Spaces:
Running
Running
Upload default.smartquant.json
Browse files- default.smartquant.json +562 -0
default.smartquant.json
ADDED
@@ -0,0 +1,562 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"blk.79.ffn_down.weight": 12,
|
3 |
+
"blk.79.ffn_gate.weight": 21,
|
4 |
+
"blk.79.attn_output.weight": 18,
|
5 |
+
"blk.79.attn_v.weight": 23,
|
6 |
+
"blk.79.attn_k.weight": 23,
|
7 |
+
"blk.79.attn_q.weight": 23,
|
8 |
+
"blk.78.ffn_up.weight": 23,
|
9 |
+
"blk.78.attn_output.weight": 16,
|
10 |
+
"blk.78.attn_v.weight": 23,
|
11 |
+
"blk.78.attn_k.weight": 23,
|
12 |
+
"blk.78.attn_q.weight": 23,
|
13 |
+
"blk.77.ffn_up.weight": 23,
|
14 |
+
"blk.77.ffn_gate.weight": 23,
|
15 |
+
"blk.77.attn_v.weight": 23,
|
16 |
+
"blk.59.ffn_gate.weight": 29,
|
17 |
+
"blk.36.attn_q.weight": 23,
|
18 |
+
"blk.45.attn_v.weight": 23,
|
19 |
+
"blk.35.ffn_gate.weight": 17,
|
20 |
+
"blk.75.attn_q.weight": 23,
|
21 |
+
"blk.22.attn_output.weight": 19,
|
22 |
+
"blk.38.ffn_up.weight": 29,
|
23 |
+
"blk.70.attn_v.weight": 23,
|
24 |
+
"blk.7.ffn_down.weight": 19,
|
25 |
+
"blk.33.attn_output.weight": 19,
|
26 |
+
"blk.32.ffn_up.weight": 17,
|
27 |
+
"blk.5.attn_v.weight": 23,
|
28 |
+
"blk.32.attn_output.weight": 19,
|
29 |
+
"blk.30.attn_q.weight": 23,
|
30 |
+
"blk.32.attn_q.weight": 23,
|
31 |
+
"blk.31.attn_q.weight": 23,
|
32 |
+
"blk.5.ffn_gate.weight": 19,
|
33 |
+
"blk.78.ffn_gate.weight": 23,
|
34 |
+
"blk.16.attn_q.weight": 23,
|
35 |
+
"blk.60.attn_output.weight": 19,
|
36 |
+
"blk.30.attn_v.weight": 23,
|
37 |
+
"blk.36.ffn_up.weight": 29,
|
38 |
+
"blk.32.ffn_down.weight": 19,
|
39 |
+
"blk.30.attn_k.weight": 23,
|
40 |
+
"blk.29.ffn_up.weight": 22,
|
41 |
+
"blk.22.attn_q.weight": 23,
|
42 |
+
"blk.40.attn_output.weight": 19,
|
43 |
+
"blk.29.attn_q.weight": 23,
|
44 |
+
"blk.28.ffn_down.weight": 19,
|
45 |
+
"blk.47.attn_k.weight": 23,
|
46 |
+
"blk.2.ffn_gate.weight": 19,
|
47 |
+
"blk.28.ffn_up.weight": 17,
|
48 |
+
"blk.29.ffn_down.weight": 16,
|
49 |
+
"blk.28.attn_v.weight": 23,
|
50 |
+
"blk.28.attn_q.weight": 23,
|
51 |
+
"blk.27.ffn_down.weight": 19,
|
52 |
+
"blk.27.attn_output.weight": 19,
|
53 |
+
"blk.5.attn_q.weight": 23,
|
54 |
+
"blk.26.ffn_down.weight": 29,
|
55 |
+
"blk.26.attn_output.weight": 19,
|
56 |
+
"blk.26.attn_k.weight": 23,
|
57 |
+
"blk.25.ffn_down.weight": 29,
|
58 |
+
"blk.17.attn_v.weight": 23,
|
59 |
+
"blk.25.ffn_gate.weight": 17,
|
60 |
+
"blk.19.ffn_up.weight": 18,
|
61 |
+
"blk.25.attn_q.weight": 23,
|
62 |
+
"blk.24.attn_output.weight": 19,
|
63 |
+
"blk.73.ffn_gate.weight": 11,
|
64 |
+
"blk.23.attn_k.weight": 23,
|
65 |
+
"blk.31.ffn_down.weight": 29,
|
66 |
+
"blk.23.attn_output.weight": 19,
|
67 |
+
"blk.26.attn_v.weight": 23,
|
68 |
+
"blk.31.attn_k.weight": 23,
|
69 |
