English
acalatrava commited on
Commit
21bb91b
1 Parent(s): a0c5f18

TinyLlama-1.1B-orca-gpt4

Browse files
README.md ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ datasets:
4
+ - cerebras/SlimPajama-627B
5
+ - bigcode/starcoderdata
6
+ - sam-mosaic/orca-gpt4-chatml
7
+ language:
8
+ - en
9
+ ---
10
+ <div align="center">
11
+
12
+ # TinyLlama-1.1B
13
+
14
+ Finetuned with ORCA-GPT4 (chatml format)
15
+
16
+ This is the MLC-LLM version from https://huggingface.co/acalatrava/TinyLlama-1.1B-orca-gpt4
17
+ </div>
18
+
19
+ https://github.com/jzhang38/TinyLlama
20
+
mlc-chat-config.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_lib": "TinyLlama-1.1b-q4f16_1",
3
+ "local_id": "TinyLlama-1.1B-orca-gpt4-q4f16_1",
4
+ "conv_template": "chatml",
5
+ "temperature": 0.7,
6
+ "repetition_penalty": 1.2,
7
+ "top_p": 0.9,
8
+ "mean_gen_len": 128,
9
+ "max_gen_len": 512,
10
+ "max_window_size": 2048,
11
+ "num_shards": 1,
12
+ "shift_fill_factor": 0.3,
13
+ "tokenizer_files": [
14
+ "tokenizer.json",
15
+ "tokenizer.model"
16
+ ],
17
+ "model_category": "llama",
18
+ "model_name": "TinyLlama-1.1B-orca-gpt4",
19
+ "vocab_size": 32000
20
+ }
ndarray-cache.json ADDED
@@ -0,0 +1,2627 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "ParamSize": 227
4
+ },
5
+ "records": [
6
+ {
7
+ "dataPath": "params_shard_0.bin",
8
+ "format": "raw-shard",
9
+ "nbytes": 32768000,
10
+ "records": [
11
+ {
12
+ "name": "param_0",
13
+ "shape": [
14
+ 32000,
15
+ 256
16
+ ],
17
+ "dtype": "uint32",
18
+ "format": "raw",
19
+ "nbytes": 32768000,
20
+ "byteOffset": 0
21
+ }
22
+ ]
23
+ },
24
+ {
25
+ "dataPath": "params_shard_1.bin",
26
+ "format": "raw-shard",
27
+ "nbytes": 31825920,
28
+ "records": [
29
+ {
30
+ "name": "param_1",
31
+ "shape": [
32
+ 32000,
33
+ 64
34
+ ],
35
+ "dtype": "float16",
36
+ "format": "raw",
37
+ "nbytes": 4096000,
38
+ "byteOffset": 0
39
+ },
40
+ {
41
+ "name": "param_2",
42
+ "shape": [
43
+ 2560,
44
+ 256
45
+ ],
46
+ "dtype": "uint32",
47
+ "format": "raw",
48
+ "nbytes": 2621440,
49
+ "byteOffset": 4096000
50
+ },
51
+ {
52
+ "name": "param_3",
53
+ "shape": [
54
+ 2560,
55
+ 64
56
+ ],
57
+ "dtype": "float16",
58
+ "format": "raw",
59
+ "nbytes": 327680,
60
+ "byteOffset": 6717440
61
+ },
62
+ {
63
+ "name": "param_4",
64
+ "shape": [
65
+ 2048,
66
+ 256
67
+ ],
68
+ "dtype": "uint32",
69
+ "format": "raw",
70
+ "nbytes": 2097152,
71
+ "byteOffset": 7045120
72
+ },
73
+ {
74
+ "name": "param_5",
75
+ "shape": [
76
+ 2048,
77
+ 64
78
+ ],
79
+ "dtype": "float16",
80
+ "format": "raw",
81
+ "nbytes": 262144,
82
+ "byteOffset": 9142272
83
+ },
84
+ {
85
+ "name": "param_6",
86
+ "shape": [
87
+ 11264,
88
+ 256
89
+ ],
90
+ "dtype": "uint32",
91
+ "format": "raw",
92
+ "nbytes": 11534336,
93
+ "byteOffset": 9404416
94
+ },
95
+ {
96
+ "name": "param_7",
97
+ "shape": [
98
+ 11264,
99
+ 64
100
+ ],
101
+ "dtype": "float16",
102
+ "format": "raw",
103
+ "nbytes": 1441792,
104
+ "byteOffset": 20938752
105
+ },
106
+ {
107
+ "name": "param_8",
108
+ "shape": [
109
+ 2048,
110
+ 704
111
+ ],
112
+ "dtype": "uint32",
113
+ "format": "raw",
114
+ "nbytes": 5767168,
115
+ "byteOffset": 22380544
116
+ },
117
+ {
118
+ "name": "param_9",
119
+ "shape": [
120
+ 2048,
121
+ 176
122
+ ],
123
+ "dtype": "float16",
124
+ "format": "raw",
125
+ "nbytes": 720896,
126
+ "byteOffset": 28147712
127
+ },
128
+ {
129
+ "name": "param_10",
130
+ "shape": [
131
+ 2048
132
+ ],
133
+ "dtype": "float16",
134
+ "format": "raw",
135
+ "nbytes": 4096,
136
+ "byteOffset": 28868608
137
+ },
138
+ {
139
+ "name": "param_11",
140
+ "shape": [
141
+ 2048
142
+ ],
143
+ "dtype": "float16",
144
+ "format": "raw",
145
+ "nbytes": 4096,
146
+ "byteOffset": 28872704
147
+ },
148
+ {
149
+ "name": "param_12",
150
+ "shape": [
151
+ 2560,
152
+ 256
153
+ ],
154
+ "dtype": "uint32",
155
+ "format": "raw",
156
+ "nbytes": 2621440,
157
+ "byteOffset": 28876800
158
+ },
159
+ {
160
+ "name": "param_13",
161
+ "shape": [
162
+ 2560,
163
+ 64
164
+ ],
165
+ "dtype": "float16",
166
+ "format": "raw",
167
+ "nbytes": 327680,
168
+ "byteOffset": 31498240
169
+ }
170
+ ]
171
+ },
172
+ {
173
+ "dataPath": "params_shard_2.bin",
174
+ "format": "raw-shard",
175
+ "nbytes": 27140096,
176
+ "records": [
177
+ {
178
+ "name": "param_14",
179
+ "shape": [
180
+ 2048,
181
+ 256
182
+ ],
183
+ "dtype": "uint32",
184
+ "format": "raw",
185
+ "nbytes": 2097152,
186
+ "byteOffset": 0
187
+ },
188
+ {
189
+ "name": "param_15",
190
+ "shape": [
191
+ 2048,
192
+ 64
193
+ ],
194
+ "dtype": "float16",
195
+ "format": "raw",
196
+ "nbytes": 262144,
197
+ "byteOffset": 2097152
198
+ },
199
+ {
200
+ "name": "param_16",
201
+ "shape": [
202
+ 11264,
203
+ 256
204
+ ],
205
+ "dtype": "uint32",
206
+ "format": "raw",
207
+ "nbytes": 11534336,
208
+ "byteOffset": 2359296
209
+ },
210
+ {
211
+ "name": "param_17",
212
+ "shape": [
213
+ 11264,
214
+ 64
215
+ ],
216
+ "dtype": "float16",
217
+ "format": "raw",
218
+ "nbytes": 1441792,
219
+ "byteOffset": 13893632
220
+ },
221
+ {
222
+ "name": "param_18",
223
+ "shape": [
224
+ 2048,
225
+ 704
226
+ ],
227
+ "dtype": "uint32",
228
+ "format": "raw",
229
+ "nbytes": 5767168,
230
+ "byteOffset": 15335424
231
+ },
232
+ {
233
+ "name": "param_19",
234
+ "shape": [
235
+ 2048,
236
+ 176
237
+ ],
238
+ "dtype": "float16",
239
+ "format": "raw",
240
+ "nbytes": 720896,
241
+ "byteOffset": 21102592
242
+ },
243
+ {
244
+ "name": "param_20",
245
+ "shape": [
246
+ 2048
247
+ ],
248
+ "dtype": "float16",
249
+ "format": "raw",
250
+ "nbytes": 4096,
251
+ "byteOffset": 21823488
252
+ },
253
+ {
254
+ "name": "param_21",
255
+ "shape": [
256
+ 2048
257
+ ],
258
+ "dtype": "float16",
259
+ "format": "raw",
260
+ "nbytes": 4096,
261
+ "byteOffset": 21827584
262
+ },
263
+ {
264
+ "name": "param_22",
265
+ "shape": [
266
+ 2560,
267
+ 256
268
+ ],
269
+ "dtype": "uint32",
270
+ "format": "raw",
271
+ "nbytes": 2621440,
272
+ "byteOffset": 21831680
273
+ },
274
+ {
275
+ "name": "param_23",
276
+ "shape": [
277
+ 2560,
278
+ 64
279
+ ],
280
+ "dtype": "float16",
281
+ "format": "raw",
282
+ "nbytes": 327680,
283
+ "byteOffset": 24453120
284
+ },
285
+ {
286
+ "name": "param_24",
287
+ "shape": [
288
+ 2048,
289
+ 256
290
+ ],
291
+ "dtype": "uint32",
292
+ "format": "raw",
293
+ "nbytes": 2097152,
294
+ "byteOffset": 24780800
295
+ },
296
+ {
297
+ "name": "param_25",
298
+ "shape": [
299
+ 2048,
300
+ 64
301
+ ],
302
+ "dtype": "float16",
303
+ "format": "raw",
304
+ "nbytes": 262144,
305
+ "byteOffset": 26877952
306
+ }
307
+ ]
308
+ },
309
+ {
310
+ "dataPath": "params_shard_3.bin",
311
+ "format": "raw-shard",
312
+ "nbytes": 24780800,
313
+ "records": [
314
+ {
315
+ "name": "param_26",
316
+ "shape": [
317
+ 11264,
318
+ 256
