Hugging Face's logo

mlx-community
/
Perception-LM-1B

Safetensors
Model card Files
xet
Community
Deekshith-Dade commited on
Commit
c46d971
·
verified ·
1 Parent(s): c7b33ab

Upload folder using huggingface_hub

Browse files
Files changed (4) hide show
  1. consolidated.safetensors +3 -0
  2. model.safetensors.index.json +484 -0
  3. params.json +43 -0
  4. tokenizer.model +3 -0
consolidated.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:74eb3f5af733f180635d68b7960b01fcfab2603fa9552a6b62fdcb43ae49a315
3
+ size 6134156483
model.safetensors.index.json ADDED
@@ -0,0 +1,484 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 6134099968
4
+ },
5
+ "weight_map": {
6
+ "layers.0.attention.wk.weight": "consolidated.safetensors",
7
+ "layers.0.attention.wo.weight": "consolidated.safetensors",
8
+ "layers.0.attention.wq.weight": "consolidated.safetensors",
9
+ "layers.0.attention.wv.weight": "consolidated.safetensors",
10
+ "layers.0.attention_norm.weight": "consolidated.safetensors",
11
+ "layers.0.feed_forward.w1.weight": "consolidated.safetensors",
12
+ "layers.0.feed_forward.w2.weight": "consolidated.safetensors",
13
+ "layers.0.feed_forward.w3.weight": "consolidated.safetensors",
14
+ "layers.0.ffn_norm.weight": "consolidated.safetensors",
15
+ "layers.1.attention.wk.weight": "consolidated.safetensors",
16
+ "layers.1.attention.wo.weight": "consolidated.safetensors",
17
+ "layers.1.attention.wq.weight": "consolidated.safetensors",
18
+ "layers.1.attention.wv.weight": "consolidated.safetensors",
19
+ "layers.1.attention_norm.weight": "consolidated.safetensors",
20
+ "layers.1.feed_forward.w1.weight": "consolidated.safetensors",
21
+ "layers.1.feed_forward.w2.weight": "consolidated.safetensors",
22
+ "layers.1.feed_forward.w3.weight": "consolidated.safetensors",
23
+ "layers.1.ffn_norm.weight": "consolidated.safetensors",
24
+ "layers.10.attention.wk.weight": "consolidated.safetensors",
25
+ "layers.10.attention.wo.weight": "consolidated.safetensors",
26
+ "layers.10.attention.wq.weight": "consolidated.safetensors",
27
+ "layers.10.attention.wv.weight": "consolidated.safetensors",
28
+ "layers.10.attention_norm.weight": "consolidated.safetensors",
29
+ "layers.10.feed_forward.w1.weight": "consolidated.safetensors",
30
+ "layers.10.feed_forward.w2.weight": "consolidated.safetensors",
31
+ "layers.10.feed_forward.w3.weight": "consolidated.safetensors",
32
+ "layers.10.ffn_norm.weight": "consolidated.safetensors",
33
+ "layers.11.attention.wk.weight": "consolidated.safetensors",
34
+ "layers.11.attention.wo.weight": "consolidated.safetensors",
35
+ "layers.11.attention.wq.weight": "consolidated.safetensors",
36
+ "layers.11.attention.wv.weight": "consolidated.safetensors",
37
+ "layers.11.attention_norm.weight": "consolidated.safetensors",
38
+ "layers.11.feed_forward.w1.weight": "consolidated.safetensors",
39
+ "layers.11.feed_forward.w2.weight": "consolidated.safetensors",
40
+ "layers.11.feed_forward.w3.weight": "consolidated.safetensors",
41
+ "layers.11.ffn_norm.weight": "consolidated.safetensors",
42
+ "layers.12.attention.wk.weight": "consolidated.safetensors",
43
+ "layers.12.attention.wo.weight": "consolidated.safetensors",
44
+ "layers.12.attention.wq.weight": "consolidated.safetensors",
45
+ "layers.12.attention.wv.weight": "consolidated.safetensors",
46
+ "layers.12.attention_norm.weight": "consolidated.safetensors",
47
+ "layers.12.feed_forward.w1.weight": "consolidated.safetensors",
48
+ "layers.12.feed_forward.w2.weight": "consolidated.safetensors",
49
+ "layers.12.feed_forward.w3.weight": "consolidated.safetensors",
50
+ "layers.12.ffn_norm.weight": "consolidated.safetensors",
51
+ "layers.13.attention.wk.weight": "consolidated.safetensors",
52
+ "layers.13.attention.wo.weight": "consolidated.safetensors",
53
+ "layers.13.attention.wq.weight": "consolidated.safetensors",
54
+ "layers.13.attention.wv.weight": "consolidated.safetensors",
55
+ "layers.13.attention_norm.weight": "consolidated.safetensors",
56
+ "layers.13.feed_forward.w1.weight": "consolidated.safetensors",
57
+ "layers.13.feed_forward.w2.weight": "consolidated.safetensors",
58
+ "layers.13.feed_forward.w3.weight": "consolidated.safetensors",
59
+ "layers.13.ffn_norm.weight": "consolidated.safetensors",
60
+ "layers.14.attention.wk.weight": "consolidated.safetensors",
61
+ "layers.14.attention.wo.weight": "consolidated.safetensors",
62
+ "layers.14.attention.wq.weight": "consolidated.safetensors",
63
+ "layers.14.attention.wv.weight": "consolidated.safetensors",
64
+ "layers.14.attention_norm.weight": "consolidated.safetensors",
65
+ "layers.14.feed_forward.w1.weight": "consolidated.safetensors",
66
+ "layers.14.feed_forward.w2.weight": "consolidated.safetensors",
67
+ "layers.14.feed_forward.w3.weight": "consolidated.safetensors",
68
+ "layers.14.ffn_norm.weight": "consolidated.safetensors",
69
+ "layers.15.attention.wk.weight": "consolidated.safetensors",
70
+ "layers.15.attention.wo.weight": "consolidated.safetensors",
71
+ "layers.15.attention.wq.weight": "consolidated.safetensors",
72
+ "layers.15.attention.wv.weight": "consolidated.safetensors",
