Hugging Face's logo

mlc-ai
/
Llama-3.2-1B-Instruct-q4f16_1-MLC

MLC-LLM
web-llm
Model card Files Community
CharlieFRuan commited on
Commit
7e6866e
•
1 Parent(s): e38c706

Upload folder using huggingface_hub

Browse files
Files changed (23) hide show
  1. ndarray-cache.json +1945 -0
  2. params_shard_0.bin +3 -0
  3. params_shard_1.bin +3 -0
  4. params_shard_10.bin +3 -0
  5. params_shard_11.bin +3 -0
  6. params_shard_12.bin +3 -0
  7. params_shard_13.bin +3 -0
  8. params_shard_14.bin +3 -0
  9. params_shard_15.bin +3 -0
  10. params_shard_16.bin +3 -0
  11. params_shard_17.bin +3 -0
  12. params_shard_18.bin +3 -0
  13. params_shard_19.bin +3 -0
  14. params_shard_2.bin +3 -0
  15. params_shard_20.bin +3 -0
  16. params_shard_21.bin +3 -0
  17. params_shard_3.bin +3 -0
  18. params_shard_4.bin +3 -0
  19. params_shard_5.bin +3 -0
  20. params_shard_6.bin +3 -0
  21. params_shard_7.bin +3 -0
  22. params_shard_8.bin +3 -0
  23. params_shard_9.bin +3 -0
ndarray-cache.json ADDED
@@ -0,0 +1,1945 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "ParamSize": 163,
4
+ "ParamBytes": 695242752.0,
5
+ "BitsPerParam": 4.500628909972241
6
+ },
7
+ "records": [
8
+ {
9
+ "dataPath": "params_shard_0.bin",
10
+ "format": "raw-shard",
11
+ "nbytes": 131334144,
12
+ "records": [
13
+ {
14
+ "name": "model.embed_tokens.q_weight",
15
+ "shape": [
16
+ 128256,
17
+ 256
18
+ ],
19
+ "dtype": "uint32",
20
+ "format": "f32-to-bf16",
21
+ "nbytes": 131334144,
22
+ "byteOffset": 0
23
+ }
24
+ ],
25
+ "md5sum": "780f65bb3be7cdc499634d8cae98c5aa"
26
+ },
27
+ {
28
+ "dataPath": "params_shard_1.bin",
29
+ "format": "raw-shard",
30
+ "nbytes": 16777216,
31
+ "records": [
32
+ {
33
+ "name": "model.layers.0.mlp.gate_up_proj.q_weight",
34
+ "shape": [
35
+ 16384,
36
+ 256
37
+ ],
38
+ "dtype": "uint32",
39
+ "format": "f32-to-bf16",
40
+ "nbytes": 16777216,
41
+ "byteOffset": 0
42
+ }
43
+ ],
44
+ "md5sum": "90cc2adb82e582599ddd8d97e5b1bcc0"
45
+ },
46
+ {
47
+ "dataPath": "params_shard_2.bin",
48
+ "format": "raw-shard",
49
+ "nbytes": 31498240,
50
+ "records": [
51
+ {
52
+ "name": "model.embed_tokens.q_scale",
53
+ "shape": [
54
+ 128256,
55
+ 64
56
+ ],
57
+ "dtype": "float16",
58
+ "format": "f32-to-bf16",
59
+ "nbytes": 16416768,
60
+ "byteOffset": 0
61
+ },
62
+ {
63
+ "name": "model.layers.0.input_layernorm.weight",
64
+ "shape": [
65
+ 2048
66
+ ],
67
+ "dtype": "float16",
68
+ "format": "f32-to-bf16",
69
+ "nbytes": 4096,
70
+ "byteOffset": 16416768
71
+ },
72
+ {
73
+ "name": "model.layers.0.mlp.down_proj.q_weight",
74
+ "shape": [
75
+ 2048,
76
+ 1024
77
+ ],
78
+ "dtype": "uint32",
79
+ "format": "f32-to-bf16",
80
+ "nbytes": 8388608,
81
+ "byteOffset": 16420864
82
+ },
83
+ {
84
+ "name": "model.layers.0.mlp.down_proj.q_scale",
85
+ "shape": [
86
+ 2048,
87
+ 256
88
+ ],
89
+ "dtype": "float16",
90
+ "format": "f32-to-bf16",
91
+ "nbytes": 1048576,
92
+ "byteOffset": 24809472
93
+ },
94
+ {
95
+ "name": "model.layers.0.mlp.gate_up_proj.q_scale",
96
+ "shape": [
97
+ 16384,
98
+ 64
99
+ ],
100
+ "dtype": "float16",
101
+ "format": "f32-to-bf16",
102
+ "nbytes": 2097152,
103
+ "byteOffset": 25858048
104
+ },
105
+ {
106
+ "name": "model.layers.0.post_attention_layernorm.weight",
107
+ "shape": [
108
+ 2048
109
+ ],
110
+ "dtype": "float16",
111
+ "format": "f32-to-bf16",
112
+ "nbytes": 4096,
113
+ "byteOffset": 27955200
114
+ },
115
+ {
116
+ "name": "model.layers.0.self_attn.qkv_proj.q_weight",
117
+ "shape": [
118
+ 3072,
119
+ 256
120
+ ],
121
+ "dtype": "uint32",
122
+ "format": "f32-to-bf16",
123
+ "nbytes": 3145728,
124
+ "byteOffset": 27959296
125
+ },
126
+ {
127
+ "name": "model.layers.0.self_attn.qkv_proj.q_scale",
128
+ "shape": [
129
+ 3072,
130
+ 64
131
+ ],
132
+ "dtype": "float16",
133
+ "format": "f32-to-bf16",
134
+ "nbytes": 393216,
135
+ "byteOffset": 31105024
136
+ }
137
+ ],
138
+ "md5sum": "d6f8e0150d3d538726eb1276cee29bd5"
139
+ },
140
+ {
141
+ "dataPath": "params_shard_3.bin",
142
+ "format": "raw-shard",
143
+ "nbytes": 30679040,
144
+ "records": [
145
+ {
146
+ "name": "model.layers.0.self_attn.o_proj.q_weight",
147
+ "shape": [
148
+ 2048,
149
+ 256
150
+ ],
151
+ "dtype": "uint32",
152
+ "format": "f32-to-bf16",
153
+ "nbytes": 2097152,
154
+ "byteOffset": 0
155
+ },
156
+ {
157
+ "name": "model.layers.0.self_attn.o_proj.q_scale",
158
+ "shape": [
159
+ 2048,
160
+ 64
161
+ ],
162
+ "dtype": "float16",
163
+ "format": "f32-to-bf16",
164
+ "nbytes": 262144,
165
+ "byteOffset": 2097152
166
+ },
167
+ {
168
+ "name": "model.layers.1.input_layernorm.weight",
169
+ "shape": [
170
+ 2048
171
+ ],
172
+ "dtype": "float16",
173
+ "format": "f32-to-bf16",
174
+ "nbytes": 4096,
175
+ "byteOffset": 2359296
176
+ },
177
+ {
178
+ "name": "model.layers.1.mlp.down_proj.q_weight",
179
+ "shape": [
180
+ 2048,
181
+ 1024
182
+ ],
183
+ "dtype": "uint32",
184
+ "format": "f32-to-bf16",
185
+ "nbytes": 8388608,
186
+ "byteOffset": 2363392
187
+ },
188
+ {
189
+ "name": "model.layers.1.mlp.down_proj.q_scale",
190
+ "shape": [
191
+ 2048,
192
+ 256
193
+ ],
194
+ "dtype": "float16",
195
+ "format": "f32-to-bf16",
196
+ "nbytes": 1048576,
197
+ "byteOffset": 10752000
198
+ },
199
+ {
200
+ "name": "model.layers.1.mlp.gate_up_proj.q_weight",
201
+ "shape": [
202
+ 16384,
203
+ 256
204
+ ],
205
+ "dtype": "uint32",
206
+ "format": "f32-to-bf16",
207
+ "nbytes": 16777216,
208
+ "byteOffset": 11800576
209
+ },
210
+ {
211
+ "name": "model.layers.1.mlp.gate_up_proj.q_scale",
212
+ "shape": [
213
+ 16384,
214
+ 64
215
+ ],
216
+ "dtype": "float16",
217
+ "format": "f32-to-bf16",
218
+ "nbytes": 2097152,
219
+ "byteOffset": 28577792
220
+ },
221
+ {
222
+ "name": "model.layers.1.post_attention_layernorm.weight",
223
+ "shape": [
224
+ 2048
225
+ ],
226
+ "dtype": "float16",
227
+ "format": "f32-to-bf16",
228
+ "nbytes": 4096,
229
+ "byteOffset": 30674944
230
+ }
