": "vision_encoder.safetensors",
252
+ "layer.18.layer_scale1.lambda1": "vision_encoder.safetensors",
253
+ "layer.18.layer_scale2.lambda1": "vision_encoder.safetensors",
254
+ "layer.18.mlp.down_proj.bias": "vision_encoder.safetensors",
255
+ "layer.18.mlp.down_proj.weight": "vision_encoder.safetensors",
256
+ "layer.18.mlp.gate_proj.bias": "vision_encoder.safetensors",
257
+ "layer.18.mlp.gate_proj.weight": "vision_encoder.safetensors",
258
+ "layer.18.mlp.up_proj.bias": "vision_encoder.safetensors",
259
+ "layer.18.mlp.up_proj.weight": "vision_encoder.safetensors",
260
+ "layer.18.norm1.bias": "vision_encoder.safetensors",
261
+ "layer.18.norm1.weight": "vision_encoder.safetensors",
262
+ "layer.18.norm2.bias": "vision_encoder.safetensors",
263
+ "layer.18.norm2.weight": "vision_encoder.safetensors",
264
+ "layer.19.attention.alpha_logit": "vision_encoder.safetensors",
265
+ "layer.19.attention.attn_base.k_proj.weight": "vision_encoder.safetensors",
266
+ "layer.19.attention.attn_base.o_proj.bias": "vision_encoder.safetensors",
267
+ "layer.19.attention.attn_base.o_proj.weight": "vision_encoder.safetensors",
268
+ "layer.19.attention.attn_base.q_proj.weight": "vision_encoder.safetensors",
269
+ "layer.19.attention.attn_base.v_proj.weight": "vision_encoder.safetensors",
270
+ "layer.19.attention.attn_plus.k_proj.weight": "vision_encoder.safetensors",
271
+ "layer.19.attention.attn_plus.o_proj.bias": "vision_encoder.safetensors",
272
+ "layer.19.attention.attn_plus.o_proj.weight": "vision_encoder.safetensors",
273
+ "layer.19.attention.attn_plus.q_proj.weight": "vision_encoder.safetensors",
274
+ "layer.19.attention.attn_plus.v_proj.weight": "vision_encoder.safetensors",
275
+ "layer.19.layer_scale1.lambda1": "vision_encoder.safetensors",
276
+ "layer.19.layer_scale2.lambda1": "vision_encoder.safetensors",
277
+ "layer.19.mlp.down_proj.bias": "vision_encoder.safetensors",
278
+ "layer.19.mlp.down_proj.weight": "vision_encoder.safetensors",
279
+ "layer.19.mlp.gate_proj.bias": "vision_encoder.safetensors",
280
+ "layer.19.mlp.gate_proj.weight": "vision_encoder.safetensors",
281
+ "layer.19.mlp.up_proj.bias": "vision_encoder.safetensors",
282
+ "layer.19.mlp.up_proj.weight": "vision_encoder.safetensors",
283
+ "layer.19.norm1.bias": "vision_encoder.safetensors",
284
+ "layer.19.norm1.weight": "vision_encoder.safetensors",
285
+ "layer.19.norm2.bias": "vision_encoder.safetensors",
286
+ "layer.19.norm2.weight": "vision_encoder.safetensors",
287
+ "layer.2.attention.alpha_logit": "vision_encoder.safetensors",
288
+ "layer.2.attention.attn_base.k_proj.weight": "vision_encoder.safetensors",
289
+ "layer.2.attention.attn_base.o_proj.bias": "vision_encoder.safetensors",
290
+ "layer.2.attention.attn_base.o_proj.weight": "vision_encoder.safetensors",
291
+ "layer.2.attention.attn_base.q_proj.weight": "vision_encoder.safetensors",
292
+ "layer.2.attention.attn_base.v_proj.weight": "vision_encoder.safetensors",
293
+ "layer.2.attention.attn_plus.k_proj.weight": "vision_encoder.safetensors",
294
+ "layer.2.attention.attn_plus.o_proj.bias": "vision_encoder.safetensors",
295
+ "layer.2.attention.attn_plus.o_proj.weight": "vision_encoder.safetensors",
296
+ "layer.2.attention.attn_plus.q_proj.weight": "vision_encoder.safetensors",
297
+ "layer.2.attention.attn_plus.v_proj.weight": "vision_encoder.safetensors",
298
+ "layer.2.layer_scale1.lambda1": "vision_encoder.safetensors",
299
+ "layer.2.layer_scale2.lambda1": "vision_encoder.safetensors",
300
+ "layer.2.mlp.down_proj.bias": "vision_encoder.safetensors",
301
+ "layer.2.mlp.down_proj.weight": "vision_encoder.safetensors",
302
+ "layer.2.mlp.gate_proj.bias": "vision_encoder.safetensors",
303
+ "layer.2.mlp.gate_proj.weight": "vision_encoder.safetensors",
304
+ "layer.2.mlp.up_proj.bias": "vision_encoder.safetensors",
305
+ "layer.2.mlp.up_proj.weight": "vision_encoder.safetensors",
306
+ "layer.2.norm1.bias": "vision_encoder.safetensors",
307
+ "layer.2.norm1.weight": "vision_encoder.safetensors",
308
+ "layer.2.norm2.bias": "vision_encoder.safetensors",
309
+ "layer.2.norm2.weight": "vision_encoder.safetensors",
310
+ "layer.20.attention.alpha_logit": "vision_encoder.safetensors",
311
+ "layer.20.attention.attn_base.k_proj.weight": "vision_encoder.safetensors",
312
+ "layer.20.attention.attn_base.o_proj.bias": "vision_encoder.safetensors",
313
+ "layer.20.attention.attn_base.o_proj.weight": "vision_encoder.safetensors",
314
+ "layer.20.attention.attn_base.q_proj.weight": "vision_encoder.safetensors",
315
+ "layer.20.attention.attn_base.v_proj.weight": "vision_encoder.safetensors",
316
+ "layer.20.attention.attn_plus.k_proj.weight": "vision_encoder.safetensors",
317
+ "layer.20.attention.attn_plus.o_proj.bias": "vision_encoder.safetensors",
318
+ "layer.20.attention.attn_plus.o_proj.weight": "vision_encoder.safetensors",
319
+ "layer.20.attention.attn_plus.q_proj.weight": "vision_encoder.safetensors",
320
+ "layer.20.attention.attn_plus.v_proj.weight": "vision_encoder.safetensors",
321
+ "layer.20.layer_scale1.lambda1": "vision_encoder.safetensors",
322
+ "layer.20.layer_scale2.lambda1": "vision_encoder.safetensors",
323
+ "layer.20.mlp.down_proj.bias": "vision_encoder.safetensors",
324
+ "layer.20.mlp.down_proj.weight": "vision_encoder.safetensors",
325
+ "layer.20.mlp.gate_proj.bias": "vision_encoder.safetensors",
326
+ "layer.20.mlp.gate_proj.weight": "vision_encoder.safetensors",
327
+ "layer.20.mlp.up_proj.bias": "vision_encoder.safetensors",
328
+ "layer.20.mlp.up_proj.weight": "vision_encoder.safetensors",
329
+ "layer.20.norm1.bias": "vision_encoder.safetensors",
330
+ "layer.20.norm1.weight": "vision_encoder.safetensors",
331
+ "layer.20.norm2.bias": "vision_encoder.safetensors",
332
+ "layer.20.norm2.weight": "vision_encoder.safetensors",
333
+ "layer.21.attention.alpha_logit": "vision_encoder.safetensors",
334
+ "layer.21.attention.attn_base.k_proj.weight": "vision_encoder.safetensors",
335
+ "layer.21.attention.attn_base.o_proj.bias": "vision_encoder.safetensors",
336
+ "layer.21.attention.attn_base.o_proj.weight": "vision_encoder.safetensors",
337