319
+ ],
320
+ "dtype": "uint32",
321
+ "format": "raw",
322
+ "nbytes": 11534336,
323
+ "byteOffset": 0
324
+ },
325
+ {
326
+ "name": "param_27",
327
+ "shape": [
328
+ 11264,
329
+ 64
330
+ ],
331
+ "dtype": "float16",
332
+ "format": "raw",
333
+ "nbytes": 1441792,
334
+ "byteOffset": 11534336
335
+ },
336
+ {
337
+ "name": "param_28",
338
+ "shape": [
339
+ 2048,
340
+ 704
341
+ ],
342
+ "dtype": "uint32",
343
+ "format": "raw",
344
+ "nbytes": 5767168,
345
+ "byteOffset": 12976128
346
+ },
347
+ {
348
+ "name": "param_29",
349
+ "shape": [
350
+ 2048,
351
+ 176
352
+ ],
353
+ "dtype": "float16",
354
+ "format": "raw",
355
+ "nbytes": 720896,
356
+ "byteOffset": 18743296
357
+ },
358
+ {
359
+ "name": "param_30",
360
+ "shape": [
361
+ 2048
362
+ ],
363
+ "dtype": "float16",
364
+ "format": "raw",
365
+ "nbytes": 4096,
366
+ "byteOffset": 19464192
367
+ },
368
+ {
369
+ "name": "param_31",
370
+ "shape": [
371
+ 2048
372
+ ],
373
+ "dtype": "float16",
374
+ "format": "raw",
375
+ "nbytes": 4096,
376
+ "byteOffset": 19468288
377
+ },
378
+ {
379
+ "name": "param_32",
380
+ "shape": [
381
+ 2560,
382
+ 256
383
+ ],
384
+ "dtype": "uint32",
385
+ "format": "raw",
386
+ "nbytes": 2621440,
387
+ "byteOffset": 19472384
388
+ },
389
+ {
390
+ "name": "param_33",
391
+ "shape": [
392
+ 2560,
393
+ 64
394
+ ],
395
+ "dtype": "float16",
396
+ "format": "raw",
397
+ "nbytes": 327680,
398
+ "byteOffset": 22093824
399
+ },
400
+ {
401
+ "name": "param_34",
402
+ "shape": [
403
+ 2048,
404
+ 256
405
+ ],
406
+ "dtype": "uint32",
407
+ "format": "raw",
408
+ "nbytes": 2097152,
409
+ "byteOffset": 22421504
410
+ },
411
+ {
412
+ "name": "param_35",
413
+ "shape": [
414
+ 2048,
415
+ 64
416
+ ],
417
+ "dtype": "float16",
418
+ "format": "raw",
419
+ "nbytes": 262144,
420
+ "byteOffset": 24518656
421
+ }
422
+ ]
423
+ },
424
+ {
425
+ "dataPath": "params_shard_4.bin",
426
+ "format": "raw-shard",
427
+ "nbytes": 24780800,
428
+ "records": [
429
+ {
430
+ "name": "param_36",
431
+ "shape": [
432
+ 11264,
433
+ 256
434
+ ],
435
+ "dtype": "uint32",
436
+ "format": "raw",
437
+ "nbytes": 11534336,
438
+ "byteOffset": 0
439
+ },
440
+ {
441
+ "name": "param_37",
442
+ "shape": [
443
+ 11264,
444
+ 64
445
+ ],
446
+ "dtype": "float16",
447
+ "format": "raw",
448
+ "nbytes": 1441792,
449
+ "byteOffset": 11534336
450
+ },
451
+ {
452
+ "name": "param_38",
453
+ "shape": [
454
+ 2048,
455
+ 704
456
+ ],
457
+ "dtype": "uint32",
458
+ "format": "raw",
459
+ "nbytes": 5767168,
460
+ "byteOffset": 12976128
461
+ },
462
+ {
463
+ "name": "param_39",
464
+ "shape": [
465
+ 2048,
466
+ 176
467
+ ],
468
+ "dtype": "float16",
469
+ "format": "raw",
470
+ "nbytes": 720896,
471
+ "byteOffset": 18743296
472
+ },
473
+ {
474
+ "name": "param_40",
475
+ "shape": [
476
+ 2048
477
+ ],
478
+ "dtype": "float16",
479
+ "format": "raw",
480
+ "nbytes": 4096,
481
+ "byteOffset": 19464192
482
+ },
483
+ {
484
+ "name": "param_41",
485
+ "shape": [
486
+ 2048
487
+ ],
488
+ "dtype": "float16",
489
+ "format": "raw",
490
+ "nbytes": 4096,
491
+ "byteOffset": 19468288
492
+ },
493
+ {
494
+ "name": "param_42",
495
+ "shape": [
496
+ 2560,
497
+ 256
498
+ ],
499
+ "dtype": "uint32",
500
+ "format": "raw",
501
+ "nbytes": 2621440,
502
+ "byteOffset": 19472384
503
+ },
504
+ {
505
+ "name": "param_43",
506
+ "shape": [
507
+ 2560,
508
+ 64
509
+ ],
510
+ "dtype": "float16",
511
+ "format": "raw",
512
+ "nbytes": 327680,
513
+ "byteOffset": 22093824
514
+ },
515
+ {
516
+ "name": "param_44",
517
+ "shape": [
518
+ 2048,
519
+ 256
520
+ ],
521
+ "dtype": "uint32",
522
+ "format": "raw",
523
+ "nbytes": 2097152,
524
+ "byteOffset": 22421504
525
+ },
526
+ {
527
+ "name": "param_45",
528
+ "shape": [
529
+ 2048,
530
+ 64
531
+ ],
532
+ "dtype": "float16",
533
+ "format": "raw",
534
+ "nbytes": 262144,
535
+ "byteOffset": 24518656
536
+ }
537
+ ]
538
+ },
539
+ {
540
+ "dataPath": "params_shard_5.bin",
541
+ "format": "raw-shard",
542
+ "nbytes": 24780800,
543
+ "records": [
544
+ {
545
+ "name": "param_46",
546
+ "shape": [
547
+ 11264,
548
+ 256
549
+ ],
550
+ "dtype": "uint32",
551
+ "format": "raw",
552
+ "nbytes": 11534336,
553
+ "byteOffset": 0
554
+ },
555
+ {
556
+ "name": "param_47",
557
+ "shape": [
558
+ 11264,
559
+ 64
560
+ ],
561
+ "dtype": "float16",
562
+ "format": "raw",
563
+ "nbytes": 1441792,
564
+ "byteOffset": 11534336
565
+ },
566
+ {
567
+ "name": "param_48",
568
+ "shape": [
569
+ 2048,
570
+ 704
571
+ ],
572
+ "dtype": "uint32",
573
+ "format": "raw",
574
+ "nbytes": 5767168,
575
+ "byteOffset": 12976128
576
+ },
577
+ {
578
+ "name": "param_49",
579
+ "shape": [
580
+ 2048,
581
+ 176
582
+ ],
583
+ "dtype": "float16",
584
+ "format": "raw",
585
+ "nbytes": 720896,
586
+ "byteOffset": 18743296
587
+ },
588
+ {
589
+ "name": "param_50",
590
+ "shape": [
591
+ 2048
592
+ ],
593
+ "dtype": "float16",
594
+ "format": "raw",
595
+ "nbytes": 4096,
596
+ "byteOffset": 19464192
597
+ },
598
+ {
599
+ "name": "param_51",
600
+ "shape": [
601
+ 2048
602
+ ],
603
+ "dtype": "float16",
604
+ "format": "raw",
605
+ "nbytes": 4096,
606
+ "byteOffset": 19468288
607
+ },
608
+ {
609
+ "name": "param_52",
610
+ "shape": [
611
+ 2560,
612
+ 256
613
+ ],
614
+ "dtype": "uint32",
615
+ "format": "raw",
616
+ "nbytes": 2621440,
617
+ "byteOffset": 19472384
618
+ },
619
+ {
620
+ "name": "param_53",
621
+ "shape": [
622
+ 2560,
623
+ 64
624
+ ],
625
+ "dtype": "float16",
626
+ "format": "raw",
627
+ "nbytes": 327680,
628
+ "byteOffset": 22093824
629
+ },
630
+ {
631
+ "name": "param_54",
632
+ "shape": [
633
+ 2048,
634
+ 256
635
+ ],
636
+ "dtype": "uint32",
637
+ "format": "raw",
638
+ "nbytes": 2097152,
639
+ "byteOffset": 22421504
640
+ },
641
+ {
642
+ "name": "param_55",
643
+ "shape": [
644
+ 2048,
645
+ 64
646
+ ],
647
+ "dtype": "float16",
648
+ "format": "raw",
649
+ "nbytes": 262144,
650
+ "byteOffset": 24518656
651
+ }
652
+ ]
653
+ },
654
+ {
655
+ "dataPath": "params_shard_6.bin",
656
+ "format": "raw-shard",
657
+ "nbytes": 24780800,
658
+ "records": [
659
+ {
660
+ "name": "param_56",
661
+ "shape": [
662
+ 11264,
663
+ 256
664
+ ],
665
+ "dtype": "uint32",
666
+ "format": "raw",
667
+ "nbytes": 11534336,
668
+ "byteOffset": 0
669
+ },
670
+ {
671
+ "name": "param_57",
672
+ "shape": [
673
+ 11264,
674
+ 64
675
+ ],
676
+ "dtype": "float16",
677
+ "format": "raw",
678
+ "nbytes": 1441792,
679
+ "byteOffset": 11534336
680
+ },
681
+ {
682
+ "name": "param_58",
683
+ "shape": [
684
+ 2048,
685
+ 704
686
+ ],
687
+ "dtype": "uint32",
688
+ "format": "raw",
689
+ "nbytes": 5767168,
690
+ "byteOffset": 12976128
691
+ },
692
+ {
693
+ "name": "param_59",
694
+ "shape": [
695
+ 2048,
696
+ 176
697
+ ],
698
+ "dtype": "float16",
699
+ "format": "raw",