+
"blk.23.attn_v.weight": 23,
|
70 |
+
"blk.22.ffn_down.weight": 19,
|
71 |
+
"blk.22.ffn_gate.weight": 16,
|
72 |
+
"blk.63.attn_k.weight": 23,
|
73 |
+
"blk.22.attn_v.weight": 23,
|
74 |
+
"blk.33.attn_q.weight": 23,
|
75 |
+
"blk.36.attn_output.weight": 19,
|
76 |
+
"blk.13.ffn_gate.weight": 29,
|
77 |
+
"blk.21.ffn_up.weight": 17,
|
78 |
+
"blk.21.attn_output.weight": 19,
|
79 |
+
"blk.41.ffn_up.weight": 29,
|
80 |
+
"blk.36.attn_k.weight": 23,
|
81 |
+
"blk.34.attn_v.weight": 23,
|
82 |
+
"blk.20.ffn_down.weight": 19,
|
83 |
+
"blk.35.attn_v.weight": 23,
|
84 |
+
"blk.25.attn_output.weight": 19,
|
85 |
+
"blk.23.ffn_down.weight": 29,
|
86 |
+
"blk.21.ffn_gate.weight": 17,
|
87 |
+
"blk.58.attn_v.weight": 23,
|
88 |
+
"blk.20.ffn_gate.weight": 22,
|
89 |
+
"blk.64.attn_k.weight": 23,
|
90 |
+
"blk.8.attn_output.weight": 19,
|
91 |
+
"blk.20.attn_v.weight": 23,
|
92 |
+
"blk.55.attn_output.weight": 19,
|
93 |
+
"blk.67.attn_k.weight": 23,
|
94 |
+
"blk.19.ffn_gate.weight": 18,
|
95 |
+
"blk.43.attn_q.weight": 23,
|
96 |
+
"blk.19.attn_k.weight": 23,
|
97 |
+
"blk.19.attn_q.weight": 23,
|
98 |
+
"blk.31.ffn_gate.weight": 17,
|
99 |
+
"blk.28.ffn_gate.weight": 17,
|
100 |
+
"blk.22.attn_k.weight": 23,
|
101 |
+
"blk.18.ffn_up.weight": 18,
|
102 |
+
"blk.43.ffn_up.weight": 29,
|
103 |
+
"blk.35.attn_output.weight": 19,
|
104 |
+
"blk.18.attn_output.weight": 19,
|
105 |
+
"blk.8.attn_q.weight": 23,
|
106 |
+
"blk.71.attn_output.weight": 29,
|
107 |
+
"blk.32.attn_v.weight": 23,
|
108 |
+
"blk.54.attn_output.weight": 19,
|
109 |
+
"blk.9.attn_output.weight": 19,
|
110 |
+
"blk.17.attn_k.weight": 23,
|
111 |
+
"blk.33.ffn_gate.weight": 17,
|
112 |
+
"blk.7.attn_output.weight": 19,
|
113 |
+
"blk.66.ffn_gate.weight": 22,
|
114 |
+
"blk.27.ffn_gate.weight": 17,
|
115 |
+
"blk.61.attn_output.weight": 19,
|
116 |
+
"blk.23.ffn_gate.weight": 29,
|
117 |
+
"blk.36.ffn_gate.weight": 29,
|
118 |
+
"blk.18.ffn_gate.weight": 18,
|
119 |
+
"blk.0.attn_q.weight": 23,
|
120 |
+
"blk.17.attn_output.weight": 19,
|
121 |
+
"blk.13.attn_q.weight": 23,
|
122 |
+
"blk.4.ffn_down.weight": 19,
|
123 |
+
"blk.9.attn_k.weight": 23,
|
124 |
+
"blk.60.ffn_down.weight": 29,
|
125 |
+
"blk.8.attn_v.weight": 23,
|
126 |
+
"blk.2.attn_v.weight": 23,
|
127 |
+
"blk.6.ffn_down.weight": 19,
|
128 |
+
"blk.3.attn_output.weight": 19,
|
129 |
+
"blk.79.ffn_up.weight": 21,
|
130 |
+
"blk.34.attn_q.weight": 23,
|
131 |
+
"blk.7.attn_k.weight": 23,
|
132 |
+
"blk.27.ffn_up.weight": 17,
|
133 |
+
"blk.25.attn_v.weight": 23,
|
134 |
+
"blk.15.ffn_up.weight": 29,
|
135 |
+
"blk.4.attn_v.weight": 23,
|
136 |
+
"blk.16.attn_k.weight": 23,
|
137 |
+
"blk.52.ffn_up.weight": 29,
|
138 |
+
"blk.5.ffn_up.weight": 19,
|
139 |
+
"blk.77.ffn_down.weight": 21,