+ "layer.21.attention.attn_base.q_proj.weight": "vision_encoder.safetensors",
338
+ "layer.21.attention.attn_base.v_proj.weight": "vision_encoder.safetensors",
339
+ "layer.21.attention.attn_plus.k_proj.weight": "vision_encoder.safetensors",
340
+ "layer.21.attention.attn_plus.o_proj.bias": "vision_encoder.safetensors",
341
+ "layer.21.attention.attn_plus.o_proj.weight": "vision_encoder.safetensors",
342
+ "layer.21.attention.attn_plus.q_proj.weight": "vision_encoder.safetensors",
343
+ "layer.21.attention.attn_plus.v_proj.weight": "vision_encoder.safetensors",
344
+ "layer.21.layer_scale1.lambda1": "vision_encoder.safetensors",
345
+ "layer.21.layer_scale2.lambda1": "vision_encoder.safetensors",
346
+ "layer.21.mlp.down_proj.bias": "vision_encoder.safetensors",
347
+ "layer.21.mlp.down_proj.weight": "vision_encoder.safetensors",
348
+ "layer.21.mlp.gate_proj.bias": "vision_encoder.safetensors",
349
+ "layer.21.mlp.gate_proj.weight": "vision_encoder.safetensors",
350
+ "layer.21.mlp.up_proj.bias": "vision_encoder.safetensors",
351
+ "layer.21.mlp.up_proj.weight": "vision_encoder.safetensors",
352
+ "layer.21.norm1.bias": "vision_encoder.safetensors",
353
+ "layer.21.norm1.weight": "vision_encoder.safetensors",
354
+ "layer.21.norm2.bias": "vision_encoder.safetensors",
355
+ "layer.21.norm2.weight": "vision_encoder.safetensors",
356
+ "layer.22.attention.alpha_logit": "vision_encoder.safetensors",
357
+ "layer.22.attention.attn_base.k_proj.weight": "vision_encoder.safetensors",
358
+ "layer.22.attention.attn_base.o_proj.bias": "vision_encoder.safetensors",
359
+ "layer.22.attention.attn_base.o_proj.weight": "vision_encoder.safetensors",
360
+ "layer.22.attention.attn_base.q_proj.weight": "vision_encoder.safetensors",
361
+ "layer.22.attention.attn_base.v_proj.weight": "vision_encoder.safetensors",
362
+ "layer.22.attention.attn_plus.k_proj.weight": "vision_encoder.safetensors",
363
+ "layer.22.attention.attn_plus.o_proj.bias": "vision_encoder.safetensors",
364
+ "layer.22.attention.attn_plus.o_proj.weight": "vision_encoder.safetensors",
365
+ "layer.22.attention.attn_plus.q_proj.weight": "vision_encoder.safetensors",
366
+ "layer.22.attention.attn_plus.v_proj.weight": "vision_encoder.safetensors",
367
+ "layer.22.layer_scale1.lambda1": "vision_encoder.safetensors",
368
+ "layer.22.layer_scale2.lambda1": "vision_encoder.safetensors",
369
+ "layer.22.mlp.down_proj.bias": "vision_encoder.safetensors",
370
+ "layer.22.mlp.down_proj.weight": "vision_encoder.safetensors",
371
+ "layer.22.mlp.gate_proj.bias": "vision_encoder.safetensors",
372
+ "layer.22.mlp.gate_proj.weight": "vision_encoder.safetensors",
373
+ "layer.22.mlp.up_proj.bias": "vision_encoder.safetensors",
374
+ "layer.22.mlp.up_proj.weight": "vision_encoder.safetensors",
375
+ "layer.22.norm1.bias": "vision_encoder.safetensors",
376
+ "layer.22.norm1.weight": "vision_encoder.safetensors",
377
+ "layer.22.norm2.bias": "vision_encoder.safetensors",
378
+ "layer.22.norm2.weight": "vision_encoder.safetensors",
379
+ "layer.23.attention.alpha_logit": "vision_encoder.safetensors",
380
+ "layer.23.attention.attn_base.k_proj.weight": "vision_encoder.safetensors",
381
+ "layer.23.attention.attn_base.o_proj.bias": "vision_encoder.safetensors",
382
+ "layer.23.attention.attn_base.o_proj.weight": "vision_encoder.safetensors",
383
+ "layer.23.attention.attn_base.q_proj.weight": "vision_encoder.safetensors",
384
+ "layer.23.attention.attn_base.v_proj.weight": "vision_encoder.safetensors",
385
+ "layer.23.attention.attn_plus.k_proj.weight": "vision_encoder.safetensors",
386
+ "layer.23.attention.attn_plus.o_proj.bias": "vision_encoder.safetensors",
387
+ "layer.23.attention.attn_plus.o_proj.weight": "vision_encoder.safetensors",
388
+ "layer.23.attention.attn_plus.q_proj.weight": "vision_encoder.safetensors",
389
+ "layer.23.attention.attn_plus.v_proj.weight": "vision_encoder.safetensors",
390
+ "layer.23.layer_scale1.lambda1": "vision_encoder.safetensors",
391
+ "layer.23.layer_scale2.lambda1": "vision_encoder.safetensors",
392
+ "layer.23.mlp.down_proj.bias": "vision_encoder.safetensors",
393
+ "layer.23.mlp.down_proj.weight": "vision_encoder.safetensors",
394
+ "layer.23.mlp.gate_proj.bias": "vision_encoder.safetensors",
395
+ "layer.23.mlp.gate_proj.weight": "vision_encoder.safetensors",
396
+ "layer.23.mlp.up_proj.bias": "vision_encoder.safetensors",
397
+ "layer.23.mlp.up_proj.weight": "vision_encoder.safetensors",
398
+ "layer.23.norm1.bias": "vision_encoder.safetensors",
399
+ "layer.23.norm1.weight": "vision_encoder.safetensors",
400
+ "layer.23.norm2.bias": "vision_encoder.safetensors",
401
+ "layer.23.norm2.weight": "vision_encoder.safetensors",
402
+ "layer.24.attention.alpha_logit": "vision_encoder.safetensors",
403
+ "layer.24.attention.attn_base.k_proj.weight": "vision_encoder.safetensors",
404
+ "layer.24.attention.attn_base.o_proj.bias": "vision_encoder.safetensors",
405
+ "layer.24.attention.attn_base.o_proj.weight": "vision_encoder.safetensors",
406
+ "layer.24.attention.attn_base.q_proj.weight": "vision_encoder.safetensors",
407
+ "layer.24.attention.attn_base.v_proj.weight": "vision_encoder.safetensors",
408
+ "layer.24.attention.attn_plus.k_proj.weight": "vision_encoder.safetensors",
409
+ "layer.24.attention.attn_plus.o_proj.bias": "vision_encoder.safetensors",
410
+ "layer.24.attention.attn_plus.o_proj.weight": "vision_encoder.safetensors",
411
+ "layer.24.attention.attn_plus.q_proj.weight": "vision_encoder.safetensors",
412
+ "layer.24.attention.attn_plus.v_proj.weight": "vision_encoder.safetensors",
413
+ "layer.24.layer_scale1.lambda1": "vision_encoder.safetensors",
414
+ "layer.24.layer_scale2.lambda1": "vision_encoder.safetensors",
415
+ "layer.24.mlp.down_proj.bias": "vision_encoder.safetensors",
416
+ "layer.24.mlp.down_proj.weight": "vision_encoder.safetensors",
417
+ "layer.24.mlp.gate_proj.bias": "vision_encoder.safetensors",
418
+ "layer.24.mlp.gate_proj.weight": "vision_encoder.safetensors",
419
+ "layer.24.mlp.up_proj.bias": "vision_encoder.safetensors",
420
+ "layer.24.mlp.up_proj.weight": "vision_encoder.safetensors",
421
+ "layer.24.norm1.bias": "vision_encoder.safetensors",