73
+ "layers.15.attention_norm.weight": "consolidated.safetensors",
74
+ "layers.15.feed_forward.w1.weight": "consolidated.safetensors",
75
+ "layers.15.feed_forward.w2.weight": "consolidated.safetensors",
76
+ "layers.15.feed_forward.w3.weight": "consolidated.safetensors",
77
+ "layers.15.ffn_norm.weight": "consolidated.safetensors",
78
+ "layers.2.attention.wk.weight": "consolidated.safetensors",
79
+ "layers.2.attention.wo.weight": "consolidated.safetensors",
80
+ "layers.2.attention.wq.weight": "consolidated.safetensors",
81
+ "layers.2.attention.wv.weight": "consolidated.safetensors",
82
+ "layers.2.attention_norm.weight": "consolidated.safetensors",
83
+ "layers.2.feed_forward.w1.weight": "consolidated.safetensors",
84
+ "layers.2.feed_forward.w2.weight": "consolidated.safetensors",
85
+ "layers.2.feed_forward.w3.weight": "consolidated.safetensors",
86
+ "layers.2.ffn_norm.weight": "consolidated.safetensors",
87
+ "layers.3.attention.wk.weight": "consolidated.safetensors",
88
+ "layers.3.attention.wo.weight": "consolidated.safetensors",
89
+ "layers.3.attention.wq.weight": "consolidated.safetensors",
90
+ "layers.3.attention.wv.weight": "consolidated.safetensors",
91
+ "layers.3.attention_norm.weight": "consolidated.safetensors",
92
+ "layers.3.feed_forward.w1.weight": "consolidated.safetensors",
93
+ "layers.3.feed_forward.w2.weight": "consolidated.safetensors",
94
+ "layers.3.feed_forward.w3.weight": "consolidated.safetensors",
95
+ "layers.3.ffn_norm.weight": "consolidated.safetensors",
96
+ "layers.4.attention.wk.weight": "consolidated.safetensors",
97
+ "layers.4.attention.wo.weight": "consolidated.safetensors",
98
+ "layers.4.attention.wq.weight": "consolidated.safetensors",
99
+ "layers.4.attention.wv.weight": "consolidated.safetensors",
100
+ "layers.4.attention_norm.weight": "consolidated.safetensors",
101
+ "layers.4.feed_forward.w1.weight": "consolidated.safetensors",
102
+ "layers.4.feed_forward.w2.weight": "consolidated.safetensors",
103
+ "layers.4.feed_forward.w3.weight": "consolidated.safetensors",
104
+ "layers.4.ffn_norm.weight": "consolidated.safetensors",
105
+ "layers.5.attention.wk.weight": "consolidated.safetensors",
106
+ "layers.5.attention.wo.weight": "consolidated.safetensors",
107
+ "layers.5.attention.wq.weight": "consolidated.safetensors",
108
+ "layers.5.attention.wv.weight": "consolidated.safetensors",
109
+ "layers.5.attention_norm.weight": "consolidated.safetensors",
110
+ "layers.5.feed_forward.w1.weight": "consolidated.safetensors",
111
+ "layers.5.feed_forward.w2.weight": "consolidated.safetensors",
112
+ "layers.5.feed_forward.w3.weight": "consolidated.safetensors",
113
+ "layers.5.ffn_norm.weight": "consolidated.safetensors",
114
+ "layers.6.attention.wk.weight": "consolidated.safetensors",
115
+ "layers.6.attention.wo.weight": "consolidated.safetensors",
116
+ "layers.6.attention.wq.weight": "consolidated.safetensors",
117
+ "layers.6.attention.wv.weight": "consolidated.safetensors",
118
+ "layers.6.attention_norm.weight": "consolidated.safetensors",
119
+ "layers.6.feed_forward.w1.weight": "consolidated.safetensors",
120
+ "layers.6.feed_forward.w2.weight": "consolidated.safetensors",
121
+ "layers.6.feed_forward.w3.weight": "consolidated.safetensors",
122
+ "layers.6.ffn_norm.weight": "consolidated.safetensors",
123
+ "layers.7.attention.wk.weight": "consolidated.safetensors",
124
+ "layers.7.attention.wo.weight": "consolidated.safetensors",
125
+ "layers.7.attention.wq.weight": "consolidated.safetensors",
126
+ "layers.7.attention.wv.weight": "consolidated.safetensors",
127
+ "layers.7.attention_norm.weight": "consolidated.safetensors",
128
+ "layers.7.feed_forward.w1.weight": "consolidated.safetensors",
129
+ "layers.7.feed_forward.w2.weight": "consolidated.safetensors",
130
+ "layers.7.feed_forward.w3.weight": "consolidated.safetensors",
131
+ "layers.7.ffn_norm.weight": "consolidated.safetensors",
132
+ "layers.8.attention.wk.weight": "consolidated.safetensors",
133
+ "layers.8.attention.wo.weight": "consolidated.safetensors",
134
+ "layers.8.attention.wq.weight": "consolidated.safetensors",
135
+ "layers.8.attention.wv.weight": "consolidated.safetensors",
136
+ "layers.8.attention_norm.weight": "consolidated.safetensors",
137
+ "layers.8.feed_forward.w1.weight": "consolidated.safetensors",
138
+ "layers.8.feed_forward.w2.weight": "consolidated.safetensors",
139
+ "layers.8.feed_forward.w3.weight": "consolidated.safetensors",
140
+ "layers.8.ffn_norm.weight": "consolidated.safetensors",
141
+ "layers.9.attention.wk.weight": "consolidated.safetensors",
142
+ "layers.9.attention.wo.weight": "consolidated.safetensors",
143
+ "layers.9.attention.wq.weight": "consolidated.safetensors",
144
+ "layers.9.attention.wv.weight": "consolidated.safetensors",
145
+ "layers.9.attention_norm.weight": "consolidated.safetensors",
146
+ "layers.9.feed_forward.w1.weight": "consolidated.safetensors",
147
+ "layers.9.feed_forward.w2.weight": "consolidated.safetensors",
148
+ "layers.9.feed_forward.w3.weight": "consolidated.safetensors",
149
+ "layers.9.ffn_norm.weight": "consolidated.safetensors",