|
140 |
+
"blk.1.attn_output.weight": 19,
|
141 |
+
"blk.41.attn_q.weight": 23,
|
142 |
+
"blk.64.attn_v.weight": 23,
|
143 |
+
"blk.63.ffn_up.weight": 29,
|
144 |
+
"blk.6.attn_q.weight": 23,
|
145 |
+
"blk.4.attn_q.weight": 23,
|
146 |
+
"blk.65.ffn_gate.weight": 16,
|
147 |
+
"blk.26.ffn_up.weight": 22,
|
148 |
+
"blk.50.attn_v.weight": 23,
|
149 |
+
"blk.14.attn_k.weight": 23,
|
150 |
+
"blk.0.ffn_down.weight": 18,
|
151 |
+
"blk.36.attn_v.weight": 23,
|
152 |
+
"blk.72.attn_output.weight": 16,
|
153 |
+
"blk.1.attn_q.weight": 23,
|
154 |
+
"blk.44.attn_v.weight": 23,
|
155 |
+
"blk.23.ffn_up.weight": 29,
|
156 |
+
"blk.66.attn_v.weight": 23,
|
157 |
+
"blk.1.ffn_down.weight": 19,
|
158 |
+
"blk.10.attn_output.weight": 19,
|
159 |
+
"blk.13.ffn_up.weight": 29,
|
160 |
+
"blk.45.attn_q.weight": 23,
|
161 |
+
"blk.15.ffn_down.weight": 19,
|
162 |
+
"blk.30.ffn_down.weight": 19,
|
163 |
+
"blk.6.attn_v.weight": 23,
|
164 |
+
"blk.53.ffn_up.weight": 29,
|
165 |
+
"blk.51.ffn_up.weight": 29,
|
166 |
+
"blk.74.attn_k.weight": 23,
|
167 |
+
"blk.23.attn_q.weight": 23,
|
168 |
+
"blk.26.attn_q.weight": 23,
|
169 |
+
"blk.77.attn_output.weight": 16,
|
170 |
+
"blk.8.ffn_down.weight": 19,
|
171 |
+
"blk.54.attn_q.weight": 23,
|
172 |
+
"blk.20.ffn_up.weight": 22,
|
173 |
+
"blk.61.ffn_up.weight": 29,
|
174 |
+
"blk.77.attn_q.weight": 23,
|
175 |
+
"blk.0.attn_output.weight": 19,
|
176 |
+
"blk.66.attn_q.weight": 23,
|
177 |
+
"blk.54.ffn_up.weight": 29,
|
178 |
+
"blk.0.attn_k.weight": 23,
|
179 |
+
"blk.6.ffn_up.weight": 19,
|
180 |
+
"blk.72.attn_v.weight": 23,
|
181 |
+
"blk.5.attn_k.weight": 23,
|
182 |
+
"blk.44.ffn_up.weight": 29,
|
183 |
+
"blk.3.ffn_gate.weight": 21,
|
184 |
+
"blk.24.ffn_down.weight": 16,
|
185 |
+
"blk.32.attn_k.weight": 23,
|
186 |
+
"blk.15.attn_output.weight": 19,
|
187 |
+
"blk.37.ffn_down.weight": 19,
|
188 |
+
"blk.3.attn_v.weight": 23,
|
189 |
+
"blk.29.attn_v.weight": 23,
|
190 |
+
"blk.21.attn_q.weight": 23,
|
191 |
+
"blk.13.ffn_down.weight": 19,
|
192 |
+
"blk.74.attn_q.weight": 23,
|
193 |
+
"blk.33.attn_v.weight": 23,
|
194 |
+
"blk.1.ffn_gate.weight": 19,
|
195 |
+
"blk.8.ffn_gate.weight": 29,
|
196 |
+
"blk.8.attn_k.weight": 23,
|
197 |
+
"blk.35.attn_k.weight": 23,
|
198 |
+
"blk.60.attn_v.weight": 23,
|
199 |
+
"blk.10.attn_k.weight": 23,
|
200 |
+
"blk.20.attn_q.weight": 23,
|
201 |
+
"blk.3.attn_q.weight": 23,
|
202 |
+
"blk.4.ffn_up.weight": 19,
|
203 |
+
"blk.7.ffn_up.weight": 29,
|
204 |
+
"blk.48.ffn_down.weight": 19,
|
205 |
+
"blk.7.attn_q.weight": 23,
|
206 |
+
"blk.63.attn_v.weight": 23,
|
207 |
+
"blk.1.attn_v.weight": 23,
|
208 |
+
"blk.76.attn_q.weight": 23,
|
209 |
+
"blk.12.attn_v.weight": 23,
|
210 |
+
"blk.0.ffn_gate.weight": 19,
|
211 |
+
