cfahlgren1 HF staff commited on
Commit
c62648b
1 Parent(s): 3486077
ndarray-cache.json ADDED
@@ -0,0 +1,2262 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "ParamSize": 194,
4
+ "ParamBytes": 723642240.0,
5
+ "BitsPerParam": 16.0
6
+ },
7
+ "records": [
8
+ {
9
+ "dataPath": "params_shard_0.bin",
10
+ "format": "raw-shard",
11
+ "nbytes": 94371840,
12
+ "records": [
13
+ {
14
+ "name": "model.embed_tokens.weight",
15
+ "shape": [
16
+ 49152,
17
+ 960
18
+ ],
19
+ "dtype": "float16",
20
+ "format": "f32-to-bf16",
21
+ "nbytes": 94371840,
22
+ "byteOffset": 0
23
+ }
24
+ ],
25
+ "md5sum": "f448befd1fb7cad85ced1b59dcebba38"
26
+ },
27
+ {
28
+ "dataPath": "params_shard_1.bin",
29
+ "format": "raw-shard",
30
+ "nbytes": 24581760,
31
+ "records": [
32
+ {
33
+ "name": "model.layers.0.input_layernorm.weight",
34
+ "shape": [
35
+ 960
36
+ ],
37
+ "dtype": "float16",
38
+ "format": "f32-to-bf16",
39
+ "nbytes": 1920,
40
+ "byteOffset": 0
41
+ },
42
+ {
43
+ "name": "model.layers.0.mlp.down_proj.weight",
44
+ "shape": [
45
+ 960,
46
+ 2560
47
+ ],
48
+ "dtype": "float16",
49
+ "format": "f32-to-bf16",
50
+ "nbytes": 4915200,
51
+ "byteOffset": 1920
52
+ },
53
+ {
54
+ "name": "model.layers.0.mlp.gate_up_proj.weight",
55
+ "shape": [
56
+ 5120,
57
+ 960
58
+ ],
59
+ "dtype": "float16",
60
+ "format": "f32-to-bf16",
61
+ "nbytes": 9830400,
62
+ "byteOffset": 4917120
63
+ },
64
+ {
65
+ "name": "model.layers.0.post_attention_layernorm.weight",
66
+ "shape": [
67
+ 960
68
+ ],
69
+ "dtype": "float16",
70
+ "format": "f32-to-bf16",
71
+ "nbytes": 1920,
72
+ "byteOffset": 14747520
73
+ },
74
+ {
75
+ "name": "model.layers.0.self_attn.qkv_proj.weight",
76
+ "shape": [
77
+ 1600,
78
+ 960
79
+ ],
80
+ "dtype": "float16",
81
+ "format": "f32-to-bf16",
82
+ "nbytes": 3072000,
83
+ "byteOffset": 14749440
84
+ },
85
+ {
86
+ "name": "model.layers.0.self_attn.o_proj.weight",
87
+ "shape": [
88
+ 960,
89
+ 960
90
+ ],
91
+ "dtype": "float16",
92
+ "format": "f32-to-bf16",
93
+ "nbytes": 1843200,
94
+ "byteOffset": 17821440
95
+ },
96
+ {
97
+ "name": "model.layers.1.input_layernorm.weight",
98
+ "shape": [
99
+ 960
100
+ ],
101
+ "dtype": "float16",
102
+ "format": "f32-to-bf16",
103
+ "nbytes": 1920,
104
+ "byteOffset": 19664640
105
+ },
106
+ {
107
+ "name": "model.layers.1.mlp.down_proj.weight",
108
+ "shape": [
109
+ 960,
110
+ 2560
111
+ ],
112
+ "dtype": "float16",
113
+ "format": "f32-to-bf16",
114
+ "nbytes": 4915200,
115
+ "byteOffset": 19666560
116
+ }
117
+ ],
118
+ "md5sum": "2b2e1873a01963818d08ae9737b3ae92"
119
+ },
120
+ {
121
+ "dataPath": "params_shard_2.bin",
122
+ "format": "raw-shard",
123
+ "nbytes": 32568960,
124
+ "records": [
125
+ {
126
+ "name": "model.layers.1.mlp.gate_up_proj.weight",
127
+ "shape": [
128
+ 5120,
129
+ 960
130
+ ],
131
+ "dtype": "float16",
132
+ "format": "f32-to-bf16",
133
+ "nbytes": 9830400,
134
+ "byteOffset": 0
135
+ },
136
+ {
137
+ "name": "model.layers.1.post_attention_layernorm.weight",
138
+ "shape": [
139
+ 960
140
+ ],
141
+ "dtype": "float16",
142
+ "format": "f32-to-bf16",
143
+ "nbytes": 1920,
144
+ "byteOffset": 9830400
145
+ },
146
+ {
147
+ "name": "model.layers.1.self_attn.qkv_proj.weight",
148
+ "shape": [
149
+ 1600,
150
+ 960
151
+ ],
152
+ "dtype": "float16",
153
+ "format": "f32-to-bf16",
154
+ "nbytes": 3072000,
155
+ "byteOffset": 9832320
156
+ },
157
+ {
158
+ "name": "model.layers.1.self_attn.o_proj.weight",
159
+ "shape": [
160
+ 960,
161
+ 960
162
+ ],
163
+ "dtype": "float16",
164
+ "format": "f32-to-bf16",
165
+ "nbytes": 1843200,
166
+ "byteOffset": 12904320
167
+ },
168
+ {
169
+ "name": "model.layers.10.input_layernorm.weight",
170
+ "shape": [
171
+ 960
172
+ ],
173
+ "dtype": "float16",
174
+ "format": "f32-to-bf16",
175
+ "nbytes": 1920,
176
+ "byteOffset": 14747520
177
+ },
178
+ {
179
+ "name": "model.layers.10.mlp.down_proj.weight",
180
+ "shape": [
181
+ 960,
182
+ 2560
183
+ ],
184
+ "dtype": "float16",
185
+ "format": "f32-to-bf16",
186
+ "nbytes": 4915200,
187
+ "byteOffset": 14749440
188
+ },
189
+ {
190
+ "name": "model.layers.10.mlp.gate_up_proj.weight",
191
+ "shape": [
192
+ 5120,
193
+ 960
194
+ ],
195
+ "dtype": "float16",
196
+ "format": "f32-to-bf16",
197
+ "nbytes": 9830400,
198
+ "byteOffset": 19664640
199
+ },
200
+ {
201
+ "name": "model.layers.10.post_attention_layernorm.weight",
202
+ "shape": [
203
+ 960
204
+ ],
205
+ "dtype": "float16",
206
+ "format": "f32-to-bf16",
207
+ "nbytes": 1920,
208
+ "byteOffset": 29495040
209
+ },
210
+ {
211
+ "name": "model.layers.10.self_attn.qkv_proj.weight",
212
+ "shape": [
213
+ 1600,
214
+ 960
215
+ ],
216
+ "dtype": "float16",
217
+ "format": "f32-to-bf16",
218
+ "nbytes": 3072000,
219
+ "byteOffset": 29496960
220
+ }
221
+ ],
222
+ "md5sum": "f3e026bf7dc600085dd3ed1e517dec2e"
223
+ },
224
+ {
225
+ "dataPath": "params_shard_3.bin",
226
+ "format": "raw-shard",
227
+ "nbytes": 26424960,
228
+ "records": [
229
+ {
230
+ "name": "model.layers.10.self_attn.o_proj.weight",
231
+ "shape": [
232
+ 960,
233
+ 960
234
+ ],
235
+ "dtype": "float16",
236
+ "format": "f32-to-bf16",
237
+ "nbytes": 1843200,
238
+ "byteOffset": 0
239
+ },
240
+ {
241
+ "name": "model.layers.11.input_layernorm.weight",
242
+ "shape": [
243
+ 960
244
+ ],
245
+ "dtype": "float16",
246
+ "format": "f32-to-bf16",
247
+ "nbytes": 1920,
248
+ "byteOffset": 1843200
249
+ },
250
+ {
251
+ "name": "model.layers.11.mlp.down_proj.weight",
252
+ "shape": [
253
+ 960,
254
+ 2560
255
+ ],
256
+ "dtype": "float16",
257
+ "format": "f32-to-bf16",
258
+ "nbytes": 4915200,
259
+ "byteOffset": 1845120
260
+ },
261
+ {
262
+ "name": "model.layers.11.mlp.gate_up_proj.weight",
263
+ "shape": [
264
+ 5120,
265
+ 960
266
+ ],
267
+ "dtype": "float16",
268
+ "format": "f32-to-bf16",
269
+ "nbytes": 9830400,
270
+ "byteOffset": 6760320
271
+ },
272
+ {
273
+ "name": "model.layers.11.post_attention_layernorm.weight",
274
+ "shape": [
275
+ 960
276
+ ],
277
+ "dtype": "float16",
278
+ "format": "f32-to-bf16",
279
+ "nbytes": 1920,
280
+ "byteOffset": 16590720
281
+ },
282
+ {
283
+ "name": "model.layers.11.self_attn.qkv_proj.weight",
284
+ "shape": [
285
+ 1600,
286
+ 960
287
+ ],
288
+ "dtype": "float16",
289
+ "format": "f32-to-bf16",
290
+ "nbytes": 3072000,
291
+ "byteOffset": 16592640
292
+ },
293
+ {
294
+ "name": "model.layers.11.self_attn.o_proj.weight",
295
+ "shape": [
296
+ 960,
297
+ 960
298
+ ],
299
+ "dtype": "float16",
300
+ "format": "f32-to-bf16",
301
+ "nbytes": 1843200,
302
+ "byteOffset": 19664640
303
+ },
304
+ {
305
+ "name": "model.layers.12.input_layernorm.weight",
306
+ "shape": [
307
+ 960
308
+ ],
309
+ "dtype": "float16",
310
+ "format": "f32-to-bf16",