150
+ "norm.weight": "consolidated.safetensors",
151
+ "tok_embeddings.weight": "consolidated.safetensors",
152
+ "vision_model.class_embedding": "consolidated.safetensors",
153
+ "vision_model.conv1.weight": "consolidated.safetensors",
154
+ "vision_model.ln_pre.bias": "consolidated.safetensors",
155
+ "vision_model.ln_pre.weight": "consolidated.safetensors",
156
+ "vision_model.positional_embedding": "consolidated.safetensors",
157
+ "vision_model.transformer.resblocks.0.attn.in_proj_bias": "consolidated.safetensors",
158
+ "vision_model.transformer.resblocks.0.attn.in_proj_weight": "consolidated.safetensors",
159
+ "vision_model.transformer.resblocks.0.attn.out_proj.bias": "consolidated.safetensors",
160
+ "vision_model.transformer.resblocks.0.attn.out_proj.weight": "consolidated.safetensors",
161
+ "vision_model.transformer.resblocks.0.ln_1.bias": "consolidated.safetensors",
162
+ "vision_model.transformer.resblocks.0.ln_1.weight": "consolidated.safetensors",
163
+ "vision_model.transformer.resblocks.0.ln_2.bias": "consolidated.safetensors",
164
+ "vision_model.transformer.resblocks.0.ln_2.weight": "consolidated.safetensors",
165
+ "vision_model.transformer.resblocks.0.ls_1.gamma": "consolidated.safetensors",
166
+ "vision_model.transformer.resblocks.0.ls_2.gamma": "consolidated.safetensors",
167
+ "vision_model.transformer.resblocks.0.mlp.c_fc.bias": "consolidated.safetensors",
168
+ "vision_model.transformer.resblocks.0.mlp.c_fc.weight": "consolidated.safetensors",
169
+ "vision_model.transformer.resblocks.0.mlp.c_proj.bias": "consolidated.safetensors",
170
+ "vision_model.transformer.resblocks.0.mlp.c_proj.weight": "consolidated.safetensors",
171
+ "vision_model.transformer.resblocks.1.attn.in_proj_bias": "consolidated.safetensors",
172
+ "vision_model.transformer.resblocks.1.attn.in_proj_weight": "consolidated.safetensors",
173
+ "vision_model.transformer.resblocks.1.attn.out_proj.bias": "consolidated.safetensors",
174
+ "vision_model.transformer.resblocks.1.attn.out_proj.weight": "consolidated.safetensors",
175
+ "vision_model.transformer.resblocks.1.ln_1.bias": "consolidated.safetensors",
176
+ "vision_model.transformer.resblocks.1.ln_1.weight": "consolidated.safetensors",
177
+ "vision_model.transformer.resblocks.1.ln_2.bias": "consolidated.safetensors",
178
+ "vision_model.transformer.resblocks.1.ln_2.weight": "consolidated.safetensors",
179
+ "vision_model.transformer.resblocks.1.ls_1.gamma": "consolidated.safetensors",
180
+ "vision_model.transformer.resblocks.1.ls_2.gamma": "consolidated.safetensors",
181
+ "vision_model.transformer.resblocks.1.mlp.c_fc.bias": "consolidated.safetensors",
182
+ "vision_model.transformer.resblocks.1.mlp.c_fc.weight": "consolidated.safetensors",
183
+ "vision_model.transformer.resblocks.1.mlp.c_proj.bias": "consolidated.safetensors",
184
+ "vision_model.transformer.resblocks.1.mlp.c_proj.weight": "consolidated.safetensors",
185
+ "vision_model.transformer.resblocks.10.attn.in_proj_bias": "consolidated.safetensors",
186
+ "vision_model.transformer.resblocks.10.attn.in_proj_weight": "consolidated.safetensors",
187
+ "vision_model.transformer.resblocks.10.attn.out_proj.bias": "consolidated.safetensors",
188
+ "vision_model.transformer.resblocks.10.attn.out_proj.weight": "consolidated.safetensors",
189
+ "vision_model.transformer.resblocks.10.ln_1.bias": "consolidated.safetensors",
190
+ "vision_model.transformer.resblocks.10.ln_1.weight": "consolidated.safetensors",
191
+ "vision_model.transformer.resblocks.10.ln_2.bias": "consolidated.safetensors",
192
+ "vision_model.transformer.resblocks.10.ln_2.weight": "consolidated.safetensors",
193
+ "vision_model.transformer.resblocks.10.ls_1.gamma": "consolidated.safetensors",
194
+ "vision_model.transformer.resblocks.10.ls_2.gamma": "consolidated.safetensors",
195
+ "vision_model.transformer.resblocks.10.mlp.c_fc.bias": "consolidated.safetensors",
196
+ "vision_model.transformer.resblocks.10.mlp.c_fc.weight": "consolidated.safetensors",
197
+ "vision_model.transformer.resblocks.10.mlp.c_proj.bias": "consolidated.safetensors",
198
+ "vision_model.transformer.resblocks.10.mlp.c_proj.weight": "consolidated.safetensors",
199
+ "vision_model.transformer.resblocks.11.attn.in_proj_bias": "consolidated.safetensors",
200
+ "vision_model.transformer.resblocks.11.attn.in_proj_weight": "consolidated.safetensors",
201
+ "vision_model.transformer.resblocks.11.attn.out_proj.bias": "consolidated.safetensors",
202
+ "vision_model.transformer.resblocks.11.attn.out_proj.weight": "consolidated.safetensors",
203
+ "vision_model.transformer.resblocks.11.ln_1.bias": "consolidated.safetensors",
204
+ "vision_model.transformer.resblocks.11.ln_1.weight": "consolidated.safetensors",
205
+ "vision_model.transformer.resblocks.11.ln_2.bias": "consolidated.safetensors",
206
+ "vision_model.transformer.resblocks.11.ln_2.weight": "consolidated.safetensors",
207
+ "vision_model.transformer.resblocks.11.ls_1.gamma": "consolidated.safetensors",
208
+ "vision_model.transformer.resblocks.11.ls_2.gamma": "consolidated.safetensors",
209