231
+ ],
232
+ "md5sum": "2760a8b4dcb645e26283def0403f4deb"
233
+ },
234
+ {
235
+ "dataPath": "params_shard_4.bin",
236
+ "format": "raw-shard",
237
+ "nbytes": 32116736,
238
+ "records": [
239
+ {
240
+ "name": "model.layers.1.self_attn.qkv_proj.q_weight",
241
+ "shape": [
242
+ 3072,
243
+ 256
244
+ ],
245
+ "dtype": "uint32",
246
+ "format": "f32-to-bf16",
247
+ "nbytes": 3145728,
248
+ "byteOffset": 0
249
+ },
250
+ {
251
+ "name": "model.layers.1.self_attn.qkv_proj.q_scale",
252
+ "shape": [
253
+ 3072,
254
+ 64
255
+ ],
256
+ "dtype": "float16",
257
+ "format": "f32-to-bf16",
258
+ "nbytes": 393216,
259
+ "byteOffset": 3145728
260
+ },
261
+ {
262
+ "name": "model.layers.1.self_attn.o_proj.q_weight",
263
+ "shape": [
264
+ 2048,
265
+ 256
266
+ ],
267
+ "dtype": "uint32",
268
+ "format": "f32-to-bf16",
269
+ "nbytes": 2097152,
270
+ "byteOffset": 3538944
271
+ },
272
+ {
273
+ "name": "model.layers.1.self_attn.o_proj.q_scale",
274
+ "shape": [
275
+ 2048,
276
+ 64
277
+ ],
278
+ "dtype": "float16",
279
+ "format": "f32-to-bf16",
280
+ "nbytes": 262144,
281
+ "byteOffset": 5636096
282
+ },
283
+ {
284
+ "name": "model.layers.10.input_layernorm.weight",
285
+ "shape": [
286
+ 2048
287
+ ],
288
+ "dtype": "float16",
289
+ "format": "f32-to-bf16",
290
+ "nbytes": 4096,
291
+ "byteOffset": 5898240
292
+ },
293
+ {
294
+ "name": "model.layers.10.mlp.down_proj.q_weight",
295
+ "shape": [
296
+ 2048,
297
+ 1024
298
+ ],
299
+ "dtype": "uint32",
300
+ "format": "f32-to-bf16",
301
+ "nbytes": 8388608,
302
+ "byteOffset": 5902336
303
+ },
304
+ {
305
+ "name": "model.layers.10.mlp.down_proj.q_scale",
306
+ "shape": [
307
+ 2048,
308
+ 256
309
+ ],
310
+ "dtype": "float16",
311
+ "format": "f32-to-bf16",
312
+ "nbytes": 1048576,
313
+ "byteOffset": 14290944
314
+ },
315
+ {
316
+ "name": "model.layers.10.mlp.gate_up_proj.q_weight",
317
+ "shape": [
318
+ 16384,
319
+ 256
320
+ ],
321
+ "dtype": "uint32",
322
+ "format": "f32-to-bf16",
323
+ "nbytes": 16777216,
324
+ "byteOffset": 15339520
325
+ }
326
+ ],
327
+ "md5sum": "d40c880de01576bbaecc3abf7f820688"
328
+ },
329
+ {
330
+ "dataPath": "params_shard_5.bin",
331
+ "format": "raw-shard",
332
+ "nbytes": 16777216,
333
+ "records": [
334
+ {
335
+ "name": "model.layers.11.mlp.gate_up_proj.q_weight",
336
+ "shape": [
337
+ 16384,
338
+ 256
339
+ ],
340
+ "dtype": "uint32",
341
+ "format": "f32-to-bf16",
342
+ "nbytes": 16777216,
343
+ "byteOffset": 0
344
+ }
345
+ ],
346
+ "md5sum": "9e3b2f06d748bb4267a77b44d8a489df"
347
+ },
348
+ {
349
+ "dataPath": "params_shard_6.bin",
350
+ "format": "raw-shard",
351
+ "nbytes": 25444352,
352
+ "records": [
353
+ {
354
+ "name": "model.layers.10.mlp.gate_up_proj.q_scale",
355
+ "shape": [
356
+ 16384,
357
+ 64
358
+ ],
359
+ "dtype": "float16",
360
+ "format": "f32-to-bf16",
361
+ "nbytes": 2097152,
362
+ "byteOffset": 0
363
+ },
364
+ {
365
+ "name": "model.layers.10.post_attention_layernorm.weight",
366
+ "shape": [
367
+ 2048
368
+ ],
369
+ "dtype": "float16",
370
+ "format": "f32-to-bf16",
371
+ "nbytes": 4096,
372
+ "byteOffset": 2097152
373
+ },
374
+ {
375
+ "name": "model.layers.10.self_attn.qkv_proj.q_weight",
376
+ "shape": [
377
+ 3072,
378
+ 256
379
+ ],
380
+ "dtype": "uint32",
381
+ "format": "f32-to-bf16",
382
+ "nbytes": 3145728,
383
+ "byteOffset": 2101248
384
+ },
385
+ {
386
+ "name": "model.layers.10.self_attn.qkv_proj.q_scale",
387
+ "shape": [
388
+ 3072,
389
+ 64
390
+ ],
391
+ "dtype": "float16",
392
+ "format": "f32-to-bf16",
393
+ "nbytes": 393216,
394
+ "byteOffset": 5246976
395
+ },
396
+ {
397
+ "name": "model.layers.10.self_attn.o_proj.q_weight",
398
+ "shape": [
399
+ 2048,
400
+ 256
401
+ ],
402
+ "dtype": "uint32",
403
+ "format": "f32-to-bf16",
404
+ "nbytes": 2097152,
405
+ "byteOffset": 5640192
406
+ },
407
+ {
408
+ "name": "model.layers.10.self_attn.o_proj.q_scale",
409
+ "shape": [
410
+ 2048,
411
+ 64
412
+ ],
413
+ "dtype": "float16",
414
+ "format": "f32-to-bf16",
415
+ "nbytes": 262144,
416
+ "byteOffset": 7737344
417
+ },
418
+ {
419
+ "name": "model.layers.11.input_layernorm.weight",
420
+ "shape": [
421
+ 2048
422
+ ],
423
+ "dtype": "float16",
424
+ "format": "f32-to-bf16",
425
+ "nbytes": 4096,
426
+ "byteOffset": 7999488
427
+ },
428
+ {
429
+ "name": "model.layers.11.mlp.down_proj.q_weight",
430
+ "shape": [
431
+ 2048,
432
+ 1024
433
+ ],
434
+ "dtype": "uint32",
435
+ "format": "f32-to-bf16",
436
+ "nbytes": 8388608,
437
+ "byteOffset": 8003584
438
+ },
439
+ {
440
+ "name": "model.layers.11.mlp.down_proj.q_scale",
441
+ "shape": [
442
+ 2048,
443
+ 256
444
+ ],
445
+ "dtype": "float16",
446
+ "format": "f32-to-bf16",
447
+ "nbytes": 1048576,
448
+ "byteOffset": 16392192
449
+ },
450
+ {
451
+ "name": "model.layers.11.mlp.gate_up_proj.q_scale",
452
+ "shape": [
453
+ 16384,
454
+ 64
455
+ ],
456
+ "dtype": "float16",
457
+ "format": "f32-to-bf16",
458
+ "nbytes": 2097152,
459
+ "byteOffset": 17440768
460
+ },
461
+ {
462
+ "name": "model.layers.11.post_attention_layernorm.weight",
463
+ "shape": [
464
+ 2048
465
+ ],
466
+ "dtype": "float16",
467
+ "format": "f32-to-bf16",
468
+ "nbytes": 4096,
469
+ "byteOffset": 19537920
470
+ },
471
+ {
472
+ "name": "model.layers.11.self_attn.qkv_proj.q_weight",
473
+ "shape": [
474
+ 3072,
475
+ 256
476
+ ],
477
+ "dtype": "uint32",
478
+ "format": "f32-to-bf16",
479
+ "nbytes": 3145728,
480
+ "byteOffset": 19542016
481
+ },
482
+ {
483
+ "name": "model.layers.11.self_attn.qkv_proj.q_scale",
484
+ "shape": [
485
+ 3072,
486
+ 64
487
+ ],
488
+ "dtype": "float16",
489
+ "format": "f32-to-bf16",
490
+ "nbytes": 393216,
491
+ "byteOffset": 22687744
492
+ },
493
+ {
494
+ "name": "model.layers.11.self_attn.o_proj.q_weight",
495
+ "shape": [
496
+ 2048,
497
+ 256
498
+ ],
499
+ "dtype": "uint32",
500
+ "format": "f32-to-bf16",
501
+ "nbytes": 2097152,
502
+ "byteOffset": 23080960
503
+ },