311
+ "nbytes": 1920,
312
+ "byteOffset": 21507840
313
+ },
314
+ {
315
+ "name": "model.layers.12.mlp.down_proj.weight",
316
+ "shape": [
317
+ 960,
318
+ 2560
319
+ ],
320
+ "dtype": "float16",
321
+ "format": "f32-to-bf16",
322
+ "nbytes": 4915200,
323
+ "byteOffset": 21509760
324
+ }
325
+ ],
326
+ "md5sum": "b436474cccb700cf6035d8ee6de7fd40"
327
+ },
328
+ {
329
+ "dataPath": "params_shard_4.bin",
330
+ "format": "raw-shard",
331
+ "nbytes": 32568960,
332
+ "records": [
333
+ {
334
+ "name": "model.layers.12.mlp.gate_up_proj.weight",
335
+ "shape": [
336
+ 5120,
337
+ 960
338
+ ],
339
+ "dtype": "float16",
340
+ "format": "f32-to-bf16",
341
+ "nbytes": 9830400,
342
+ "byteOffset": 0
343
+ },
344
+ {
345
+ "name": "model.layers.12.post_attention_layernorm.weight",
346
+ "shape": [
347
+ 960
348
+ ],
349
+ "dtype": "float16",
350
+ "format": "f32-to-bf16",
351
+ "nbytes": 1920,
352
+ "byteOffset": 9830400
353
+ },
354
+ {
355
+ "name": "model.layers.12.self_attn.qkv_proj.weight",
356
+ "shape": [
357
+ 1600,
358
+ 960
359
+ ],
360
+ "dtype": "float16",
361
+ "format": "f32-to-bf16",
362
+ "nbytes": 3072000,
363
+ "byteOffset": 9832320
364
+ },
365
+ {
366
+ "name": "model.layers.12.self_attn.o_proj.weight",
367
+ "shape": [
368
+ 960,
369
+ 960
370
+ ],
371
+ "dtype": "float16",
372
+ "format": "f32-to-bf16",
373
+ "nbytes": 1843200,
374
+ "byteOffset": 12904320
375
+ },
376
+ {
377
+ "name": "model.layers.13.input_layernorm.weight",
378
+ "shape": [
379
+ 960
380
+ ],
381
+ "dtype": "float16",
382
+ "format": "f32-to-bf16",
383
+ "nbytes": 1920,
384
+ "byteOffset": 14747520
385
+ },
386
+ {
387
+ "name": "model.layers.13.mlp.down_proj.weight",
388
+ "shape": [
389
+ 960,
390
+ 2560
391
+ ],
392
+ "dtype": "float16",
393
+ "format": "f32-to-bf16",
394
+ "nbytes": 4915200,
395
+ "byteOffset": 14749440
396
+ },
397
+ {
398
+ "name": "model.layers.13.mlp.gate_up_proj.weight",
399
+ "shape": [
400
+ 5120,
401
+ 960
402
+ ],
403
+ "dtype": "float16",
404
+ "format": "f32-to-bf16",
405
+ "nbytes": 9830400,
406
+ "byteOffset": 19664640
407
+ },
408
+ {
409
+ "name": "model.layers.13.post_attention_layernorm.weight",
410
+ "shape": [
411
+ 960
412
+ ],
413
+ "dtype": "float16",
414
+ "format": "f32-to-bf16",
415
+ "nbytes": 1920,
416
+ "byteOffset": 29495040
417
+ },
418
+ {
419
+ "name": "model.layers.13.self_attn.qkv_proj.weight",
420
+ "shape": [
421
+ 1600,
422
+ 960
423
+ ],
424
+ "dtype": "float16",
425
+ "format": "f32-to-bf16",
426
+ "nbytes": 3072000,
427
+ "byteOffset": 29496960
428
+ }
429
+ ],
430
+ "md5sum": "7e2a8c918c5a37e4f7eb6f30c22e158f"
431
+ },
432
+ {
433
+ "dataPath": "params_shard_5.bin",
434
+ "format": "raw-shard",
435
+ "nbytes": 26424960,
436
+ "records": [
437
+ {
438
+ "name": "model.layers.13.self_attn.o_proj.weight",
439
+ "shape": [
440
+ 960,
441
+ 960
442
+ ],
443
+ "dtype": "float16",
444
+ "format": "f32-to-bf16",
445
+ "nbytes": 1843200,
446
+ "byteOffset": 0
447
+ },
448
+ {
449
+ "name": "model.layers.14.input_layernorm.weight",
450
+ "shape": [
451
+ 960
452
+ ],
453
+ "dtype": "float16",
454
+ "format": "f32-to-bf16",
455
+ "nbytes": 1920,
456
+ "byteOffset": 1843200
457
+ },
458
+ {
459
+ "name": "model.layers.14.mlp.down_proj.weight",
460
+ "shape": [
461
+ 960,
462
+ 2560
463
+ ],
464
+ "dtype": "float16",
465
+ "format": "f32-to-bf16",
466
+ "nbytes": 4915200,
467
+ "byteOffset": 1845120
468
+ },
469
+ {
470
+ "name": "model.layers.14.mlp.gate_up_proj.weight",
471
+ "shape": [
472
+ 5120,
473
+ 960
474
+ ],
475
+ "dtype": "float16",
476
+ "format": "f32-to-bf16",
477
+ "nbytes": 9830400,
478
+ "byteOffset": 6760320
479
+ },
480
+ {
481
+ "name": "model.layers.14.post_attention_layernorm.weight",
482
+ "shape": [
483
+ 960
484
+ ],
485
+ "dtype": "float16",
486
+ "format": "f32-to-bf16",
487
+ "nbytes": 1920,
488
+ "byteOffset": 16590720
489
+ },
490
+ {
491
+ "name": "model.layers.14.self_attn.qkv_proj.weight",
492
+ "shape": [
493
+ 1600,
494
+ 960
495
+ ],
496
+ "dtype": "float16",
497
+ "format": "f32-to-bf16",
498
+ "nbytes": 3072000,
499
+ "byteOffset": 16592640
500
+ },
501
+ {
502
+ "name": "model.layers.14.self_attn.o_proj.weight",
503
+ "shape": [
504
+ 960,
505
+ 960
506
+ ],
507
+ "dtype": "float16",
508
+ "format": "f32-to-bf16",
509
+ "nbytes": 1843200,
510
+ "byteOffset": 19664640
511
+ },
512
+ {
513
+ "name": "model.layers.15.input_layernorm.weight",
514
+ "shape": [
515
+ 960
516
+ ],
517
+ "dtype": "float16",
518
+ "format": "f32-to-bf16",
519
+ "nbytes": 1920,
520
+ "byteOffset": 21507840
521
+ },
522
+ {
523
+ "name": "model.layers.15.mlp.down_proj.weight",
524
+ "shape": [
525
+ 960,
526
+ 2560
527
+ ],
528
+ "dtype": "float16",
529
+ "format": "f32-to-bf16",
530
+ "nbytes": 4915200,
531
+ "byteOffset": 21509760
532
+ }
533
+ ],
534
+ "md5sum": "fe556a3c8ee15c7bb5961868ff03f079"
535
+ },
536
+ {
537
+ "dataPath": "params_shard_6.bin",
538
+ "format": "raw-shard",
539
+ "nbytes": 32568960,
540
+ "records": [
541
+ {
542
+ "name": "model.layers.15.mlp.gate_up_proj.weight",
543
+ "shape": [
544
+ 5120,
545
+ 960
546
+ ],
547
+ "dtype": "float16",
548
+ "format": "f32-to-bf16",
549
+ "nbytes": 9830400,
550
+ "byteOffset": 0
551
+ },
552
+ {
553
+ "name": "model.layers.15.post_attention_layernorm.weight",
554
+ "shape": [
555
+ 960
556
+ ],
557
+ "dtype": "float16",
558
+ "format": "f32-to-bf16",
559
+ "nbytes": 1920,
560
+ "byteOffset": 9830400
561
+ },
562
+ {
563
+ "name": "model.layers.15.self_attn.qkv_proj.weight",
564
+ "shape": [
565
+ 1600,
566
+ 960
567
+ ],
568
+ "dtype": "float16",
569
+ "format": "f32-to-bf16",
570
+ "nbytes": 3072000,
571
+ "byteOffset": 9832320
572
+ },
573
+ {
574
+ "name": "model.layers.15.self_attn.o_proj.weight",
575
+ "shape": [
576
+ 960,
577
+ 960
578
+ ],
579
+ "dtype": "float16",
580
+ "format": "f32-to-bf16",
581
+ "nbytes": 1843200,
582
+ "byteOffset": 12904320
583
+ },
584
+ {
585
+ "name": "model.layers.16.input_layernorm.weight",
586
+ "shape": [
587
+ 960
588
+ ],
589
+ "dtype": "float16",
590
+ "format": "f32-to-bf16",
591
+ "nbytes": 1920,
592
+ "byteOffset": 14747520
593
+ },
594
+ {
595
+ "name": "model.layers.16.mlp.down_proj.weight",
596
+ "shape": [
597
+ 960,
598
+ 2560
599
+ ],
600
+ "dtype": "float16",
601
+ "format": "f32-to-bf16",
602
+ "nbytes": 4915200,
603
+ "byteOffset": 14749440
604
+ },
605
+ {
606
+ "name": "model.layers.16.mlp.gate_up_proj.weight",
607
+ "shape": [
608
+ 5120,
609
+ 960
610
+ ],
611
+ "dtype": "float16",
612
+ "format": "f32-to-bf16",
613
+ "nbytes": 9830400,
614
+ "byteOffset": 19664640
615
+ },
616
+ {
617
+ "name": "model.layers.16.post_attention_layernorm.weight",
618
+ "shape": [
619
+ 960
620
+ ],
621