700
+ "nbytes": 720896,
701
+ "byteOffset": 18743296
702
+ },
703
+ {
704
+ "name": "param_60",
705
+ "shape": [
706
+ 2048
707
+ ],
708
+ "dtype": "float16",
709
+ "format": "raw",
710
+ "nbytes": 4096,
711
+ "byteOffset": 19464192
712
+ },
713
+ {
714
+ "name": "param_61",
715
+ "shape": [
716
+ 2048
717
+ ],
718
+ "dtype": "float16",
719
+ "format": "raw",
720
+ "nbytes": 4096,
721
+ "byteOffset": 19468288
722
+ },
723
+ {
724
+ "name": "param_62",
725
+ "shape": [
726
+ 2560,
727
+ 256
728
+ ],
729
+ "dtype": "uint32",
730
+ "format": "raw",
731
+ "nbytes": 2621440,
732
+ "byteOffset": 19472384
733
+ },
734
+ {
735
+ "name": "param_63",
736
+ "shape": [
737
+ 2560,
738
+ 64
739
+ ],
740
+ "dtype": "float16",
741
+ "format": "raw",
742
+ "nbytes": 327680,
743
+ "byteOffset": 22093824
744
+ },
745
+ {
746
+ "name": "param_64",
747
+ "shape": [
748
+ 2048,
749
+ 256
750
+ ],
751
+ "dtype": "uint32",
752
+ "format": "raw",
753
+ "nbytes": 2097152,
754
+ "byteOffset": 22421504
755
+ },
756
+ {
757
+ "name": "param_65",
758
+ "shape": [
759
+ 2048,
760
+ 64
761
+ ],
762
+ "dtype": "float16",
763
+ "format": "raw",
764
+ "nbytes": 262144,
765
+ "byteOffset": 24518656
766
+ }
767
+ ]
768
+ },
769
+ {
770
+ "dataPath": "params_shard_7.bin",
771
+ "format": "raw-shard",
772
+ "nbytes": 24780800,
773
+ "records": [
774
+ {
775
+ "name": "param_66",
776
+ "shape": [
777
+ 11264,
778
+ 256
779
+ ],
780
+ "dtype": "uint32",
781
+ "format": "raw",
782
+ "nbytes": 11534336,
783
+ "byteOffset": 0
784
+ },
785
+ {
786
+ "name": "param_67",
787
+ "shape": [
788
+ 11264,
789
+ 64
790
+ ],
791
+ "dtype": "float16",
792
+ "format": "raw",
793
+ "nbytes": 1441792,
794
+ "byteOffset": 11534336
795
+ },
796
+ {
797
+ "name": "param_68",
798
+ "shape": [
799
+ 2048,
800
+ 704
801
+ ],
802
+ "dtype": "uint32",
803
+ "format": "raw",
804
+ "nbytes": 5767168,
805
+ "byteOffset": 12976128
806
+ },
807
+ {
808
+ "name": "param_69",
809
+ "shape": [
810
+ 2048,
811
+ 176
812
+ ],
813
+ "dtype": "float16",
814
+ "format": "raw",
815
+ "nbytes": 720896,
816
+ "byteOffset": 18743296
817
+ },
818
+ {
819
+ "name": "param_70",
820
+ "shape": [
821
+ 2048
822
+ ],
823
+ "dtype": "float16",
824
+ "format": "raw",
825
+ "nbytes": 4096,
826
+ "byteOffset": 19464192
827
+ },
828
+ {
829
+ "name": "param_71",
830
+ "shape": [
831
+ 2048
832
+ ],
833
+ "dtype": "float16",
834
+ "format": "raw",
835
+ "nbytes": 4096,
836
+ "byteOffset": 19468288
837
+ },
838
+ {
839
+ "name": "param_72",
840
+ "shape": [
841
+ 2560,
842
+ 256
843
+ ],
844
+ "dtype": "uint32",
845
+ "format": "raw",
846
+ "nbytes": 2621440,
847
+ "byteOffset": 19472384
848
+ },
849
+ {
850
+ "name": "param_73",
851
+ "shape": [
852
+ 2560,
853
+ 64
854
+ ],
855
+ "dtype": "float16",
856
+ "format": "raw",
857
+ "nbytes": 327680,
858
+ "byteOffset": 22093824
859
+ },
860
+ {
861
+ "name": "param_74",
862
+ "shape": [
863
+ 2048,
864
+ 256
865
+ ],
866
+ "dtype": "uint32",
867
+ "format": "raw",
868
+ "nbytes": 2097152,
869
+ "byteOffset": 22421504
870
+ },
871
+ {
872
+ "name": "param_75",
873
+ "shape": [
874
+ 2048,
875
+ 64
876
+ ],
877
+ "dtype": "float16",
878
+ "format": "raw",
879
+ "nbytes": 262144,
880
+ "byteOffset": 24518656
881
+ }
882
+ ]
883
+ },
884
+ {
885
+ "dataPath": "params_shard_8.bin",
886
+ "format": "raw-shard",
887
+ "nbytes": 24780800,
888
+ "records": [
889
+ {
890
+ "name": "param_76",
891
+ "shape": [
892
+ 11264,
893
+ 256
894
+ ],
895
+ "dtype": "uint32",
896
+ "format": "raw",
897
+ "nbytes": 11534336,
898
+ "byteOffset": 0
899
+ },
900
+ {
901
+ "name": "param_77",
902
+ "shape": [
903
+ 11264,
904
+ 64
905
+ ],
906
+ "dtype": "float16",
907
+ "format": "raw",
908
+ "nbytes": 1441792,
909
+ "byteOffset": 11534336
910
+ },
911
+ {
912
+ "name": "param_78",
913
+ "shape": [
914
+ 2048,
915
+ 704
916
+ ],
917
+ "dtype": "uint32",
918
+ "format": "raw",
919
+ "nbytes": 5767168,
920
+ "byteOffset": 12976128
921
+ },
922
+ {
923
+ "name": "param_79",
924
+ "shape": [
925
+ 2048,
926
+ 176
927
+ ],
928
+ "dtype": "float16",
929
+ "format": "raw",
930
+ "nbytes": 720896,
931
+ "byteOffset": 18743296
932
+ },
933
+ {
934
+ "name": "param_80",
935
+ "shape": [
936
+ 2048
937
+ ],
938
+ "dtype": "float16",
939
+ "format": "raw",
940
+ "nbytes": 4096,
941
+ "byteOffset": 19464192
942
+ },
943
+ {
944
+ "name": "param_81",
945
+ "shape": [
946
+ 2048
947
+ ],
948
+ "dtype": "float16",
949
+ "format": "raw",
950
+ "nbytes": 4096,
951
+ "byteOffset": 19468288
952
+ },
953
+ {
954
+ "name": "param_82",
955
+ "shape": [
956
+ 2560,
957
+ 256
958
+ ],
959
+ "dtype": "uint32",
960
+ "format": "raw",
961
+ "nbytes": 2621440,
962
+ "byteOffset": 19472384
963
+ },
964
+ {
965
+ "name": "param_83",
966
+ "shape": [
967
+ 2560,
968
+ 64
969
+ ],
970
+ "dtype": "float16",
971
+ "format": "raw",
972
+ "nbytes": 327680,
973
+ "byteOffset": 22093824
974
+ },
975
+ {
976
+ "name": "param_84",
977
+ "shape": [
978
+ 2048,
979
+ 256
980
+ ],
981
+ "dtype": "uint32",
982
+ "format": "raw",
983
+ "nbytes": 2097152,
984
+ "byteOffset": 22421504
985
+ },
986
+ {
987
+ "name": "param_85",
988
+ "shape": [
989
+ 2048,
990
+ 64
991
+ ],
992
+ "dtype": "float16",
993
+ "format": "raw",
994
+ "nbytes": 262144,
995
+ "byteOffset": 24518656
996
+ }
997
+ ]
998
+ },
999
+ {
1000
+ "dataPath": "params_shard_9.bin",
1001
+ "format": "raw-shard",
1002
+ "nbytes": 24780800,
1003
+ "records": [
1004
+ {
1005
+ "name": "param_86",
1006
+ "shape": [
1007
+ 11264,
1008
+ 256
1009
+ ],
1010
+ "dtype": "uint32",
1011
+ "format": "raw",
1012
+ "nbytes": 11534336,
1013
+ "byteOffset": 0
1014
+ },
1015
+ {
1016
+ "name": "param_87",
1017
+ "shape": [
1018
+ 11264,
1019
+ 64
1020
+ ],
1021
+ "dtype": "float16",
1022
+ "format": "raw",
1023
+ "nbytes": 1441792,
1024
+ "byteOffset": 11534336
1025
+ },
1026
+ {
1027
+ "name": "param_88",
1028
+ "shape": [
1029
+ 2048,
1030
+ 704
1031
+ ],
1032
+ "dtype": "uint32",
1033
+ "format": "raw",
1034
+ "nbytes": 5767168,
1035
+ "byteOffset": 12976128
1036
+ },
1037
+ {
1038
+ "name": "param_89",
1039
+ "shape": [
1040
+ 2048,
1041
+ 176
1042
+ ],
1043
+ "dtype": "float16",
1044
+ "format": "raw",
1045
+ "nbytes": 720896,
1046
+ "byteOffset": 18743296
1047
+ },
1048
+ {
1049
+ "name": "param_90",
1050
+ "shape": [
1051
+ 2048
1052
+ ],
1053
+ "dtype": "float16",
1054
+ "format": "raw",
1055
+ "nbytes": 4096,
1056
+ "byteOffset": 19464192
1057
+ },
1058
+ {
1059
+ "name": "param_91",
1060
+ "shape": [
1061
+ 2048
1062
+ ],
1063
+ "dtype": "float16",
1064
+ "format": "raw",
1065
+ "nbytes": 4096,
1066
+ "byteOffset": 19468288
1067
+ },
1068
+ {
1069
+ "name": "param_92",
1070
+ "shape": [
1071
+ 2560,
1072
+ 256
1073
+ ],
1074
+ "dtype": "uint32",
1075
+ "format": "raw",