504
+ {
505
+ "name": "model.layers.11.self_attn.o_proj.q_scale",
506
+ "shape": [
507
+ 2048,
508
+ 64
509
+ ],
510
+ "dtype": "float16",
511
+ "format": "f32-to-bf16",
512
+ "nbytes": 262144,
513
+ "byteOffset": 25178112
514
+ },
515
+ {
516
+ "name": "model.layers.12.input_layernorm.weight",
517
+ "shape": [
518
+ 2048
519
+ ],
520
+ "dtype": "float16",
521
+ "format": "f32-to-bf16",
522
+ "nbytes": 4096,
523
+ "byteOffset": 25440256
524
+ }
525
+ ],
526
+ "md5sum": "cba2e299b4275ebc7054f8ad16b7cdb9"
527
+ },
528
+ {
529
+ "dataPath": "params_shard_7.bin",
530
+ "format": "raw-shard",
531
+ "nbytes": 31854592,
532
+ "records": [
533
+ {
534
+ "name": "model.layers.12.mlp.down_proj.q_weight",
535
+ "shape": [
536
+ 2048,
537
+ 1024
538
+ ],
539
+ "dtype": "uint32",
540
+ "format": "f32-to-bf16",
541
+ "nbytes": 8388608,
542
+ "byteOffset": 0
543
+ },
544
+ {
545
+ "name": "model.layers.12.mlp.down_proj.q_scale",
546
+ "shape": [
547
+ 2048,
548
+ 256
549
+ ],
550
+ "dtype": "float16",
551
+ "format": "f32-to-bf16",
552
+ "nbytes": 1048576,
553
+ "byteOffset": 8388608
554
+ },
555
+ {
556
+ "name": "model.layers.12.mlp.gate_up_proj.q_weight",
557
+ "shape": [
558
+ 16384,
559
+ 256
560
+ ],
561
+ "dtype": "uint32",
562
+ "format": "f32-to-bf16",
563
+ "nbytes": 16777216,
564
+ "byteOffset": 9437184
565
+ },
566
+ {
567
+ "name": "model.layers.12.mlp.gate_up_proj.q_scale",
568
+ "shape": [
569
+ 16384,
570
+ 64
571
+ ],
572
+ "dtype": "float16",
573
+ "format": "f32-to-bf16",
574
+ "nbytes": 2097152,
575
+ "byteOffset": 26214400
576
+ },
577
+ {
578
+ "name": "model.layers.12.post_attention_layernorm.weight",
579
+ "shape": [
580
+ 2048
581
+ ],
582
+ "dtype": "float16",
583
+ "format": "f32-to-bf16",
584
+ "nbytes": 4096,
585
+ "byteOffset": 28311552
586
+ },
587
+ {
588
+ "name": "model.layers.12.self_attn.qkv_proj.q_weight",
589
+ "shape": [
590
+ 3072,
591
+ 256
592
+ ],
593
+ "dtype": "uint32",
594
+ "format": "f32-to-bf16",
595
+ "nbytes": 3145728,
596
+ "byteOffset": 28315648
597
+ },
598
+ {
599
+ "name": "model.layers.12.self_attn.qkv_proj.q_scale",
600
+ "shape": [
601
+ 3072,
602
+ 64
603
+ ],
604
+ "dtype": "float16",
605
+ "format": "f32-to-bf16",
606
+ "nbytes": 393216,
607
+ "byteOffset": 31461376
608
+ }
609
+ ],
610
+ "md5sum": "2bc3abb018d2d9db8a546c24c33abbc2"
611
+ },
612
+ {
613
+ "dataPath": "params_shard_8.bin",
614
+ "format": "raw-shard",
615
+ "nbytes": 30679040,
616
+ "records": [
617
+ {
618
+ "name": "model.layers.12.self_attn.o_proj.q_weight",
619
+ "shape": [
620
+ 2048,
621
+ 256
622
+ ],
623
+ "dtype": "uint32",
624
+ "format": "f32-to-bf16",
625
+ "nbytes": 2097152,
626
+ "byteOffset": 0
627
+ },
628
+ {
629
+ "name": "model.layers.12.self_attn.o_proj.q_scale",
630
+ "shape": [
631
+ 2048,
632
+ 64
633
+ ],
634
+ "dtype": "float16",
635
+ "format": "f32-to-bf16",
636
+ "nbytes": 262144,
637
+ "byteOffset": 2097152
638
+ },
639
+ {
640
+ "name": "model.layers.13.input_layernorm.weight",
641
+ "shape": [
642
+ 2048
643
+ ],
644
+ "dtype": "float16",
645
+ "format": "f32-to-bf16",
646
+ "nbytes": 4096,
647
+ "byteOffset": 2359296
648
+ },
649
+ {
650
+ "name": "model.layers.13.mlp.down_proj.q_weight",
651
+ "shape": [
652
+ 2048,
653
+ 1024
654
+ ],
655
+ "dtype": "uint32",
656
+ "format": "f32-to-bf16",
657
+ "nbytes": 8388608,
658
+ "byteOffset": 2363392
659
+ },
660
+ {
661
+ "name": "model.layers.13.mlp.down_proj.q_scale",
662
+ "shape": [
663
+ 2048,
664
+ 256
665
+ ],
666
+ "dtype": "float16",
667
+ "format": "f32-to-bf16",
668
+ "nbytes": 1048576,
669
+ "byteOffset": 10752000
670
+ },
671
+ {
672
+ "name": "model.layers.13.mlp.gate_up_proj.q_weight",
673
+ "shape": [
674
+ 16384,
675
+ 256
676
+ ],
677
+ "dtype": "uint32",
678
+ "format": "f32-to-bf16",
679
+ "nbytes": 16777216,
680
+ "byteOffset": 11800576
681
+ },
682
+ {
683
+ "name": "model.layers.13.mlp.gate_up_proj.q_scale",
684
+ "shape": [
685
+ 16384,
686
+ 64
687
+ ],
688
+ "dtype": "float16",
689
+ "format": "f32-to-bf16",
690
+ "nbytes": 2097152,
691
+ "byteOffset": 28577792
692
+ },
693
+ {
694
+ "name": "model.layers.13.post_attention_layernorm.weight",
695
+ "shape": [
696
+ 2048
697
+ ],
698
+ "dtype": "float16",
699
+ "format": "f32-to-bf16",
700
+ "nbytes": 4096,
701
+ "byteOffset": 30674944
702
+ }
703
+ ],
704
+ "md5sum": "4f50bccecdd051f00c706d50db12cda6"
705
+ },
706
+ {
707
+ "dataPath": "params_shard_9.bin",
708
+ "format": "raw-shard",
709
+ "nbytes": 32116736,
710
+ "records": [
711
+ {
712
+ "name": "model.layers.13.self_attn.qkv_proj.q_weight",
713
+ "shape": [
714
+ 3072,
715
+ 256
716
+ ],
717
+ "dtype": "uint32",
718
+ "format": "f32-to-bf16",
719
+ "nbytes": 3145728,
720
+ "byteOffset": 0
721
+ },
722
+ {
723
+ "name": "model.layers.13.self_attn.qkv_proj.q_scale",
724
+ "shape": [
725
+ 3072,
726
+ 64
727
+ ],
728
+ "dtype": "float16",
729
+ "format": "f32-to-bf16",
730
+ "nbytes": 393216,
731
+ "byteOffset": 3145728
732
+ },
733
+ {
734
+ "name": "model.layers.13.self_attn.o_proj.q_weight",
735
+ "shape": [
736
+ 2048,
737
+ 256
738
+ ],
739
+ "dtype": "uint32",
740
+ "format": "f32-to-bf16",
741
+ "nbytes": 2097152,
742
+ "byteOffset": 3538944
743
+ },
744
+ {
745
+ "name": "model.layers.13.self_attn.o_proj.q_scale",
746
+ "shape": [
747
+ 2048,
748
+ 64
749
+ ],
750
+ "dtype": "float16",
751
+ "format": "f32-to-bf16",
752
+ "nbytes": 262144,
753
+ "byteOffset": 5636096
754
+ },
755
+ {
756
+ "name": "model.layers.14.input_layernorm.weight",
757
+ "shape": [
758
+ 2048
759
+ ],
760
+ "dtype": "float16",
761
+ "format": "f32-to-bf16",
762
+ "nbytes": 4096,
763
+ "byteOffset": 5898240
764
+ },
765
+ {
766
+ "name": "model.layers.14.mlp.down_proj.q_weight",
767
+ "shape": [
768
+ 2048,
769
+ 1024
770
+ ],
771
+ "dtype": "uint32",
772
+ "format": "f32-to-bf16",
773
+ "nbytes": 8388608,
774
+ "byteOffset": 5902336
775
+ },
776
+ {
777