422
+ "layer.24.norm1.weight": "vision_encoder.safetensors",
423
+ "layer.24.norm2.bias": "vision_encoder.safetensors",
424
+ "layer.24.norm2.weight": "vision_encoder.safetensors",
425
+ "layer.25.attention.alpha_logit": "vision_encoder.safetensors",
426
+ "layer.25.attention.attn_base.k_proj.weight": "vision_encoder.safetensors",
427
+ "layer.25.attention.attn_base.o_proj.bias": "vision_encoder.safetensors",
428
+ "layer.25.attention.attn_base.o_proj.weight": "vision_encoder.safetensors",
429
+ "layer.25.attention.attn_base.q_proj.weight": "vision_encoder.safetensors",
430
+ "layer.25.attention.attn_base.v_proj.weight": "vision_encoder.safetensors",
431
+ "layer.25.attention.attn_plus.k_proj.weight": "vision_encoder.safetensors",
432
+ "layer.25.attention.attn_plus.o_proj.bias": "vision_encoder.safetensors",
433
+ "layer.25.attention.attn_plus.o_proj.weight": "vision_encoder.safetensors",
434
+ "layer.25.attention.attn_plus.q_proj.weight": "vision_encoder.safetensors",
435
+ "layer.25.attention.attn_plus.v_proj.weight": "vision_encoder.safetensors",
436
+ "layer.25.layer_scale1.lambda1": "vision_encoder.safetensors",
437
+ "layer.25.layer_scale2.lambda1": "vision_encoder.safetensors",
438
+ "layer.25.mlp.down_proj.bias": "vision_encoder.safetensors",
439
+ "layer.25.mlp.down_proj.weight": "vision_encoder.safetensors",
440
+ "layer.25.mlp.gate_proj.bias": "vision_encoder.safetensors",
441
+ "layer.25.mlp.gate_proj.weight": "vision_encoder.safetensors",
442
+ "layer.25.mlp.up_proj.bias": "vision_encoder.safetensors",
443
+ "layer.25.mlp.up_proj.weight": "vision_encoder.safetensors",
444
+ "layer.25.norm1.bias": "vision_encoder.safetensors",
445
+ "layer.25.norm1.weight": "vision_encoder.safetensors",
446
+ "layer.25.norm2.bias": "vision_encoder.safetensors",
447
+ "layer.25.norm2.weight": "vision_encoder.safetensors",
448
+ "layer.26.attention.alpha_logit": "vision_encoder.safetensors",
449
+ "layer.26.attention.attn_base.k_proj.weight": "vision_encoder.safetensors",
450
+ "layer.26.attention.attn_base.o_proj.bias": "vision_encoder.safetensors",
451
+ "layer.26.attention.attn_base.o_proj.weight": "vision_encoder.safetensors",
452
+ "layer.26.attention.attn_base.q_proj.weight": "vision_encoder.safetensors",
453
+ "layer.26.attention.attn_base.v_proj.weight": "vision_encoder.safetensors",
454
+ "layer.26.attention.attn_plus.k_proj.weight": "vision_encoder.safetensors",
455
+ "layer.26.attention.attn_plus.o_proj.bias": "vision_encoder.safetensors",
456
+ "layer.26.attention.attn_plus.o_proj.weight": "vision_encoder.safetensors",
457
+ "layer.26.attention.attn_plus.q_proj.weight": "vision_encoder.safetensors",
458
+ "layer.26.attention.attn_plus.v_proj.weight": "vision_encoder.safetensors",
459
+ "layer.26.layer_scale1.lambda1": "vision_encoder.safetensors",
460
+ "layer.26.layer_scale2.lambda1": "vision_encoder.safetensors",
461
+ "layer.26.mlp.down_proj.bias": "vision_encoder.safetensors",
462
+ "layer.26.mlp.down_proj.weight": "vision_encoder.safetensors",
463
+ "layer.26.mlp.gate_proj.bias": "vision_encoder.safetensors",
464
+ "layer.26.mlp.gate_proj.weight": "vision_encoder.safetensors",
465
+ "layer.26.mlp.up_proj.bias": "vision_encoder.safetensors",
466
+ "layer.26.mlp.up_proj.weight": "vision_encoder.safetensors",
467
+ "layer.26.norm1.bias": "vision_encoder.safetensors",
468
+ "layer.26.norm1.weight": "vision_encoder.safetensors",
469
+ "layer.26.norm2.bias": "vision_encoder.safetensors",
470
+ "layer.26.norm2.weight": "vision_encoder.safetensors",
471
+ "layer.27.attention.alpha_logit": "vision_encoder.safetensors",
472
+ "layer.27.attention.attn_base.k_proj.weight": "vision_encoder.safetensors",
473
+ "layer.27.attention.attn_base.o_proj.bias": "vision_encoder.safetensors",
474
+ "layer.27.attention.attn_base.o_proj.weight": "vision_encoder.safetensors",
475
+ "layer.27.attention.attn_base.q_proj.weight": "vision_encoder.safetensors",
476
+ "layer.27.attention.attn_base.v_proj.weight": "vision_encoder.safetensors",
477
+ "layer.27.attention.attn_plus.k_proj.weight": "vision_encoder.safetensors",
478
+ "layer.27.attention.attn_plus.o_proj.bias": "vision_encoder.safetensors",
479
+ "layer.27.attention.attn_plus.o_proj.weight": "vision_encoder.safetensors",
480
+ "layer.27.attention.attn_plus.q_proj.weight": "vision_encoder.safetensors",
481
+ "layer.27.attention.attn_plus.v_proj.weight": "vision_encoder.safetensors",
482
+ "layer.27.layer_scale1.lambda1": "vision_encoder.safetensors",
483
+ "layer.27.layer_scale2.lambda1": "vision_encoder.safetensors",
484
+ "layer.27.mlp.down_proj.bias": "vision_encoder.safetensors",
485
+ "layer.27.mlp.down_proj.weight": "vision_encoder.safetensors",
486
+ "layer.27.mlp.gate_proj.bias": "vision_encoder.safetensors",
487
+ "layer.27.mlp.gate_proj.weight": "vision_encoder.safetensors",
488
+ "layer.27.mlp.up_proj.bias": "vision_encoder.safetensors",
489
+ "layer.27.mlp.up_proj.weight": "vision_encoder.safetensors",
490
+ "layer.27.norm1.bias": "vision_encoder.safetensors",
491
+ "layer.27.norm1.weight": "vision_encoder.safetensors",
492
+ "layer.27.norm2.bias": "vision_encoder.safetensors",
493
+ "layer.27.norm2.weight": "vision_encoder.safetensors",
494
+ "layer.28.attention.alpha_logit": "vision_encoder.safetensors",
495
+ "layer.28.attention.attn_base.k_proj.weight": "vision_encoder.safetensors",
496
+ "layer.28.attention.attn_base.o_proj.bias": "vision_encoder.safetensors",
497
+ "layer.28.attention.attn_base.o_proj.weight": "vision_encoder.safetensors",
498
+ "layer.28.attention.attn_base.q_proj.weight": "vision_encoder.safetensors",
499
+ "layer.28.attention.attn_base.v_proj.weight": "vision_encoder.safetensors",
500
+ "layer.28.attention.attn_plus.k_proj.weight": "vision_encoder.safetensors",
501
+ "layer.28.attention.attn_plus.o_proj.bias": "vision_encoder.safetensors",
502
+ "layer.28.attention.attn_plus.o_proj.weight": "vision_encoder.safetensors",
503
+ "layer.28.attention.attn_plus.q_proj.weight": "vision_encoder.safetensors",
504
+ "layer.28.attention.attn_plus.v_proj.weight": "vision_encoder.safetensors",
505
+ "layer.28.layer_scale1.lambda1": "vision_encoder.safetensors",