+ "dtype": "float16",
622
+ "format": "f32-to-bf16",
623
+ "nbytes": 1920,
624
+ "byteOffset": 29495040
625
+ },
626
+ {
627
+ "name": "model.layers.16.self_attn.qkv_proj.weight",
628
+ "shape": [
629
+ 1600,
630
+ 960
631
+ ],
632
+ "dtype": "float16",
633
+ "format": "f32-to-bf16",
634
+ "nbytes": 3072000,
635
+ "byteOffset": 29496960
636
+ }
637
+ ],
638
+ "md5sum": "773636b5b252d5bf6671548a37b510f4"
639
+ },
640
+ {
641
+ "dataPath": "params_shard_7.bin",
642
+ "format": "raw-shard",
643
+ "nbytes": 26424960,
644
+ "records": [
645
+ {
646
+ "name": "model.layers.16.self_attn.o_proj.weight",
647
+ "shape": [
648
+ 960,
649
+ 960
650
+ ],
651
+ "dtype": "float16",
652
+ "format": "f32-to-bf16",
653
+ "nbytes": 1843200,
654
+ "byteOffset": 0
655
+ },
656
+ {
657
+ "name": "model.layers.17.input_layernorm.weight",
658
+ "shape": [
659
+ 960
660
+ ],
661
+ "dtype": "float16",
662
+ "format": "f32-to-bf16",
663
+ "nbytes": 1920,
664
+ "byteOffset": 1843200
665
+ },
666
+ {
667
+ "name": "model.layers.17.mlp.down_proj.weight",
668
+ "shape": [
669
+ 960,
670
+ 2560
671
+ ],
672
+ "dtype": "float16",
673
+ "format": "f32-to-bf16",
674
+ "nbytes": 4915200,
675
+ "byteOffset": 1845120
676
+ },
677
+ {
678
+ "name": "model.layers.17.mlp.gate_up_proj.weight",
679
+ "shape": [
680
+ 5120,
681
+ 960
682
+ ],
683
+ "dtype": "float16",
684
+ "format": "f32-to-bf16",
685
+ "nbytes": 9830400,
686
+ "byteOffset": 6760320
687
+ },
688
+ {
689
+ "name": "model.layers.17.post_attention_layernorm.weight",
690
+ "shape": [
691
+ 960
692
+ ],
693
+ "dtype": "float16",
694
+ "format": "f32-to-bf16",
695
+ "nbytes": 1920,
696
+ "byteOffset": 16590720
697
+ },
698
+ {
699
+ "name": "model.layers.17.self_attn.qkv_proj.weight",
700
+ "shape": [
701
+ 1600,
702
+ 960
703
+ ],
704
+ "dtype": "float16",
705
+ "format": "f32-to-bf16",
706
+ "nbytes": 3072000,
707
+ "byteOffset": 16592640
708
+ },
709
+ {
710
+ "name": "model.layers.17.self_attn.o_proj.weight",
711
+ "shape": [
712
+ 960,
713
+ 960
714
+ ],
715
+ "dtype": "float16",
716
+ "format": "f32-to-bf16",
717
+ "nbytes": 1843200,
718
+ "byteOffset": 19664640
719
+ },
720
+ {
721
+ "name": "model.layers.18.input_layernorm.weight",
722
+ "shape": [
723
+ 960
724
+ ],
725
+ "dtype": "float16",
726
+ "format": "f32-to-bf16",
727
+ "nbytes": 1920,
728
+ "byteOffset": 21507840
729
+ },
730
+ {
731
+ "name": "model.layers.18.mlp.down_proj.weight",
732
+ "shape": [
733
+ 960,
734
+ 2560
735
+ ],
736
+ "dtype": "float16",
737
+ "format": "f32-to-bf16",
738
+ "nbytes": 4915200,
739
+ "byteOffset": 21509760
740
+ }
741
+ ],
742
+ "md5sum": "5ce6123e4827dace6867b4b984d0fed2"
743
+ },
744
+ {
745
+ "dataPath": "params_shard_8.bin",
746
+ "format": "raw-shard",
747
+ "nbytes": 32568960,
748
+ "records": [
749
+ {
750
+ "name": "model.layers.18.mlp.gate_up_proj.weight",
751
+ "shape": [
752
+ 5120,
753
+ 960
754
+ ],
755
+ "dtype": "float16",
756
+ "format": "f32-to-bf16",
757
+ "nbytes": 9830400,
758
+ "byteOffset": 0
759
+ },
760
+ {
761
+ "name": "model.layers.18.post_attention_layernorm.weight",
762
+ "shape": [
763
+ 960
764
+ ],
765
+ "dtype": "float16",
766
+ "format": "f32-to-bf16",
767
+ "nbytes": 1920,
768
+ "byteOffset": 9830400
769
+ },
770
+ {
771
+ "name": "model.layers.18.self_attn.qkv_proj.weight",
772
+ "shape": [
773
+ 1600,
774
+ 960
775
+ ],
776
+ "dtype": "float16",
777
+ "format": "f32-to-bf16",
778
+ "nbytes": 3072000,
779
+ "byteOffset": 9832320
780
+ },
781
+ {
782
+ "name": "model.layers.18.self_attn.o_proj.weight",
783
+ "shape": [
784
+ 960,
785
+ 960
786
+ ],
787
+ "dtype": "float16",
788
+ "format": "f32-to-bf16",
789
+ "nbytes": 1843200,
790
+ "byteOffset": 12904320
791
+ },
792
+ {
793
+ "name": "model.layers.19.input_layernorm.weight",
794
+ "shape": [
795
+ 960
796
+ ],
797
+ "dtype": "float16",
798
+ "format": "f32-to-bf16",
799
+ "nbytes": 1920,
800
+ "byteOffset": 14747520
801
+ },
802
+ {
803
+ "name": "model.layers.19.mlp.down_proj.weight",
804
+ "shape": [
805
+ 960,
806
+ 2560
807
+ ],
808
+ "dtype": "float16",
809
+ "format": "f32-to-bf16",
810
+ "nbytes": 4915200,
811
+ "byteOffset": 14749440
812
+ },
813
+ {
814
+ "name": "model.layers.19.mlp.gate_up_proj.weight",
815
+ "shape": [
816
+ 5120,
817
+ 960
818
+ ],
819
+ "dtype": "float16",
820
+ "format": "f32-to-bf16",
821
+ "nbytes": 9830400,
822
+ "byteOffset": 19664640
823
+ },
824
+ {
825
+ "name": "model.layers.19.post_attention_layernorm.weight",
826
+ "shape": [
827
+ 960
828
+ ],
829
+ "dtype": "float16",
830
+ "format": "f32-to-bf16",
831
+ "nbytes": 1920,
832
+ "byteOffset": 29495040
833
+ },
834
+ {
835
+ "name": "model.layers.19.self_attn.qkv_proj.weight",
836
+ "shape": [
837
+ 1600,
838
+ 960
839
+ ],
840
+ "dtype": "float16",
841
+ "format": "f32-to-bf16",
842
+ "nbytes": 3072000,
843
+ "byteOffset": 29496960
844
+ }
845
+ ],
846
+ "md5sum": "957c1a4d9608ba124906199b18b46fde"
847
+ },
848
+ {
849
+ "dataPath": "params_shard_9.bin",
850
+ "format": "raw-shard",
851
+ "nbytes": 26424960,
852
+ "records": [
853
+ {
854
+ "name": "model.layers.19.self_attn.o_proj.weight",
855
+ "shape": [
856
+ 960,
857
+ 960
858
+ ],
859
+ "dtype": "float16",
860
+ "format": "f32-to-bf16",
861
+ "nbytes": 1843200,
862
+ "byteOffset": 0
863
+ },
864
+ {
865
+ "name": "model.layers.2.input_layernorm.weight",
866
+ "shape": [
867
+ 960
868
+ ],
869
+ "dtype": "float16",
870
+ "format": "f32-to-bf16",
871
+ "nbytes": 1920,
872
+ "byteOffset": 1843200
873
+ },
874
+ {
875
+ "name": "model.layers.2.mlp.down_proj.weight",
876
+ "shape": [
877
+ 960,
878
+ 2560
879
+ ],
880
+ "dtype": "float16",
881
+ "format": "f32-to-bf16",
882
+ "nbytes": 4915200,
883
+ "byteOffset": 1845120
884
+ },
885
+ {
886
+ "name": "model.layers.2.mlp.gate_up_proj.weight",
887
+ "shape": [
888
+ 5120,
889
+ 960
890
+ ],
891
+ "dtype": "float16",
892
+ "format": "f32-to-bf16",
893
+ "nbytes": 9830400,
894
+ "byteOffset": 6760320
895
+ },
896
+ {
897
+ "name": "model.layers.2.post_attention_layernorm.weight",
898
+ "shape": [
899
+ 960
900
+ ],
901
+ "dtype": "float16",
902
+ "format": "f32-to-bf16",
903
+ "nbytes": 1920,
904
+ "byteOffset": 16590720
905
+ },
906
+ {
907
+ "name": "model.layers.2.self_attn.qkv_proj.weight",
908
+ "shape": [
909
+ 1600,
910
+ 960
911
+ ],
912
+ "dtype": "float16",
913
+ "format": "f32-to-bf16",
914
+ "nbytes": 3072000,
915
+ "byteOffset": 16592640
916
+ },
917
+ {
918
+ "name": "model.layers.2.self_attn.o_proj.weight",
919
+ "shape": [
920
+ 960,
921
+ 960
922
+ ],
923
+ "dtype": "float16",
924
+ "format": "f32-to-bf16",
925
+ "nbytes": 1843200,
926
+ "byteOffset": 19664640
927
+ },
928
+ {
929
+ "name": "model.layers.20.input_layernorm.weight",
930