1076
+ "nbytes": 2621440,
1077
+ "byteOffset": 19472384
1078
+ },
1079
+ {
1080
+ "name": "param_93",
1081
+ "shape": [
1082
+ 2560,
1083
+ 64
1084
+ ],
1085
+ "dtype": "float16",
1086
+ "format": "raw",
1087
+ "nbytes": 327680,
1088
+ "byteOffset": 22093824
1089
+ },
1090
+ {
1091
+ "name": "param_94",
1092
+ "shape": [
1093
+ 2048,
1094
+ 256
1095
+ ],
1096
+ "dtype": "uint32",
1097
+ "format": "raw",
1098
+ "nbytes": 2097152,
1099
+ "byteOffset": 22421504
1100
+ },
1101
+ {
1102
+ "name": "param_95",
1103
+ "shape": [
1104
+ 2048,
1105
+ 64
1106
+ ],
1107
+ "dtype": "float16",
1108
+ "format": "raw",
1109
+ "nbytes": 262144,
1110
+ "byteOffset": 24518656
1111
+ }
1112
+ ]
1113
+ },
1114
+ {
1115
+ "dataPath": "params_shard_10.bin",
1116
+ "format": "raw-shard",
1117
+ "nbytes": 24780800,
1118
+ "records": [
1119
+ {
1120
+ "name": "param_96",
1121
+ "shape": [
1122
+ 11264,
1123
+ 256
1124
+ ],
1125
+ "dtype": "uint32",
1126
+ "format": "raw",
1127
+ "nbytes": 11534336,
1128
+ "byteOffset": 0
1129
+ },
1130
+ {
1131
+ "name": "param_97",
1132
+ "shape": [
1133
+ 11264,
1134
+ 64
1135
+ ],
1136
+ "dtype": "float16",
1137
+ "format": "raw",
1138
+ "nbytes": 1441792,
1139
+ "byteOffset": 11534336
1140
+ },
1141
+ {
1142
+ "name": "param_98",
1143
+ "shape": [
1144
+ 2048,
1145
+ 704
1146
+ ],
1147
+ "dtype": "uint32",
1148
+ "format": "raw",
1149
+ "nbytes": 5767168,
1150
+ "byteOffset": 12976128
1151
+ },
1152
+ {
1153
+ "name": "param_99",
1154
+ "shape": [
1155
+ 2048,
1156
+ 176
1157
+ ],
1158
+ "dtype": "float16",
1159
+ "format": "raw",
1160
+ "nbytes": 720896,
1161
+ "byteOffset": 18743296
1162
+ },
1163
+ {
1164
+ "name": "param_100",
1165
+ "shape": [
1166
+ 2048
1167
+ ],
1168
+ "dtype": "float16",
1169
+ "format": "raw",
1170
+ "nbytes": 4096,
1171
+ "byteOffset": 19464192
1172
+ },
1173
+ {
1174
+ "name": "param_101",
1175
+ "shape": [
1176
+ 2048
1177
+ ],
1178
+ "dtype": "float16",
1179
+ "format": "raw",
1180
+ "nbytes": 4096,
1181
+ "byteOffset": 19468288
1182
+ },
1183
+ {
1184
+ "name": "param_102",
1185
+ "shape": [
1186
+ 2560,
1187
+ 256
1188
+ ],
1189
+ "dtype": "uint32",
1190
+ "format": "raw",
1191
+ "nbytes": 2621440,
1192
+ "byteOffset": 19472384
1193
+ },
1194
+ {
1195
+ "name": "param_103",
1196
+ "shape": [
1197
+ 2560,
1198
+ 64
1199
+ ],
1200
+ "dtype": "float16",
1201
+ "format": "raw",
1202
+ "nbytes": 327680,
1203
+ "byteOffset": 22093824
1204
+ },
1205
+ {
1206
+ "name": "param_104",
1207
+ "shape": [
1208
+ 2048,
1209
+ 256
1210
+ ],
1211
+ "dtype": "uint32",
1212
+ "format": "raw",
1213
+ "nbytes": 2097152,
1214
+ "byteOffset": 22421504
1215
+ },
1216
+ {
1217
+ "name": "param_105",
1218
+ "shape": [
1219
+ 2048,
1220
+ 64
1221
+ ],
1222
+ "dtype": "float16",
1223
+ "format": "raw",
1224
+ "nbytes": 262144,
1225
+ "byteOffset": 24518656
1226
+ }
1227
+ ]
1228
+ },
1229
+ {
1230
+ "dataPath": "params_shard_11.bin",
1231
+ "format": "raw-shard",
1232
+ "nbytes": 24780800,
1233
+ "records": [
1234
+ {
1235
+ "name": "param_106",
1236
+ "shape": [
1237
+ 11264,
1238
+ 256
1239
+ ],
1240
+ "dtype": "uint32",
1241
+ "format": "raw",
1242
+ "nbytes": 11534336,
1243
+ "byteOffset": 0
1244
+ },
1245
+ {
1246
+ "name": "param_107",
1247
+ "shape": [
1248
+ 11264,
1249
+ 64
1250
+ ],
1251
+ "dtype": "float16",
1252
+ "format": "raw",
1253
+ "nbytes": 1441792,
1254
+ "byteOffset": 11534336
1255
+ },
1256
+ {
1257
+ "name": "param_108",
1258
+ "shape": [
1259
+ 2048,
1260
+ 704
1261
+ ],
1262
+ "dtype": "uint32",
1263
+ "format": "raw",
1264
+ "nbytes": 5767168,
1265
+ "byteOffset": 12976128
1266
+ },
1267
+ {
1268
+ "name": "param_109",
1269
+ "shape": [
1270
+ 2048,
1271
+ 176
1272
+ ],
1273
+ "dtype": "float16",
1274
+ "format": "raw",
1275
+ "nbytes": 720896,
1276
+ "byteOffset": 18743296
1277
+ },
1278
+ {
1279
+ "name": "param_110",
1280
+ "shape": [
1281
+ 2048
1282
+ ],
1283
+ "dtype": "float16",
1284
+ "format": "raw",
1285
+ "nbytes": 4096,
1286
+ "byteOffset": 19464192
1287
+ },
1288
+ {
1289
+ "name": "param_111",
1290
+ "shape": [
1291
+ 2048
1292
+ ],
1293
+ "dtype": "float16",
1294
+ "format": "raw",
1295
+ "nbytes": 4096,
1296
+ "byteOffset": 19468288
1297
+ },
1298
+ {
1299
+ "name": "param_112",
1300
+ "shape": [
1301
+ 2560,
1302
+ 256
1303
+ ],
1304
+ "dtype": "uint32",
1305
+ "format": "raw",
1306
+ "nbytes": 2621440,
1307
+ "byteOffset": 19472384
1308
+ },
1309
+ {
1310
+ "name": "param_113",
1311
+ "shape": [
1312
+ 2560,
1313
+ 64
1314
+ ],
1315
+ "dtype": "float16",
1316
+ "format": "raw",
1317
+ "nbytes": 327680,
1318
+ "byteOffset": 22093824
1319
+ },
1320
+ {
1321
+ "name": "param_114",
1322
+ "shape": [
1323
+ 2048,
1324
+ 256
1325
+ ],
1326
+ "dtype": "uint32",
1327
+ "format": "raw",
1328
+ "nbytes": 2097152,
1329
+ "byteOffset": 22421504
1330
+ },
1331
+ {
1332
+ "name": "param_115",
1333
+ "shape": [
1334
+ 2048,
1335
+ 64
1336
+ ],
1337
+ "dtype": "float16",
1338
+ "format": "raw",
1339
+ "nbytes": 262144,
1340
+ "byteOffset": 24518656
1341
+ }
1342
+ ]
1343
+ },
1344
+ {
1345
+ "dataPath": "params_shard_12.bin",
1346
+ "format": "raw-shard",
1347
+ "nbytes": 24780800,
1348
+ "records": [
1349
+ {
1350
+ "name": "param_116",
1351
+ "shape": [
1352
+ 11264,
1353
+ 256
1354
+ ],
1355
+ "dtype": "uint32",
1356
+ "format": "raw",
1357
+ "nbytes": 11534336,
1358
+ "byteOffset": 0
1359
+ },
1360
+ {
1361
+ "name": "param_117",
1362
+ "shape": [
1363
+ 11264,
1364
+ 64
1365
+ ],
1366
+ "dtype": "float16",
1367
+ "format": "raw",
1368
+ "nbytes": 1441792,
1369
+ "byteOffset": 11534336
1370
+ },
1371
+ {
1372
+ "name": "param_118",
1373
+ "shape": [
1374
+ 2048,
1375
+ 704
1376
+ ],
1377
+ "dtype": "uint32",
1378
+ "format": "raw",
1379
+ "nbytes": 5767168,
1380
+ "byteOffset": 12976128
1381
+ },
1382
+ {
1383
+ "name": "param_119",
1384
+ "shape": [
1385
+ 2048,
1386
+ 176
1387
+ ],
1388
+ "dtype": "float16",
1389
+ "format": "raw",
1390
+ "nbytes": 720896,
1391
+ "byteOffset": 18743296
1392
+ },
1393
+ {
1394
+ "name": "param_120",
1395
+ "shape": [
1396
+ 2048
1397
+ ],
1398
+ "dtype": "float16",
1399
+ "format": "raw",
1400
+ "nbytes": 4096,
1401
+ "byteOffset": 19464192
1402
+ },
1403
+ {
1404
+ "name": "param_121",
1405
+ "shape": [
1406
+ 2048
1407
+ ],
1408
+ "dtype": "float16",
1409
+ "format": "raw",
1410
+ "nbytes": 4096,
1411
+ "byteOffset": 19468288
1412
+ },
1413
+ {
1414
+ "name": "param_122",
1415
+ "shape": [
1416
+ 2560,
1417
+ 256
1418
+ ],
1419
+ "dtype": "uint32",
1420
+ "format": "raw",
1421
+ "nbytes": 2621440,
1422
+ "byteOffset": 19472384
1423
+ },
1424
+ {
1425
+ "name": "param_123",
1426
+ "shape": [
1427
+ 2560,
1428
+ 64
1429
+ ],
1430
+ "dtype": "float16",
1431
+ "format": "raw",
1432
+ "nbytes": 327680,
1433
+ "byteOffset": 22093824