506
+ "layer.28.layer_scale2.lambda1": "vision_encoder.safetensors",
507
+ "layer.28.mlp.down_proj.bias": "vision_encoder.safetensors",
508
+ "layer.28.mlp.down_proj.weight": "vision_encoder.safetensors",
509
+ "layer.28.mlp.gate_proj.bias": "vision_encoder.safetensors",
510
+ "layer.28.mlp.gate_proj.weight": "vision_encoder.safetensors",
511
+ "layer.28.mlp.up_proj.bias": "vision_encoder.safetensors",
512
+ "layer.28.mlp.up_proj.weight": "vision_encoder.safetensors",
513
+ "layer.28.norm1.bias": "vision_encoder.safetensors",
514
+ "layer.28.norm1.weight": "vision_encoder.safetensors",
515
+ "layer.28.norm2.bias": "vision_encoder.safetensors",
516
+ "layer.28.norm2.weight": "vision_encoder.safetensors",
517
+ "layer.29.attention.alpha_logit": "vision_encoder.safetensors",
518
+ "layer.29.attention.attn_base.k_proj.weight": "vision_encoder.safetensors",
519
+ "layer.29.attention.attn_base.o_proj.bias": "vision_encoder.safetensors",
520
+ "layer.29.attention.attn_base.o_proj.weight": "vision_encoder.safetensors",
521
+ "layer.29.attention.attn_base.q_proj.weight": "vision_encoder.safetensors",
522
+ "layer.29.attention.attn_base.v_proj.weight": "vision_encoder.safetensors",
523
+ "layer.29.attention.attn_plus.k_proj.weight": "vision_encoder.safetensors",
524
+ "layer.29.attention.attn_plus.o_proj.bias": "vision_encoder.safetensors",
525
+ "layer.29.attention.attn_plus.o_proj.weight": "vision_encoder.safetensors",
526
+ "layer.29.attention.attn_plus.q_proj.weight": "vision_encoder.safetensors",
527
+ "layer.29.attention.attn_plus.v_proj.weight": "vision_encoder.safetensors",
528
+ "layer.29.layer_scale1.lambda1": "vision_encoder.safetensors",
529
+ "layer.29.layer_scale2.lambda1": "vision_encoder.safetensors",
530
+ "layer.29.mlp.down_proj.bias": "vision_encoder.safetensors",
531
+ "layer.29.mlp.down_proj.weight": "vision_encoder.safetensors",
532
+ "layer.29.mlp.gate_proj.bias": "vision_encoder.safetensors",
533
+ "layer.29.mlp.gate_proj.weight": "vision_encoder.safetensors",
534
+ "layer.29.mlp.up_proj.bias": "vision_encoder.safetensors",
535
+ "layer.29.mlp.up_proj.weight": "vision_encoder.safetensors",
536
+ "layer.29.norm1.bias": "vision_encoder.safetensors",
537
+ "layer.29.norm1.weight": "vision_encoder.safetensors",
538
+ "layer.29.norm2.bias": "vision_encoder.safetensors",
539
+ "layer.29.norm2.weight": "vision_encoder.safetensors",
540
+ "layer.3.attention.alpha_logit": "vision_encoder.safetensors",
541
+ "layer.3.attention.attn_base.k_proj.weight": "vision_encoder.safetensors",
542
+ "layer.3.attention.attn_base.o_proj.bias": "vision_encoder.safetensors",
543
+ "layer.3.attention.attn_base.o_proj.weight": "vision_encoder.safetensors",
544
+ "layer.3.attention.attn_base.q_proj.weight": "vision_encoder.safetensors",
545
+ "layer.3.attention.attn_base.v_proj.weight": "vision_encoder.safetensors",
546
+ "layer.3.attention.attn_plus.k_proj.weight": "vision_encoder.safetensors",
547
+ "layer.3.attention.attn_plus.o_proj.bias": "vision_encoder.safetensors",
548
+ "layer.3.attention.attn_plus.o_proj.weight": "vision_encoder.safetensors",
549
+ "layer.3.attention.attn_plus.q_proj.weight": "vision_encoder.safetensors",
550
+ "layer.3.attention.attn_plus.v_proj.weight": "vision_encoder.safetensors",
551
+ "layer.3.layer_scale1.lambda1": "vision_encoder.safetensors",
552
+ "layer.3.layer_scale2.lambda1": "vision_encoder.safetensors",
553
+ "layer.3.mlp.down_proj.bias": "vision_encoder.safetensors",
554
+ "layer.3.mlp.down_proj.weight": "vision_encoder.safetensors",
555
+ "layer.3.mlp.gate_proj.bias": "vision_encoder.safetensors",
556
+ "layer.3.mlp.gate_proj.weight": "vision_encoder.safetensors",
557
+ "layer.3.mlp.up_proj.bias": "vision_encoder.safetensors",
558
+ "layer.3.mlp.up_proj.weight": "vision_encoder.safetensors",
559
+ "layer.3.norm1.bias": "vision_encoder.safetensors",
560
+ "layer.3.norm1.weight": "vision_encoder.safetensors",
561
+ "layer.3.norm2.bias": "vision_encoder.safetensors",
562
+ "layer.3.norm2.weight": "vision_encoder.safetensors",
563
+ "layer.30.attention.alpha_logit": "vision_encoder.safetensors",
564
+ "layer.30.attention.attn_base.k_proj.weight": "vision_encoder.safetensors",
565
+ "layer.30.attention.attn_base.o_proj.bias": "vision_encoder.safetensors",
566
+ "layer.30.attention.attn_base.o_proj.weight": "vision_encoder.safetensors",
567
+ "layer.30.attention.attn_base.q_proj.weight": "vision_encoder.safetensors",
568
+ "layer.30.attention.attn_base.v_proj.weight": "vision_encoder.safetensors",
569
+ "layer.30.attention.attn_plus.k_proj.weight": "vision_encoder.safetensors",
570
+ "layer.30.attention.attn_plus.o_proj.bias": "vision_encoder.safetensors",
571
+ "layer.30.attention.attn_plus.o_proj.weight": "vision_encoder.safetensors",
572
+ "layer.30.attention.attn_plus.q_proj.weight": "vision_encoder.safetensors",
573
+ "layer.30.attention.attn_plus.v_proj.weight": "vision_encoder.safetensors",
574
+ "layer.30.layer_scale1.lambda1": "vision_encoder.safetensors",
575
+ "layer.30.layer_scale2.lambda1": "vision_encoder.safetensors",
576
+ "layer.30.mlp.down_proj.bias": "vision_encoder.safetensors",
577
+ "layer.30.mlp.down_proj.weight": "vision_encoder.safetensors",
578
+ "layer.30.mlp.gate_proj.bias": "vision_encoder.safetensors",
579
+ "layer.30.mlp.gate_proj.weight": "vision_encoder.safetensors",
580
+ "layer.30.mlp.up_proj.bias": "vision_encoder.safetensors",
581
+ "layer.30.mlp.up_proj.weight": "vision_encoder.safetensors",
582
+ "layer.30.norm1.bias": "vision_encoder.safetensors",
583
+ "layer.30.norm1.weight": "vision_encoder.safetensors",
584
+ "layer.30.norm2.bias": "vision_encoder.safetensors",
585
+ "layer.30.norm2.weight": "vision_encoder.safetensors",
586
+ "layer.31.attention.alpha_logit": "vision_encoder.safetensors",
587
+ "layer.31.attention.attn_base.k_proj.weight": "vision_encoder.safetensors",
588
+ "layer.31.attention.attn_base.o_proj.bias": "vision_encoder.safetensors",
589
+ "layer.31.attention.attn_base.o_proj.weight": "vision_encoder.safetensors",
590
+ "layer.31.attention.attn_base.q_proj.weight": "vision_encoder.safetensors",
591
