Hugging Face's logo

theunlikely
/
BAGEL-RecA-DF11

Safetensors
qwen2
Model card Files
xet
Community
theunlikely commited on
Commit
6dd012a
·
verified ·
1 Parent(s): da6830c

Update config.json

Browse files
Files changed (1) hide show
  1. config.json +189 -0
config.json CHANGED
@@ -0,0 +1,189 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dfloat11_config": {
3
+ "bytes_per_thread": 8,
4
+ "pattern_dict": {
5
+ "language_model\\.model\\.embed_tokens": [],
6
+ "language_model\\.model\\.layers.\\d+": [
7
+ "self_attn.q_proj",
8
+ "self_attn.k_proj",
9
+ "self_attn.v_proj",
10
+ "self_attn.o_proj",
11
+ "self_attn.q_proj_moe_gen",
12
+ "self_attn.k_proj_moe_gen",
13
+ "self_attn.v_proj_moe_gen",
14
+ "self_attn.o_proj_moe_gen",
15
+ "mlp.gate_proj",
16
+ "mlp.up_proj",
17
+ "mlp.down_proj",
18
+ "mlp_moe_gen.gate_proj",
19
+ "mlp_moe_gen.up_proj",
20
+ "mlp_moe_gen.down_proj"
21
+ ],
22
+ "language_model\\.lm_head": [],
23
+ "vit_model": [
24
+ "vision_model.encoder.layers.0.self_attn.k_proj",
25
+ "vision_model.encoder.layers.0.self_attn.v_proj",
26
+ "vision_model.encoder.layers.0.self_attn.q_proj",
27
+ "vision_model.encoder.layers.0.self_attn.out_proj",
28
+ "vision_model.encoder.layers.0.mlp.fc1",
29
+ "vision_model.encoder.layers.0.mlp.fc2",
30
+ "vision_model.encoder.layers.1.self_attn.k_proj",
31
+ "vision_model.encoder.layers.1.self_attn.v_proj",
32
+ "vision_model.encoder.layers.1.self_attn.q_proj",
33
+ "vision_model.encoder.layers.1.self_attn.out_proj",
34
+ "vision_model.encoder.layers.1.mlp.fc1",
35
+ "vision_model.encoder.layers.1.mlp.fc2",
36
+ "vision_model.encoder.layers.2.self_attn.k_proj",
37
+ "vision_model.encoder.layers.2.self_attn.v_proj",
38
+ "vision_model.encoder.layers.2.self_attn.q_proj",
39
+ "vision_model.encoder.layers.2.self_attn.out_proj",
40
+ "vision_model.encoder.layers.2.mlp.fc1",
41
+ "vision_model.encoder.layers.2.mlp.fc2",
42
+ "vision_model.encoder.layers.3.self_attn.k_proj",
43
+ "vision_model.encoder.layers.3.self_attn.v_proj",
44
+ "vision_model.encoder.layers.3.self_attn.q_proj",
45
+ "vision_model.encoder.layers.3.self_attn.out_proj",
46
+ "vision_model.encoder.layers.3.mlp.fc1",
47
+ "vision_model.encoder.layers.3.mlp.fc2",
48
+ "vision_model.encoder.layers.4.self_attn.k_proj",
49
+ "vision_model.encoder.layers.4.self_attn.v_proj",
50
+ "vision_model.encoder.layers.4.self_attn.q_proj",
51
+ "vision_model.encoder.layers.4.self_attn.out_proj",
52
+ "vision_model.encoder.layers.4.mlp.fc1",
53
+ "vision_model.encoder.layers.4.mlp.fc2",
54
+ "vision_model.encoder.layers.5.self_attn.k_proj",
55
+ "vision_model.encoder.layers.5.self_attn.v_proj",
56
+ "vision_model.encoder.layers.5.self_attn.q_proj",
57
+ "vision_model.encoder.layers.5.self_attn.out_proj",
58
+ "vision_model.encoder.layers.5.mlp.fc1",
59
+ "vision_model.encoder.layers.5.mlp.fc2",
60
+ "vision_model.encoder.layers.6.self_attn.k_proj",
61
+ "vision_model.encoder.layers.6.self_attn.v_proj",
62
+ "vision_model.encoder.layers.6.self_attn.q_proj",
63
+ "vision_model.encoder.layers.6.self_attn.out_proj",
64
+ "vision_model.encoder.layers.6.mlp.fc1",
65
+ "vision_model.encoder.layers.6.mlp.fc2",
66