1434
+ },
1435
+ {
1436
+ "name": "param_124",
1437
+ "shape": [
1438
+ 2048,
1439
+ 256
1440
+ ],
1441
+ "dtype": "uint32",
1442
+ "format": "raw",
1443
+ "nbytes": 2097152,
1444
+ "byteOffset": 22421504
1445
+ },
1446
+ {
1447
+ "name": "param_125",
1448
+ "shape": [
1449
+ 2048,
1450
+ 64
1451
+ ],
1452
+ "dtype": "float16",
1453
+ "format": "raw",
1454
+ "nbytes": 262144,
1455
+ "byteOffset": 24518656
1456
+ }
1457
+ ]
1458
+ },
1459
+ {
1460
+ "dataPath": "params_shard_13.bin",
1461
+ "format": "raw-shard",
1462
+ "nbytes": 24780800,
1463
+ "records": [
1464
+ {
1465
+ "name": "param_126",
1466
+ "shape": [
1467
+ 11264,
1468
+ 256
1469
+ ],
1470
+ "dtype": "uint32",
1471
+ "format": "raw",
1472
+ "nbytes": 11534336,
1473
+ "byteOffset": 0
1474
+ },
1475
+ {
1476
+ "name": "param_127",
1477
+ "shape": [
1478
+ 11264,
1479
+ 64
1480
+ ],
1481
+ "dtype": "float16",
1482
+ "format": "raw",
1483
+ "nbytes": 1441792,
1484
+ "byteOffset": 11534336
1485
+ },
1486
+ {
1487
+ "name": "param_128",
1488
+ "shape": [
1489
+ 2048,
1490
+ 704
1491
+ ],
1492
+ "dtype": "uint32",
1493
+ "format": "raw",
1494
+ "nbytes": 5767168,
1495
+ "byteOffset": 12976128
1496
+ },
1497
+ {
1498
+ "name": "param_129",
1499
+ "shape": [
1500
+ 2048,
1501
+ 176
1502
+ ],
1503
+ "dtype": "float16",
1504
+ "format": "raw",
1505
+ "nbytes": 720896,
1506
+ "byteOffset": 18743296
1507
+ },
1508
+ {
1509
+ "name": "param_130",
1510
+ "shape": [
1511
+ 2048
1512
+ ],
1513
+ "dtype": "float16",
1514
+ "format": "raw",
1515
+ "nbytes": 4096,
1516
+ "byteOffset": 19464192
1517
+ },
1518
+ {
1519
+ "name": "param_131",
1520
+ "shape": [
1521
+ 2048
1522
+ ],
1523
+ "dtype": "float16",
1524
+ "format": "raw",
1525
+ "nbytes": 4096,
1526
+ "byteOffset": 19468288
1527
+ },
1528
+ {
1529
+ "name": "param_132",
1530
+ "shape": [
1531
+ 2560,
1532
+ 256
1533
+ ],
1534
+ "dtype": "uint32",
1535
+ "format": "raw",
1536
+ "nbytes": 2621440,
1537
+ "byteOffset": 19472384
1538
+ },
1539
+ {
1540
+ "name": "param_133",
1541
+ "shape": [
1542
+ 2560,
1543
+ 64
1544
+ ],
1545
+ "dtype": "float16",
1546
+ "format": "raw",
1547
+ "nbytes": 327680,
1548
+ "byteOffset": 22093824
1549
+ },
1550
+ {
1551
+ "name": "param_134",
1552
+ "shape": [
1553
+ 2048,
1554
+ 256
1555
+ ],
1556
+ "dtype": "uint32",
1557
+ "format": "raw",
1558
+ "nbytes": 2097152,
1559
+ "byteOffset": 22421504
1560
+ },
1561
+ {
1562
+ "name": "param_135",
1563
+ "shape": [
1564
+ 2048,
1565
+ 64
1566
+ ],
1567
+ "dtype": "float16",
1568
+ "format": "raw",
1569
+ "nbytes": 262144,
1570
+ "byteOffset": 24518656
1571
+ }
1572
+ ]
1573
+ },
1574
+ {
1575
+ "dataPath": "params_shard_14.bin",
1576
+ "format": "raw-shard",
1577
+ "nbytes": 24780800,
1578
+ "records": [
1579
+ {
1580
+ "name": "param_136",
1581
+ "shape": [
1582
+ 11264,
1583
+ 256
1584
+ ],
1585
+ "dtype": "uint32",
1586
+ "format": "raw",
1587
+ "nbytes": 11534336,
1588
+ "byteOffset": 0
1589
+ },
1590
+ {
1591
+ "name": "param_137",
1592
+ "shape": [
1593
+ 11264,
1594
+ 64
1595
+ ],
1596
+ "dtype": "float16",
1597
+ "format": "raw",
1598
+ "nbytes": 1441792,
1599
+ "byteOffset": 11534336
1600
+ },
1601
+ {
1602
+ "name": "param_138",
1603
+ "shape": [
1604
+ 2048,
1605
+ 704
1606
+ ],
1607
+ "dtype": "uint32",
1608
+ "format": "raw",
1609
+ "nbytes": 5767168,
1610
+ "byteOffset": 12976128
1611
+ },
1612
+ {
1613
+ "name": "param_139",
1614
+ "shape": [
1615
+ 2048,
1616
+ 176
1617
+ ],
1618
+ "dtype": "float16",
1619
+ "format": "raw",
1620
+ "nbytes": 720896,
1621
+ "byteOffset": 18743296
1622
+ },
1623
+ {
1624
+ "name": "param_140",
1625
+ "shape": [
1626
+ 2048
1627
+ ],
1628
+ "dtype": "float16",
1629
+ "format": "raw",
1630
+ "nbytes": 4096,
1631
+ "byteOffset": 19464192
1632
+ },
1633
+ {
1634
+ "name": "param_141",
1635
+ "shape": [
1636
+ 2048
1637
+ ],
1638
+ "dtype": "float16",
1639
+ "format": "raw",
1640
+ "nbytes": 4096,
1641
+ "byteOffset": 19468288
1642
+ },
1643
+ {
1644
+ "name": "param_142",
1645
+ "shape": [
1646
+ 2560,
1647
+ 256
1648
+ ],
1649
+ "dtype": "uint32",
1650
+ "format": "raw",
1651
+ "nbytes": 2621440,
1652
+ "byteOffset": 19472384
1653
+ },
1654
+ {
1655
+ "name": "param_143",
1656
+ "shape": [
1657
+ 2560,
1658
+ 64
1659
+ ],
1660
+ "dtype": "float16",
1661
+ "format": "raw",
1662
+ "nbytes": 327680,
1663
+ "byteOffset": 22093824
1664
+ },
1665
+ {
1666
+ "name": "param_144",
1667
+ "shape": [
1668
+ 2048,
1669
+ 256
1670
+ ],
1671
+ "dtype": "uint32",
1672
+ "format": "raw",
1673
+ "nbytes": 2097152,
1674
+ "byteOffset": 22421504
1675
+ },
1676
+ {
1677
+ "name": "param_145",
1678
+ "shape": [
1679
+ 2048,
1680
+ 64
1681
+ ],
1682
+ "dtype": "float16",
1683
+ "format": "raw",
1684
+ "nbytes": 262144,
1685
+ "byteOffset": 24518656
1686
+ }
1687
+ ]
1688
+ },
1689
+ {
1690
+ "dataPath": "params_shard_15.bin",
1691
+ "format": "raw-shard",
1692
+ "nbytes": 24780800,
1693
+ "records": [
1694
+ {
1695
+ "name": "param_146",
1696
+ "shape": [
1697
+ 11264,
1698
+ 256
1699
+ ],
1700
+ "dtype": "uint32",
1701
+ "format": "raw",
1702
+ "nbytes": 11534336,
1703
+ "byteOffset": 0
1704
+ },
1705
+ {
1706
+ "name": "param_147",
1707
+ "shape": [
1708
+ 11264,
1709
+ 64
1710
+ ],
1711
+ "dtype": "float16",
1712
+ "format": "raw",
1713
+ "nbytes": 1441792,
1714
+ "byteOffset": 11534336
1715
+ },
1716
+ {
1717
+ "name": "param_148",
1718
+ "shape": [
1719
+ 2048,
1720
+ 704
1721
+ ],
1722
+ "dtype": "uint32",
1723
+ "format": "raw",
1724
+ "nbytes": 5767168,
1725
+ "byteOffset": 12976128
1726
+ },
1727
+ {
1728
+ "name": "param_149",
1729
+ "shape": [
1730
+ 2048,
1731
+ 176
1732
+ ],
1733
+ "dtype": "float16",
1734
+ "format": "raw",
1735
+ "nbytes": 720896,
1736
+ "byteOffset": 18743296
1737
+ },
1738
+ {
1739
+ "name": "param_150",
1740
+ "shape": [
1741
+ 2048
1742
+ ],
1743
+ "dtype": "float16",
1744
+ "format": "raw",
1745
+ "nbytes": 4096,
1746
+ "byteOffset": 19464192
1747
+ },
1748
+ {
1749
+ "name": "param_151",
1750
+ "shape": [
1751
+ 2048
1752
+ ],
1753
+ "dtype": "float16",
1754
+ "format": "raw",
1755
+ "nbytes": 4096,
1756
+ "byteOffset": 19468288
1757
+ },
1758
+ {
1759
+ "name": "param_152",
1760
+ "shape": [
1761
+ 2560,
1762
+ 256
1763
+ ],
1764
+ "dtype": "uint32",
1765
+ "format": "raw",
1766
+ "nbytes": 2621440,
1767
+ "byteOffset": 19472384
1768
+ },
1769
+ {
1770
+ "name": "param_153",
1771
+ "shape": [
1772
+ 2560,
1773
+ 64
1774
+ ],
1775
+ "dtype": "float16",
1776
+ "format": "raw",
1777
+ "nbytes": 327680,
1778
+ "byteOffset": 22093824
1779
+ },
1780
+ {
1781
+ "name": "param_154",
1782
+ "shape": [
1783
+ 2048,
1784
+ 256
1785
+ ],
1786
+ "dtype": "uint32",
1787
+ "format": "raw",
1788
+ "nbytes": 2097152,
1789
+ "byteOffset": 22421504
1790
+ },
1791
+ {
1792
+ "name": "param_155",