+ "vision_model.transformer.resblocks.11.mlp.c_fc.bias": "consolidated.safetensors",
210
+ "vision_model.transformer.resblocks.11.mlp.c_fc.weight": "consolidated.safetensors",
211
+ "vision_model.transformer.resblocks.11.mlp.c_proj.bias": "consolidated.safetensors",
212
+ "vision_model.transformer.resblocks.11.mlp.c_proj.weight": "consolidated.safetensors",
213
+ "vision_model.transformer.resblocks.12.attn.in_proj_bias": "consolidated.safetensors",
214
+ "vision_model.transformer.resblocks.12.attn.in_proj_weight": "consolidated.safetensors",
215
+ "vision_model.transformer.resblocks.12.attn.out_proj.bias": "consolidated.safetensors",
216
+ "vision_model.transformer.resblocks.12.attn.out_proj.weight": "consolidated.safetensors",
217
+ "vision_model.transformer.resblocks.12.ln_1.bias": "consolidated.safetensors",
218
+ "vision_model.transformer.resblocks.12.ln_1.weight": "consolidated.safetensors",
219
+ "vision_model.transformer.resblocks.12.ln_2.bias": "consolidated.safetensors",
220
+ "vision_model.transformer.resblocks.12.ln_2.weight": "consolidated.safetensors",
221
+ "vision_model.transformer.resblocks.12.ls_1.gamma": "consolidated.safetensors",
222
+ "vision_model.transformer.resblocks.12.ls_2.gamma": "consolidated.safetensors",
223
+ "vision_model.transformer.resblocks.12.mlp.c_fc.bias": "consolidated.safetensors",
224
+ "vision_model.transformer.resblocks.12.mlp.c_fc.weight": "consolidated.safetensors",
225
+ "vision_model.transformer.resblocks.12.mlp.c_proj.bias": "consolidated.safetensors",
226
+ "vision_model.transformer.resblocks.12.mlp.c_proj.weight": "consolidated.safetensors",
227
+ "vision_model.transformer.resblocks.13.attn.in_proj_bias": "consolidated.safetensors",
228
+ "vision_model.transformer.resblocks.13.attn.in_proj_weight": "consolidated.safetensors",
229
+ "vision_model.transformer.resblocks.13.attn.out_proj.bias": "consolidated.safetensors",
230
+ "vision_model.transformer.resblocks.13.attn.out_proj.weight": "consolidated.safetensors",
231
+ "vision_model.transformer.resblocks.13.ln_1.bias": "consolidated.safetensors",
232
+ "vision_model.transformer.resblocks.13.ln_1.weight": "consolidated.safetensors",
233
+ "vision_model.transformer.resblocks.13.ln_2.bias": "consolidated.safetensors",
234
+ "vision_model.transformer.resblocks.13.ln_2.weight": "consolidated.safetensors",
235
+ "vision_model.transformer.resblocks.13.ls_1.gamma": "consolidated.safetensors",
236
+ "vision_model.transformer.resblocks.13.ls_2.gamma": "consolidated.safetensors",
237
+ "vision_model.transformer.resblocks.13.mlp.c_fc.bias": "consolidated.safetensors",
238
+ "vision_model.transformer.resblocks.13.mlp.c_fc.weight": "consolidated.safetensors",
239
+ "vision_model.transformer.resblocks.13.mlp.c_proj.bias": "consolidated.safetensors",
240
+ "vision_model.transformer.resblocks.13.mlp.c_proj.weight": "consolidated.safetensors",
241
+ "vision_model.transformer.resblocks.14.attn.in_proj_bias": "consolidated.safetensors",
242
+ "vision_model.transformer.resblocks.14.attn.in_proj_weight": "consolidated.safetensors",
243
+ "vision_model.transformer.resblocks.14.attn.out_proj.bias": "consolidated.safetensors",
244
+ "vision_model.transformer.resblocks.14.attn.out_proj.weight": "consolidated.safetensors",
245
+ "vision_model.transformer.resblocks.14.ln_1.bias": "consolidated.safetensors",
246
+ "vision_model.transformer.resblocks.14.ln_1.weight": "consolidated.safetensors",
247
+ "vision_model.transformer.resblocks.14.ln_2.bias": "consolidated.safetensors",
248
+ "vision_model.transformer.resblocks.14.ln_2.weight": "consolidated.safetensors",
249
+ "vision_model.transformer.resblocks.14.ls_1.gamma": "consolidated.safetensors",
250
+ "vision_model.transformer.resblocks.14.ls_2.gamma": "consolidated.safetensors",
251
+ "vision_model.transformer.resblocks.14.mlp.c_fc.bias": "consolidated.safetensors",
252
+ "vision_model.transformer.resblocks.14.mlp.c_fc.weight": "consolidated.safetensors",
253
+ "vision_model.transformer.resblocks.14.mlp.c_proj.bias": "consolidated.safetensors",
254
+ "vision_model.transformer.resblocks.14.mlp.c_proj.weight": "consolidated.safetensors",
255
+ "vision_model.transformer.resblocks.15.attn.in_proj_bias": "consolidated.safetensors",
256
+ "vision_model.transformer.resblocks.15.attn.in_proj_weight": "consolidated.safetensors",
257
+ "vision_model.transformer.resblocks.15.attn.out_proj.bias": "consolidated.safetensors",
258
+ "vision_model.transformer.resblocks.15.attn.out_proj.weight": "consolidated.safetensors",
259
+ "vision_model.transformer.resblocks.15.ln_1.bias": "consolidated.safetensors",
260
+ "vision_model.transformer.resblocks.15.ln_1.weight": "consolidated.safetensors",
261
+ "vision_model.transformer.resblocks.15.ln_2.bias": "consolidated.safetensors",
262
+ "vision_model.transformer.resblocks.15.ln_2.weight": "consolidated.safetensors",
263
+ "vision_model.transformer.resblocks.15.ls_1.gamma": "consolidated.safetensors",
264
+ "vision_model.transformer.resblocks.15.ls_2.gamma": "consolidated.safetensors",
265
+ "vision_model.transformer.resblocks.15.mlp.c_fc.bias": "consolidated.safetensors",
266