+ "vision_model.encoder.layers.7.self_attn.k_proj",
67
+ "vision_model.encoder.layers.7.self_attn.v_proj",
68
+ "vision_model.encoder.layers.7.self_attn.q_proj",
69
+ "vision_model.encoder.layers.7.self_attn.out_proj",
70
+ "vision_model.encoder.layers.7.mlp.fc1",
71
+ "vision_model.encoder.layers.7.mlp.fc2",
72
+ "vision_model.encoder.layers.8.self_attn.k_proj",
73
+ "vision_model.encoder.layers.8.self_attn.v_proj",
74
+ "vision_model.encoder.layers.8.self_attn.q_proj",
75
+ "vision_model.encoder.layers.8.self_attn.out_proj",
76
+ "vision_model.encoder.layers.8.mlp.fc1",
77
+ "vision_model.encoder.layers.8.mlp.fc2",
78
+ "vision_model.encoder.layers.9.self_attn.k_proj",
79
+ "vision_model.encoder.layers.9.self_attn.v_proj",
80
+ "vision_model.encoder.layers.9.self_attn.q_proj",
81
+ "vision_model.encoder.layers.9.self_attn.out_proj",
82
+ "vision_model.encoder.layers.9.mlp.fc1",
83
+ "vision_model.encoder.layers.9.mlp.fc2",
84
+ "vision_model.encoder.layers.10.self_attn.k_proj",
85
+ "vision_model.encoder.layers.10.self_attn.v_proj",
86
+ "vision_model.encoder.layers.10.self_attn.q_proj",
87
+ "vision_model.encoder.layers.10.self_attn.out_proj",
88
+ "vision_model.encoder.layers.10.mlp.fc1",
89
+ "vision_model.encoder.layers.10.mlp.fc2",
90
+ "vision_model.encoder.layers.11.self_attn.k_proj",
91
+ "vision_model.encoder.layers.11.self_attn.v_proj",
92
+ "vision_model.encoder.layers.11.self_attn.q_proj",
93
+ "vision_model.encoder.layers.11.self_attn.out_proj",
94
+ "vision_model.encoder.layers.11.mlp.fc1",
95
+ "vision_model.encoder.layers.11.mlp.fc2",
96
+ "vision_model.encoder.layers.12.self_attn.k_proj",
97
+ "vision_model.encoder.layers.12.self_attn.v_proj",
98
+ "vision_model.encoder.layers.12.self_attn.q_proj",
99
+ "vision_model.encoder.layers.12.self_attn.out_proj",
100
+ "vision_model.encoder.layers.12.mlp.fc1",
101
+ "vision_model.encoder.layers.12.mlp.fc2",
102
+ "vision_model.encoder.layers.13.self_attn.k_proj",
103
+ "vision_model.encoder.layers.13.self_attn.v_proj",
104
+ "vision_model.encoder.layers.13.self_attn.q_proj",
105
+ "vision_model.encoder.layers.13.self_attn.out_proj",
106
+ "vision_model.encoder.layers.13.mlp.fc1",
107
+ "vision_model.encoder.layers.13.mlp.fc2",
108
+ "vision_model.encoder.layers.14.self_attn.k_proj",
109
+ "vision_model.encoder.layers.14.self_attn.v_proj",
110
+ "vision_model.encoder.layers.14.self_attn.q_proj",
111
+ "vision_model.encoder.layers.14.self_attn.out_proj",
112
+ "vision_model.encoder.layers.14.mlp.fc1",
113
+ "vision_model.encoder.layers.14.mlp.fc2",
114
+ "vision_model.encoder.layers.15.self_attn.k_proj",
115
+ "vision_model.encoder.layers.15.self_attn.v_proj",
116
+ "vision_model.encoder.layers.15.self_attn.q_proj",
117
+ "vision_model.encoder.layers.15.self_attn.out_proj",
118
+ "vision_model.encoder.layers.15.mlp.fc1",
119
+ "vision_model.encoder.layers.15.mlp.fc2",
120
+ "vision_model.encoder.layers.16.self_attn.k_proj",
121
+ "vision_model.encoder.layers.16.self_attn.v_proj",
122
+ "vision_model.encoder.layers.16.self_attn.q_proj",
123
+ "vision_model.encoder.layers.16.self_attn.out_proj",
124
+ "vision_model.encoder.layers.16.mlp.fc1",
125