1793
+ "shape": [
1794
+ 2048,
1795
+ 64
1796
+ ],
1797
+ "dtype": "float16",
1798
+ "format": "raw",
1799
+ "nbytes": 262144,
1800
+ "byteOffset": 24518656
1801
+ }
1802
+ ]
1803
+ },
1804
+ {
1805
+ "dataPath": "params_shard_16.bin",
1806
+ "format": "raw-shard",
1807
+ "nbytes": 24780800,
1808
+ "records": [
1809
+ {
1810
+ "name": "param_156",
1811
+ "shape": [
1812
+ 11264,
1813
+ 256
1814
+ ],
1815
+ "dtype": "uint32",
1816
+ "format": "raw",
1817
+ "nbytes": 11534336,
1818
+ "byteOffset": 0
1819
+ },
1820
+ {
1821
+ "name": "param_157",
1822
+ "shape": [
1823
+ 11264,
1824
+ 64
1825
+ ],
1826
+ "dtype": "float16",
1827
+ "format": "raw",
1828
+ "nbytes": 1441792,
1829
+ "byteOffset": 11534336
1830
+ },
1831
+ {
1832
+ "name": "param_158",
1833
+ "shape": [
1834
+ 2048,
1835
+ 704
1836
+ ],
1837
+ "dtype": "uint32",
1838
+ "format": "raw",
1839
+ "nbytes": 5767168,
1840
+ "byteOffset": 12976128
1841
+ },
1842
+ {
1843
+ "name": "param_159",
1844
+ "shape": [
1845
+ 2048,
1846
+ 176
1847
+ ],
1848
+ "dtype": "float16",
1849
+ "format": "raw",
1850
+ "nbytes": 720896,
1851
+ "byteOffset": 18743296
1852
+ },
1853
+ {
1854
+ "name": "param_160",
1855
+ "shape": [
1856
+ 2048
1857
+ ],
1858
+ "dtype": "float16",
1859
+ "format": "raw",
1860
+ "nbytes": 4096,
1861
+ "byteOffset": 19464192
1862
+ },
1863
+ {
1864
+ "name": "param_161",
1865
+ "shape": [
1866
+ 2048
1867
+ ],
1868
+ "dtype": "float16",
1869
+ "format": "raw",
1870
+ "nbytes": 4096,
1871
+ "byteOffset": 19468288
1872
+ },
1873
+ {
1874
+ "name": "param_162",
1875
+ "shape": [
1876
+ 2560,
1877
+ 256
1878
+ ],
1879
+ "dtype": "uint32",
1880
+ "format": "raw",
1881
+ "nbytes": 2621440,
1882
+ "byteOffset": 19472384
1883
+ },
1884
+ {
1885
+ "name": "param_163",
1886
+ "shape": [
1887
+ 2560,
1888
+ 64
1889
+ ],
1890
+ "dtype": "float16",
1891
+ "format": "raw",
1892
+ "nbytes": 327680,
1893
+ "byteOffset": 22093824
1894
+ },
1895
+ {
1896
+ "name": "param_164",
1897
+ "shape": [
1898
+ 2048,
1899
+ 256
1900
+ ],
1901
+ "dtype": "uint32",
1902
+ "format": "raw",
1903
+ "nbytes": 2097152,
1904
+ "byteOffset": 22421504
1905
+ },
1906
+ {
1907
+ "name": "param_165",
1908
+ "shape": [
1909
+ 2048,
1910
+ 64
1911
+ ],
1912
+ "dtype": "float16",
1913
+ "format": "raw",
1914
+ "nbytes": 262144,
1915
+ "byteOffset": 24518656
1916
+ }
1917
+ ]
1918
+ },
1919
+ {
1920
+ "dataPath": "params_shard_17.bin",
1921
+ "format": "raw-shard",
1922
+ "nbytes": 24780800,
1923
+ "records": [
1924
+ {
1925
+ "name": "param_166",
1926
+ "shape": [
1927
+ 11264,
1928
+ 256
1929
+ ],
1930
+ "dtype": "uint32",
1931
+ "format": "raw",
1932
+ "nbytes": 11534336,
1933
+ "byteOffset": 0
1934
+ },
1935
+ {
1936
+ "name": "param_167",
1937
+ "shape": [
1938
+ 11264,
1939
+ 64
1940
+ ],
1941
+ "dtype": "float16",
1942
+ "format": "raw",
1943
+ "nbytes": 1441792,
1944
+ "byteOffset": 11534336
1945
+ },
1946
+ {
1947
+ "name": "param_168",
1948
+ "shape": [
1949
+ 2048,
1950
+ 704
1951
+ ],
1952
+ "dtype": "uint32",
1953
+ "format": "raw",
1954
+ "nbytes": 5767168,
1955
+ "byteOffset": 12976128
1956
+ },
1957
+ {
1958
+ "name": "param_169",
1959
+ "shape": [
1960
+ 2048,
1961
+ 176
1962
+ ],
1963
+ "dtype": "float16",
1964
+ "format": "raw",
1965
+ "nbytes": 720896,
1966
+ "byteOffset": 18743296
1967
+ },
1968
+ {
1969
+ "name": "param_170",
1970
+ "shape": [
1971
+ 2048
1972
+ ],
1973
+ "dtype": "float16",
1974
+ "format": "raw",
1975
+ "nbytes": 4096,
1976
+ "byteOffset": 19464192
1977
+ },
1978
+ {
1979
+ "name": "param_171",
1980
+ "shape": [
1981
+ 2048
1982
+ ],
1983
+ "dtype": "float16",
1984
+ "format": "raw",
1985
+ "nbytes": 4096,
1986
+ "byteOffset": 19468288
1987
+ },
1988
+ {
1989
+ "name": "param_172",
1990
+ "shape": [
1991
+ 2560,
1992
+ 256
1993
+ ],
1994
+ "dtype": "uint32",
1995
+ "format": "raw",
1996
+ "nbytes": 2621440,
1997
+ "byteOffset": 19472384
1998
+ },
1999
+ {
2000
+ "name": "param_173",
2001
+ "shape": [
2002
+ 2560,
2003
+ 64
2004
+ ],
2005
+ "dtype": "float16",
2006
+ "format": "raw",
2007
+ "nbytes": 327680,
2008
+ "byteOffset": 22093824
2009
+ },
2010
+ {
2011
+ "name": "param_174",
2012
+ "shape": [
2013
+ 2048,
2014
+ 256
2015
+ ],
2016
+ "dtype": "uint32",
2017
+ "format": "raw",
2018
+ "nbytes": 2097152,
2019
+ "byteOffset": 22421504
2020
+ },
2021
+ {
2022
+ "name": "param_175",
2023
+ "shape": [
2024
+ 2048,
2025
+ 64
2026
+ ],
2027
+ "dtype": "float16",
2028
+ "format": "raw",
2029
+ "nbytes": 262144,
2030
+ "byteOffset": 24518656
2031
+ }
2032
+ ]
2033
+ },
2034
+ {
2035
+ "dataPath": "params_shard_18.bin",
2036
+ "format": "raw-shard",
2037
+ "nbytes": 24780800,
2038
+ "records": [
2039
+ {
2040
+ "name": "param_176",
2041
+ "shape": [
2042
+ 11264,
2043
+ 256
2044
+ ],
2045
+ "dtype": "uint32",
2046
+ "format": "raw",
2047
+ "nbytes": 11534336,
2048
+ "byteOffset": 0
2049
+ },
2050
+ {
2051
+ "name": "param_177",
2052
+ "shape": [
2053
+ 11264,
2054
+ 64
2055
+ ],
2056
+ "dtype": "float16",
2057
+ "format": "raw",
2058
+ "nbytes": 1441792,
2059
+ "byteOffset": 11534336
2060
+ },
2061
+ {
2062
+ "name": "param_178",
2063
+ "shape": [
2064
+ 2048,
2065
+ 704
2066
+ ],
2067
+ "dtype": "uint32",
2068
+ "format": "raw",
2069
+ "nbytes": 5767168,
2070
+ "byteOffset": 12976128
2071
+ },
2072
+ {
2073
+ "name": "param_179",
2074
+ "shape": [
2075
+ 2048,
2076
+ 176
2077
+ ],
2078
+ "dtype": "float16",
2079
+ "format": "raw",
2080
+ "nbytes": 720896,
2081
+ "byteOffset": 18743296
2082
+ },
2083
+ {
2084
+ "name": "param_180",
2085
+ "shape": [
2086
+ 2048
2087
+ ],
2088
+ "dtype": "float16",
2089
+ "format": "raw",
2090
+ "nbytes": 4096,
2091
+ "byteOffset": 19464192
2092
+ },
2093
+ {
2094
+ "name": "param_181",
2095
+ "shape": [
2096
+ 2048
2097
+ ],
2098
+ "dtype": "float16",
2099
+ "format": "raw",
2100
+ "nbytes": 4096,
2101
+ "byteOffset": 19468288
2102
+ },
2103
+ {
2104
+ "name": "param_182",
2105
+ "shape": [
2106
+ 2560,
2107
+ 256
2108
+ ],
2109
+ "dtype": "uint32",
2110
+ "format": "raw",
2111
+ "nbytes": 2621440,
2112
+ "byteOffset": 19472384
2113
+ },
2114
+ {
2115
+ "name": "param_183",
2116
+ "shape": [
2117
+ 2560,
2118
+ 64
2119
+ ],
2120
+ "dtype": "float16",
2121
+ "format": "raw",
2122
+ "nbytes": 327680,
2123
+ "byteOffset": 22093824
2124
+ },
2125
+ {
2126
+ "name": "param_184",
2127
+ "shape": [
2128
+ 2048,
2129
+ 256
2130
+ ],
2131
+ "dtype": "uint32",
2132
+ "format": "raw",
2133
+ "nbytes": 2097152,
2134
+ "byteOffset": 22421504
2135
+ },
2136
+ {
2137
+ "name": "param_185",
2138
+ "shape": [
2139
+ 2048,
2140
+ 64
2141
+ ],
2142
+ "dtype": "float16",
2143
+ "format": "raw",
2144
+ "nbytes": 262144,
2145
+ "byteOffset": 24518656
2146
+ }
2147
+ ]
2148
+ },
2149
+ {
2150
+ "dataPath": "params_shard_19.bin",