+ "vision_model.transformer.resblocks.15.mlp.c_fc.weight": "consolidated.safetensors",
267
+ "vision_model.transformer.resblocks.15.mlp.c_proj.bias": "consolidated.safetensors",
268
+ "vision_model.transformer.resblocks.15.mlp.c_proj.weight": "consolidated.safetensors",
269
+ "vision_model.transformer.resblocks.16.attn.in_proj_bias": "consolidated.safetensors",
270
+ "vision_model.transformer.resblocks.16.attn.in_proj_weight": "consolidated.safetensors",
271
+ "vision_model.transformer.resblocks.16.attn.out_proj.bias": "consolidated.safetensors",
272
+ "vision_model.transformer.resblocks.16.attn.out_proj.weight": "consolidated.safetensors",
273
+ "vision_model.transformer.resblocks.16.ln_1.bias": "consolidated.safetensors",
274
+ "vision_model.transformer.resblocks.16.ln_1.weight": "consolidated.safetensors",
275
+ "vision_model.transformer.resblocks.16.ln_2.bias": "consolidated.safetensors",
276
+ "vision_model.transformer.resblocks.16.ln_2.weight": "consolidated.safetensors",
277
+ "vision_model.transformer.resblocks.16.ls_1.gamma": "consolidated.safetensors",
278
+ "vision_model.transformer.resblocks.16.ls_2.gamma": "consolidated.safetensors",
279
+ "vision_model.transformer.resblocks.16.mlp.c_fc.bias": "consolidated.safetensors",
280
+ "vision_model.transformer.resblocks.16.mlp.c_fc.weight": "consolidated.safetensors",
281
+ "vision_model.transformer.resblocks.16.mlp.c_proj.bias": "consolidated.safetensors",
282
+ "vision_model.transformer.resblocks.16.mlp.c_proj.weight": "consolidated.safetensors",
283
+ "vision_model.transformer.resblocks.17.attn.in_proj_bias": "consolidated.safetensors",
284
+ "vision_model.transformer.resblocks.17.attn.in_proj_weight": "consolidated.safetensors",
285
+ "vision_model.transformer.resblocks.17.attn.out_proj.bias": "consolidated.safetensors",
286
+ "vision_model.transformer.resblocks.17.attn.out_proj.weight": "consolidated.safetensors",
287
+ "vision_model.transformer.resblocks.17.ln_1.bias": "consolidated.safetensors",
288
+ "vision_model.transformer.resblocks.17.ln_1.weight": "consolidated.safetensors",
289
+ "vision_model.transformer.resblocks.17.ln_2.bias": "consolidated.safetensors",
290
+ "vision_model.transformer.resblocks.17.ln_2.weight": "consolidated.safetensors",
291
+ "vision_model.transformer.resblocks.17.ls_1.gamma": "consolidated.safetensors",
292
+ "vision_model.transformer.resblocks.17.ls_2.gamma": "consolidated.safetensors",
293
+ "vision_model.transformer.resblocks.17.mlp.c_fc.bias": "consolidated.safetensors",
294
+ "vision_model.transformer.resblocks.17.mlp.c_fc.weight": "consolidated.safetensors",
295
+ "vision_model.transformer.resblocks.17.mlp.c_proj.bias": "consolidated.safetensors",
296
+ "vision_model.transformer.resblocks.17.mlp.c_proj.weight": "consolidated.safetensors",
297
+ "vision_model.transformer.resblocks.18.attn.in_proj_bias": "consolidated.safetensors",
298
+ "vision_model.transformer.resblocks.18.attn.in_proj_weight": "consolidated.safetensors",
299
+ "vision_model.transformer.resblocks.18.attn.out_proj.bias": "consolidated.safetensors",
300
+ "vision_model.transformer.resblocks.18.attn.out_proj.weight": "consolidated.safetensors",
301
+ "vision_model.transformer.resblocks.18.ln_1.bias": "consolidated.safetensors",
302
+ "vision_model.transformer.resblocks.18.ln_1.weight": "consolidated.safetensors",
303
+ "vision_model.transformer.resblocks.18.ln_2.bias": "consolidated.safetensors",
304
+ "vision_model.transformer.resblocks.18.ln_2.weight": "consolidated.safetensors",
305
+ "vision_model.transformer.resblocks.18.ls_1.gamma": "consolidated.safetensors",
306
+ "vision_model.transformer.resblocks.18.ls_2.gamma": "consolidated.safetensors",
307
+ "vision_model.transformer.resblocks.18.mlp.c_fc.bias": "consolidated.safetensors",
308
+ "vision_model.transformer.resblocks.18.mlp.c_fc.weight": "consolidated.safetensors",
309
+ "vision_model.transformer.resblocks.18.mlp.c_proj.bias": "consolidated.safetensors",
310
+ "vision_model.transformer.resblocks.18.mlp.c_proj.weight": "consolidated.safetensors",
311
+ "vision_model.transformer.resblocks.19.attn.in_proj_bias": "consolidated.safetensors",
312
+ "vision_model.transformer.resblocks.19.attn.in_proj_weight": "consolidated.safetensors",
313
+ "vision_model.transformer.resblocks.19.attn.out_proj.bias": "consolidated.safetensors",
314
+ "vision_model.transformer.resblocks.19.attn.out_proj.weight": "consolidated.safetensors",
315
+ "vision_model.transformer.resblocks.19.ln_1.bias": "consolidated.safetensors",
316
+ "vision_model.transformer.resblocks.19.ln_1.weight": "consolidated.safetensors",
317
+ "vision_model.transformer.resblocks.19.ln_2.bias": "consolidated.safetensors",
318
+ "vision_model.transformer.resblocks.19.ln_2.weight": "consolidated.safetensors",
319
+ "vision_model.transformer.resblocks.19.ls_1.gamma": "consolidated.safetensors",
320
+ "vision_model.transformer.resblocks.19.ls_2.gamma": "consolidated.safetensors",
321
+ "vision_model.transformer.resblocks.19.mlp.c_fc.bias": "consolidated.safetensors",
322
+ "vision_model.transformer.resblocks.19.mlp.c_fc.weight": "consolidated.safetensors",
323