"blk.76.attn_v.weight": 23,
|
212 |
+
"blk.2.ffn_up.weight": 19,
|
213 |
+
"blk.56.ffn_gate.weight": 29,
|
214 |
+
"blk.10.ffn_gate.weight": 19,
|
215 |
+
"blk.10.ffn_up.weight": 19,
|
216 |
+
"blk.34.attn_output.weight": 19,
|
217 |
+
"blk.4.attn_k.weight": 23,
|
218 |
+
"blk.3.ffn_down.weight": 23,
|
219 |
+
"blk.62.attn_k.weight": 23,
|
220 |
+
"blk.17.ffn_down.weight": 19,
|
221 |
+
"blk.27.attn_q.weight": 23,
|
222 |
+
"blk.58.ffn_down.weight": 19,
|
223 |
+
"blk.33.ffn_up.weight": 17,
|
224 |
+
"blk.35.attn_q.weight": 23,
|
225 |
+
"blk.11.ffn_down.weight": 19,
|
226 |
+
"blk.52.attn_output.weight": 19,
|
227 |
+
"blk.4.ffn_gate.weight": 19,
|
228 |
+
"blk.2.attn_q.weight": 23,
|
229 |
+
"blk.20.attn_output.weight": 19,
|
230 |
+
"blk.16.attn_output.weight": 19,
|
231 |
+
"blk.33.attn_k.weight": 23,
|
232 |
+
"blk.59.attn_v.weight": 23,
|
233 |
+
"blk.24.ffn_gate.weight": 29,
|
234 |
+
"blk.46.attn_v.weight": 23,
|
235 |
+
"blk.34.ffn_up.weight": 17,
|
236 |
+
"blk.59.attn_q.weight": 23,
|
237 |
+
"blk.43.attn_k.weight": 23,
|
238 |
+
"blk.28.attn_k.weight": 23,
|
239 |
+
"blk.43.attn_v.weight": 23,
|
240 |
+
"blk.74.ffn_up.weight": 23,
|
241 |
+
"blk.9.attn_v.weight": 23,
|
242 |
+
"blk.11.attn_output.weight": 19,
|
243 |
+
"blk.18.ffn_down.weight": 21,
|
244 |
+
"blk.17.ffn_up.weight": 22,
|
245 |
+
"blk.9.ffn_gate.weight": 29,
|
246 |
+
"blk.58.attn_q.weight": 23,
|
247 |
+
"blk.14.attn_output.weight": 19,
|
248 |
+
"blk.12.ffn_up.weight": 29,
|
249 |
+
"blk.30.attn_output.weight": 19,
|
250 |
+
"blk.28.attn_output.weight": 19,
|
251 |
+
"blk.14.attn_q.weight": 23,
|
252 |
+
"blk.24.attn_k.weight": 23,
|
253 |
+
"blk.25.attn_k.weight": 23,
|
254 |
+
"blk.61.ffn_down.weight": 19,
|
255 |
+
"blk.16.ffn_gate.weight": 16,
|
256 |
+
"blk.4.attn_output.weight": 19,
|
257 |
+
"blk.62.ffn_down.weight": 29,
|
258 |
+
"blk.69.attn_k.weight": 23,
|
259 |
+
"blk.9.ffn_down.weight": 19,
|
260 |
+
"blk.5.attn_output.weight": 19,
|
261 |
+
"blk.10.attn_v.weight": 23,
|
262 |
+
"blk.13.attn_v.weight": 23,
|
263 |
+
"blk.10.ffn_down.weight": 19,
|
264 |
+
"blk.21.ffn_down.weight": 19,
|
265 |
+
"blk.19.ffn_down.weight": 19,
|
266 |
+
"blk.54.ffn_down.weight": 19,
|
267 |
+
"blk.25.ffn_up.weight": 17,
|
268 |
+
"blk.63.ffn_down.weight": 29,
|
269 |
+
"blk.11.attn_v.weight": 23,
|
270 |
+
"blk.35.ffn_down.weight": 19,
|
271 |
+
"blk.11.ffn_gate.weight": 19,
|
272 |
+
"blk.77.attn_k.weight": 23,
|
273 |
+
"blk.61.attn_k.weight": 23,
|
274 |
+
"blk.24.ffn_up.weight": 29,
|
275 |
+
"blk.11.ffn_up.weight": 19,
|
276 |
+
"blk.21.attn_k.weight": 23,
|
277 |
+
"blk.16.ffn_down.weight": 19,
|
278 |
+
"blk.17.attn_q.weight": 23,
|
279 |
+
"blk.31.ffn_up.weight": 17,
|
280 |
+
"blk.57.attn_k.weight": 23,
|
281 |
+