+ "layer.31.attention.attn_base.v_proj.weight": "vision_encoder.safetensors",
592
+ "layer.31.attention.attn_plus.k_proj.weight": "vision_encoder.safetensors",
593
+ "layer.31.attention.attn_plus.o_proj.bias": "vision_encoder.safetensors",
594
+ "layer.31.attention.attn_plus.o_proj.weight": "vision_encoder.safetensors",
595
+ "layer.31.attention.attn_plus.q_proj.weight": "vision_encoder.safetensors",
596
+ "layer.31.attention.attn_plus.v_proj.weight": "vision_encoder.safetensors",
597
+ "layer.31.layer_scale1.lambda1": "vision_encoder.safetensors",
598
+ "layer.31.layer_scale2.lambda1": "vision_encoder.safetensors",
599
+ "layer.31.mlp.down_proj.bias": "vision_encoder.safetensors",
600
+ "layer.31.mlp.down_proj.weight": "vision_encoder.safetensors",
601
+ "layer.31.mlp.gate_proj.bias": "vision_encoder.safetensors",
602
+ "layer.31.mlp.gate_proj.weight": "vision_encoder.safetensors",
603
+ "layer.31.mlp.up_proj.bias": "vision_encoder.safetensors",
604
+ "layer.31.mlp.up_proj.weight": "vision_encoder.safetensors",
605
+ "layer.31.norm1.bias": "vision_encoder.safetensors",
606
+ "layer.31.norm1.weight": "vision_encoder.safetensors",
607
+ "layer.31.norm2.bias": "vision_encoder.safetensors",
608
+ "layer.31.norm2.weight": "vision_encoder.safetensors",
609
+ "layer.32.attention.alpha_logit": "vision_encoder.safetensors",
610
+ "layer.32.attention.attn_base.k_proj.weight": "vision_encoder.safetensors",
611
+ "layer.32.attention.attn_base.o_proj.bias": "vision_encoder.safetensors",
612
+ "layer.32.attention.attn_base.o_proj.weight": "vision_encoder.safetensors",
613
+ "layer.32.attention.attn_base.q_proj.weight": "vision_encoder.safetensors",
614
+ "layer.32.attention.attn_base.v_proj.weight": "vision_encoder.safetensors",
615
+ "layer.32.attention.attn_plus.k_proj.weight": "vision_encoder.safetensors",
616
+ "layer.32.attention.attn_plus.o_proj.bias": "vision_encoder.safetensors",
617
+ "layer.32.attention.attn_plus.o_proj.weight": "vision_encoder.safetensors",
618
+ "layer.32.attention.attn_plus.q_proj.weight": "vision_encoder.safetensors",
619
+ "layer.32.attention.attn_plus.v_proj.weight": "vision_encoder.safetensors",
620
+ "layer.32.layer_scale1.lambda1": "vision_encoder.safetensors",
621
+ "layer.32.layer_scale2.lambda1": "vision_encoder.safetensors",
622
+ "layer.32.mlp.down_proj.bias": "vision_encoder.safetensors",
623
+ "layer.32.mlp.down_proj.weight": "vision_encoder.safetensors",
624
+ "layer.32.mlp.gate_proj.bias": "vision_encoder.safetensors",
625
+ "layer.32.mlp.gate_proj.weight": "vision_encoder.safetensors",
626
+ "layer.32.mlp.up_proj.bias": "vision_encoder.safetensors",
627
+ "layer.32.mlp.up_proj.weight": "vision_encoder.safetensors",
628
+ "layer.32.norm1.bias": "vision_encoder.safetensors",
629
+ "layer.32.norm1.weight": "vision_encoder.safetensors",
630
+ "layer.32.norm2.bias": "vision_encoder.safetensors",
631
+ "layer.32.norm2.weight": "vision_encoder.safetensors",
632
+ "layer.33.attention.alpha_logit": "vision_encoder.safetensors",
633
+ "layer.33.attention.attn_base.k_proj.weight": "vision_encoder.safetensors",
634
+ "layer.33.attention.attn_base.o_proj.bias": "vision_encoder.safetensors",
635
+ "layer.33.attention.attn_base.o_proj.weight": "vision_encoder.safetensors",
636
+ "layer.33.attention.attn_base.q_proj.weight": "vision_encoder.safetensors",
637
+ "layer.33.attention.attn_base.v_proj.weight": "vision_encoder.safetensors",
638
+ "layer.33.attention.attn_plus.k_proj.weight": "vision_encoder.safetensors",
639
+ "layer.33.attention.attn_plus.o_proj.bias": "vision_encoder.safetensors",
640
+ "layer.33.attention.attn_plus.o_proj.weight": "vision_encoder.safetensors",
641
+ "layer.33.attention.attn_plus.q_proj.weight": "vision_encoder.safetensors",
642
+ "layer.33.attention.attn_plus.v_proj.weight": "vision_encoder.safetensors",
643
+ "layer.33.layer_scale1.lambda1": "vision_encoder.safetensors",
644
+ "layer.33.layer_scale2.lambda1": "vision_encoder.safetensors",
645
+ "layer.33.mlp.down_proj.bias": "vision_encoder.safetensors",
646
+ "layer.33.mlp.down_proj.weight": "vision_encoder.safetensors",
647
+ "layer.33.mlp.gate_proj.bias": "vision_encoder.safetensors",
648
+ "layer.33.mlp.gate_proj.weight": "vision_encoder.safetensors",
649
+ "layer.33.mlp.up_proj.bias": "vision_encoder.safetensors",
650
+ "layer.33.mlp.up_proj.weight": "vision_encoder.safetensors",
651
+ "layer.33.norm1.bias": "vision_encoder.safetensors",
652
+ "layer.33.norm1.weight": "vision_encoder.safetensors",
653
+ "layer.33.norm2.bias": "vision_encoder.safetensors",
654
+ "layer.33.norm2.weight": "vision_encoder.safetensors",
655
+ "layer.34.attention.alpha_logit": "vision_encoder.safetensors",
656
+ "layer.34.attention.attn_base.k_proj.weight": "vision_encoder.safetensors",
657
+ "layer.34.attention.attn_base.o_proj.bias": "vision_encoder.safetensors",
658
+ "layer.34.attention.attn_base.o_proj.weight": "vision_encoder.safetensors",
659
+ "layer.34.attention.attn_base.q_proj.weight": "vision_encoder.safetensors",
660
+ "layer.34.attention.attn_base.v_proj.weight": "vision_encoder.safetensors",
661
+ "layer.34.attention.attn_plus.k_proj.weight": "vision_encoder.safetensors",
662
+ "layer.34.attention.attn_plus.o_proj.bias": "vision_encoder.safetensors",
663
+ "layer.34.attention.attn_plus.o_proj.weight": "vision_encoder.safetensors",
664
+ "layer.34.attention.attn_plus.q_proj.weight": "vision_encoder.safetensors",
665
+ "layer.34.attention.attn_plus.v_proj.weight": "vision_encoder.safetensors",
666
+ "layer.34.layer_scale1.lambda1": "vision_encoder.safetensors",
667
+ "layer.34.layer_scale2.lambda1": "vision_encoder.safetensors",
668
+ "layer.34.mlp.down_proj.bias": "vision_encoder.safetensors",
669
+ "layer.34.mlp.down_proj.weight": "vision_encoder.safetensors",
670
+ "layer.34.mlp.gate_proj.bias": "vision_encoder.safetensors",
671
+ "layer.34.mlp.gate_proj.weight": "vision_encoder.safetensors",
672
+ "layer.34.mlp.up_proj.bias": "vision_encoder.safetensors",
673
+ "layer.34.mlp.up_proj.weight": "vision_encoder.safetensors",
674
+ "layer.34.norm1.bias": "vision_encoder.safetensors",
675
+ "layer.34.norm1.weight": "vision_encoder.safetensors",
676