+ "shape": [
931
+ 960
932
+ ],
933
+ "dtype": "float16",
934
+ "format": "f32-to-bf16",
935
+ "nbytes": 1920,
936
+ "byteOffset": 21507840
937
+ },
938
+ {
939
+ "name": "model.layers.20.mlp.down_proj.weight",
940
+ "shape": [
941
+ 960,
942
+ 2560
943
+ ],
944
+ "dtype": "float16",
945
+ "format": "f32-to-bf16",
946
+ "nbytes": 4915200,
947
+ "byteOffset": 21509760
948
+ }
949
+ ],
950
+ "md5sum": "d9f9e115188182a8d41fcfba304c124c"
951
+ },
952
+ {
953
+ "dataPath": "params_shard_10.bin",
954
+ "format": "raw-shard",
955
+ "nbytes": 32568960,
956
+ "records": [
957
+ {
958
+ "name": "model.layers.20.mlp.gate_up_proj.weight",
959
+ "shape": [
960
+ 5120,
961
+ 960
962
+ ],
963
+ "dtype": "float16",
964
+ "format": "f32-to-bf16",
965
+ "nbytes": 9830400,
966
+ "byteOffset": 0
967
+ },
968
+ {
969
+ "name": "model.layers.20.post_attention_layernorm.weight",
970
+ "shape": [
971
+ 960
972
+ ],
973
+ "dtype": "float16",
974
+ "format": "f32-to-bf16",
975
+ "nbytes": 1920,
976
+ "byteOffset": 9830400
977
+ },
978
+ {
979
+ "name": "model.layers.20.self_attn.qkv_proj.weight",
980
+ "shape": [
981
+ 1600,
982
+ 960
983
+ ],
984
+ "dtype": "float16",
985
+ "format": "f32-to-bf16",
986
+ "nbytes": 3072000,
987
+ "byteOffset": 9832320
988
+ },
989
+ {
990
+ "name": "model.layers.20.self_attn.o_proj.weight",
991
+ "shape": [
992
+ 960,
993
+ 960
994
+ ],
995
+ "dtype": "float16",
996
+ "format": "f32-to-bf16",
997
+ "nbytes": 1843200,
998
+ "byteOffset": 12904320
999
+ },
1000
+ {
1001
+ "name": "model.layers.21.input_layernorm.weight",
1002
+ "shape": [
1003
+ 960
1004
+ ],
1005
+ "dtype": "float16",
1006
+ "format": "f32-to-bf16",
1007
+ "nbytes": 1920,
1008
+ "byteOffset": 14747520
1009
+ },
1010
+ {
1011
+ "name": "model.layers.21.mlp.down_proj.weight",
1012
+ "shape": [
1013
+ 960,
1014
+ 2560
1015
+ ],
1016
+ "dtype": "float16",
1017
+ "format": "f32-to-bf16",
1018
+ "nbytes": 4915200,
1019
+ "byteOffset": 14749440
1020
+ },
1021
+ {
1022
+ "name": "model.layers.21.mlp.gate_up_proj.weight",
1023
+ "shape": [
1024
+ 5120,
1025
+ 960
1026
+ ],
1027
+ "dtype": "float16",
1028
+ "format": "f32-to-bf16",
1029
+ "nbytes": 9830400,
1030
+ "byteOffset": 19664640
1031
+ },
1032
+ {
1033
+ "name": "model.layers.21.post_attention_layernorm.weight",
1034
+ "shape": [
1035
+ 960
1036
+ ],
1037
+ "dtype": "float16",
1038
+ "format": "f32-to-bf16",
1039
+ "nbytes": 1920,
1040
+ "byteOffset": 29495040
1041
+ },
1042
+ {
1043
+ "name": "model.layers.21.self_attn.qkv_proj.weight",
1044
+ "shape": [
1045
+ 1600,
1046
+ 960
1047
+ ],
1048
+ "dtype": "float16",
1049
+ "format": "f32-to-bf16",
1050
+ "nbytes": 3072000,
1051
+ "byteOffset": 29496960
1052
+ }
1053
+ ],
1054
+ "md5sum": "0373b6c096825c5ab145f7190bfa5fad"
1055
+ },
1056
+ {
1057
+ "dataPath": "params_shard_11.bin",
1058
+ "format": "raw-shard",
1059
+ "nbytes": 26424960,
1060
+ "records": [
1061
+ {
1062
+ "name": "model.layers.21.self_attn.o_proj.weight",
1063
+ "shape": [
1064
+ 960,
1065
+ 960
1066
+ ],
1067
+ "dtype": "float16",
1068
+ "format": "f32-to-bf16",
1069
+ "nbytes": 1843200,
1070
+ "byteOffset": 0
1071
+ },
1072
+ {
1073
+ "name": "model.layers.22.input_layernorm.weight",
1074
+ "shape": [
1075
+ 960
1076
+ ],
1077
+ "dtype": "float16",
1078
+ "format": "f32-to-bf16",
1079
+ "nbytes": 1920,
1080
+ "byteOffset": 1843200
1081
+ },
1082
+ {
1083
+ "name": "model.layers.22.mlp.down_proj.weight",
1084
+ "shape": [
1085
+ 960,
1086
+ 2560
1087
+ ],
1088
+ "dtype": "float16",
1089
+ "format": "f32-to-bf16",
1090
+ "nbytes": 4915200,
1091
+ "byteOffset": 1845120
1092
+ },
1093
+ {
1094
+ "name": "model.layers.22.mlp.gate_up_proj.weight",
1095
+ "shape": [
1096
+ 5120,
1097
+ 960
1098
+ ],
1099
+ "dtype": "float16",
1100
+ "format": "f32-to-bf16",
1101
+ "nbytes": 9830400,
1102
+ "byteOffset": 6760320
1103
+ },
1104
+ {
1105
+ "name": "model.layers.22.post_attention_layernorm.weight",
1106
+ "shape": [
1107
+ 960
1108
+ ],
1109
+ "dtype": "float16",
1110
+ "format": "f32-to-bf16",
1111
+ "nbytes": 1920,
1112
+ "byteOffset": 16590720
1113
+ },
1114
+ {
1115
+ "name": "model.layers.22.self_attn.qkv_proj.weight",
1116
+ "shape": [
1117
+ 1600,
1118
+ 960
1119
+ ],
1120
+ "dtype": "float16",
1121
+ "format": "f32-to-bf16",
1122
+ "nbytes": 3072000,
1123
+ "byteOffset": 16592640
1124
+ },
1125
+ {
1126
+ "name": "model.layers.22.self_attn.o_proj.weight",
1127
+ "shape": [
1128
+ 960,
1129
+ 960
1130
+ ],
1131
+ "dtype": "float16",
1132
+ "format": "f32-to-bf16",
1133
+ "nbytes": 1843200,
1134
+ "byteOffset": 19664640
1135
+ },
1136
+ {
1137
+ "name": "model.layers.23.input_layernorm.weight",
1138
+ "shape": [
1139
+ 960
1140
+ ],
1141
+ "dtype": "float16",
1142
+ "format": "f32-to-bf16",
1143
+ "nbytes": 1920,
1144
+ "byteOffset": 21507840
1145
+ },
1146
+ {
1147
+ "name": "model.layers.23.mlp.down_proj.weight",
1148
+ "shape": [
1149
+ 960,
1150
+ 2560
1151
+ ],
1152
+ "dtype": "float16",
1153
+ "format": "f32-to-bf16",
1154
+ "nbytes": 4915200,
1155
+ "byteOffset": 21509760
1156
+ }
1157
+ ],
1158
+ "md5sum": "cf420c31a892cefc6bec3a8a4a16687a"
1159
+ },
1160
+ {
1161
+ "dataPath": "params_shard_12.bin",
1162
+ "format": "raw-shard",
1163
+ "nbytes": 32568960,
1164
+ "records": [
1165
+ {
1166
+ "name": "model.layers.23.mlp.gate_up_proj.weight",
1167
+ "shape": [
1168
+ 5120,
1169
+ 960
1170
+ ],
1171
+ "dtype": "float16",
1172
+ "format": "f32-to-bf16",
1173
+ "nbytes": 9830400,
1174
+ "byteOffset": 0
1175
+ },
1176
+ {
1177
+ "name": "model.layers.23.post_attention_layernorm.weight",
1178
+ "shape": [
1179
+ 960
1180
+ ],
1181
+ "dtype": "float16",
1182
+ "format": "f32-to-bf16",
1183
+ "nbytes": 1920,
1184
+ "byteOffset": 9830400
1185
+ },
1186
+ {
1187
+ "name": "model.layers.23.self_attn.qkv_proj.weight",
1188
+ "shape": [
1189
+ 1600,
1190
+ 960
1191
+ ],
1192
+ "dtype": "float16",
1193
+ "format": "f32-to-bf16",
1194
+ "nbytes": 3072000,
1195
+ "byteOffset": 9832320
1196
+ },
1197
+ {
1198
+ "name": "model.layers.23.self_attn.o_proj.weight",
1199
+ "shape": [
1200
+ 960,
1201
+ 960
1202
+ ],
1203
+ "dtype": "float16",
1204
+ "format": "f32-to-bf16",
1205
+ "nbytes": 1843200,
1206
+ "byteOffset": 12904320
1207
+ },
1208
+ {
1209
+ "name": "model.layers.24.input_layernorm.weight",
1210
+ "shape": [
1211
+ 960
1212
+ ],
1213
+ "dtype": "float16",
1214
+ "format": "f32-to-bf16",
1215
+ "nbytes": 1920,
1216
+ "byteOffset": 14747520
1217
+ },
1218
+ {
1219
+ "name": "model.layers.24.mlp.down_proj.weight",
1220
+ "shape": [
1221
+ 960,
1222
+ 2560
1223
+ ],
1224
+ "dtype": "float16",
1225
+ "format": "f32-to-bf16",
1226
+ "nbytes": 4915200,
1227
+ "byteOffset": 14749440
1228
+ },
1229
+ {
1230