+ "name": "model.layers.14.mlp.down_proj.q_scale",
778
+ "shape": [
779
+ 2048,
780
+ 256
781
+ ],
782
+ "dtype": "float16",
783
+ "format": "f32-to-bf16",
784
+ "nbytes": 1048576,
785
+ "byteOffset": 14290944
786
+ },
787
+ {
788
+ "name": "model.layers.14.mlp.gate_up_proj.q_weight",
789
+ "shape": [
790
+ 16384,
791
+ 256
792
+ ],
793
+ "dtype": "uint32",
794
+ "format": "f32-to-bf16",
795
+ "nbytes": 16777216,
796
+ "byteOffset": 15339520
797
+ }
798
+ ],
799
+ "md5sum": "2b7c430520a6507126e0bbb9b31bb7ee"
800
+ },
801
+ {
802
+ "dataPath": "params_shard_10.bin",
803
+ "format": "raw-shard",
804
+ "nbytes": 16777216,
805
+ "records": [
806
+ {
807
+ "name": "model.layers.15.mlp.gate_up_proj.q_weight",
808
+ "shape": [
809
+ 16384,
810
+ 256
811
+ ],
812
+ "dtype": "uint32",
813
+ "format": "f32-to-bf16",
814
+ "nbytes": 16777216,
815
+ "byteOffset": 0
816
+ }
817
+ ],
818
+ "md5sum": "043ab15ecb84e64c808edbfe036ae595"
819
+ },
820
+ {
821
+ "dataPath": "params_shard_11.bin",
822
+ "format": "raw-shard",
823
+ "nbytes": 25444352,
824
+ "records": [
825
+ {
826
+ "name": "model.layers.14.mlp.gate_up_proj.q_scale",
827
+ "shape": [
828
+ 16384,
829
+ 64
830
+ ],
831
+ "dtype": "float16",
832
+ "format": "f32-to-bf16",
833
+ "nbytes": 2097152,
834
+ "byteOffset": 0
835
+ },
836
+ {
837
+ "name": "model.layers.14.post_attention_layernorm.weight",
838
+ "shape": [
839
+ 2048
840
+ ],
841
+ "dtype": "float16",
842
+ "format": "f32-to-bf16",
843
+ "nbytes": 4096,
844
+ "byteOffset": 2097152
845
+ },
846
+ {
847
+ "name": "model.layers.14.self_attn.qkv_proj.q_weight",
848
+ "shape": [
849
+ 3072,
850
+ 256
851
+ ],
852
+ "dtype": "uint32",
853
+ "format": "f32-to-bf16",
854
+ "nbytes": 3145728,
855
+ "byteOffset": 2101248
856
+ },
857
+ {
858
+ "name": "model.layers.14.self_attn.qkv_proj.q_scale",
859
+ "shape": [
860
+ 3072,
861
+ 64
862
+ ],
863
+ "dtype": "float16",
864
+ "format": "f32-to-bf16",
865
+ "nbytes": 393216,
866
+ "byteOffset": 5246976
867
+ },
868
+ {
869
+ "name": "model.layers.14.self_attn.o_proj.q_weight",
870
+ "shape": [
871
+ 2048,
872
+ 256
873
+ ],
874
+ "dtype": "uint32",
875
+ "format": "f32-to-bf16",
876
+ "nbytes": 2097152,
877
+ "byteOffset": 5640192
878
+ },
879
+ {
880
+ "name": "model.layers.14.self_attn.o_proj.q_scale",
881
+ "shape": [
882
+ 2048,
883
+ 64
884
+ ],
885
+ "dtype": "float16",
886
+ "format": "f32-to-bf16",
887
+ "nbytes": 262144,
888
+ "byteOffset": 7737344
889
+ },
890
+ {
891
+ "name": "model.layers.15.input_layernorm.weight",
892
+ "shape": [
893
+ 2048
894
+ ],
895
+ "dtype": "float16",
896
+ "format": "f32-to-bf16",
897
+ "nbytes": 4096,
898
+ "byteOffset": 7999488
899
+ },
900
+ {
901
+ "name": "model.layers.15.mlp.down_proj.q_weight",
902
+ "shape": [
903
+ 2048,
904
+ 1024
905
+ ],
906
+ "dtype": "uint32",
907
+ "format": "f32-to-bf16",
908
+ "nbytes": 8388608,
909
+ "byteOffset": 8003584
910
+ },
911
+ {
912
+ "name": "model.layers.15.mlp.down_proj.q_scale",
913
+ "shape": [
914
+ 2048,
915
+ 256
916
+ ],
917
+ "dtype": "float16",
918
+ "format": "f32-to-bf16",
919
+ "nbytes": 1048576,
920
+ "byteOffset": 16392192
921
+ },
922
+ {
923
+ "name": "model.layers.15.mlp.gate_up_proj.q_scale",
924
+ "shape": [
925
+ 16384,
926
+ 64
927
+ ],
928
+ "dtype": "float16",
929
+ "format": "f32-to-bf16",
930
+ "nbytes": 2097152,
931
+ "byteOffset": 17440768
932
+ },
933
+ {
934
+ "name": "model.layers.15.post_attention_layernorm.weight",
935
+ "shape": [
936
+ 2048
937
+ ],
938
+ "dtype": "float16",
939
+ "format": "f32-to-bf16",
940
+ "nbytes": 4096,
941
+ "byteOffset": 19537920
942
+ },
943
+ {
944
+ "name": "model.layers.15.self_attn.qkv_proj.q_weight",
945
+ "shape": [
946
+ 3072,
947
+ 256
948
+ ],
949
+ "dtype": "uint32",
950
+ "format": "f32-to-bf16",
951
+ "nbytes": 3145728,
952
+ "byteOffset": 19542016
953
+ },
954
+ {
955
+ "name": "model.layers.15.self_attn.qkv_proj.q_scale",
956
+ "shape": [
957
+ 3072,
958
+ 64
959
+ ],
960
+ "dtype": "float16",
961
+ "format": "f32-to-bf16",
962
+ "nbytes": 393216,
963
+ "byteOffset": 22687744
964
+ },
965
+ {
966
+ "name": "model.layers.15.self_attn.o_proj.q_weight",
967
+ "shape": [
968
+ 2048,
969
+ 256
970
+ ],
971
+ "dtype": "uint32",
972
+ "format": "f32-to-bf16",
973
+ "nbytes": 2097152,
974
+ "byteOffset": 23080960
975
+ },
976
+ {
977
+ "name": "model.layers.15.self_attn.o_proj.q_scale",
978
+ "shape": [
979
+ 2048,
980
+ 64
981
+ ],
982
+ "dtype": "float16",
983
+ "format": "f32-to-bf16",
984
+ "nbytes": 262144,
985
+ "byteOffset": 25178112
986
+ },
987
+ {
988
+ "name": "model.layers.2.input_layernorm.weight",
989
+ "shape": [
990
+ 2048
991
+ ],
992
+ "dtype": "float16",
993
+ "format": "f32-to-bf16",
994
+ "nbytes": 4096,
995
+ "byteOffset": 25440256
996
+ }
997
+ ],
998
+ "md5sum": "542c8599d640c38ae6f192a742e235d0"
999
+ },
1000
+ {
1001
+ "dataPath": "params_shard_12.bin",
1002
+ "format": "raw-shard",
1003
+ "nbytes": 31854592,
1004
+ "records": [
1005
+ {
1006
+ "name": "model.layers.2.mlp.down_proj.q_weight",
1007
+ "shape": [
1008
+ 2048,
1009
+ 1024
1010
+ ],
1011
+ "dtype": "uint32",
1012
+ "format": "f32-to-bf16",
1013
+ "nbytes": 8388608,
1014
+ "byteOffset": 0
1015
+ },
1016
+ {
1017
+ "name": "model.layers.2.mlp.down_proj.q_scale",
1018
+ "shape": [
1019
+ 2048,
1020
+ 256
1021
+ ],
1022
+ "dtype": "float16",
1023
+ "format": "f32-to-bf16",
1024
+ "nbytes": 1048576,
1025
+ "byteOffset": 8388608
1026
+ },
1027
+ {
1028
+ "name": "model.layers.2.mlp.gate_up_proj.q_weight",
1029
+ "shape": [
1030
+ 16384,
1031
+ 256
1032
+ ],
1033
+ "dtype": "uint32",
1034
+ "format": "f32-to-bf16",
1035
+ "nbytes": 16777216,
1036
+ "byteOffset": 9437184
1037
+ },
1038
+ {
1039
+ "name": "model.layers.2.mlp.gate_up_proj.q_scale",
1040
+ "shape": [
1041
+ 16384,
1042
+ 64
1043
+ ],
1044
+ "dtype": "float16",
1045
+ "format": "f32-to-bf16",
1046
+ "nbytes": 2097152,
1047