2151
+ "format": "raw-shard",
2152
+ "nbytes": 24780800,
2153
+ "records": [
2154
+ {
2155
+ "name": "param_186",
2156
+ "shape": [
2157
+ 11264,
2158
+ 256
2159
+ ],
2160
+ "dtype": "uint32",
2161
+ "format": "raw",
2162
+ "nbytes": 11534336,
2163
+ "byteOffset": 0
2164
+ },
2165
+ {
2166
+ "name": "param_187",
2167
+ "shape": [
2168
+ 11264,
2169
+ 64
2170
+ ],
2171
+ "dtype": "float16",
2172
+ "format": "raw",
2173
+ "nbytes": 1441792,
2174
+ "byteOffset": 11534336
2175
+ },
2176
+ {
2177
+ "name": "param_188",
2178
+ "shape": [
2179
+ 2048,
2180
+ 704
2181
+ ],
2182
+ "dtype": "uint32",
2183
+ "format": "raw",
2184
+ "nbytes": 5767168,
2185
+ "byteOffset": 12976128
2186
+ },
2187
+ {
2188
+ "name": "param_189",
2189
+ "shape": [
2190
+ 2048,
2191
+ 176
2192
+ ],
2193
+ "dtype": "float16",
2194
+ "format": "raw",
2195
+ "nbytes": 720896,
2196
+ "byteOffset": 18743296
2197
+ },
2198
+ {
2199
+ "name": "param_190",
2200
+ "shape": [
2201
+ 2048
2202
+ ],
2203
+ "dtype": "float16",
2204
+ "format": "raw",
2205
+ "nbytes": 4096,
2206
+ "byteOffset": 19464192
2207
+ },
2208
+ {
2209
+ "name": "param_191",
2210
+ "shape": [
2211
+ 2048
2212
+ ],
2213
+ "dtype": "float16",
2214
+ "format": "raw",
2215
+ "nbytes": 4096,
2216
+ "byteOffset": 19468288
2217
+ },
2218
+ {
2219
+ "name": "param_192",
2220
+ "shape": [
2221
+ 2560,
2222
+ 256
2223
+ ],
2224
+ "dtype": "uint32",
2225
+ "format": "raw",
2226
+ "nbytes": 2621440,
2227
+ "byteOffset": 19472384
2228
+ },
2229
+ {
2230
+ "name": "param_193",
2231
+ "shape": [
2232
+ 2560,
2233
+ 64
2234
+ ],
2235
+ "dtype": "float16",
2236
+ "format": "raw",
2237
+ "nbytes": 327680,
2238
+ "byteOffset": 22093824
2239
+ },
2240
+ {
2241
+ "name": "param_194",
2242
+ "shape": [
2243
+ 2048,
2244
+ 256
2245
+ ],
2246
+ "dtype": "uint32",
2247
+ "format": "raw",
2248
+ "nbytes": 2097152,
2249
+ "byteOffset": 22421504
2250
+ },
2251
+ {
2252
+ "name": "param_195",
2253
+ "shape": [
2254
+ 2048,
2255
+ 64
2256
+ ],
2257
+ "dtype": "float16",
2258
+ "format": "raw",
2259
+ "nbytes": 262144,
2260
+ "byteOffset": 24518656
2261
+ }
2262
+ ]
2263
+ },
2264
+ {
2265
+ "dataPath": "params_shard_20.bin",
2266
+ "format": "raw-shard",
2267
+ "nbytes": 24780800,
2268
+ "records": [
2269
+ {
2270
+ "name": "param_196",
2271
+ "shape": [
2272
+ 11264,
2273
+ 256
2274
+ ],
2275
+ "dtype": "uint32",
2276
+ "format": "raw",
2277
+ "nbytes": 11534336,
2278
+ "byteOffset": 0
2279
+ },
2280
+ {
2281
+ "name": "param_197",
2282
+ "shape": [
2283
+ 11264,
2284
+ 64
2285
+ ],
2286
+ "dtype": "float16",
2287
+ "format": "raw",
2288
+ "nbytes": 1441792,
2289
+ "byteOffset": 11534336
2290
+ },
2291
+ {
2292
+ "name": "param_198",
2293
+ "shape": [
2294
+ 2048,
2295
+ 704
2296
+ ],
2297
+ "dtype": "uint32",
2298
+ "format": "raw",
2299
+ "nbytes": 5767168,
2300
+ "byteOffset": 12976128
2301
+ },
2302
+ {
2303
+ "name": "param_199",
2304
+ "shape": [
2305
+ 2048,
2306
+ 176
2307
+ ],
2308
+ "dtype": "float16",
2309
+ "format": "raw",
2310
+ "nbytes": 720896,
2311
+ "byteOffset": 18743296
2312
+ },
2313
+ {
2314
+ "name": "param_200",
2315
+ "shape": [
2316
+ 2048
2317
+ ],
2318
+ "dtype": "float16",
2319
+ "format": "raw",
2320
+ "nbytes": 4096,
2321
+ "byteOffset": 19464192
2322
+ },
2323
+ {
2324
+ "name": "param_201",
2325
+ "shape": [
2326
+ 2048
2327
+ ],
2328
+ "dtype": "float16",
2329
+ "format": "raw",
2330
+ "nbytes": 4096,
2331
+ "byteOffset": 19468288
2332
+ },
2333
+ {
2334
+ "name": "param_202",
2335
+ "shape": [
2336
+ 2560,
2337
+ 256
2338
+ ],
2339
+ "dtype": "uint32",
2340
+ "format": "raw",
2341
+ "nbytes": 2621440,
2342
+ "byteOffset": 19472384
2343
+ },
2344
+ {
2345
+ "name": "param_203",
2346
+ "shape": [
2347
+ 2560,
2348
+ 64
2349
+ ],
2350
+ "dtype": "float16",
2351
+ "format": "raw",
2352
+ "nbytes": 327680,
2353
+ "byteOffset": 22093824
2354
+ },
2355
+ {
2356
+ "name": "param_204",
2357
+ "shape": [
2358
+ 2048,
2359
+ 256
2360
+ ],
2361
+ "dtype": "uint32",
2362
+ "format": "raw",
2363
+ "nbytes": 2097152,
2364
+ "byteOffset": 22421504
2365
+ },
2366
+ {
2367
+ "name": "param_205",
2368
+ "shape": [
2369
+ 2048,
2370
+ 64
2371
+ ],
2372
+ "dtype": "float16",
2373
+ "format": "raw",
2374
+ "nbytes": 262144,
2375
+ "byteOffset": 24518656
2376
+ }
2377
+ ]
2378
+ },
2379
+ {
2380
+ "dataPath": "params_shard_21.bin",
2381
+ "format": "raw-shard",
2382
+ "nbytes": 24780800,
2383
+ "records": [
2384
+ {
2385
+ "name": "param_206",
2386
+ "shape": [
2387
+ 11264,
2388
+ 256
2389
+ ],
2390
+ "dtype": "uint32",
2391
+ "format": "raw",
2392
+ "nbytes": 11534336,
2393
+ "byteOffset": 0
2394
+ },
2395
+ {
2396
+ "name": "param_207",
2397
+ "shape": [
2398
+ 11264,
2399
+ 64
2400
+ ],
2401
+ "dtype": "float16",
2402
+ "format": "raw",
2403
+ "nbytes": 1441792,
2404
+ "byteOffset": 11534336
2405
+ },
2406
+ {
2407
+ "name": "param_208",
2408
+ "shape": [
2409
+ 2048,
2410
+ 704
2411
+ ],
2412
+ "dtype": "uint32",
2413
+ "format": "raw",
2414
+ "nbytes": 5767168,
2415
+ "byteOffset": 12976128
2416
+ },
2417
+ {
2418
+ "name": "param_209",
2419
+ "shape": [
2420
+ 2048,
2421
+ 176
2422
+ ],
2423
+ "dtype": "float16",
2424
+ "format": "raw",
2425
+ "nbytes": 720896,
2426
+ "byteOffset": 18743296
2427
+ },
2428
+ {
2429
+ "name": "param_210",
2430
+ "shape": [
2431
+ 2048
2432
+ ],
2433
+ "dtype": "float16",
2434
+ "format": "raw",
2435
+ "nbytes": 4096,
2436
+ "byteOffset": 19464192
2437
+ },
2438
+ {
2439
+ "name": "param_211",
2440
+ "shape": [
2441
+ 2048
2442
+ ],
2443
+ "dtype": "float16",
2444
+ "format": "raw",
2445
+ "nbytes": 4096,
2446
+ "byteOffset": 19468288
2447
+ },
2448
+ {
2449
+ "name": "param_212",
2450
+ "shape": [
2451
+ 2560,
2452
+ 256
2453
+ ],
2454
+ "dtype": "uint32",
2455
+ "format": "raw",
2456
+ "nbytes": 2621440,
2457
+ "byteOffset": 19472384
2458
+ },
2459
+ {
2460
+ "name": "param_213",
2461
+ "shape": [
2462
+ 2560,
2463
+ 64
2464
+ ],
2465
+ "dtype": "float16",
2466
+ "format": "raw",
2467
+ "nbytes": 327680,
2468
+ "byteOffset": 22093824
2469
+ },
2470
+ {
2471
+ "name": "param_214",
2472
+ "shape": [
2473
+ 2048,
2474
+ 256
2475
+ ],
2476
+ "dtype": "uint32",
2477
+ "format": "raw",
2478
+ "nbytes": 2097152,
2479
+ "byteOffset": 22421504
2480
+ },
2481
+ {
2482
+ "name": "param_215",
2483
+ "shape": [
2484
+ 2048,
2485
+ 64
2486
+ ],
2487
+ "dtype": "float16",
2488
+ "format": "raw",
2489
+ "nbytes": 262144,
2490
+ "byteOffset": 24518656
2491
+ }
2492
+ ]
2493
+ },
2494
+ {
2495
+ "dataPath": "params_shard_22.bin",
2496
+ "format": "raw-shard",
2497
+ "nbytes": 32768000,
2498
+ "records": [
2499
+ {
2500
+ "name": "param_223",
2501
+ "shape": [
2502
+ 32000,
2503
+ 256
2504
+ ],
2505
+ "dtype": "uint32",
2506
+ "format": "raw",
2507
+ "nbytes": 32768000,
2508