+ "layer.34.norm2.bias": "vision_encoder.safetensors",
677
+ "layer.34.norm2.weight": "vision_encoder.safetensors",
678
+ "layer.35.attention.alpha_logit": "vision_encoder.safetensors",
679
+ "layer.35.attention.attn_base.k_proj.weight": "vision_encoder.safetensors",
680
+ "layer.35.attention.attn_base.o_proj.bias": "vision_encoder.safetensors",
681
+ "layer.35.attention.attn_base.o_proj.weight": "vision_encoder.safetensors",
682
+ "layer.35.attention.attn_base.q_proj.weight": "vision_encoder.safetensors",
683
+ "layer.35.attention.attn_base.v_proj.weight": "vision_encoder.safetensors",
684
+ "layer.35.attention.attn_plus.k_proj.weight": "vision_encoder.safetensors",
685
+ "layer.35.attention.attn_plus.o_proj.bias": "vision_encoder.safetensors",
686
+ "layer.35.attention.attn_plus.o_proj.weight": "vision_encoder.safetensors",
687
+ "layer.35.attention.attn_plus.q_proj.weight": "vision_encoder.safetensors",
688
+ "layer.35.attention.attn_plus.v_proj.weight": "vision_encoder.safetensors",
689
+ "layer.35.layer_scale1.lambda1": "vision_encoder.safetensors",
690
+ "layer.35.layer_scale2.lambda1": "vision_encoder.safetensors",
691
+ "layer.35.mlp.down_proj.bias": "vision_encoder.safetensors",
692
+ "layer.35.mlp.down_proj.weight": "vision_encoder.safetensors",
693
+ "layer.35.mlp.gate_proj.bias": "vision_encoder.safetensors",
694
+ "layer.35.mlp.gate_proj.weight": "vision_encoder.safetensors",
695
+ "layer.35.mlp.up_proj.bias": "vision_encoder.safetensors",
696
+ "layer.35.mlp.up_proj.weight": "vision_encoder.safetensors",
697
+ "layer.35.norm1.bias": "vision_encoder.safetensors",
698
+ "layer.35.norm1.weight": "vision_encoder.safetensors",
699
+ "layer.35.norm2.bias": "vision_encoder.safetensors",
700
+ "layer.35.norm2.weight": "vision_encoder.safetensors",
701
+ "layer.36.attention.alpha_logit": "vision_encoder.safetensors",
702
+ "layer.36.attention.attn_base.k_proj.weight": "vision_encoder.safetensors",
703
+ "layer.36.attention.attn_base.o_proj.bias": "vision_encoder.safetensors",
704
+ "layer.36.attention.attn_base.o_proj.weight": "vision_encoder.safetensors",
705
+ "layer.36.attention.attn_base.q_proj.weight": "vision_encoder.safetensors",
706
+ "layer.36.attention.attn_base.v_proj.weight": "vision_encoder.safetensors",
707
+ "layer.36.attention.attn_plus.k_proj.weight": "vision_encoder.safetensors",
708
+ "layer.36.attention.attn_plus.o_proj.bias": "vision_encoder.safetensors",
709
+ "layer.36.attention.attn_plus.o_proj.weight": "vision_encoder.safetensors",
710
+ "layer.36.attention.attn_plus.q_proj.weight": "vision_encoder.safetensors",
711
+ "layer.36.attention.attn_plus.v_proj.weight": "vision_encoder.safetensors",
712
+ "layer.36.layer_scale1.lambda1": "vision_encoder.safetensors",
713
+ "layer.36.layer_scale2.lambda1": "vision_encoder.safetensors",
714
+ "layer.36.mlp.down_proj.bias": "vision_encoder.safetensors",
715
+ "layer.36.mlp.down_proj.weight": "vision_encoder.safetensors",
716
+ "layer.36.mlp.gate_proj.bias": "vision_encoder.safetensors",
717
+ "layer.36.mlp.gate_proj.weight": "vision_encoder.safetensors",
718
+ "layer.36.mlp.up_proj.bias": "vision_encoder.safetensors",
719
+ "layer.36.mlp.up_proj.weight": "vision_encoder.safetensors",
720
+ "layer.36.norm1.bias": "vision_encoder.safetensors",
721
+ "layer.36.norm1.weight": "vision_encoder.safetensors",
722
+ "layer.36.norm2.bias": "vision_encoder.safetensors",
723
+ "layer.36.norm2.weight": "vision_encoder.safetensors",
724
+ "layer.37.attention.alpha_logit": "vision_encoder.safetensors",
725
+ "layer.37.attention.attn_base.k_proj.weight": "vision_encoder.safetensors",
726
+ "layer.37.attention.attn_base.o_proj.bias": "vision_encoder.safetensors",
727
+ "layer.37.attention.attn_base.o_proj.weight": "vision_encoder.safetensors",
728
+ "layer.37.attention.attn_base.q_proj.weight": "vision_encoder.safetensors",
729
+ "layer.37.attention.attn_base.v_proj.weight": "vision_encoder.safetensors",
730
+ "layer.37.attention.attn_plus.k_proj.weight": "vision_encoder.safetensors",
731
+ "layer.37.attention.attn_plus.o_proj.bias": "vision_encoder.safetensors",
732
+ "layer.37.attention.attn_plus.o_proj.weight": "vision_encoder.safetensors",
733
+ "layer.37.attention.attn_plus.q_proj.weight": "vision_encoder.safetensors",
734
+ "layer.37.attention.attn_plus.v_proj.weight": "vision_encoder.safetensors",
735
+ "layer.37.layer_scale1.lambda1": "vision_encoder.safetensors",
736
+ "layer.37.layer_scale2.lambda1": "vision_encoder.safetensors",
737
+ "layer.37.mlp.down_proj.bias": "vision_encoder.safetensors",
738
+ "layer.37.mlp.down_proj.weight": "vision_encoder.safetensors",
739
+ "layer.37.mlp.gate_proj.bias": "vision_encoder.safetensors",
740
+ "layer.37.mlp.gate_proj.weight": "vision_encoder.safetensors",
741
+ "layer.37.mlp.up_proj.bias": "vision_encoder.safetensors",
742
+ "layer.37.mlp.up_proj.weight": "vision_encoder.safetensors",
743
+ "layer.37.norm1.bias": "vision_encoder.safetensors",
744
+ "layer.37.norm1.weight": "vision_encoder.safetensors",
745
+ "layer.37.norm2.bias": "vision_encoder.safetensors",
746
+ "layer.37.norm2.weight": "vision_encoder.safetensors",
747
+ "layer.38.attention.alpha_logit": "vision_encoder.safetensors",
748
+ "layer.38.attention.attn_base.k_proj.weight": "vision_encoder.safetensors",
749
+ "layer.38.attention.attn_base.o_proj.bias": "vision_encoder.safetensors",
750
+ "layer.38.attention.attn_base.o_proj.weight": "vision_encoder.safetensors",
751
+ "layer.38.attention.attn_base.q_proj.weight": "vision_encoder.safetensors",
752
+ "layer.38.attention.attn_base.v_proj.weight": "vision_encoder.safetensors",
753
+ "layer.38.attention.attn_plus.k_proj.weight": "vision_encoder.safetensors",
754
+ "layer.38.attention.attn_plus.o_proj.bias": "vision_encoder.safetensors",
755
+ "layer.38.attention.attn_plus.o_proj.weight": "vision_encoder.safetensors",
756
+ "layer.38.attention.attn_plus.q_proj.weight": "vision_encoder.safetensors",
757
+ "layer.38.attention.attn_plus.v_proj.weight": "vision_encoder.safetensors",
758
+ "layer.38.layer_scale1.lambda1": "vision_encoder.safetensors",
759
+ "layer.38.layer_scale2.lambda1": "vision_encoder.safetensors",
760