+ "vision_model.transformer.resblocks.19.mlp.c_proj.bias": "consolidated.safetensors",
324
+ "vision_model.transformer.resblocks.19.mlp.c_proj.weight": "consolidated.safetensors",
325
+ "vision_model.transformer.resblocks.2.attn.in_proj_bias": "consolidated.safetensors",
326
+ "vision_model.transformer.resblocks.2.attn.in_proj_weight": "consolidated.safetensors",
327
+ "vision_model.transformer.resblocks.2.attn.out_proj.bias": "consolidated.safetensors",
328
+ "vision_model.transformer.resblocks.2.attn.out_proj.weight": "consolidated.safetensors",
329
+ "vision_model.transformer.resblocks.2.ln_1.bias": "consolidated.safetensors",
330
+ "vision_model.transformer.resblocks.2.ln_1.weight": "consolidated.safetensors",
331
+ "vision_model.transformer.resblocks.2.ln_2.bias": "consolidated.safetensors",
332
+ "vision_model.transformer.resblocks.2.ln_2.weight": "consolidated.safetensors",
333
+ "vision_model.transformer.resblocks.2.ls_1.gamma": "consolidated.safetensors",
334
+ "vision_model.transformer.resblocks.2.ls_2.gamma": "consolidated.safetensors",
335
+ "vision_model.transformer.resblocks.2.mlp.c_fc.bias": "consolidated.safetensors",
336
+ "vision_model.transformer.resblocks.2.mlp.c_fc.weight": "consolidated.safetensors",
337
+ "vision_model.transformer.resblocks.2.mlp.c_proj.bias": "consolidated.safetensors",
338
+ "vision_model.transformer.resblocks.2.mlp.c_proj.weight": "consolidated.safetensors",
339
+ "vision_model.transformer.resblocks.20.attn.in_proj_bias": "consolidated.safetensors",
340
+ "vision_model.transformer.resblocks.20.attn.in_proj_weight": "consolidated.safetensors",
341
+ "vision_model.transformer.resblocks.20.attn.out_proj.bias": "consolidated.safetensors",
342
+ "vision_model.transformer.resblocks.20.attn.out_proj.weight": "consolidated.safetensors",
343
+ "vision_model.transformer.resblocks.20.ln_1.bias": "consolidated.safetensors",
344
+ "vision_model.transformer.resblocks.20.ln_1.weight": "consolidated.safetensors",
345
+ "vision_model.transformer.resblocks.20.ln_2.bias": "consolidated.safetensors",
346
+ "vision_model.transformer.resblocks.20.ln_2.weight": "consolidated.safetensors",
347
+ "vision_model.transformer.resblocks.20.ls_1.gamma": "consolidated.safetensors",
348
+ "vision_model.transformer.resblocks.20.ls_2.gamma": "consolidated.safetensors",
349
+ "vision_model.transformer.resblocks.20.mlp.c_fc.bias": "consolidated.safetensors",
350
+ "vision_model.transformer.resblocks.20.mlp.c_fc.weight": "consolidated.safetensors",
351
+ "vision_model.transformer.resblocks.20.mlp.c_proj.bias": "consolidated.safetensors",
352
+ "vision_model.transformer.resblocks.20.mlp.c_proj.weight": "consolidated.safetensors",
353
+ "vision_model.transformer.resblocks.21.attn.in_proj_bias": "consolidated.safetensors",
354
+ "vision_model.transformer.resblocks.21.attn.in_proj_weight": "consolidated.safetensors",
355
+ "vision_model.transformer.resblocks.21.attn.out_proj.bias": "consolidated.safetensors",
356
+ "vision_model.transformer.resblocks.21.attn.out_proj.weight": "consolidated.safetensors",
357
+ "vision_model.transformer.resblocks.21.ln_1.bias": "consolidated.safetensors",
358
+ "vision_model.transformer.resblocks.21.ln_1.weight": "consolidated.safetensors",
359
+ "vision_model.transformer.resblocks.21.ln_2.bias": "consolidated.safetensors",
360
+ "vision_model.transformer.resblocks.21.ln_2.weight": "consolidated.safetensors",
361
+ "vision_model.transformer.resblocks.21.ls_1.gamma": "consolidated.safetensors",
362
+ "vision_model.transformer.resblocks.21.ls_2.gamma": "consolidated.safetensors",
363
+ "vision_model.transformer.resblocks.21.mlp.c_fc.bias": "consolidated.safetensors",
364
+ "vision_model.transformer.resblocks.21.mlp.c_fc.weight": "consolidated.safetensors",
365
+ "vision_model.transformer.resblocks.21.mlp.c_proj.bias": "consolidated.safetensors",
366
+ "vision_model.transformer.resblocks.21.mlp.c_proj.weight": "consolidated.safetensors",
367
+ "vision_model.transformer.resblocks.22.attn.in_proj_bias": "consolidated.safetensors",
368
+ "vision_model.transformer.resblocks.22.attn.in_proj_weight": "consolidated.safetensors",
369
+ "vision_model.transformer.resblocks.22.attn.out_proj.bias": "consolidated.safetensors",
370
+ "vision_model.transformer.resblocks.22.attn.out_proj.weight": "consolidated.safetensors",
371
+ "vision_model.transformer.resblocks.22.ln_1.bias": "consolidated.safetensors",
372
+ "vision_model.transformer.resblocks.22.ln_1.weight": "consolidated.safetensors",
373
+ "vision_model.transformer.resblocks.22.ln_2.bias": "consolidated.safetensors",
374
+ "vision_model.transformer.resblocks.22.ln_2.weight": "consolidated.safetensors",
375
+ "vision_model.transformer.resblocks.22.ls_1.gamma": "consolidated.safetensors",
376
+ "vision_model.transformer.resblocks.22.ls_2.gamma": "consolidated.safetensors",
377
+ "vision_model.transformer.resblocks.22.mlp.c_fc.bias": "consolidated.safetensors",
378
+ "vision_model.transformer.resblocks.22.mlp.c_fc.weight": "consolidated.safetensors",
379
+ "vision_model.transformer.resblocks.22.mlp.c_proj.bias": "consolidated.safetensors",
380
+ "vision_model.transformer.resblocks.22.mlp.c_proj.weight": "consolidated.safetensors",
