Hugging Face's logo

ArthurZ
/
roberta-large-sharded

Feature Extraction
Transformers
TensorFlow
roberta
generated_from_keras_callback
Inference Endpoints
Model card Files Community
roberta-large-sharded
File size: 35,735 Bytes 
718d3b7
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
 
{
  "metadata": {
    "total_size": 1421438976
  },
  "weight_map": {
    "roberta/embeddings/LayerNorm/beta:0": "tf_model-00002-of-00002.h5",
    "roberta/embeddings/LayerNorm/gamma:0": "tf_model-00002-of-00002.h5",
    "roberta/embeddings/position_embeddings/embeddings:0": "tf_model-00002-of-00002.h5",
    "roberta/embeddings/token_type_embeddings/embeddings:0": "tf_model-00002-of-00002.h5",
    "roberta/embeddings/word_embeddings/weight:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._0/attention/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._0/attention/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._0/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._0/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._0/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._0/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._0/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._0/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._0/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._0/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._0/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._0/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._0/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._0/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._0/output/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._0/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._1/attention/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._1/attention/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._1/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._1/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._1/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._1/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._1/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._1/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._1/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._1/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._1/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._1/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._1/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._1/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._1/output/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._1/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._10/attention/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._10/attention/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._10/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._10/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._10/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._10/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._10/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._10/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._10/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._10/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._10/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._10/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._10/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._10/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._10/output/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._10/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._11/attention/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._11/attention/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._11/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._11/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._11/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._11/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._11/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._11/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._11/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._11/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._11/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._11/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._11/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._11/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._11/output/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._11/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._12/attention/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._12/attention/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._12/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._12/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._12/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._12/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._12/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._12/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._12/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._12/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._12/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._12/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._12/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._12/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._12/output/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._12/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._13/attention/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._13/attention/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._13/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._13/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._13/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._13/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._13/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._13/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._13/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._13/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._13/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._13/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._13/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._13/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._13/output/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._13/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._14/attention/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._14/attention/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._14/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._14/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._14/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._14/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._14/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._14/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._14/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._14/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._14/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._14/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._14/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._14/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._14/output/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._14/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._15/attention/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._15/attention/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._15/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._15/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._15/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._15/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._15/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._15/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._15/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._15/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._15/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._15/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._15/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._15/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._15/output/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._15/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._16/attention/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._16/attention/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._16/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._16/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._16/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._16/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._16/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._16/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._16/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._16/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._16/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._16/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._16/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._16/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._16/output/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._16/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._17/attention/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._17/attention/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._17/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._17/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._17/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._17/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._17/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._17/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._17/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._17/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._17/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._17/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._17/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._17/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._17/output/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._17/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._18/attention/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._18/attention/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._18/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._18/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._18/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._18/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._18/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._18/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._18/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._18/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._18/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._18/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._18/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._18/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._18/output/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._18/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._19/attention/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._19/attention/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._19/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._19/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._19/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._19/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._19/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._19/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._19/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._19/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._19/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._19/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._19/output/LayerNorm/beta:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._19/output/LayerNorm/gamma:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._19/output/dense/bias:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._19/output/dense/kernel:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._2/attention/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._2/attention/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._2/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._2/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._2/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._2/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._2/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._2/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._2/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._2/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._2/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._2/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._2/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._2/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._2/output/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._2/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._20/attention/output/LayerNorm/beta:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._20/attention/output/LayerNorm/gamma:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._20/attention/output/dense/bias:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._20/attention/output/dense/kernel:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._20/attention/self/key/bias:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._20/attention/self/key/kernel:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._20/attention/self/query/bias:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._20/attention/self/query/kernel:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._20/attention/self/value/bias:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._20/attention/self/value/kernel:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._20/intermediate/dense/bias:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._20/intermediate/dense/kernel:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._20/output/LayerNorm/beta:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._20/output/LayerNorm/gamma:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._20/output/dense/bias:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._20/output/dense/kernel:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._21/attention/output/LayerNorm/beta:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._21/attention/output/LayerNorm/gamma:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._21/attention/output/dense/bias:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._21/attention/output/dense/kernel:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._21/attention/self/key/bias:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._21/attention/self/key/kernel:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._21/attention/self/query/bias:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._21/attention/self/query/kernel:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._21/attention/self/value/bias:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._21/attention/self/value/kernel:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._21/intermediate/dense/bias:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._21/intermediate/dense/kernel:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._21/output/LayerNorm/beta:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._21/output/LayerNorm/gamma:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._21/output/dense/bias:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._21/output/dense/kernel:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._22/attention/output/LayerNorm/beta:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._22/attention/output/LayerNorm/gamma:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._22/attention/output/dense/bias:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._22/attention/output/dense/kernel:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._22/attention/self/key/bias:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._22/attention/self/key/kernel:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._22/attention/self/query/bias:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._22/attention/self/query/kernel:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._22/attention/self/value/bias:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._22/attention/self/value/kernel:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._22/intermediate/dense/bias:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._22/intermediate/dense/kernel:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._22/output/LayerNorm/beta:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._22/output/LayerNorm/gamma:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._22/output/dense/bias:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._22/output/dense/kernel:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._23/attention/output/LayerNorm/beta:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._23/attention/output/LayerNorm/gamma:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._23/attention/output/dense/bias:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._23/attention/output/dense/kernel:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._23/attention/self/key/bias:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._23/attention/self/key/kernel:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._23/attention/self/query/bias:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._23/attention/self/query/kernel:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._23/attention/self/value/bias:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._23/attention/self/value/kernel:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._23/intermediate/dense/bias:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._23/intermediate/dense/kernel:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._23/output/LayerNorm/beta:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._23/output/LayerNorm/gamma:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._23/output/dense/bias:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._23/output/dense/kernel:0": "tf_model-00002-of-00002.h5",
    "roberta/encoder/layer_._3/attention/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._3/attention/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._3/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._3/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._3/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._3/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._3/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._3/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._3/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._3/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._3/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._3/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._3/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._3/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._3/output/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._3/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._4/attention/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._4/attention/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._4/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._4/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._4/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._4/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._4/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._4/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._4/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._4/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._4/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._4/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._4/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._4/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._4/output/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._4/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._5/attention/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._5/attention/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._5/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._5/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._5/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._5/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._5/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._5/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._5/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._5/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._5/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._5/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._5/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._5/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._5/output/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._5/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._6/attention/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._6/attention/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._6/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._6/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._6/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._6/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._6/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._6/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._6/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._6/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._6/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._6/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._6/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._6/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._6/output/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._6/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._7/attention/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._7/attention/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._7/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._7/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._7/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._7/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._7/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._7/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._7/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._7/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._7/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._7/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._7/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._7/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._7/output/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._7/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._8/attention/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._8/attention/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._8/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._8/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._8/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._8/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._8/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._8/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._8/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._8/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._8/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._8/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._8/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._8/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._8/output/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._8/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._9/attention/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._9/attention/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._9/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._9/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._9/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._9/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._9/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._9/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._9/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._9/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._9/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._9/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._9/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._9/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._9/output/dense/bias:0": "tf_model-00001-of-00002.h5",
    "roberta/encoder/layer_._9/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
    "roberta/pooler/dense/bias:0": "tf_model-00002-of-00002.h5",
    "roberta/pooler/dense/kernel:0": "tf_model-00002-of-00002.h5"
  }
}