+ "layer.38.mlp.down_proj.bias": "vision_encoder.safetensors",
761
+ "layer.38.mlp.down_proj.weight": "vision_encoder.safetensors",
762
+ "layer.38.mlp.gate_proj.bias": "vision_encoder.safetensors",
763
+ "layer.38.mlp.gate_proj.weight": "vision_encoder.safetensors",
764
+ "layer.38.mlp.up_proj.bias": "vision_encoder.safetensors",
765
+ "layer.38.mlp.up_proj.weight": "vision_encoder.safetensors",
766
+ "layer.38.norm1.bias": "vision_encoder.safetensors",
767
+ "layer.38.norm1.weight": "vision_encoder.safetensors",
768
+ "layer.38.norm2.bias": "vision_encoder.safetensors",
769
+ "layer.38.norm2.weight": "vision_encoder.safetensors",
770
+ "layer.39.attention.alpha_logit": "vision_encoder.safetensors",
771
+ "layer.39.attention.attn_base.k_proj.weight": "vision_encoder.safetensors",
772
+ "layer.39.attention.attn_base.o_proj.bias": "vision_encoder.safetensors",
773
+ "layer.39.attention.attn_base.o_proj.weight": "vision_encoder.safetensors",
774
+ "layer.39.attention.attn_base.q_proj.weight": "vision_encoder.safetensors",
775
+ "layer.39.attention.attn_base.v_proj.weight": "vision_encoder.safetensors",
776
+ "layer.39.attention.attn_plus.k_proj.weight": "vision_encoder.safetensors",
777
+ "layer.39.attention.attn_plus.o_proj.bias": "vision_encoder.safetensors",
778
+ "layer.39.attention.attn_plus.o_proj.weight": "vision_encoder.safetensors",
779
+ "layer.39.attention.attn_plus.q_proj.weight": "vision_encoder.safetensors",
780
+ "layer.39.attention.attn_plus.v_proj.weight": "vision_encoder.safetensors",
781
+ "layer.39.layer_scale1.lambda1": "vision_encoder.safetensors",
782
+ "layer.39.layer_scale2.lambda1": "vision_encoder.safetensors",
783
+ "layer.39.mlp.down_proj.bias": "vision_encoder.safetensors",
784
+ "layer.39.mlp.down_proj.weight": "vision_encoder.safetensors",
785
+ "layer.39.mlp.gate_proj.bias": "vision_encoder.safetensors",
786
+ "layer.39.mlp.gate_proj.weight": "vision_encoder.safetensors",
787
+ "layer.39.mlp.up_proj.bias": "vision_encoder.safetensors",
788
+ "layer.39.mlp.up_proj.weight": "vision_encoder.safetensors",
789
+ "layer.39.norm1.bias": "vision_encoder.safetensors",
790
+ "layer.39.norm1.weight": "vision_encoder.safetensors",
791
+ "layer.39.norm2.bias": "vision_encoder.safetensors",
792
+ "layer.39.norm2.weight": "vision_encoder.safetensors",
793
+ "layer.4.attention.alpha_logit": "vision_encoder.safetensors",
794
+ "layer.4.attention.attn_base.k_proj.weight": "vision_encoder.safetensors",
795
+ "layer.4.attention.attn_base.o_proj.bias": "vision_encoder.safetensors",
796
+ "layer.4.attention.attn_base.o_proj.weight": "vision_encoder.safetensors",
797
+ "layer.4.attention.attn_base.q_proj.weight": "vision_encoder.safetensors",
798
+ "layer.4.attention.attn_base.v_proj.weight": "vision_encoder.safetensors",
799
+ "layer.4.attention.attn_plus.k_proj.weight": "vision_encoder.safetensors",
800
+ "layer.4.attention.attn_plus.o_proj.bias": "vision_encoder.safetensors",
801
+ "layer.4.attention.attn_plus.o_proj.weight": "vision_encoder.safetensors",
802
+ "layer.4.attention.attn_plus.q_proj.weight": "vision_encoder.safetensors",
803
+ "layer.4.attention.attn_plus.v_proj.weight": "vision_encoder.safetensors",
804
+ "layer.4.layer_scale1.lambda1": "vision_encoder.safetensors",
805
+ "layer.4.layer_scale2.lambda1": "vision_encoder.safetensors",
806
+ "layer.4.mlp.down_proj.bias": "vision_encoder.safetensors",
807
+ "layer.4.mlp.down_proj.weight": "vision_encoder.safetensors",
808
+ "layer.4.mlp.gate_proj.bias": "vision_encoder.safetensors",
809
+ "layer.4.mlp.gate_proj.weight": "vision_encoder.safetensors",
810
+ "layer.4.mlp.up_proj.bias": "vision_encoder.safetensors",
811
+ "layer.4.mlp.up_proj.weight": "vision_encoder.safetensors",
812
+ "layer.4.norm1.bias": "vision_encoder.safetensors",
813
+ "layer.4.norm1.weight": "vision_encoder.safetensors",
814
+ "layer.4.norm2.bias": "vision_encoder.safetensors",
815
+ "layer.4.norm2.weight": "vision_encoder.safetensors",
816
+ "layer.5.attention.alpha_logit": "vision_encoder.safetensors",
817
+ "layer.5.attention.attn_base.k_proj.weight": "vision_encoder.safetensors",
818
+ "layer.5.attention.attn_base.o_proj.bias": "vision_encoder.safetensors",
819
+ "layer.5.attention.attn_base.o_proj.weight": "vision_encoder.safetensors",
820
+ "layer.5.attention.attn_base.q_proj.weight": "vision_encoder.safetensors",
821
+ "layer.5.attention.attn_base.v_proj.weight": "vision_encoder.safetensors",
822
+ "layer.5.attention.attn_plus.k_proj.weight": "vision_encoder.safetensors",
823
+ "layer.5.attention.attn_plus.o_proj.bias": "vision_encoder.safetensors",
824
+ "layer.5.attention.attn_plus.o_proj.weight": "vision_encoder.safetensors",
825
+ "layer.5.attention.attn_plus.q_proj.weight": "vision_encoder.safetensors",
826
+ "layer.5.attention.attn_plus.v_proj.weight": "vision_encoder.safetensors",
827
+ "layer.5.layer_scale1.lambda1": "vision_encoder.safetensors",
828
+ "layer.5.layer_scale2.lambda1": "vision_encoder.safetensors",
829
+ "layer.5.mlp.down_proj.bias": "vision_encoder.safetensors",
830
+ "layer.5.mlp.down_proj.weight": "vision_encoder.safetensors",
831
+ "layer.5.mlp.gate_proj.bias": "vision_encoder.safetensors",
832
+ "layer.5.mlp.gate_proj.weight": "vision_encoder.safetensors",
833
+ "layer.5.mlp.up_proj.bias": "vision_encoder.safetensors",
834
+ "layer.5.mlp.up_proj.weight": "vision_encoder.safetensors",
835
+ "layer.5.norm1.bias": "vision_encoder.safetensors",
836
+ "layer.5.norm1.weight": "vision_encoder.safetensors",
837
+ "layer.5.norm2.bias": "vision_encoder.safetensors",
838
+ "layer.5.norm2.weight": "vision_encoder.safetensors",
839
+ "layer.6.attention.alpha_logit": "vision_encoder.safetensors",
840
+ "layer.6.attention.attn_base.k_proj.weight": "vision_encoder.safetensors",
841
+ "layer.6.attention.attn_base.o_proj.bias": "vision_encoder.safetensors",
842
+ "layer.6.attention.attn_base.o_proj.weight": "vision_encoder.safetensors",
843
+ "layer.6.attention.attn_base.q_proj.weight": "vision_encoder.safetensors",
844
+ "layer.6.attention.attn_base.v_proj.weight": "vision_encoder.safetensors",
845
+ "layer.6.attention.attn_plus.k_proj.weight": "vision_encoder.safetensors",