"blk.68.attn_output.weight": 19,
|
282 |
+
"blk.3.attn_k.weight": 23,
|
283 |
+
"blk.12.attn_k.weight": 23,
|
284 |
+
"blk.70.ffn_gate.weight": 21,
|
285 |
+
"blk.12.ffn_gate.weight": 29,
|
286 |
+
"blk.78.ffn_down.weight": 23,
|
287 |
+
"blk.5.ffn_down.weight": 19,
|
288 |
+
"blk.12.ffn_down.weight": 19,
|
289 |
+
"blk.15.ffn_gate.weight": 29,
|
290 |
+
"blk.13.attn_k.weight": 23,
|
291 |
+
"blk.13.attn_output.weight": 19,
|
292 |
+
"blk.34.ffn_down.weight": 29,
|
293 |
+
"blk.56.attn_output.weight": 19,
|
294 |
+
"blk.68.ffn_up.weight": 18,
|
295 |
+
"blk.29.ffn_gate.weight": 22,
|
296 |
+
"blk.24.attn_v.weight": 23,
|
297 |
+
"blk.67.attn_output.weight": 19,
|
298 |
+
"blk.14.attn_v.weight": 23,
|
299 |
+
"blk.14.ffn_gate.weight": 29,
|
300 |
+
"blk.40.attn_q.weight": 23,
|
301 |
+
"blk.10.attn_q.weight": 23,
|
302 |
+
"blk.15.attn_v.weight": 23,
|
303 |
+
"blk.34.attn_k.weight": 23,
|
304 |
+
"blk.16.ffn_up.weight": 16,
|
305 |
+
"blk.53.attn_output.weight": 19,
|
306 |
+
"blk.50.ffn_gate.weight": 29,
|
307 |
+
"blk.31.attn_v.weight": 23,
|
308 |
+
"blk.30.ffn_gate.weight": 17,
|
309 |
+
"blk.17.ffn_gate.weight": 22,
|
310 |
+
"blk.38.attn_v.weight": 23,
|
311 |
+
"blk.72.ffn_gate.weight": 11,
|
312 |
+
"blk.18.attn_k.weight": 23,
|
313 |
+
"blk.11.attn_k.weight": 23,
|
314 |
+
"blk.6.attn_output.weight": 19,
|
315 |
+
"blk.58.ffn_gate.weight": 29,
|
316 |
+
"blk.18.attn_q.weight": 23,
|
317 |
+
"blk.42.attn_v.weight": 23,
|
318 |
+
"blk.36.ffn_down.weight": 19,
|
319 |
+
"blk.37.attn_q.weight": 23,
|
320 |
+
"blk.37.attn_output.weight": 19,
|
321 |
+
"blk.37.ffn_gate.weight": 29,
|
322 |
+
"blk.8.ffn_up.weight": 29,
|
323 |
+
"blk.37.ffn_up.weight": 29,
|
324 |
+
"blk.39.attn_v.weight": 23,
|
325 |
+
"blk.38.attn_q.weight": 23,
|
326 |
+
"blk.52.attn_k.weight": 23,
|
327 |
+
"blk.38.attn_output.weight": 19,
|
328 |
+
"blk.24.attn_q.weight": 23,
|
329 |
+
"blk.76.attn_k.weight": 23,
|
330 |
+
"blk.38.ffn_gate.weight": 29,
|
331 |
+
"blk.75.ffn_down.weight": 18,
|
332 |
+
"blk.27.attn_v.weight": 23,
|
333 |
+
"blk.38.ffn_down.weight": 29,
|
334 |
+
"blk.54.ffn_gate.weight": 29,
|
335 |
+
"blk.39.attn_q.weight": 23,
|
336 |
+
"blk.39.attn_k.weight": 23,
|
337 |
+
"blk.39.attn_output.weight": 19,
|
338 |
+
"blk.39.ffn_gate.weight": 29,
|
339 |
+
"blk.9.attn_q.weight": 23,
|
340 |
+
"blk.39.ffn_up.weight": 29,
|
341 |
+
"blk.34.ffn_gate.weight": 17,
|
342 |
+
"blk.39.ffn_down.weight": 19,
|
343 |
+
"blk.40.attn_k.weight": 23,
|
344 |
+
"blk.29.attn_k.weight": 23,
|
345 |
+
"blk.48.ffn_gate.weight": 29,
|
346 |
+
"blk.40.attn_v.weight": 23,
|
347 |
+
"blk.40.ffn_gate.weight": 29,
|
348 |
+
"blk.62.ffn_up.weight": 29,
|
349 |
+
"blk.40.ffn_up.weight": 29,
|
350 |
+
"blk.2.ffn_down.weight": 19,
|
351 |
+
"blk.41.attn_k.weight": 23,