+ "name": "model.layers.24.mlp.gate_up_proj.weight",
1231
+ "shape": [
1232
+ 5120,
1233
+ 960
1234
+ ],
1235
+ "dtype": "float16",
1236
+ "format": "f32-to-bf16",
1237
+ "nbytes": 9830400,
1238
+ "byteOffset": 19664640
1239
+ },
1240
+ {
1241
+ "name": "model.layers.24.post_attention_layernorm.weight",
1242
+ "shape": [
1243
+ 960
1244
+ ],
1245
+ "dtype": "float16",
1246
+ "format": "f32-to-bf16",
1247
+ "nbytes": 1920,
1248
+ "byteOffset": 29495040
1249
+ },
1250
+ {
1251
+ "name": "model.layers.24.self_attn.qkv_proj.weight",
1252
+ "shape": [
1253
+ 1600,
1254
+ 960
1255
+ ],
1256
+ "dtype": "float16",
1257
+ "format": "f32-to-bf16",
1258
+ "nbytes": 3072000,
1259
+ "byteOffset": 29496960
1260
+ }
1261
+ ],
1262
+ "md5sum": "f87cfd0e10e02607d638b09aad0ecaa4"
1263
+ },
1264
+ {
1265
+ "dataPath": "params_shard_13.bin",
1266
+ "format": "raw-shard",
1267
+ "nbytes": 26424960,
1268
+ "records": [
1269
+ {
1270
+ "name": "model.layers.24.self_attn.o_proj.weight",
1271
+ "shape": [
1272
+ 960,
1273
+ 960
1274
+ ],
1275
+ "dtype": "float16",
1276
+ "format": "f32-to-bf16",
1277
+ "nbytes": 1843200,
1278
+ "byteOffset": 0
1279
+ },
1280
+ {
1281
+ "name": "model.layers.25.input_layernorm.weight",
1282
+ "shape": [
1283
+ 960
1284
+ ],
1285
+ "dtype": "float16",
1286
+ "format": "f32-to-bf16",
1287
+ "nbytes": 1920,
1288
+ "byteOffset": 1843200
1289
+ },
1290
+ {
1291
+ "name": "model.layers.25.mlp.down_proj.weight",
1292
+ "shape": [
1293
+ 960,
1294
+ 2560
1295
+ ],
1296
+ "dtype": "float16",
1297
+ "format": "f32-to-bf16",
1298
+ "nbytes": 4915200,
1299
+ "byteOffset": 1845120
1300
+ },
1301
+ {
1302
+ "name": "model.layers.25.mlp.gate_up_proj.weight",
1303
+ "shape": [
1304
+ 5120,
1305
+ 960
1306
+ ],
1307
+ "dtype": "float16",
1308
+ "format": "f32-to-bf16",
1309
+ "nbytes": 9830400,
1310
+ "byteOffset": 6760320
1311
+ },
1312
+ {
1313
+ "name": "model.layers.25.post_attention_layernorm.weight",
1314
+ "shape": [
1315
+ 960
1316
+ ],
1317
+ "dtype": "float16",
1318
+ "format": "f32-to-bf16",
1319
+ "nbytes": 1920,
1320
+ "byteOffset": 16590720
1321
+ },
1322
+ {
1323
+ "name": "model.layers.25.self_attn.qkv_proj.weight",
1324
+ "shape": [
1325
+ 1600,
1326
+ 960
1327
+ ],
1328
+ "dtype": "float16",
1329
+ "format": "f32-to-bf16",
1330
+ "nbytes": 3072000,
1331
+ "byteOffset": 16592640
1332
+ },
1333
+ {
1334
+ "name": "model.layers.25.self_attn.o_proj.weight",
1335
+ "shape": [
1336
+ 960,
1337
+ 960
1338
+ ],
1339
+ "dtype": "float16",
1340
+ "format": "f32-to-bf16",
1341
+ "nbytes": 1843200,
1342
+ "byteOffset": 19664640
1343
+ },
1344
+ {
1345
+ "name": "model.layers.26.input_layernorm.weight",
1346
+ "shape": [
1347
+ 960
1348
+ ],
1349
+ "dtype": "float16",
1350
+ "format": "f32-to-bf16",
1351
+ "nbytes": 1920,
1352
+ "byteOffset": 21507840
1353
+ },
1354
+ {
1355
+ "name": "model.layers.26.mlp.down_proj.weight",
1356
+ "shape": [
1357
+ 960,
1358
+ 2560
1359
+ ],
1360
+ "dtype": "float16",
1361
+ "format": "f32-to-bf16",
1362
+ "nbytes": 4915200,
1363
+ "byteOffset": 21509760
1364
+ }
1365
+ ],
1366
+ "md5sum": "f321ad5591871768015ca1bf95e11496"
1367
+ },
1368
+ {
1369
+ "dataPath": "params_shard_14.bin",
1370
+ "format": "raw-shard",
1371
+ "nbytes": 32568960,
1372
+ "records": [
1373
+ {
1374
+ "name": "model.layers.26.mlp.gate_up_proj.weight",
1375
+ "shape": [
1376
+ 5120,
1377
+ 960
1378
+ ],
1379
+ "dtype": "float16",
1380
+ "format": "f32-to-bf16",
1381
+ "nbytes": 9830400,
1382
+ "byteOffset": 0
1383
+ },
1384
+ {
1385
+ "name": "model.layers.26.post_attention_layernorm.weight",
1386
+ "shape": [
1387
+ 960
1388
+ ],
1389
+ "dtype": "float16",
1390
+ "format": "f32-to-bf16",
1391
+ "nbytes": 1920,
1392
+ "byteOffset": 9830400
1393
+ },
1394
+ {
1395
+ "name": "model.layers.26.self_attn.qkv_proj.weight",
1396
+ "shape": [
1397
+ 1600,
1398
+ 960
1399
+ ],
1400
+ "dtype": "float16",
1401
+ "format": "f32-to-bf16",
1402
+ "nbytes": 3072000,
1403
+ "byteOffset": 9832320
1404
+ },
1405
+ {
1406
+ "name": "model.layers.26.self_attn.o_proj.weight",
1407
+ "shape": [
1408
+ 960,
1409
+ 960
1410
+ ],
1411
+ "dtype": "float16",
1412
+ "format": "f32-to-bf16",
1413
+ "nbytes": 1843200,
1414
+ "byteOffset": 12904320
1415
+ },
1416
+ {
1417
+ "name": "model.layers.27.input_layernorm.weight",
1418
+ "shape": [
1419
+ 960
1420
+ ],
1421
+ "dtype": "float16",
1422
+ "format": "f32-to-bf16",
1423
+ "nbytes": 1920,
1424
+ "byteOffset": 14747520
1425
+ },
1426
+ {
1427
+ "name": "model.layers.27.mlp.down_proj.weight",
1428
+ "shape": [
1429
+ 960,
1430
+ 2560
1431
+ ],
1432
+ "dtype": "float16",
1433
+ "format": "f32-to-bf16",
1434
+ "nbytes": 4915200,
1435
+ "byteOffset": 14749440
1436
+ },
1437
+ {
1438
+ "name": "model.layers.27.mlp.gate_up_proj.weight",
1439
+ "shape": [
1440
+ 5120,
1441
+ 960
1442
+ ],
1443
+ "dtype": "float16",
1444
+ "format": "f32-to-bf16",
1445
+ "nbytes": 9830400,
1446
+ "byteOffset": 19664640
1447
+ },
1448
+ {
1449
+ "name": "model.layers.27.post_attention_layernorm.weight",
1450
+ "shape": [
1451
+ 960
1452
+ ],
1453
+ "dtype": "float16",
1454
+ "format": "f32-to-bf16",
1455
+ "nbytes": 1920,
1456
+ "byteOffset": 29495040
1457
+ },
1458
+ {
1459
+ "name": "model.layers.27.self_attn.qkv_proj.weight",
1460
+ "shape": [
1461
+ 1600,
1462
+ 960
1463
+ ],
1464
+ "dtype": "float16",
1465
+ "format": "f32-to-bf16",
1466
+ "nbytes": 3072000,
1467
+ "byteOffset": 29496960
1468
+ }
1469
+ ],
1470
+ "md5sum": "3d619394bd76f30960bbea02e921fbaa"
1471
+ },
1472
+ {
1473
+ "dataPath": "params_shard_15.bin",
1474
+ "format": "raw-shard",
1475
+ "nbytes": 26424960,
1476
+ "records": [
1477
+ {
1478
+ "name": "model.layers.27.self_attn.o_proj.weight",
1479
+ "shape": [
1480
+ 960,
1481
+ 960
1482
+ ],
1483
+ "dtype": "float16",
1484
+ "format": "f32-to-bf16",
1485
+ "nbytes": 1843200,
1486
+ "byteOffset": 0
1487
+ },
1488
+ {
1489
+ "name": "model.layers.28.input_layernorm.weight",
1490
+ "shape": [
1491
+ 960
1492
+ ],
1493
+ "dtype": "float16",
1494
+ "format": "f32-to-bf16",
1495
+ "nbytes": 1920,
1496
+ "byteOffset": 1843200
1497
+ },
1498
+ {
1499
+ "name": "model.layers.28.mlp.down_proj.weight",
1500
+ "shape": [
1501
+ 960,
1502
+ 2560
1503
+ ],
1504
+ "dtype": "float16",
1505
+ "format": "f32-to-bf16",
1506
+ "nbytes": 4915200,
1507
+ "byteOffset": 1845120
1508
+ },
1509
+ {
1510
+ "name": "model.layers.28.mlp.gate_up_proj.weight",
1511
+ "shape": [
1512
+ 5120,
1513
+ 960
1514
+ ],
1515
+ "dtype": "float16",
1516
+ "format": "f32-to-bf16",
1517
+ "nbytes": 9830400,
1518
+ "byteOffset": 6760320
1519
+ },
1520
+ {
1521
+ "name": "model.layers.28.post_attention_layernorm.weight",
1522
+ "shape": [
1523
+ 960
1524
+ ],
1525
+ "dtype": "float16",
1526