+ "byteOffset": 26214400
1048
+ },
1049
+ {
1050
+ "name": "model.layers.2.post_attention_layernorm.weight",
1051
+ "shape": [
1052
+ 2048
1053
+ ],
1054
+ "dtype": "float16",
1055
+ "format": "f32-to-bf16",
1056
+ "nbytes": 4096,
1057
+ "byteOffset": 28311552
1058
+ },
1059
+ {
1060
+ "name": "model.layers.2.self_attn.qkv_proj.q_weight",
1061
+ "shape": [
1062
+ 3072,
1063
+ 256
1064
+ ],
1065
+ "dtype": "uint32",
1066
+ "format": "f32-to-bf16",
1067
+ "nbytes": 3145728,
1068
+ "byteOffset": 28315648
1069
+ },
1070
+ {
1071
+ "name": "model.layers.2.self_attn.qkv_proj.q_scale",
1072
+ "shape": [
1073
+ 3072,
1074
+ 64
1075
+ ],
1076
+ "dtype": "float16",
1077
+ "format": "f32-to-bf16",
1078
+ "nbytes": 393216,
1079
+ "byteOffset": 31461376
1080
+ }
1081
+ ],
1082
+ "md5sum": "b4cd5568e2877aabade1a08c4f9f8e20"
1083
+ },
1084
+ {
1085
+ "dataPath": "params_shard_13.bin",
1086
+ "format": "raw-shard",
1087
+ "nbytes": 30679040,
1088
+ "records": [
1089
+ {
1090
+ "name": "model.layers.2.self_attn.o_proj.q_weight",
1091
+ "shape": [
1092
+ 2048,
1093
+ 256
1094
+ ],
1095
+ "dtype": "uint32",
1096
+ "format": "f32-to-bf16",
1097
+ "nbytes": 2097152,
1098
+ "byteOffset": 0
1099
+ },
1100
+ {
1101
+ "name": "model.layers.2.self_attn.o_proj.q_scale",
1102
+ "shape": [
1103
+ 2048,
1104
+ 64
1105
+ ],
1106
+ "dtype": "float16",
1107
+ "format": "f32-to-bf16",
1108
+ "nbytes": 262144,
1109
+ "byteOffset": 2097152
1110
+ },
1111
+ {
1112
+ "name": "model.layers.3.input_layernorm.weight",
1113
+ "shape": [
1114
+ 2048
1115
+ ],
1116
+ "dtype": "float16",
1117
+ "format": "f32-to-bf16",
1118
+ "nbytes": 4096,
1119
+ "byteOffset": 2359296
1120
+ },
1121
+ {
1122
+ "name": "model.layers.3.mlp.down_proj.q_weight",
1123
+ "shape": [
1124
+ 2048,
1125
+ 1024
1126
+ ],
1127
+ "dtype": "uint32",
1128
+ "format": "f32-to-bf16",
1129
+ "nbytes": 8388608,
1130
+ "byteOffset": 2363392
1131
+ },
1132
+ {
1133
+ "name": "model.layers.3.mlp.down_proj.q_scale",
1134
+ "shape": [
1135
+ 2048,
1136
+ 256
1137
+ ],
1138
+ "dtype": "float16",
1139
+ "format": "f32-to-bf16",
1140
+ "nbytes": 1048576,
1141
+ "byteOffset": 10752000
1142
+ },
1143
+ {
1144
+ "name": "model.layers.3.mlp.gate_up_proj.q_weight",
1145
+ "shape": [
1146
+ 16384,
1147
+ 256
1148
+ ],
1149
+ "dtype": "uint32",
1150
+ "format": "f32-to-bf16",
1151
+ "nbytes": 16777216,
1152
+ "byteOffset": 11800576
1153
+ },
1154
+ {
1155
+ "name": "model.layers.3.mlp.gate_up_proj.q_scale",
1156
+ "shape": [
1157
+ 16384,
1158
+ 64
1159
+ ],
1160
+ "dtype": "float16",
1161
+ "format": "f32-to-bf16",
1162
+ "nbytes": 2097152,
1163
+ "byteOffset": 28577792
1164
+ },
1165
+ {
1166
+ "name": "model.layers.3.post_attention_layernorm.weight",
1167
+ "shape": [
1168
+ 2048
1169
+ ],
1170
+ "dtype": "float16",
1171
+ "format": "f32-to-bf16",
1172
+ "nbytes": 4096,
1173
+ "byteOffset": 30674944
1174
+ }
1175
+ ],
1176
+ "md5sum": "cbce2e42eb5932d41213ab17b83daed4"
1177
+ },
1178
+ {
1179
+ "dataPath": "params_shard_14.bin",
1180
+ "format": "raw-shard",
1181
+ "nbytes": 32116736,
1182
+ "records": [
1183
+ {
1184
+ "name": "model.layers.3.self_attn.qkv_proj.q_weight",
1185
+ "shape": [
1186
+ 3072,
1187
+ 256
1188
+ ],
1189
+ "dtype": "uint32",
1190
+ "format": "f32-to-bf16",
1191
+ "nbytes": 3145728,
1192
+ "byteOffset": 0
1193
+ },
1194
+ {
1195
+ "name": "model.layers.3.self_attn.qkv_proj.q_scale",
1196
+ "shape": [
1197
+ 3072,
1198
+ 64
1199
+ ],
1200
+ "dtype": "float16",
1201
+ "format": "f32-to-bf16",
1202
+ "nbytes": 393216,
1203
+ "byteOffset": 3145728
1204
+ },
1205
+ {
1206
+ "name": "model.layers.3.self_attn.o_proj.q_weight",
1207
+ "shape": [
1208
+ 2048,
1209
+ 256
1210
+ ],
1211
+ "dtype": "uint32",
1212
+ "format": "f32-to-bf16",
1213
+ "nbytes": 2097152,
1214
+ "byteOffset": 3538944
1215
+ },
1216
+ {
1217
+ "name": "model.layers.3.self_attn.o_proj.q_scale",
1218
+ "shape": [
1219
+ 2048,
1220
+ 64
1221
+ ],
1222
+ "dtype": "float16",
1223
+ "format": "f32-to-bf16",
1224
+ "nbytes": 262144,
1225
+ "byteOffset": 5636096
1226
+ },
1227
+ {
1228
+ "name": "model.layers.4.input_layernorm.weight",
1229
+ "shape": [
1230
+ 2048
1231
+ ],
1232
+ "dtype": "float16",
1233
+ "format": "f32-to-bf16",
1234
+ "nbytes": 4096,
1235
+ "byteOffset": 5898240
1236
+ },
1237
+ {
1238
+ "name": "model.layers.4.mlp.down_proj.q_weight",
1239
+ "shape": [
1240
+ 2048,
1241
+ 1024
1242
+ ],
1243
+ "dtype": "uint32",
1244
+ "format": "f32-to-bf16",
1245
+ "nbytes": 8388608,
1246
+ "byteOffset": 5902336
1247
+ },
1248
+ {
1249
+ "name": "model.layers.4.mlp.down_proj.q_scale",
1250
+ "shape": [
1251
+ 2048,
1252
+ 256
1253
+ ],
1254
+ "dtype": "float16",
1255
+ "format": "f32-to-bf16",
1256
+ "nbytes": 1048576,
1257
+ "byteOffset": 14290944
1258
+ },
1259
+ {
1260
+ "name": "model.layers.4.mlp.gate_up_proj.q_weight",
1261
+ "shape": [
1262
+ 16384,
1263
+ 256
1264
+ ],
1265
+ "dtype": "uint32",
1266
+ "format": "f32-to-bf16",
1267
+ "nbytes": 16777216,
1268
+ "byteOffset": 15339520
1269
+ }
1270
+ ],
1271
+ "md5sum": "d7f189698bd8fc3341223af3b608f4d5"
1272
+ },
1273
+ {
1274
+ "dataPath": "params_shard_15.bin",
1275
+ "format": "raw-shard",
1276
+ "nbytes": 16777216,
1277
+ "records": [
1278
+ {
1279
+ "name": "model.layers.5.mlp.gate_up_proj.q_weight",
1280
+ "shape": [
1281
+ 16384,
1282
+ 256
1283
+ ],
1284
+ "dtype": "uint32",
1285
+ "format": "f32-to-bf16",
1286
+ "nbytes": 16777216,
1287
+ "byteOffset": 0
1288
+ }
1289
+ ],
1290
+ "md5sum": "77dc740bbecc7ab456e27ba134d17034"
1291
+ },
1292
+ {
1293
+ "dataPath": "params_shard_16.bin",
1294
+ "format": "raw-shard",
1295
+ "nbytes": 25444352,
1296
+ "records": [
1297
+ {
1298
+ "name": "model.layers.4.mlp.gate_up_proj.q_scale",
1299
+ "shape": [
1300
+ 16384,
1301
+ 64
1302
+ ],
1303
+ "dtype": "float16",
1304
+ "format": "f32-to-bf16",
1305
+ "nbytes": 2097152,
1306
+ "byteOffset": 0
1307
+ },