|
352 |
+
"blk.76.ffn_down.weight": 18,
|
353 |
+
"blk.41.attn_v.weight": 23,
|
354 |
+
"blk.65.attn_v.weight": 23,
|
355 |
+
"blk.41.attn_output.weight": 19,
|
356 |
+
"blk.48.attn_output.weight": 19,
|
357 |
+
"blk.76.ffn_gate.weight": 23,
|
358 |
+
"blk.60.ffn_up.weight": 29,
|
359 |
+
"blk.2.attn_k.weight": 23,
|
360 |
+
"blk.42.attn_output.weight": 19,
|
361 |
+
"blk.44.ffn_gate.weight": 29,
|
362 |
+
"blk.37.attn_v.weight": 23,
|
363 |
+
"blk.42.ffn_gate.weight": 29,
|
364 |
+
"blk.42.ffn_up.weight": 29,
|
365 |
+
"blk.41.ffn_gate.weight": 29,
|
366 |
+
"blk.42.ffn_down.weight": 19,
|
367 |
+
"blk.59.attn_output.weight": 19,
|
368 |
+
"blk.43.attn_output.weight": 19,
|
369 |
+
"blk.65.attn_k.weight": 23,
|
370 |
+
"blk.43.ffn_gate.weight": 29,
|
371 |
+
"blk.43.ffn_down.weight": 19,
|
372 |
+
"blk.6.ffn_gate.weight": 19,
|
373 |
+
"blk.44.attn_q.weight": 23,
|
374 |
+
"blk.44.attn_k.weight": 23,
|
375 |
+
"blk.44.attn_output.weight": 19,
|
376 |
+
"blk.44.ffn_down.weight": 19,
|
377 |
+
"blk.45.attn_k.weight": 23,
|
378 |
+
"blk.45.attn_output.weight": 19,
|
379 |
+
"blk.45.ffn_up.weight": 29,
|
380 |
+
"blk.7.attn_v.weight": 23,
|
381 |
+
"blk.30.ffn_up.weight": 17,
|
382 |
+
"blk.45.ffn_down.weight": 16,
|
383 |
+
"blk.46.attn_q.weight": 23,
|
384 |
+
"blk.1.ffn_up.weight": 19,
|
385 |
+
"blk.15.attn_k.weight": 23,
|
386 |
+
"blk.46.attn_k.weight": 23,
|
387 |
+
"blk.18.attn_v.weight": 23,
|
388 |
+
"blk.66.attn_output.weight": 19,
|
389 |
+
"blk.14.ffn_down.weight": 19,
|
390 |
+
"blk.41.ffn_down.weight": 19,
|
391 |
+
"blk.46.attn_output.weight": 19,
|
392 |
+
"blk.38.attn_k.weight": 23,
|
393 |
+
"blk.75.attn_k.weight": 23,
|
394 |
+
"blk.12.attn_q.weight": 23,
|
395 |
+
"blk.46.ffn_gate.weight": 29,
|
396 |
+
"blk.46.ffn_up.weight": 29,
|
397 |
+
"blk.46.ffn_down.weight": 19,
|
398 |
+
"blk.55.attn_v.weight": 23,
|
399 |
+
"blk.37.attn_k.weight": 23,
|
400 |
+
"blk.47.attn_v.weight": 23,
|
401 |
+
"blk.71.attn_q.weight": 23,
|
402 |
+
"blk.47.attn_output.weight": 19,
|
403 |
+
"blk.47.ffn_gate.weight": 29,
|
404 |
+
"blk.47.ffn_up.weight": 29,
|
405 |
+
"blk.47.ffn_down.weight": 19,
|
406 |
+
"blk.55.attn_k.weight": 23,
|
407 |
+
"blk.56.attn_q.weight": 23,
|
408 |
+
"blk.72.attn_q.weight": 23,
|
409 |
+
"blk.21.attn_v.weight": 23,
|
410 |
+
"blk.48.attn_q.weight": 23,
|
411 |
+
"blk.48.attn_k.weight": 23,
|
412 |
+
"blk.48.attn_v.weight": 23,
|
413 |
+
"blk.49.attn_k.weight": 23,
|
414 |
+
"blk.49.attn_v.weight": 23,
|
415 |
+
"blk.26.ffn_gate.weight": 22,
|
416 |
+
"blk.19.attn_output.weight": 19,
|
417 |
+
"blk.49.attn_output.weight": 19,
|
418 |
+
"blk.42.attn_q.weight": 23,
|
419 |
+
"blk.49.ffn_gate.weight": 29,
|
420 |
+
"blk.49.ffn_up.weight": 29,
|
421 |
+
"blk.49.ffn_down.weight": 19,