381
+ "vision_model.transformer.resblocks.3.attn.in_proj_bias": "consolidated.safetensors",
382
+ "vision_model.transformer.resblocks.3.attn.in_proj_weight": "consolidated.safetensors",
383
+ "vision_model.transformer.resblocks.3.attn.out_proj.bias": "consolidated.safetensors",
384
+ "vision_model.transformer.resblocks.3.attn.out_proj.weight": "consolidated.safetensors",
385
+ "vision_model.transformer.resblocks.3.ln_1.bias": "consolidated.safetensors",
386
+ "vision_model.transformer.resblocks.3.ln_1.weight": "consolidated.safetensors",
387
+ "vision_model.transformer.resblocks.3.ln_2.bias": "consolidated.safetensors",
388
+ "vision_model.transformer.resblocks.3.ln_2.weight": "consolidated.safetensors",
389
+ "vision_model.transformer.resblocks.3.ls_1.gamma": "consolidated.safetensors",
390
+ "vision_model.transformer.resblocks.3.ls_2.gamma": "consolidated.safetensors",
391
+ "vision_model.transformer.resblocks.3.mlp.c_fc.bias": "consolidated.safetensors",
392
+ "vision_model.transformer.resblocks.3.mlp.c_fc.weight": "consolidated.safetensors",
393
+ "vision_model.transformer.resblocks.3.mlp.c_proj.bias": "consolidated.safetensors",
394
+ "vision_model.transformer.resblocks.3.mlp.c_proj.weight": "consolidated.safetensors",
395
+ "vision_model.transformer.resblocks.4.attn.in_proj_bias": "consolidated.safetensors",
396
+ "vision_model.transformer.resblocks.4.attn.in_proj_weight": "consolidated.safetensors",
397
+ "vision_model.transformer.resblocks.4.attn.out_proj.bias": "consolidated.safetensors",
398
+ "vision_model.transformer.resblocks.4.attn.out_proj.weight": "consolidated.safetensors",
399
+ "vision_model.transformer.resblocks.4.ln_1.bias": "consolidated.safetensors",
400
+ "vision_model.transformer.resblocks.4.ln_1.weight": "consolidated.safetensors",
401
+ "vision_model.transformer.resblocks.4.ln_2.bias": "consolidated.safetensors",
402
+ "vision_model.transformer.resblocks.4.ln_2.weight": "consolidated.safetensors",
403
+ "vision_model.transformer.resblocks.4.ls_1.gamma": "consolidated.safetensors",
404
+ "vision_model.transformer.resblocks.4.ls_2.gamma": "consolidated.safetensors",
405
+ "vision_model.transformer.resblocks.4.mlp.c_fc.bias": "consolidated.safetensors",
406
+ "vision_model.transformer.resblocks.4.mlp.c_fc.weight": "consolidated.safetensors",
407
+ "vision_model.transformer.resblocks.4.mlp.c_proj.bias": "consolidated.safetensors",
408
+ "vision_model.transformer.resblocks.4.mlp.c_proj.weight": "consolidated.safetensors",
409
+ "vision_model.transformer.resblocks.5.attn.in_proj_bias": "consolidated.safetensors",
410
+ "vision_model.transformer.resblocks.5.attn.in_proj_weight": "consolidated.safetensors",
411
+ "vision_model.transformer.resblocks.5.attn.out_proj.bias": "consolidated.safetensors",
412
+ "vision_model.transformer.resblocks.5.attn.out_proj.weight": "consolidated.safetensors",
413
+ "vision_model.transformer.resblocks.5.ln_1.bias": "consolidated.safetensors",
414
+ "vision_model.transformer.resblocks.5.ln_1.weight": "consolidated.safetensors",
415
+ "vision_model.transformer.resblocks.5.ln_2.bias": "consolidated.safetensors",
416
+ "vision_model.transformer.resblocks.5.ln_2.weight": "consolidated.safetensors",
417
+ "vision_model.transformer.resblocks.5.ls_1.gamma": "consolidated.safetensors",
418
+ "vision_model.transformer.resblocks.5.ls_2.gamma": "consolidated.safetensors",
419
+ "vision_model.transformer.resblocks.5.mlp.c_fc.bias": "consolidated.safetensors",
420
+ "vision_model.transformer.resblocks.5.mlp.c_fc.weight": "consolidated.safetensors",
421
+ "vision_model.transformer.resblocks.5.mlp.c_proj.bias": "consolidated.safetensors",
422
+ "vision_model.transformer.resblocks.5.mlp.c_proj.weight": "consolidated.safetensors",
423
+ "vision_model.transformer.resblocks.6.attn.in_proj_bias": "consolidated.safetensors",
424
+ "vision_model.transformer.resblocks.6.attn.in_proj_weight": "consolidated.safetensors",
425
+ "vision_model.transformer.resblocks.6.attn.out_proj.bias": "consolidated.safetensors",
426
+ "vision_model.transformer.resblocks.6.attn.out_proj.weight": "consolidated.safetensors",
427
+ "vision_model.transformer.resblocks.6.ln_1.bias": "consolidated.safetensors",
428
+ "vision_model.transformer.resblocks.6.ln_1.weight": "consolidated.safetensors",
429
+ "vision_model.transformer.resblocks.6.ln_2.bias": "consolidated.safetensors",
430
+ "vision_model.transformer.resblocks.6.ln_2.weight": "consolidated.safetensors",
431
+ "vision_model.transformer.resblocks.6.ls_1.gamma": "consolidated.safetensors",
432
+ "vision_model.transformer.resblocks.6.ls_2.gamma": "consolidated.safetensors",
433
+ "vision_model.transformer.resblocks.6.mlp.c_fc.bias": "consolidated.safetensors",
434
+ "vision_model.transformer.resblocks.6.mlp.c_fc.weight": "consolidated.safetensors",
435
+ "vision_model.transformer.resblocks.6.mlp.c_proj.bias": "consolidated.safetensors",
436
+ "vision_model.transformer.resblocks.6.mlp.c_proj.weight": "consolidated.safetensors",
437
+ "vision_model.transformer.resblocks.7.attn.in_proj_bias": "consolidated.safetensors",
438