1308
+ {
1309
+ "name": "model.layers.4.post_attention_layernorm.weight",
1310
+ "shape": [
1311
+ 2048
1312
+ ],
1313
+ "dtype": "float16",
1314
+ "format": "f32-to-bf16",
1315
+ "nbytes": 4096,
1316
+ "byteOffset": 2097152
1317
+ },
1318
+ {
1319
+ "name": "model.layers.4.self_attn.qkv_proj.q_weight",
1320
+ "shape": [
1321
+ 3072,
1322
+ 256
1323
+ ],
1324
+ "dtype": "uint32",
1325
+ "format": "f32-to-bf16",
1326
+ "nbytes": 3145728,
1327
+ "byteOffset": 2101248
1328
+ },
1329
+ {
1330
+ "name": "model.layers.4.self_attn.qkv_proj.q_scale",
1331
+ "shape": [
1332
+ 3072,
1333
+ 64
1334
+ ],
1335
+ "dtype": "float16",
1336
+ "format": "f32-to-bf16",
1337
+ "nbytes": 393216,
1338
+ "byteOffset": 5246976
1339
+ },
1340
+ {
1341
+ "name": "model.layers.4.self_attn.o_proj.q_weight",
1342
+ "shape": [
1343
+ 2048,
1344
+ 256
1345
+ ],
1346
+ "dtype": "uint32",
1347
+ "format": "f32-to-bf16",
1348
+ "nbytes": 2097152,
1349
+ "byteOffset": 5640192
1350
+ },
1351
+ {
1352
+ "name": "model.layers.4.self_attn.o_proj.q_scale",
1353
+ "shape": [
1354
+ 2048,
1355
+ 64
1356
+ ],
1357
+ "dtype": "float16",
1358
+ "format": "f32-to-bf16",
1359
+ "nbytes": 262144,
1360
+ "byteOffset": 7737344
1361
+ },
1362
+ {
1363
+ "name": "model.layers.5.input_layernorm.weight",
1364
+ "shape": [
1365
+ 2048
1366
+ ],
1367
+ "dtype": "float16",
1368
+ "format": "f32-to-bf16",
1369
+ "nbytes": 4096,
1370
+ "byteOffset": 7999488
1371
+ },
1372
+ {
1373
+ "name": "model.layers.5.mlp.down_proj.q_weight",
1374
+ "shape": [
1375
+ 2048,
1376
+ 1024
1377
+ ],
1378
+ "dtype": "uint32",
1379
+ "format": "f32-to-bf16",
1380
+ "nbytes": 8388608,
1381
+ "byteOffset": 8003584
1382
+ },
1383
+ {
1384
+ "name": "model.layers.5.mlp.down_proj.q_scale",
1385
+ "shape": [
1386
+ 2048,
1387
+ 256
1388
+ ],
1389
+ "dtype": "float16",
1390
+ "format": "f32-to-bf16",
1391
+ "nbytes": 1048576,
1392
+ "byteOffset": 16392192
1393
+ },
1394
+ {
1395
+ "name": "model.layers.5.mlp.gate_up_proj.q_scale",
1396
+ "shape": [
1397
+ 16384,
1398
+ 64
1399
+ ],
1400
+ "dtype": "float16",
1401
+ "format": "f32-to-bf16",
1402
+ "nbytes": 2097152,
1403
+ "byteOffset": 17440768
1404
+ },
1405
+ {
1406
+ "name": "model.layers.5.post_attention_layernorm.weight",
1407
+ "shape": [
1408
+ 2048
1409
+ ],
1410
+ "dtype": "float16",
1411
+ "format": "f32-to-bf16",
1412
+ "nbytes": 4096,
1413
+ "byteOffset": 19537920
1414
+ },
1415
+ {
1416
+ "name": "model.layers.5.self_attn.qkv_proj.q_weight",
1417
+ "shape": [
1418
+ 3072,
1419
+ 256
1420
+ ],
1421
+ "dtype": "uint32",
1422
+ "format": "f32-to-bf16",
1423
+ "nbytes": 3145728,
1424
+ "byteOffset": 19542016
1425
+ },
1426
+ {
1427
+ "name": "model.layers.5.self_attn.qkv_proj.q_scale",
1428
+ "shape": [
1429
+ 3072,
1430
+ 64
1431
+ ],
1432
+ "dtype": "float16",
1433
+ "format": "f32-to-bf16",
1434
+ "nbytes": 393216,
1435
+ "byteOffset": 22687744
1436
+ },
1437
+ {
1438
+ "name": "model.layers.5.self_attn.o_proj.q_weight",
1439
+ "shape": [
1440
+ 2048,
1441
+ 256
1442
+ ],
1443
+ "dtype": "uint32",
1444
+ "format": "f32-to-bf16",
1445
+ "nbytes": 2097152,
1446
+ "byteOffset": 23080960
1447
+ },
1448
+ {
1449
+ "name": "model.layers.5.self_attn.o_proj.q_scale",
1450
+ "shape": [
1451
+ 2048,
1452
+ 64
1453
+ ],
1454
+ "dtype": "float16",
1455
+ "format": "f32-to-bf16",
1456
+ "nbytes": 262144,
1457
+ "byteOffset": 25178112
1458
+ },
1459
+ {
1460
+ "name": "model.layers.6.input_layernorm.weight",
1461
+ "shape": [
1462
+ 2048
1463
+ ],
1464
+ "dtype": "float16",
1465
+ "format": "f32-to-bf16",
1466
+ "nbytes": 4096,
1467
+ "byteOffset": 25440256
1468
+ }
1469
+ ],
1470
+ "md5sum": "03cb14edc29900dbc711ebb144ad3705"
1471
+ },
1472
+ {
1473
+ "dataPath": "params_shard_17.bin",
1474
+ "format": "raw-shard",
1475
+ "nbytes": 31854592,
1476
+ "records": [
1477
+ {
1478
+ "name": "model.layers.6.mlp.down_proj.q_weight",
1479
+ "shape": [
1480
+ 2048,
1481
+ 1024
1482
+ ],
1483
+ "dtype": "uint32",
1484
+ "format": "f32-to-bf16",
1485
+ "nbytes": 8388608,
1486
+ "byteOffset": 0
1487
+ },
1488
+ {
1489
+ "name": "model.layers.6.mlp.down_proj.q_scale",
1490
+ "shape": [
1491
+ 2048,
1492
+ 256
1493
+ ],
1494
+ "dtype": "float16",
1495
+ "format": "f32-to-bf16",
1496
+ "nbytes": 1048576,
1497
+ "byteOffset": 8388608
1498
+ },
1499
+ {
1500
+ "name": "model.layers.6.mlp.gate_up_proj.q_weight",
1501
+ "shape": [
1502
+ 16384,
1503
+ 256
1504
+ ],
1505
+ "dtype": "uint32",
1506
+ "format": "f32-to-bf16",
1507
+ "nbytes": 16777216,
1508
+ "byteOffset": 9437184
1509
+ },
1510
+ {
1511
+ "name": "model.layers.6.mlp.gate_up_proj.q_scale",
1512
+ "shape": [
1513
+ 16384,
1514
+ 64
1515
+ ],
1516
+ "dtype": "float16",
1517
+ "format": "f32-to-bf16",
1518
+ "nbytes": 2097152,
1519
+ "byteOffset": 26214400
1520
+ },
1521
+ {
1522
+ "name": "model.layers.6.post_attention_layernorm.weight",
1523
+ "shape": [
1524
+ 2048
1525
+ ],
1526
+ "dtype": "float16",
1527
+ "format": "f32-to-bf16",
1528
+ "nbytes": 4096,
1529
+ "byteOffset": 28311552
1530
+ },
1531
+ {
1532
+ "name": "model.layers.6.self_attn.qkv_proj.q_weight",
1533
+ "shape": [
1534
+ 3072,
1535
+ 256
1536
+ ],
1537
+ "dtype": "uint32",
1538
+ "format": "f32-to-bf16",
1539
+ "nbytes": 3145728,
1540
+ "byteOffset": 28315648
1541
+ },
1542
+ {
1543
+ "name": "model.layers.6.self_attn.qkv_proj.q_scale",
1544
+ "shape": [
1545
+ 3072,
1546
+ 64
1547
+ ],
1548
+ "dtype": "float16",
1549
+ "format": "f32-to-bf16",
1550
+ "nbytes": 393216,
1551
+ "byteOffset": 31461376
1552
+ }
1553
+ ],
1554
+ "md5sum": "e8ac271894487d2d5d0785ccace377cd"
1555
+ },
1556
+ {
1557
+ "dataPath": "params_shard_18.bin",
1558
+ "format": "raw-shard",
1559
+ "nbytes": 30679040,
1560
+ "records": [
1561
+ {
1562
+ "name": "model.layers.6.self_attn.o_proj.q_weight",
1563
+ "shape": [
1564
+ 2048,
1565
+ 256
1566
+ ],
1567
+ "dtype": "uint32",
1568
+ "format": "f32-to-bf16",
1569