|
422 |
+
"blk.55.attn_q.weight": 23,
|
423 |
+
"blk.54.attn_v.weight": 23,
|
424 |
+
"blk.50.attn_q.weight": 23,
|
425 |
+
"blk.70.attn_q.weight": 23,
|
426 |
+
"blk.45.ffn_gate.weight": 29,
|
427 |
+
"blk.50.attn_output.weight": 19,
|
428 |
+
"blk.50.ffn_down.weight": 19,
|
429 |
+
"blk.51.attn_q.weight": 23,
|
430 |
+
"blk.56.ffn_down.weight": 19,
|
431 |
+
"blk.19.attn_v.weight": 23,
|
432 |
+
"blk.51.attn_v.weight": 23,
|
433 |
+
"blk.49.attn_q.weight": 23,
|
434 |
+
"blk.51.ffn_gate.weight": 29,
|
435 |
+
"blk.75.attn_output.weight": 22,
|
436 |
+
"blk.51.ffn_down.weight": 19,
|
437 |
+
"blk.72.attn_k.weight": 23,
|
438 |
+
"blk.52.attn_q.weight": 23,
|
439 |
+
"blk.52.attn_v.weight": 23,
|
440 |
+
"blk.52.ffn_gate.weight": 29,
|
441 |
+
"blk.53.attn_q.weight": 23,
|
442 |
+
"blk.57.attn_v.weight": 23,
|
443 |
+
"blk.20.attn_k.weight": 23,
|
444 |
+
"blk.53.attn_k.weight": 23,
|
445 |
+
"blk.58.attn_output.weight": 19,
|
446 |
+
"blk.53.attn_v.weight": 23,
|
447 |
+
"blk.53.ffn_gate.weight": 29,
|
448 |
+
"blk.57.ffn_down.weight": 19,
|
449 |
+
"blk.53.ffn_down.weight": 19,
|
450 |
+
"blk.29.attn_output.weight": 19,
|
451 |
+
"blk.22.ffn_up.weight": 16,
|
452 |
+
"blk.55.ffn_gate.weight": 29,
|
453 |
+
"blk.55.ffn_up.weight": 29,
|
454 |
+
"blk.55.ffn_down.weight": 29,
|
455 |
+
"blk.35.ffn_up.weight": 17,
|
456 |
+
"blk.71.ffn_down.weight": 22,
|
457 |
+
"blk.67.attn_q.weight": 23,
|
458 |
+
"blk.56.attn_k.weight": 23,
|
459 |
+
"blk.50.attn_k.weight": 23,
|
460 |
+
"blk.56.attn_v.weight": 23,
|
461 |
+
"blk.47.attn_q.weight": 23,
|
462 |
+
"blk.56.ffn_up.weight": 29,
|
463 |
+
"blk.57.attn_q.weight": 23,
|
464 |
+
"blk.27.attn_k.weight": 23,
|
465 |
+
"blk.48.ffn_up.weight": 29,
|
466 |
+
"blk.62.attn_q.weight": 23,
|
467 |
+
"blk.73.attn_k.weight": 23,
|
468 |
+
"blk.60.attn_q.weight": 23,
|
469 |
+
"blk.42.attn_k.weight": 23,
|
470 |
+
"blk.57.attn_output.weight": 19,
|
471 |
+
"blk.74.ffn_gate.weight": 23,
|
472 |
+
"blk.58.ffn_up.weight": 29,
|
473 |
+
"blk.57.ffn_gate.weight": 29,
|
474 |
+
"blk.57.ffn_up.weight": 29,
|
475 |
+
"blk.52.ffn_down.weight": 19,
|
476 |
+
"blk.58.attn_k.weight": 23,
|
477 |
+
"blk.59.attn_k.weight": 23,
|
478 |
+
"blk.66.attn_k.weight": 23,
|
479 |
+
"blk.11.attn_q.weight": 23,
|
480 |
+
"blk.33.ffn_down.weight": 19,
|
481 |
+
"blk.59.ffn_up.weight": 29,
|
482 |
+
"blk.31.attn_output.weight": 19,
|
483 |
+
"blk.60.attn_k.weight": 23,
|
484 |
+
"blk.60.ffn_gate.weight": 29,
|
485 |
+
"blk.69.attn_output.weight": 19,
|
486 |
+
"blk.73.ffn_down.weight": 21,
|
487 |
+
"blk.3.ffn_up.weight": 21,
|
488 |
+
"blk.61.attn_q.weight": 23,
|
489 |
+
"blk.0.attn_v.weight": 23,
|
490 |
+
"blk.61.attn_v.weight": 23,
|
491 |
+
"blk.7.ffn_gate.weight": 29,
|
492 |
+