846
+ "layer.6.attention.attn_plus.o_proj.bias": "vision_encoder.safetensors",
847
+ "layer.6.attention.attn_plus.o_proj.weight": "vision_encoder.safetensors",
848
+ "layer.6.attention.attn_plus.q_proj.weight": "vision_encoder.safetensors",
849
+ "layer.6.attention.attn_plus.v_proj.weight": "vision_encoder.safetensors",
850
+ "layer.6.layer_scale1.lambda1": "vision_encoder.safetensors",
851
+ "layer.6.layer_scale2.lambda1": "vision_encoder.safetensors",
852
+ "layer.6.mlp.down_proj.bias": "vision_encoder.safetensors",
853
+ "layer.6.mlp.down_proj.weight": "vision_encoder.safetensors",
854
+ "layer.6.mlp.gate_proj.bias": "vision_encoder.safetensors",
855
+ "layer.6.mlp.gate_proj.weight": "vision_encoder.safetensors",
856
+ "layer.6.mlp.up_proj.bias": "vision_encoder.safetensors",
857
+ "layer.6.mlp.up_proj.weight": "vision_encoder.safetensors",
858
+ "layer.6.norm1.bias": "vision_encoder.safetensors",
859
+ "layer.6.norm1.weight": "vision_encoder.safetensors",
860
+ "layer.6.norm2.bias": "vision_encoder.safetensors",
861
+ "layer.6.norm2.weight": "vision_encoder.safetensors",
862
+ "layer.7.attention.alpha_logit": "vision_encoder.safetensors",
863
+ "layer.7.attention.attn_base.k_proj.weight": "vision_encoder.safetensors",
864
+ "layer.7.attention.attn_base.o_proj.bias": "vision_encoder.safetensors",
865
+ "layer.7.attention.attn_base.o_proj.weight": "vision_encoder.safetensors",
866
+ "layer.7.attention.attn_base.q_proj.weight": "vision_encoder.safetensors",
867
+ "layer.7.attention.attn_base.v_proj.weight": "vision_encoder.safetensors",
868
+ "layer.7.attention.attn_plus.k_proj.weight": "vision_encoder.safetensors",
869
+ "layer.7.attention.attn_plus.o_proj.bias": "vision_encoder.safetensors",
870
+ "layer.7.attention.attn_plus.o_proj.weight": "vision_encoder.safetensors",
871
+ "layer.7.attention.attn_plus.q_proj.weight": "vision_encoder.safetensors",
872
+ "layer.7.attention.attn_plus.v_proj.weight": "vision_encoder.safetensors",
873
+ "layer.7.layer_scale1.lambda1": "vision_encoder.safetensors",
874
+ "layer.7.layer_scale2.lambda1": "vision_encoder.safetensors",
875
+ "layer.7.mlp.down_proj.bias": "vision_encoder.safetensors",
876
+ "layer.7.mlp.down_proj.weight": "vision_encoder.safetensors",
877
+ "layer.7.mlp.gate_proj.bias": "vision_encoder.safetensors",
878
+ "layer.7.mlp.gate_proj.weight": "vision_encoder.safetensors",
879
+ "layer.7.mlp.up_proj.bias": "vision_encoder.safetensors",
880
+ "layer.7.mlp.up_proj.weight": "vision_encoder.safetensors",
881
+ "layer.7.norm1.bias": "vision_encoder.safetensors",
882
+ "layer.7.norm1.weight": "vision_encoder.safetensors",
883
+ "layer.7.norm2.bias": "vision_encoder.safetensors",
884
+ "layer.7.norm2.weight": "vision_encoder.safetensors",
885
+ "layer.8.attention.alpha_logit": "vision_encoder.safetensors",
886
+ "layer.8.attention.attn_base.k_proj.weight": "vision_encoder.safetensors",
887
+ "layer.8.attention.attn_base.o_proj.bias": "vision_encoder.safetensors",
888
+ "layer.8.attention.attn_base.o_proj.weight": "vision_encoder.safetensors",
889
+ "layer.8.attention.attn_base.q_proj.weight": "vision_encoder.safetensors",
890
+ "layer.8.attention.attn_base.v_proj.weight": "vision_encoder.safetensors",
891
+ "layer.8.attention.attn_plus.k_proj.weight": "vision_encoder.safetensors",
892
+ "layer.8.attention.attn_plus.o_proj.bias": "vision_encoder.safetensors",
893
+ "layer.8.attention.attn_plus.o_proj.weight": "vision_encoder.safetensors",
894
+ "layer.8.attention.attn_plus.q_proj.weight": "vision_encoder.safetensors",
895
+ "layer.8.attention.attn_plus.v_proj.weight": "vision_encoder.safetensors",
896
+ "layer.8.layer_scale1.lambda1": "vision_encoder.safetensors",
897
+ "layer.8.layer_scale2.lambda1": "vision_encoder.safetensors",
898
+ "layer.8.mlp.down_proj.bias": "vision_encoder.safetensors",
899
+ "layer.8.mlp.down_proj.weight": "vision_encoder.safetensors",
900
+ "layer.8.mlp.gate_proj.bias": "vision_encoder.safetensors",
901
+ "layer.8.mlp.gate_proj.weight": "vision_encoder.safetensors",
902
+ "layer.8.mlp.up_proj.bias": "vision_encoder.safetensors",
903
+ "layer.8.mlp.up_proj.weight": "vision_encoder.safetensors",
904
+ "layer.8.norm1.bias": "vision_encoder.safetensors",
905
+ "layer.8.norm1.weight": "vision_encoder.safetensors",
906
+ "layer.8.norm2.bias": "vision_encoder.safetensors",
907
+ "layer.8.norm2.weight": "vision_encoder.safetensors",
908
+ "layer.9.attention.alpha_logit": "vision_encoder.safetensors",
909
+ "layer.9.attention.attn_base.k_proj.weight": "vision_encoder.safetensors",
910
+ "layer.9.attention.attn_base.o_proj.bias": "vision_encoder.safetensors",
911
+ "layer.9.attention.attn_base.o_proj.weight": "vision_encoder.safetensors",
912
+ "layer.9.attention.attn_base.q_proj.weight": "vision_encoder.safetensors",
913
+ "layer.9.attention.attn_base.v_proj.weight": "vision_encoder.safetensors",
914
+ "layer.9.attention.attn_plus.k_proj.weight": "vision_encoder.safetensors",
915
+ "layer.9.attention.attn_plus.o_proj.bias": "vision_encoder.safetensors",
916
+ "layer.9.attention.attn_plus.o_proj.weight": "vision_encoder.safetensors",
917
+ "layer.9.attention.attn_plus.q_proj.weight": "vision_encoder.safetensors",
918
+ "layer.9.attention.attn_plus.v_proj.weight": "vision_encoder.safetensors",
919
+ "layer.9.layer_scale1.lambda1": "vision_encoder.safetensors",
920
+ "layer.9.layer_scale2.lambda1": "vision_encoder.safetensors",
921
+ "layer.9.mlp.down_proj.bias": "vision_encoder.safetensors",
922
+ "layer.9.mlp.down_proj.weight": "vision_encoder.safetensors",
923
+ "layer.9.mlp.gate_proj.bias": "vision_encoder.safetensors",
924
+ "layer.9.mlp.gate_proj.weight": "vision_encoder.safetensors",
925
+ "layer.9.mlp.up_proj.bias": "vision_encoder.safetensors",
926
+ "layer.9.mlp.up_proj.weight": "vision_encoder.safetensors",
927
+ "layer.9.norm1.bias": "vision_encoder.safetensors",
928
+ "layer.9.norm1.weight": "vision_encoder.safetensors",
929
+ "layer.9.norm2.bias": "vision_encoder.safetensors",
930
+ "layer.9.norm2.weight": "vision_encoder.safetensors",
931
+ "norm.bias": "vision_encoder.safetensors",
932
+ "norm.weight": "vision_encoder.safetensors"
933
+ }
934
+ }