+ "nbytes": 2097152,
1570
+ "byteOffset": 0
1571
+ },
1572
+ {
1573
+ "name": "model.layers.6.self_attn.o_proj.q_scale",
1574
+ "shape": [
1575
+ 2048,
1576
+ 64
1577
+ ],
1578
+ "dtype": "float16",
1579
+ "format": "f32-to-bf16",
1580
+ "nbytes": 262144,
1581
+ "byteOffset": 2097152
1582
+ },
1583
+ {
1584
+ "name": "model.layers.7.input_layernorm.weight",
1585
+ "shape": [
1586
+ 2048
1587
+ ],
1588
+ "dtype": "float16",
1589
+ "format": "f32-to-bf16",
1590
+ "nbytes": 4096,
1591
+ "byteOffset": 2359296
1592
+ },
1593
+ {
1594
+ "name": "model.layers.7.mlp.down_proj.q_weight",
1595
+ "shape": [
1596
+ 2048,
1597
+ 1024
1598
+ ],
1599
+ "dtype": "uint32",
1600
+ "format": "f32-to-bf16",
1601
+ "nbytes": 8388608,
1602
+ "byteOffset": 2363392
1603
+ },
1604
+ {
1605
+ "name": "model.layers.7.mlp.down_proj.q_scale",
1606
+ "shape": [
1607
+ 2048,
1608
+ 256
1609
+ ],
1610
+ "dtype": "float16",
1611
+ "format": "f32-to-bf16",
1612
+ "nbytes": 1048576,
1613
+ "byteOffset": 10752000
1614
+ },
1615
+ {
1616
+ "name": "model.layers.7.mlp.gate_up_proj.q_weight",
1617
+ "shape": [
1618
+ 16384,
1619
+ 256
1620
+ ],
1621
+ "dtype": "uint32",
1622
+ "format": "f32-to-bf16",
1623
+ "nbytes": 16777216,
1624
+ "byteOffset": 11800576
1625
+ },
1626
+ {
1627
+ "name": "model.layers.7.mlp.gate_up_proj.q_scale",
1628
+ "shape": [
1629
+ 16384,
1630
+ 64
1631
+ ],
1632
+ "dtype": "float16",
1633
+ "format": "f32-to-bf16",
1634
+ "nbytes": 2097152,
1635
+ "byteOffset": 28577792
1636
+ },
1637
+ {
1638
+ "name": "model.layers.7.post_attention_layernorm.weight",
1639
+ "shape": [
1640
+ 2048
1641
+ ],
1642
+ "dtype": "float16",
1643
+ "format": "f32-to-bf16",
1644
+ "nbytes": 4096,
1645
+ "byteOffset": 30674944
1646
+ }
1647
+ ],
1648
+ "md5sum": "ba970b7d1de3b1f9319ab0083b0f5dbe"
1649
+ },
1650
+ {
1651
+ "dataPath": "params_shard_19.bin",
1652
+ "format": "raw-shard",
1653
+ "nbytes": 32116736,
1654
+ "records": [
1655
+ {
1656
+ "name": "model.layers.7.self_attn.qkv_proj.q_weight",
1657
+ "shape": [
1658
+ 3072,
1659
+ 256
1660
+ ],
1661
+ "dtype": "uint32",
1662
+ "format": "f32-to-bf16",
1663
+ "nbytes": 3145728,
1664
+ "byteOffset": 0
1665
+ },
1666
+ {
1667
+ "name": "model.layers.7.self_attn.qkv_proj.q_scale",
1668
+ "shape": [
1669
+ 3072,
1670
+ 64
1671
+ ],
1672
+ "dtype": "float16",
1673
+ "format": "f32-to-bf16",
1674
+ "nbytes": 393216,
1675
+ "byteOffset": 3145728
1676
+ },
1677
+ {
1678
+ "name": "model.layers.7.self_attn.o_proj.q_weight",
1679
+ "shape": [
1680
+ 2048,
1681
+ 256
1682
+ ],
1683
+ "dtype": "uint32",
1684
+ "format": "f32-to-bf16",
1685
+ "nbytes": 2097152,
1686
+ "byteOffset": 3538944
1687
+ },
1688
+ {
1689
+ "name": "model.layers.7.self_attn.o_proj.q_scale",
1690
+ "shape": [
1691
+ 2048,
1692
+ 64
1693
+ ],
1694
+ "dtype": "float16",
1695
+ "format": "f32-to-bf16",
1696
+ "nbytes": 262144,
1697
+ "byteOffset": 5636096
1698
+ },
1699
+ {
1700
+ "name": "model.layers.8.input_layernorm.weight",
1701
+ "shape": [
1702
+ 2048
1703
+ ],
1704
+ "dtype": "float16",
1705
+ "format": "f32-to-bf16",
1706
+ "nbytes": 4096,
1707
+ "byteOffset": 5898240
1708
+ },
1709
+ {
1710
+ "name": "model.layers.8.mlp.down_proj.q_weight",
1711
+ "shape": [
1712
+ 2048,
1713
+ 1024
1714
+ ],
1715
+ "dtype": "uint32",
1716
+ "format": "f32-to-bf16",
1717
+ "nbytes": 8388608,
1718
+ "byteOffset": 5902336
1719
+ },
1720
+ {
1721
+ "name": "model.layers.8.mlp.down_proj.q_scale",
1722
+ "shape": [
1723
+ 2048,
1724
+ 256
1725
+ ],
1726
+ "dtype": "float16",
1727
+ "format": "f32-to-bf16",
1728
+ "nbytes": 1048576,
1729
+ "byteOffset": 14290944
1730
+ },
1731
+ {
1732
+ "name": "model.layers.8.mlp.gate_up_proj.q_weight",
1733
+ "shape": [
1734
+ 16384,
1735
+ 256
1736
+ ],
1737
+ "dtype": "uint32",
1738
+ "format": "f32-to-bf16",
1739
+ "nbytes": 16777216,
1740
+ "byteOffset": 15339520
1741
+ }
1742
+ ],
1743
+ "md5sum": "a2a7115d600451c223e73542f33715c2"
1744
+ },
1745
+ {
1746
+ "dataPath": "params_shard_20.bin",
1747
+ "format": "raw-shard",
1748
+ "nbytes": 16777216,
1749
+ "records": [
1750
+ {
1751
+ "name": "model.layers.9.mlp.gate_up_proj.q_weight",
1752
+ "shape": [
1753
+ 16384,
1754
+ 256
1755
+ ],
1756
+ "dtype": "uint32",
1757
+ "format": "f32-to-bf16",
1758
+ "nbytes": 16777216,
1759
+ "byteOffset": 0
1760
+ }
1761
+ ],
1762
+ "md5sum": "e4a1b907f3c17eb89b314cd1d09f27cd"
1763
+ },
1764
+ {
1765
+ "dataPath": "params_shard_21.bin",
1766
+ "format": "raw-shard",
1767
+ "nbytes": 25444352,
1768
+ "records": [
1769
+ {
1770
+ "name": "model.layers.8.mlp.gate_up_proj.q_scale",
1771
+ "shape": [
1772
+ 16384,
1773
+ 64
1774
+ ],
1775
+ "dtype": "float16",
1776
+ "format": "f32-to-bf16",
1777
+ "nbytes": 2097152,
1778
+ "byteOffset": 0
1779
+ },
1780
+ {
1781
+ "name": "model.layers.8.post_attention_layernorm.weight",
1782
+ "shape": [
1783
+ 2048
1784
+ ],
1785
+ "dtype": "float16",
1786
+ "format": "f32-to-bf16",
1787
+ "nbytes": 4096,
1788
+ "byteOffset": 2097152
1789
+ },
1790
+ {
1791
+ "name": "model.layers.8.self_attn.qkv_proj.q_weight",
1792
+ "shape": [
1793
+ 3072,
1794
+ 256
1795
+ ],
1796
+ "dtype": "uint32",
1797
+ "format": "f32-to-bf16",
1798
+ "nbytes": 3145728,
1799
+ "byteOffset": 2101248
1800
+ },
1801
+ {
1802
+ "name": "model.layers.8.self_attn.qkv_proj.q_scale",
1803
+ "shape": [
1804
+ 3072,
1805
+ 64
1806
+ ],
1807
+ "dtype": "float16",
1808
+ "format": "f32-to-bf16",
1809
+ "nbytes": 393216,
1810
+ "byteOffset": 5246976
1811
+ },
1812
+ {
1813
+ "name": "model.layers.8.self_attn.o_proj.q_weight",
1814
+ "shape": [
1815
+ 2048,
1816
+ 256
1817
+ ],
1818
+ "dtype": "uint32",
1819
+ "format": "f32-to-bf16",
1820
+ "nbytes": 2097152,
1821
+ "byteOffset": 5640192
1822
+ },
1823
+ {
1824
+ "name": "model.layers.8.self_attn.o_proj.q_scale",
1825
+ "shape": [
1826
+ 2048,
1827
+ 64
1828
+ ],
1829
+ "dtype": "float16",
1830
+ "format": "f32-to-bf16",