"blk.61.ffn_gate.weight": 29,
|
493 |
+
"blk.75.ffn_gate.weight": 23,
|
494 |
+
"blk.62.attn_v.weight": 23,
|
495 |
+
"blk.54.attn_k.weight": 23,
|
496 |
+
"blk.75.ffn_up.weight": 23,
|
497 |
+
"blk.62.attn_output.weight": 19,
|
498 |
+
"blk.62.ffn_gate.weight": 29,
|
499 |
+
"blk.63.attn_q.weight": 23,
|
500 |
+
"blk.40.ffn_down.weight": 19,
|
501 |
+
"blk.14.ffn_up.weight": 29,
|
502 |
+
"blk.63.attn_output.weight": 19,
|
503 |
+
"blk.6.attn_k.weight": 23,
|
504 |
+
"blk.63.ffn_gate.weight": 29,
|
505 |
+
"blk.64.attn_q.weight": 23,
|
506 |
+
"blk.64.attn_output.weight": 19,
|
507 |
+
"blk.64.ffn_gate.weight": 22,
|
508 |
+
"blk.59.ffn_down.weight": 19,
|
509 |
+
"blk.64.ffn_up.weight": 22,
|
510 |
+
"blk.64.ffn_down.weight": 19,
|
511 |
+
"blk.16.attn_v.weight": 23,
|
512 |
+
"blk.65.attn_q.weight": 23,
|
513 |
+
"blk.65.attn_output.weight": 19,
|
514 |
+
"blk.32.ffn_gate.weight": 17,
|
515 |
+
"blk.68.ffn_gate.weight": 18,
|
516 |
+
"blk.65.ffn_up.weight": 16,
|
517 |
+
"blk.67.ffn_gate.weight": 22,
|
518 |
+
"blk.65.ffn_down.weight": 19,
|
519 |
+
"blk.66.ffn_up.weight": 22,
|
520 |
+
"blk.66.ffn_down.weight": 29,
|
521 |
+
"blk.71.ffn_gate.weight": 21,
|
522 |
+
"blk.67.attn_v.weight": 23,
|
523 |
+
"blk.68.attn_q.weight": 23,
|
524 |
+
"blk.67.ffn_down.weight": 19,
|
525 |
+
"blk.68.attn_k.weight": 23,
|
526 |
+
"blk.68.attn_v.weight": 23,
|
527 |
+
"blk.68.ffn_down.weight": 29,
|
528 |
+
"blk.50.ffn_up.weight": 29,
|
529 |
+
"blk.69.attn_q.weight": 23,
|
530 |
+
"blk.69.ffn_gate.weight": 18,
|
531 |
+
"blk.69.ffn_up.weight": 18,
|
532 |
+
"blk.69.ffn_down.weight": 29,
|
533 |
+
"blk.0.ffn_up.weight": 19,
|
534 |
+
"blk.69.attn_v.weight": 23,
|
535 |
+
"blk.74.attn_output.weight": 17,
|
536 |
+
"blk.12.attn_output.weight": 19,
|
537 |
+
"blk.9.ffn_up.weight": 29,
|
538 |
+
"blk.70.attn_k.weight": 23,
|
539 |
+
"blk.70.attn_output.weight": 29,
|
540 |
+
"blk.70.ffn_up.weight": 21,
|
541 |
+
"blk.2.attn_output.weight": 19,
|
542 |
+
"blk.70.ffn_down.weight": 17,
|
543 |
+
"blk.71.attn_k.weight": 23,
|
544 |
+
"blk.71.attn_v.weight": 23,
|
545 |
+
"blk.71.ffn_up.weight": 21,
|
546 |
+
"blk.72.ffn_up.weight": 11,
|
547 |
+
"blk.51.attn_output.weight": 19,
|
548 |
+
"blk.72.ffn_down.weight": 29,
|
549 |
+
"blk.15.attn_q.weight": 23,
|
550 |
+
"blk.73.attn_q.weight": 23,
|
551 |
+
"blk.73.attn_v.weight": 23,
|
552 |
+
"blk.73.attn_output.weight": 29,
|
553 |
+
"blk.67.ffn_up.weight": 22,
|
554 |
+
"blk.73.ffn_up.weight": 11,
|
555 |
+
"blk.74.attn_v.weight": 23,
|
556 |
+
"blk.1.attn_k.weight": 23,
|
557 |
+
"blk.74.ffn_down.weight": 18,
|
558 |
+
"blk.75.attn_v.weight": 23,
|
559 |
+
"blk.76.attn_output.weight": 16,
|
560 |
+
"blk.51.attn_k.weight": 23,
|
561 |
+
"blk.76.ffn_up.weight": 23
|
562 |
+
}
|