+ "byteOffset": 0
2509
+ }
2510
+ ]
2511
+ },
2512
+ {
2513
+ "dataPath": "params_shard_23.bin",
2514
+ "format": "raw-shard",
2515
+ "nbytes": 24096768,
2516
+ "records": [
2517
+ {
2518
+ "name": "param_216",
2519
+ "shape": [
2520
+ 11264,
2521
+ 256
2522
+ ],
2523
+ "dtype": "uint32",
2524
+ "format": "raw",
2525
+ "nbytes": 11534336,
2526
+ "byteOffset": 0
2527
+ },
2528
+ {
2529
+ "name": "param_217",
2530
+ "shape": [
2531
+ 11264,
2532
+ 64
2533
+ ],
2534
+ "dtype": "float16",
2535
+ "format": "raw",
2536
+ "nbytes": 1441792,
2537
+ "byteOffset": 11534336
2538
+ },
2539
+ {
2540
+ "name": "param_218",
2541
+ "shape": [
2542
+ 2048,
2543
+ 704
2544
+ ],
2545
+ "dtype": "uint32",
2546
+ "format": "raw",
2547
+ "nbytes": 5767168,
2548
+ "byteOffset": 12976128
2549
+ },
2550
+ {
2551
+ "name": "param_219",
2552
+ "shape": [
2553
+ 2048,
2554
+ 176
2555
+ ],
2556
+ "dtype": "float16",
2557
+ "format": "raw",
2558
+ "nbytes": 720896,
2559
+ "byteOffset": 18743296
2560
+ },
2561
+ {
2562
+ "name": "param_220",
2563
+ "shape": [
2564
+ 2048
2565
+ ],
2566
+ "dtype": "float16",
2567
+ "format": "raw",
2568
+ "nbytes": 4096,
2569
+ "byteOffset": 19464192
2570
+ },
2571
+ {
2572
+ "name": "param_221",
2573
+ "shape": [
2574
+ 2048
2575
+ ],
2576
+ "dtype": "float16",
2577
+ "format": "raw",
2578
+ "nbytes": 4096,
2579
+ "byteOffset": 19468288
2580
+ },
2581
+ {
2582
+ "name": "param_222",
2583
+ "shape": [
2584
+ 2048
2585
+ ],
2586
+ "dtype": "float16",
2587
+ "format": "raw",
2588
+ "nbytes": 4096,
2589
+ "byteOffset": 19472384
2590
+ },
2591
+ {
2592
+ "name": "param_224",
2593
+ "shape": [
2594
+ 32000,
2595
+ 64
2596
+ ],
2597
+ "dtype": "float16",
2598
+ "format": "raw",
2599
+ "nbytes": 4096000,
2600
+ "byteOffset": 19476480
2601
+ },
2602
+ {
2603
+ "name": "param_225",
2604
+ "shape": [
2605
+ 2048,
2606
+ 64
2607
+ ],
2608
+ "dtype": "float16",
2609
+ "format": "raw",
2610
+ "nbytes": 262144,
2611
+ "byteOffset": 23572480
2612
+ },
2613
+ {
2614
+ "name": "param_226",
2615
+ "shape": [
2616
+ 2048,
2617
+ 64
2618
+ ],
2619
+ "dtype": "float16",
2620
+ "format": "raw",
2621
+ "nbytes": 262144,
2622
+ "byteOffset": 23834624
2623
+ }
2624
+ ]
2625
+ }
2626
+ ]
2627
+ }
params_shard_0.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c62123f4857b06a8e7902e6e830b3129621fbca0305db11b87dbd6eb7a5becaa
3
+ size 32768000
params_shard_1.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a44ab8a2873f44022ba5fd1af589a0efe41fc60a7a49d39df68867bde03a1d21
3
+ size 31825920
params_shard_10.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:738ab6092b51e8078a5d47f52339fbb06871e2cc49600bcee978dd1cb81da43c
3
+ size 24780800
params_shard_11.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5ee05a7979fc1e1167d570b4e04497dff4bf0f5a0a185d4ec5962c8421de62d7
3
+ size 24780800
params_shard_12.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b2bf1437da70f499eb91e48754f48ed78b764534f200e5004dde0391bf553afe
3
+ size 24780800
params_shard_13.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f57fe02a694e1f64f1c77398889a106bba2c2f4224ccfd752f088a894578307e
3
+ size 24780800
params_shard_14.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dab5ce52dda38ae9332cd9136980f2f9e8c0463a73a5df5f051ce2b7dee26ce2
3
+ size 24780800
params_shard_15.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6b8205a42ae3c2680d92c480a68da192297884a8dbb4684d8201974483041c17
3
+ size 24780800
params_shard_16.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fffd60d00a7f9dd3a0b3522ab37d11e65e7830ef6bd81c519d57290b467ddca4
3
+ size 24780800
params_shard_17.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:46b79872a3ae6225f694209528941d80f34382de6a9bb28eb979fdd732d39ee2
3
+ size 24780800
params_shard_18.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e1f2c2e7c772330f911804f1426f29596c63af246d29b4be7434322cceaf6d41
3
+ size 24780800
params_shard_19.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ab1c10d418a298c7b9eeef89bc5f2e0ec175312d40930c19cd71ffdf89179686
3
+ size 24780800
params_shard_2.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:747d56ae7c7900fc76843cf578ce0a742d6024ffc8fedd3b5e088b552a4b2b54
3
+ size 27140096
params_shard_20.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:30cfb25d7bf2401844e373ffe5c7598b1dc3eccfbc050d4518c0d737d23f8db8
3
+ size 24780800
params_shard_21.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:144fbb584d46dcf610eb021fbae350d04e8486e7ea97999d001de9b6f38ae6a3
3
+ size 24780800
params_shard_22.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d8e36a476452c5c4d4f84242560f16f722b8f27ea757d4ec35c6e6477a5e61b9
3
+ size 32768000
params_shard_23.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dd9fc12d67a1f436981e0c3d886e0a42049c521e2258e9a9948948a10f023af3
3
+ size 24096768
params_shard_3.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6014dfbe7940e3c6b4eaa19a613e799f83b827400fa7c27ab1aa0c1048d2c1fc
3
+ size 24780800
params_shard_4.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e53e653d0a11c4505d5741aee1b18afa83278afee1544f128b7c8da9290ce148
3
+ size 24780800
params_shard_5.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:89b8d404f30a6676f7b78998c3d3fe13b1c1d0bd51537dd9c54999b0dbbea045
3
+ size 24780800
params_shard_6.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b577d93e96a05a263b114d5cf55443c3fb48398b499fd7d84a1e32f0b48b9a15
3
+ size 24780800
params_shard_7.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3f47fde0e0be2013f4d2ae3829c33aedccb860812e7239f60b2535743dcd2136
3
+ size 24780800
params_shard_8.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e45efbd5d50df42293ba11699e5a0837aad7d4ee25b829c8882a542e100709fc
3
+ size 24780800
params_shard_9.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d12dc7020a7489f32e9658a671567d7283436820b8aedbb1acd53e3c2d5b3058
3
+ size 24780800
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
3
+ size 499723
tokenizer_config.json ADDED
@@ -0,0 +1,33 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "__type": "AddedToken",
4
+ "content": "<s>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false
9
+ },
10
+ "clean_up_tokenization_spaces": false,
11
+ "eos_token": {
12
+ "__type": "AddedToken",
13
+ "content": "</s>",
14
+ "lstrip": false,
15
+ "normalized": false,
16
+ "rstrip": false,
17
+ "single_word": false
18
+ },
19
+ "legacy": false,
20
+ "model_max_length": 1000000000000000019884624838656,
21
+ "pad_token": null,
22
+ "padding_side": "right",
23
+ "sp_model_kwargs": {},
24
+ "tokenizer_class": "LlamaTokenizer",
25
+ "unk_token": {
26
+ "__type": "AddedToken",
27
+ "content": "<unk>",
28
+ "lstrip": false,
29
+ "normalized": false,
30
+ "rstrip": false,
31
+ "single_word": false
32
+ }
33
+ }