1831
+ "nbytes": 262144,
1832
+ "byteOffset": 7737344
1833
+ },
1834
+ {
1835
+ "name": "model.layers.9.input_layernorm.weight",
1836
+ "shape": [
1837
+ 2048
1838
+ ],
1839
+ "dtype": "float16",
1840
+ "format": "f32-to-bf16",
1841
+ "nbytes": 4096,
1842
+ "byteOffset": 7999488
1843
+ },
1844
+ {
1845
+ "name": "model.layers.9.mlp.down_proj.q_weight",
1846
+ "shape": [
1847
+ 2048,
1848
+ 1024
1849
+ ],
1850
+ "dtype": "uint32",
1851
+ "format": "f32-to-bf16",
1852
+ "nbytes": 8388608,
1853
+ "byteOffset": 8003584
1854
+ },
1855
+ {
1856
+ "name": "model.layers.9.mlp.down_proj.q_scale",
1857
+ "shape": [
1858
+ 2048,
1859
+ 256
1860
+ ],
1861
+ "dtype": "float16",
1862
+ "format": "f32-to-bf16",
1863
+ "nbytes": 1048576,
1864
+ "byteOffset": 16392192
1865
+ },
1866
+ {
1867
+ "name": "model.layers.9.mlp.gate_up_proj.q_scale",
1868
+ "shape": [
1869
+ 16384,
1870
+ 64
1871
+ ],
1872
+ "dtype": "float16",
1873
+ "format": "f32-to-bf16",
1874
+ "nbytes": 2097152,
1875
+ "byteOffset": 17440768
1876
+ },
1877
+ {
1878
+ "name": "model.layers.9.post_attention_layernorm.weight",
1879
+ "shape": [
1880
+ 2048
1881
+ ],
1882
+ "dtype": "float16",
1883
+ "format": "f32-to-bf16",
1884
+ "nbytes": 4096,
1885
+ "byteOffset": 19537920
1886
+ },
1887
+ {
1888
+ "name": "model.layers.9.self_attn.qkv_proj.q_weight",
1889
+ "shape": [
1890
+ 3072,
1891
+ 256
1892
+ ],
1893
+ "dtype": "uint32",
1894
+ "format": "f32-to-bf16",
1895
+ "nbytes": 3145728,
1896
+ "byteOffset": 19542016
1897
+ },
1898
+ {
1899
+ "name": "model.layers.9.self_attn.qkv_proj.q_scale",
1900
+ "shape": [
1901
+ 3072,
1902
+ 64
1903
+ ],
1904
+ "dtype": "float16",
1905
+ "format": "f32-to-bf16",
1906
+ "nbytes": 393216,
1907
+ "byteOffset": 22687744
1908
+ },
1909
+ {
1910
+ "name": "model.layers.9.self_attn.o_proj.q_weight",
1911
+ "shape": [
1912
+ 2048,
1913
+ 256
1914
+ ],
1915
+ "dtype": "uint32",
1916
+ "format": "f32-to-bf16",
1917
+ "nbytes": 2097152,
1918
+ "byteOffset": 23080960
1919
+ },
1920
+ {
1921
+ "name": "model.layers.9.self_attn.o_proj.q_scale",
1922
+ "shape": [
1923
+ 2048,
1924
+ 64
1925
+ ],
1926
+ "dtype": "float16",
1927
+ "format": "f32-to-bf16",
1928
+ "nbytes": 262144,
1929
+ "byteOffset": 25178112
1930
+ },
1931
+ {
1932
+ "name": "model.norm.weight",
1933
+ "shape": [
1934
+ 2048
1935
+ ],
1936
+ "dtype": "float16",
1937
+ "format": "f32-to-bf16",
1938
+ "nbytes": 4096,
1939
+ "byteOffset": 25440256
1940
+ }
1941
+ ],
1942
+ "md5sum": "cb6d8c6f94323219bafcc299b20bf8a8"
1943
+ }
1944
+ ]
1945
+ }
params_shard_0.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:17f0334f88adb9a6a23cc702b1fd895e6a159d164eae2520358c8c1d3fccc9ed
3
+ size 131334144
params_shard_1.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:76b9e11e8fdb82d6d3edbba334bbf59f6fd080e8138baf6bca956af9c886c956
3
+ size 16777216
params_shard_10.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bf02d53bd89d7c7f46b2f64d9d851cdf1b4a4dd1b873369e633e97056307f8d5
3
+ size 16777216
params_shard_11.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:578a5f48dac3bc367c8ad2b3da30148686366f1b5d626de2e4083a115a985e67
3
+ size 25444352
params_shard_12.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1237ea88c9640ad3498cb59222c2a3a1d5ec47fa91a52c409f1ad9b53a4ada59
3
+ size 31854592
params_shard_13.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:42e95969cf75ffce16661b77704cdb556dcc499070b026ce8f350235348e5584
3
+ size 30679040
params_shard_14.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d7c96577cee90e7254a18af800480c249e0860b3b887e0497647032ea4eb1c59
3
+ size 32116736
params_shard_15.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:88cf9a9d4a903b765e17528be0462be1abd6b328de02b830b8d37e32deb55fbc
3
+ size 16777216
params_shard_16.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aac1472bd13261a475c06e3aaa5fc87dd78b5a2c97a2cdac7ace244158e5035e
3
+ size 25444352
params_shard_17.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e8b201b70ba99337c84fe9611e18f1e48564a273168cf5cdb6fcb173dfec749c
3
+ size 31854592
params_shard_18.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ed30fdf696b521b2ed18d91933d965f9aaecf23601d1f5b0a215598cc2bfae69
3
+ size 30679040
params_shard_19.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d93f253fdf638757eb0003d7d9b0f75ae7c9c4e36323c5df5f486dcde9790249
3
+ size 32116736
params_shard_2.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:203ba6b3283b604a8200a1f06051f85c5a9003a606ba89c2f5d47ce9d08ee2fa
3
+ size 31498240
params_shard_20.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2bcd8a5d7a3841cc9ded6b3d46fe63adb845dbabf83ee08164b8676161e8cd6a
3
+ size 16777216
params_shard_21.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2b0072e9cbdc7698cf2eefe9355e2281b1a817f58894f1b3b2f15e103de799c3
3
+ size 25444352
params_shard_3.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e1bc66ba5102f6971c0ff430234405a3a5831f13ff88301f02094fe50cb0764a
3
+ size 30679040
params_shard_4.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:19d1b44307b23060f5f3acee5a0b5abf22fcfea4a3d2e8f4469d00aeec5a4ce3
3
+ size 32116736
params_shard_5.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aa2a7f2ec3a7f3ac0c42e698bec4a3ce30e8ebfabe1f9d7f4bede1ddbb92e1e8
3
+ size 16777216
params_shard_6.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2a2d03e4a918dbd3ac1899fb155147013959edbd5cf233947d6900936fc7a419
3
+ size 25444352
params_shard_7.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0918287f31f10dd760ed405ca10571dce9fcb621f42978cd3b5c0d2dce6b4304
3
+ size 31854592
params_shard_8.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a34bdd87a4d0ccb5baec1e5bfbd66bb407d04bb3d8d966d53c8bff9ebcae5093
3
+ size 30679040
params_shard_9.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a9b2a284c794fc23919a74210ec5cda44f9068126e37ad33c7547c812954cd01
3
+ size 32116736