+ "format": "f32-to-bf16",
1527
+ "nbytes": 1920,
1528
+ "byteOffset": 16590720
1529
+ },
1530
+ {
1531
+ "name": "model.layers.28.self_attn.qkv_proj.weight",
1532
+ "shape": [
1533
+ 1600,
1534
+ 960
1535
+ ],
1536
+ "dtype": "float16",
1537
+ "format": "f32-to-bf16",
1538
+ "nbytes": 3072000,
1539
+ "byteOffset": 16592640
1540
+ },
1541
+ {
1542
+ "name": "model.layers.28.self_attn.o_proj.weight",
1543
+ "shape": [
1544
+ 960,
1545
+ 960
1546
+ ],
1547
+ "dtype": "float16",
1548
+ "format": "f32-to-bf16",
1549
+ "nbytes": 1843200,
1550
+ "byteOffset": 19664640
1551
+ },
1552
+ {
1553
+ "name": "model.layers.29.input_layernorm.weight",
1554
+ "shape": [
1555
+ 960
1556
+ ],
1557
+ "dtype": "float16",
1558
+ "format": "f32-to-bf16",
1559
+ "nbytes": 1920,
1560
+ "byteOffset": 21507840
1561
+ },
1562
+ {
1563
+ "name": "model.layers.29.mlp.down_proj.weight",
1564
+ "shape": [
1565
+ 960,
1566
+ 2560
1567
+ ],
1568
+ "dtype": "float16",
1569
+ "format": "f32-to-bf16",
1570
+ "nbytes": 4915200,
1571
+ "byteOffset": 21509760
1572
+ }
1573
+ ],
1574
+ "md5sum": "61454b4565b014fa9138fbd9e7c69368"
1575
+ },
1576
+ {
1577
+ "dataPath": "params_shard_16.bin",
1578
+ "format": "raw-shard",
1579
+ "nbytes": 32568960,
1580
+ "records": [
1581
+ {
1582
+ "name": "model.layers.29.mlp.gate_up_proj.weight",
1583
+ "shape": [
1584
+ 5120,
1585
+ 960
1586
+ ],
1587
+ "dtype": "float16",
1588
+ "format": "f32-to-bf16",
1589
+ "nbytes": 9830400,
1590
+ "byteOffset": 0
1591
+ },
1592
+ {
1593
+ "name": "model.layers.29.post_attention_layernorm.weight",
1594
+ "shape": [
1595
+ 960
1596
+ ],
1597
+ "dtype": "float16",
1598
+ "format": "f32-to-bf16",
1599
+ "nbytes": 1920,
1600
+ "byteOffset": 9830400
1601
+ },
1602
+ {
1603
+ "name": "model.layers.29.self_attn.qkv_proj.weight",
1604
+ "shape": [
1605
+ 1600,
1606
+ 960
1607
+ ],
1608
+ "dtype": "float16",
1609
+ "format": "f32-to-bf16",
1610
+ "nbytes": 3072000,
1611
+ "byteOffset": 9832320
1612
+ },
1613
+ {
1614
+ "name": "model.layers.29.self_attn.o_proj.weight",
1615
+ "shape": [
1616
+ 960,
1617
+ 960
1618
+ ],
1619
+ "dtype": "float16",
1620
+ "format": "f32-to-bf16",
1621
+ "nbytes": 1843200,
1622
+ "byteOffset": 12904320
1623
+ },
1624
+ {
1625
+ "name": "model.layers.3.input_layernorm.weight",
1626
+ "shape": [
1627
+ 960
1628
+ ],
1629
+ "dtype": "float16",
1630
+ "format": "f32-to-bf16",
1631
+ "nbytes": 1920,
1632
+ "byteOffset": 14747520
1633
+ },
1634
+ {
1635
+ "name": "model.layers.3.mlp.down_proj.weight",
1636
+ "shape": [
1637
+ 960,
1638
+ 2560
1639
+ ],
1640
+ "dtype": "float16",
1641
+ "format": "f32-to-bf16",
1642
+ "nbytes": 4915200,
1643
+ "byteOffset": 14749440
1644
+ },
1645
+ {
1646
+ "name": "model.layers.3.mlp.gate_up_proj.weight",
1647
+ "shape": [
1648
+ 5120,
1649
+ 960
1650
+ ],
1651
+ "dtype": "float16",
1652
+ "format": "f32-to-bf16",
1653
+ "nbytes": 9830400,
1654
+ "byteOffset": 19664640
1655
+ },
1656
+ {
1657
+ "name": "model.layers.3.post_attention_layernorm.weight",
1658
+ "shape": [
1659
+ 960
1660
+ ],
1661
+ "dtype": "float16",
1662
+ "format": "f32-to-bf16",
1663
+ "nbytes": 1920,
1664
+ "byteOffset": 29495040
1665
+ },
1666
+ {
1667
+ "name": "model.layers.3.self_attn.qkv_proj.weight",
1668
+ "shape": [
1669
+ 1600,
1670
+ 960
1671
+ ],
1672
+ "dtype": "float16",
1673
+ "format": "f32-to-bf16",
1674
+ "nbytes": 3072000,
1675
+ "byteOffset": 29496960
1676
+ }
1677
+ ],
1678
+ "md5sum": "14661838d59ec5d5a3f14b4eb12b6ab2"
1679
+ },
1680
+ {
1681
+ "dataPath": "params_shard_17.bin",
1682
+ "format": "raw-shard",
1683
+ "nbytes": 26424960,
1684
+ "records": [
1685
+ {
1686
+ "name": "model.layers.3.self_attn.o_proj.weight",
1687
+ "shape": [
1688
+ 960,
1689
+ 960
1690
+ ],
1691
+ "dtype": "float16",
1692
+ "format": "f32-to-bf16",
1693
+ "nbytes": 1843200,
1694
+ "byteOffset": 0
1695
+ },
1696
+ {
1697
+ "name": "model.layers.30.input_layernorm.weight",
1698
+ "shape": [
1699
+ 960
1700
+ ],
1701
+ "dtype": "float16",
1702
+ "format": "f32-to-bf16",
1703
+ "nbytes": 1920,
1704
+ "byteOffset": 1843200
1705
+ },
1706
+ {
1707
+ "name": "model.layers.30.mlp.down_proj.weight",
1708
+ "shape": [
1709
+ 960,
1710
+ 2560
1711
+ ],
1712
+ "dtype": "float16",
1713
+ "format": "f32-to-bf16",
1714
+ "nbytes": 4915200,
1715
+ "byteOffset": 1845120
1716
+ },
1717
+ {
1718
+ "name": "model.layers.30.mlp.gate_up_proj.weight",
1719
+ "shape": [
1720
+ 5120,
1721
+ 960
1722
+ ],
1723
+ "dtype": "float16",
1724
+ "format": "f32-to-bf16",
1725
+ "nbytes": 9830400,
1726
+ "byteOffset": 6760320
1727
+ },
1728
+ {
1729
+ "name": "model.layers.30.post_attention_layernorm.weight",
1730
+ "shape": [
1731
+ 960
1732
+ ],
1733
+ "dtype": "float16",
1734
+ "format": "f32-to-bf16",
1735
+ "nbytes": 1920,
1736
+ "byteOffset": 16590720
1737
+ },
1738
+ {
1739
+ "name": "model.layers.30.self_attn.qkv_proj.weight",
1740
+ "shape": [
1741
+ 1600,
1742
+ 960
1743
+ ],
1744
+ "dtype": "float16",
1745
+ "format": "f32-to-bf16",
1746
+ "nbytes": 3072000,
1747
+ "byteOffset": 16592640
1748
+ },
1749
+ {
1750
+ "name": "model.layers.30.self_attn.o_proj.weight",
1751
+ "shape": [
1752
+ 960,
1753
+ 960
1754
+ ],
1755
+ "dtype": "float16",
1756
+ "format": "f32-to-bf16",
1757
+ "nbytes": 1843200,
1758
+ "byteOffset": 19664640
1759
+ },
1760
+ {
1761
+ "name": "model.layers.31.input_layernorm.weight",
1762
+ "shape": [
1763
+ 960
1764
+ ],
1765
+ "dtype": "float16",
1766
+ "format": "f32-to-bf16",
1767
+ "nbytes": 1920,
1768
+ "byteOffset": 21507840
1769
+ },
1770
+ {
1771
+ "name": "model.layers.31.mlp.down_proj.weight",
1772
+ "shape": [
1773
+ 960,
1774
+ 2560
1775
+ ],
1776
+ "dtype": "float16",
1777
+ "format": "f32-to-bf16",
1778
+ "nbytes": 4915200,
1779
+ "byteOffset": 21509760
1780
+ }
1781
+ ],
1782
+ "md5sum": "88079b225dd9cce4a36210dc2928237e"
1783
+ },
1784
+ {
1785
+ "dataPath": "params_shard_18.bin",
1786
+ "format": "raw-shard",
1787
+ "nbytes": 32568960,
1788
+ "records": [
1789
+ {
1790
+ "name": "model.layers.31.mlp.gate_up_proj.weight",
1791
+ "shape": [
1792
+ 5120,
1793
+ 960
1794
+ ],
1795
+ "dtype": "float16",
1796
+ "format": "f32-to-bf16",
1797
+ "nbytes": 9830400,
1798
+ "byteOffset": 0
1799
+ },
1800
+ {
1801
+ "name": "model.layers.31.post_attention_layernorm.weight",
1802
+ "shape": [
1803
+ 960
1804
+ ],
1805
+ "dtype": "float16",
1806
+ "format": "f32-to-bf16",
1807
+ "nbytes": 1920,
1808
+ "byteOffset": 9830400
1809
+ },
1810
+ {
1811
+ "name": "model.layers.31.self_attn.qkv_proj.weight",
1812
+ "shape": [
1813
+ 1600,
1814
+ 960
1815
+ ],
1816
+ "dtype": "float16",
1817
+ "format": "f32-to-bf16",
1818
+ "nbytes": 3072000,
1819
+ "byteOffset": 9832320
1820
+ },
1821
+ {
1822