+ "vision_model.encoder.layers.16.mlp.fc2",
126
+ "vision_model.encoder.layers.17.self_attn.k_proj",
127
+ "vision_model.encoder.layers.17.self_attn.v_proj",
128
+ "vision_model.encoder.layers.17.self_attn.q_proj",
129
+ "vision_model.encoder.layers.17.self_attn.out_proj",
130
+ "vision_model.encoder.layers.17.mlp.fc1",
131
+ "vision_model.encoder.layers.17.mlp.fc2",
132
+ "vision_model.encoder.layers.18.self_attn.k_proj",
133
+ "vision_model.encoder.layers.18.self_attn.v_proj",
134
+ "vision_model.encoder.layers.18.self_attn.q_proj",
135
+ "vision_model.encoder.layers.18.self_attn.out_proj",
136
+ "vision_model.encoder.layers.18.mlp.fc1",
137
+ "vision_model.encoder.layers.18.mlp.fc2",
138
+ "vision_model.encoder.layers.19.self_attn.k_proj",
139
+ "vision_model.encoder.layers.19.self_attn.v_proj",
140
+ "vision_model.encoder.layers.19.self_attn.q_proj",
141
+ "vision_model.encoder.layers.19.self_attn.out_proj",
142
+ "vision_model.encoder.layers.19.mlp.fc1",
143
+ "vision_model.encoder.layers.19.mlp.fc2",
144
+ "vision_model.encoder.layers.20.self_attn.k_proj",
145
+ "vision_model.encoder.layers.20.self_attn.v_proj",
146
+ "vision_model.encoder.layers.20.self_attn.q_proj",
147
+ "vision_model.encoder.layers.20.self_attn.out_proj",
148
+ "vision_model.encoder.layers.20.mlp.fc1",
149
+ "vision_model.encoder.layers.20.mlp.fc2",
150
+ "vision_model.encoder.layers.21.self_attn.k_proj",
151
+ "vision_model.encoder.layers.21.self_attn.v_proj",
152
+ "vision_model.encoder.layers.21.self_attn.q_proj",
153
+ "vision_model.encoder.layers.21.self_attn.out_proj",
154
+ "vision_model.encoder.layers.21.mlp.fc1",
155
+ "vision_model.encoder.layers.21.mlp.fc2",
156
+ "vision_model.encoder.layers.22.self_attn.k_proj",
157
+ "vision_model.encoder.layers.22.self_attn.v_proj",
158
+ "vision_model.encoder.layers.22.self_attn.q_proj",
159
+ "vision_model.encoder.layers.22.self_attn.out_proj",
160
+ "vision_model.encoder.layers.22.mlp.fc1",
161
+ "vision_model.encoder.layers.22.mlp.fc2",
162
+ "vision_model.encoder.layers.23.self_attn.k_proj",
163
+ "vision_model.encoder.layers.23.self_attn.v_proj",
164
+ "vision_model.encoder.layers.23.self_attn.q_proj",
165
+ "vision_model.encoder.layers.23.self_attn.out_proj",
166
+ "vision_model.encoder.layers.23.mlp.fc1",
167
+ "vision_model.encoder.layers.23.mlp.fc2",
168
+ "vision_model.encoder.layers.24.self_attn.k_proj",
169
+ "vision_model.encoder.layers.24.self_attn.v_proj",
170
+ "vision_model.encoder.layers.24.self_attn.q_proj",
171
+ "vision_model.encoder.layers.24.self_attn.out_proj",
172
+ "vision_model.encoder.layers.24.mlp.fc1",
173
+ "vision_model.encoder.layers.24.mlp.fc2",
174
+ "vision_model.encoder.layers.25.self_attn.k_proj",
175
+ "vision_model.encoder.layers.25.self_attn.v_proj",
176
+ "vision_model.encoder.layers.25.self_attn.q_proj",
177
+ "vision_model.encoder.layers.25.self_attn.out_proj",
178
+ "vision_model.encoder.layers.25.mlp.fc1",
179
+ "vision_model.encoder.layers.25.mlp.fc2",
180
+ "vision_model.embeddings.patch_embedding"
181
+ ]
182
+ },
183
+ "threads_per_block": [
184
+ 512
185
+ ],
186
+ "version": "0.2.0"
187
+ },
188
+ "model_type": "qwen2"
189
+ }