+ "vision_model.transformer.resblocks.7.attn.in_proj_weight": "consolidated.safetensors",
439
+ "vision_model.transformer.resblocks.7.attn.out_proj.bias": "consolidated.safetensors",
440
+ "vision_model.transformer.resblocks.7.attn.out_proj.weight": "consolidated.safetensors",
441
+ "vision_model.transformer.resblocks.7.ln_1.bias": "consolidated.safetensors",
442
+ "vision_model.transformer.resblocks.7.ln_1.weight": "consolidated.safetensors",
443
+ "vision_model.transformer.resblocks.7.ln_2.bias": "consolidated.safetensors",
444
+ "vision_model.transformer.resblocks.7.ln_2.weight": "consolidated.safetensors",
445
+ "vision_model.transformer.resblocks.7.ls_1.gamma": "consolidated.safetensors",
446
+ "vision_model.transformer.resblocks.7.ls_2.gamma": "consolidated.safetensors",
447
+ "vision_model.transformer.resblocks.7.mlp.c_fc.bias": "consolidated.safetensors",
448
+ "vision_model.transformer.resblocks.7.mlp.c_fc.weight": "consolidated.safetensors",
449
+ "vision_model.transformer.resblocks.7.mlp.c_proj.bias": "consolidated.safetensors",
450
+ "vision_model.transformer.resblocks.7.mlp.c_proj.weight": "consolidated.safetensors",
451
+ "vision_model.transformer.resblocks.8.attn.in_proj_bias": "consolidated.safetensors",
452
+ "vision_model.transformer.resblocks.8.attn.in_proj_weight": "consolidated.safetensors",
453
+ "vision_model.transformer.resblocks.8.attn.out_proj.bias": "consolidated.safetensors",
454
+ "vision_model.transformer.resblocks.8.attn.out_proj.weight": "consolidated.safetensors",
455
+ "vision_model.transformer.resblocks.8.ln_1.bias": "consolidated.safetensors",
456
+ "vision_model.transformer.resblocks.8.ln_1.weight": "consolidated.safetensors",
457
+ "vision_model.transformer.resblocks.8.ln_2.bias": "consolidated.safetensors",
458
+ "vision_model.transformer.resblocks.8.ln_2.weight": "consolidated.safetensors",
459
+ "vision_model.transformer.resblocks.8.ls_1.gamma": "consolidated.safetensors",
460
+ "vision_model.transformer.resblocks.8.ls_2.gamma": "consolidated.safetensors",
461
+ "vision_model.transformer.resblocks.8.mlp.c_fc.bias": "consolidated.safetensors",
462
+ "vision_model.transformer.resblocks.8.mlp.c_fc.weight": "consolidated.safetensors",
463
+ "vision_model.transformer.resblocks.8.mlp.c_proj.bias": "consolidated.safetensors",
464
+ "vision_model.transformer.resblocks.8.mlp.c_proj.weight": "consolidated.safetensors",
465
+ "vision_model.transformer.resblocks.9.attn.in_proj_bias": "consolidated.safetensors",
466
+ "vision_model.transformer.resblocks.9.attn.in_proj_weight": "consolidated.safetensors",
467
+ "vision_model.transformer.resblocks.9.attn.out_proj.bias": "consolidated.safetensors",
468
+ "vision_model.transformer.resblocks.9.attn.out_proj.weight": "consolidated.safetensors",
469
+ "vision_model.transformer.resblocks.9.ln_1.bias": "consolidated.safetensors",
470
+ "vision_model.transformer.resblocks.9.ln_1.weight": "consolidated.safetensors",
471
+ "vision_model.transformer.resblocks.9.ln_2.bias": "consolidated.safetensors",
472
+ "vision_model.transformer.resblocks.9.ln_2.weight": "consolidated.safetensors",
473
+ "vision_model.transformer.resblocks.9.ls_1.gamma": "consolidated.safetensors",
474
+ "vision_model.transformer.resblocks.9.ls_2.gamma": "consolidated.safetensors",
475
+ "vision_model.transformer.resblocks.9.mlp.c_fc.bias": "consolidated.safetensors",
476
+ "vision_model.transformer.resblocks.9.mlp.c_fc.weight": "consolidated.safetensors",
477
+ "vision_model.transformer.resblocks.9.mlp.c_proj.bias": "consolidated.safetensors",
478
+ "vision_model.transformer.resblocks.9.mlp.c_proj.weight": "consolidated.safetensors",
479
+ "vision_projector.projector.0.bias": "consolidated.safetensors",
480
+ "vision_projector.projector.0.weight": "consolidated.safetensors",
481
+ "vision_projector.projector.2.bias": "consolidated.safetensors",
482
+ "vision_projector.projector.2.weight": "consolidated.safetensors"
483
+ }
484
+ }
params.json ADDED
@@ -0,0 +1,43 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "distributed": {
3
+ "model_dtype": "bf16"
4
+ },
5
+ "model": {
6
+ "dim": 2048,
7
+ "n_layers": 16,
8
+ "n_heads": 32,
9
+ "n_kv_heads": 8,
10
+ "vocab_size": 128256,
11
+ "ffn_dim_multiplier": 1.5,
12
+ "multiple_of": 256,
13
+ "norm_eps": 1e-05,
14
+ "rope_theta": 500000.0,
15
+ "use_scaled_rope": true,
16
+ "rope_scale_factor": 32,
17
+ "high_freq_factor": 4,
18
+ "max_seqlen": 11520,
19
+ "pooling_ratio": 2,
20
+ "weight_tying": true,
21
+ "vision_model": {
22
+ "image_size": 448,
23
+ "patch_size": 14,
24
+ "width": 1024,
25
+ "layers": 23,
26
+ "heads": 16,
27
+ "use_cls_token": true,
28
+ "use_abs_posemb": true,
29
+ "ls_init_value": 0.1,
30
+ "drop_path": 0.1,
31
+ "mlp_ratio": 4.0,
32
+ "use_ln_post": false,
33
+ "pool_type": "none"
34
+ }
35
+ },
36
+ "data": {
37
+ "tokenizer_name": "plmchat",
38
+ "tokenizer_path": "tokenizer.model",
39
+ "vision_input_type": "thumb+tile",
40
+ "max_num_tiles": 36,
41
+ "max_video_frames": 32
42
+ }
43
+ }
tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:82e9d31979e92ab929cd544440f129d9ecd797b69e327f80f17e1c50d5551b55
3
+ size 2183982