+ "name": "model.layers.31.self_attn.o_proj.weight",
1823
+ "shape": [
1824
+ 960,
1825
+ 960
1826
+ ],
1827
+ "dtype": "float16",
1828
+ "format": "f32-to-bf16",
1829
+ "nbytes": 1843200,
1830
+ "byteOffset": 12904320
1831
+ },
1832
+ {
1833
+ "name": "model.layers.4.input_layernorm.weight",
1834
+ "shape": [
1835
+ 960
1836
+ ],
1837
+ "dtype": "float16",
1838
+ "format": "f32-to-bf16",
1839
+ "nbytes": 1920,
1840
+ "byteOffset": 14747520
1841
+ },
1842
+ {
1843
+ "name": "model.layers.4.mlp.down_proj.weight",
1844
+ "shape": [
1845
+ 960,
1846
+ 2560
1847
+ ],
1848
+ "dtype": "float16",
1849
+ "format": "f32-to-bf16",
1850
+ "nbytes": 4915200,
1851
+ "byteOffset": 14749440
1852
+ },
1853
+ {
1854
+ "name": "model.layers.4.mlp.gate_up_proj.weight",
1855
+ "shape": [
1856
+ 5120,
1857
+ 960
1858
+ ],
1859
+ "dtype": "float16",
1860
+ "format": "f32-to-bf16",
1861
+ "nbytes": 9830400,
1862
+ "byteOffset": 19664640
1863
+ },
1864
+ {
1865
+ "name": "model.layers.4.post_attention_layernorm.weight",
1866
+ "shape": [
1867
+ 960
1868
+ ],
1869
+ "dtype": "float16",
1870
+ "format": "f32-to-bf16",
1871
+ "nbytes": 1920,
1872
+ "byteOffset": 29495040
1873
+ },
1874
+ {
1875
+ "name": "model.layers.4.self_attn.qkv_proj.weight",
1876
+ "shape": [
1877
+ 1600,
1878
+ 960
1879
+ ],
1880
+ "dtype": "float16",
1881
+ "format": "f32-to-bf16",
1882
+ "nbytes": 3072000,
1883
+ "byteOffset": 29496960
1884
+ }
1885
+ ],
1886
+ "md5sum": "57becef245b88912c429e4709c303eb4"
1887
+ },
1888
+ {
1889
+ "dataPath": "params_shard_19.bin",
1890
+ "format": "raw-shard",
1891
+ "nbytes": 26424960,
1892
+ "records": [
1893
+ {
1894
+ "name": "model.layers.4.self_attn.o_proj.weight",
1895
+ "shape": [
1896
+ 960,
1897
+ 960
1898
+ ],
1899
+ "dtype": "float16",
1900
+ "format": "f32-to-bf16",
1901
+ "nbytes": 1843200,
1902
+ "byteOffset": 0
1903
+ },
1904
+ {
1905
+ "name": "model.layers.5.input_layernorm.weight",
1906
+ "shape": [
1907
+ 960
1908
+ ],
1909
+ "dtype": "float16",
1910
+ "format": "f32-to-bf16",
1911
+ "nbytes": 1920,
1912
+ "byteOffset": 1843200
1913
+ },
1914
+ {
1915
+ "name": "model.layers.5.mlp.down_proj.weight",
1916
+ "shape": [
1917
+ 960,
1918
+ 2560
1919
+ ],
1920
+ "dtype": "float16",
1921
+ "format": "f32-to-bf16",
1922
+ "nbytes": 4915200,
1923
+ "byteOffset": 1845120
1924
+ },
1925
+ {
1926
+ "name": "model.layers.5.mlp.gate_up_proj.weight",
1927
+ "shape": [
1928
+ 5120,
1929
+ 960
1930
+ ],
1931
+ "dtype": "float16",
1932
+ "format": "f32-to-bf16",
1933
+ "nbytes": 9830400,
1934
+ "byteOffset": 6760320
1935
+ },
1936
+ {
1937
+ "name": "model.layers.5.post_attention_layernorm.weight",
1938
+ "shape": [
1939
+ 960
1940
+ ],
1941
+ "dtype": "float16",
1942
+ "format": "f32-to-bf16",
1943
+ "nbytes": 1920,
1944
+ "byteOffset": 16590720
1945
+ },
1946
+ {
1947
+ "name": "model.layers.5.self_attn.qkv_proj.weight",
1948
+ "shape": [
1949
+ 1600,
1950
+ 960
1951
+ ],
1952
+ "dtype": "float16",
1953
+ "format": "f32-to-bf16",
1954
+ "nbytes": 3072000,
1955
+ "byteOffset": 16592640
1956
+ },
1957
+ {
1958
+ "name": "model.layers.5.self_attn.o_proj.weight",
1959
+ "shape": [
1960
+ 960,
1961
+ 960
1962
+ ],
1963
+ "dtype": "float16",
1964
+ "format": "f32-to-bf16",
1965
+ "nbytes": 1843200,
1966
+ "byteOffset": 19664640
1967
+ },
1968
+ {
1969
+ "name": "model.layers.6.input_layernorm.weight",
1970
+ "shape": [
1971
+ 960
1972
+ ],
1973
+ "dtype": "float16",
1974
+ "format": "f32-to-bf16",
1975
+ "nbytes": 1920,
1976
+ "byteOffset": 21507840
1977
+ },
1978
+ {
1979
+ "name": "model.layers.6.mlp.down_proj.weight",
1980
+ "shape": [
1981
+ 960,
1982
+ 2560
1983
+ ],
1984
+ "dtype": "float16",
1985
+ "format": "f32-to-bf16",
1986
+ "nbytes": 4915200,
1987
+ "byteOffset": 21509760
1988
+ }
1989
+ ],
1990
+ "md5sum": "881b0ead4f94f9e0bedd61111bff01c8"
1991
+ },
1992
+ {
1993
+ "dataPath": "params_shard_20.bin",
1994
+ "format": "raw-shard",
1995
+ "nbytes": 32568960,
1996
+ "records": [
1997
+ {
1998
+ "name": "model.layers.6.mlp.gate_up_proj.weight",
1999
+ "shape": [
2000
+ 5120,
2001
+ 960
2002
+ ],
2003
+ "dtype": "float16",
2004
+ "format": "f32-to-bf16",
2005
+ "nbytes": 9830400,
2006
+ "byteOffset": 0
2007
+ },
2008
+ {
2009
+ "name": "model.layers.6.post_attention_layernorm.weight",
2010
+ "shape": [
2011
+ 960
2012
+ ],
2013
+ "dtype": "float16",
2014
+ "format": "f32-to-bf16",
2015
+ "nbytes": 1920,
2016
+ "byteOffset": 9830400
2017
+ },
2018
+ {
2019
+ "name": "model.layers.6.self_attn.qkv_proj.weight",
2020
+ "shape": [
2021
+ 1600,
2022
+ 960
2023
+ ],
2024
+ "dtype": "float16",
2025
+ "format": "f32-to-bf16",
2026
+ "nbytes": 3072000,
2027
+ "byteOffset": 9832320
2028
+ },
2029
+ {
2030
+ "name": "model.layers.6.self_attn.o_proj.weight",
2031
+ "shape": [
2032
+ 960,
2033
+ 960
2034
+ ],
2035
+ "dtype": "float16",
2036
+ "format": "f32-to-bf16",
2037
+ "nbytes": 1843200,
2038
+ "byteOffset": 12904320
2039
+ },
2040
+ {
2041
+ "name": "model.layers.7.input_layernorm.weight",
2042
+ "shape": [
2043
+ 960
2044
+ ],
2045
+ "dtype": "float16",
2046
+ "format": "f32-to-bf16",
2047
+ "nbytes": 1920,
2048
+ "byteOffset": 14747520
2049
+ },
2050
+ {
2051
+ "name": "model.layers.7.mlp.down_proj.weight",
2052
+ "shape": [
2053
+ 960,
2054
+ 2560
2055
+ ],
2056
+ "dtype": "float16",
2057
+ "format": "f32-to-bf16",
2058
+ "nbytes": 4915200,
2059
+ "byteOffset": 14749440
2060
+ },
2061
+ {
2062
+ "name": "model.layers.7.mlp.gate_up_proj.weight",
2063
+ "shape": [
2064
+ 5120,
2065
+ 960
2066
+ ],
2067
+ "dtype": "float16",
2068
+ "format": "f32-to-bf16",
2069
+ "nbytes": 9830400,
2070
+ "byteOffset": 19664640
2071
+ },
2072
+ {
2073
+ "name": "model.layers.7.post_attention_layernorm.weight",
2074
+ "shape": [
2075
+ 960
2076
+ ],
2077
+ "dtype": "float16",
2078
+ "format": "f32-to-bf16",
2079
+ "nbytes": 1920,
2080
+ "byteOffset": 29495040
2081
+ },
2082
+ {
2083
+ "name": "model.layers.7.self_attn.qkv_proj.weight",
2084
+ "shape": [
2085
+ 1600,
2086
+ 960
2087
+ ],
2088
+ "dtype": "float16",
2089
+ "format": "f32-to-bf16",
2090
+ "nbytes": 3072000,
2091
+ "byteOffset": 29496960
2092
+ }
2093
+ ],
2094
+ "md5sum": "0c569299428453d29097642084273bc8"
2095
+ },
2096
+ {
2097
+ "dataPath": "params_shard_21.bin",
2098
+ "format": "raw-shard",
2099
+ "nbytes": 26424960,
2100
+ "records": [
2101
+ {
2102
+ "name": "model.layers.7.self_attn.o_proj.weight",
2103
+ "shape": [
2104
+ 960,
2105
+ 960
2106
+ ],
2107
+ "dtype": "float16",
2108
+ "format": "f32-to-bf16",
2109
+ "nbytes": 1843200,
2110
+ "byteOffset": 0
2111
+ },
2112
+ {
2113
+ "name": "model.layers.8.input_layernorm.weight",
2114
+ "shape": [
2115
+ 960
2116
+ ],
2117
+ "dtype": "float16",
2118
+ "format": "f32-to-bf16",
2119
+ "nbytes": 1920,
2120
+ "byteOffset": 1843200
2121
+ },
2122
+ {
2123
+ "name": "model.layers.8.mlp.down_proj.weight",
2124
+ "shape": [
2125
+ 960,
2126
+ 2560
2127
+ ],
2128
+ "dtype": "float16",
2129
+ "format": "f32-to-bf16",
2130
+ "nbytes": 4915200,
2131
+ "byteOffset": 1845120
2132
+ },
2133
+ {
2134
+ "name": "model.layers.8.mlp.gate_up_proj.weight",
2135
+ "shape": [
2136
+ 5120,
2137
+ 960
2138
+ ],
2139
+ "dtype": "float16",
2140
+ "format": "f32-to-bf16",
2141
+ "nbytes": 9830400,
2142
+ "byteOffset": 6760320
2143
+ },
2144
+ {
2145
+ "name": "model.layers.8.post_attention_layernorm.weight",
2146
+ "shape": [
2147
+ 960
2148
+ ],
2149
+ "dtype": "float16",
2150
+ "format": "f32-to-bf16",
2151
+ "nbytes": 1920,
2152
+ "byteOffset": 16590720
2153
+ },
2154
+ {
2155
+ "name": "model.layers.8.self_attn.qkv_proj.weight",
2156
+ "shape": [
2157
+ 1600,
2158
+ 960
2159
+ ],
2160
+ "dtype": "float16",
2161
+ "format": "f32-to-bf16",
2162
+ "nbytes": 3072000,
2163
+ "byteOffset": 16592640
2164
+ },
2165
+ {
2166
+ "name": "model.layers.8.self_attn.o_proj.weight",
2167
+ "shape": [
2168
+ 960,
2169
+ 960
2170
+ ],
2171
+ "dtype": "float16",
2172
+ "format": "f32-to-bf16",
2173
+ "nbytes": 1843200,
2174
+ "byteOffset": 19664640
2175
+ },
2176
+ {
2177
+ "name": "model.layers.9.input_layernorm.weight",
2178
+ "shape": [
2179
+ 960
2180
+ ],
2181
+ "dtype": "float16",
2182
+ "format": "f32-to-bf16",
2183
+ "nbytes": 1920,
2184
+ "byteOffset": 21507840
2185
+ },
2186
+ {
2187
+ "name": "model.layers.9.mlp.down_proj.weight",
2188
+ "shape": [
2189
+ 960,
2190
+ 2560
2191
+ ],
2192
+ "dtype": "float16",
2193
+ "format": "f32-to-bf16",
2194
+ "nbytes": 4915200,
2195
+ "byteOffset": 21509760
2196
+ }
2197
+ ],
2198
+ "md5sum": "3dc093fea3171ec80bf5766503add8d0"
2199
+ },
2200
+ {
2201
+ "dataPath": "params_shard_22.bin",
2202
+ "format": "raw-shard",
2203
+ "nbytes": 14749440,
2204
+ "records": [
2205
+ {
2206
+ "name": "model.layers.9.mlp.gate_up_proj.weight",
2207
+ "shape": [
2208
+ 5120,
2209
+ 960
2210
+ ],
2211
+ "dtype": "float16",
2212
+ "format": "f32-to-bf16",
2213
+ "nbytes": 9830400,
2214
+ "byteOffset": 0
2215
+ },
2216
+ {
2217
+ "name": "model.layers.9.post_attention_layernorm.weight",
2218
+ "shape": [
2219
+ 960
2220
+ ],
2221
+ "dtype": "float16",
2222
+ "format": "f32-to-bf16",
2223
+ "nbytes": 1920,
2224
+ "byteOffset": 9830400
2225
+ },
2226
+ {
2227
+ "name": "model.layers.9.self_attn.qkv_proj.weight",
2228
+ "shape": [
2229
+ 1600,
2230
+ 960
2231
+ ],
2232
+ "dtype": "float16",
2233
+ "format": "f32-to-bf16",
2234
+ "nbytes": 3072000,
2235
+ "byteOffset": 9832320
2236
+ },
2237
+ {
2238
+ "name": "model.layers.9.self_attn.o_proj.weight",
2239
+ "shape": [
2240
+ 960,
2241
+ 960
2242
+ ],
2243
+ "dtype": "float16",
2244
+ "format": "f32-to-bf16",
2245
+ "nbytes": 1843200,
2246
+ "byteOffset": 12904320
2247
+ },
2248
+ {
2249
+ "name": "model.norm.weight",
2250
+ "shape": [
2251
+ 960
2252
+ ],
2253
+ "dtype": "float16",
2254
+ "format": "f32-to-bf16",
2255
+ "nbytes": 1920,
2256
+ "byteOffset": 14747520
2257
+ }
2258
+ ],
2259
+ "md5sum": "a572b33a6e464b171fadf32d102a0c72"
2260
+ }
2261
+ ]
2262
+ }
params_shard_0.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d02e1b8b920fb1ac376216c6cf685aefed335045e51973dff31bd06aabd4cf09
3
+ size 94371840
params_shard_1.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:962bc6b68c2a52ab376962df5fc0f932ee52d870d10b53e8bf44b74501b34a4d
3
+ size 24581760
params_shard_10.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a64cdead895e66c9f7fddda44106f3611d87f20b1d61896e6f90353aa205ec02
3
+ size 32568960
params_shard_11.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6e72bf3e6626e249e3459cca8b687ee281d822fa93967686fdc92b6fcf98e385
3
+ size 26424960
params_shard_12.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2ef7848e0e0a2f7960cf34f6633155af5566b0ded466d71d947cbbbe82688886
3
+ size 32568960
params_shard_13.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:27f6ecd9335b5bfdf91394c00b3b32e465373228d9124c408e553269711898ca
3
+ size 26424960
params_shard_14.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fdf9f49437ca8226f0763646e5b263db19cfc73d2aa0ba0c9ac543153bd0efbd
3
+ size 32568960
params_shard_15.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4d17c1fbfcc9dca34b4e901eb3e8ed51cca936efa562a5e7e7b49dd833e27725
3
+ size 26424960
params_shard_16.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:20ff1a9920ac6398cb279f8e39bbf2127a0d3f85f7d0b0cf1ce5a8c073f0dd66
3
+ size 32568960
params_shard_17.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:55d0338ffa5256369b4dd9d87a1dce35edbbc50df47fa102de6b8928e2f694e2
3
+ size 26424960
params_shard_18.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4bd09a38bf4cebcb05cafa2b6f8b65c5c9c3ae9fdbb6815c5fbf085393ccff38
3
+ size 32568960
params_shard_19.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:01c94c9563d2fe200654f5ce1bd0361762d260dc5ae8755ba7ef659389134c9c
3
+ size 26424960
params_shard_2.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:15d718b48132aa5539e7244d8c3bedc057de456b09e7746cff3c66717f102cce
3
+ size 32568960
params_shard_20.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e087d33cfb3782fba6e8f5d85bfea4adcc2055f02a96e7a43d9038a285bedac2
3
+ size 32568960
params_shard_21.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e9bbebb15a017d869bc2fc7e353745f9579fe9d9e9f8dffbdf12765d845822bc
3
+ size 26424960
params_shard_22.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5c79a08d458fe6c8b96f9bd5b7e4635c0fcf9385d4fd27cec23ffea1529aadc9
3
+ size 14749440
params_shard_3.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ee94d26b0180f6715461269e360954ef46e9e5dc607791a50406872579c2e84c
3
+ size 26424960
params_shard_4.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c8a84aa47ff154247a3e36c927bfdf743e2be124424e551cdf497267ee0321b6
3
+ size 32568960
params_shard_5.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9604cff380aac0b59298e68f0907c8d65bfe4c870d21761ea175067c68bedb31
3
+ size 26424960
params_shard_6.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:33553491a6516cf595691e0e3b06adb89d2aac7cc8c667b716fc6852186767fe
3
+ size 32568960
params_shard_7.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:89cf6d217cf4dd9ce47b772ccbcce553d8b2187e5b7e3b67a7ff8465dcc9ca67
3
+ size 26424960
params_shard_8.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1c560515d84e42c3f5bd4e772a29c9580300d53c9148bbf516fb49f1a8c0d6f4
3
+ size 32568960
params_shard_9.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:81b28c91685ac0bd864194965b596b091427fec159fb41b9bbf6dd1640071509
3
+ size 26424960