Hugging Face's logo

ArthurZ
/
roberta-large-sharded

Feature Extraction
Transformers
TensorFlow
roberta
generated_from_keras_callback
Inference Endpoints
Model card Files Community
ArthurZ HF staff commited on
Commit
718d3b7
•
1 Parent(s): 3d49c59

add model

Browse files
Files changed (3) hide show
  1. tf_model.h5 → tf_model-00001-of-00002.h5 +2 -2
  2. tf_model-00002-of-00002.h5 +3 -0
  3. tf_model.h5.index.json +398 -0
tf_model.h5 → tf_model-00001-of-00002.h5 RENAMED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:1afa914a0b05bd7397511b1d84c8cbe18a441fbcafcc1336ece549473712c7d1
3
- size 1421953544
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:974b1bea1838635597d551675e8a593a7bbb37c43c00c65654aef21d56f8cd3c
3
+ size 991323488
tf_model-00002-of-00002.h5 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d9795b878859a5eaaf0a421e13923e3994cb6ccfd89dc42bca7e38bf0f78bfd2
3
+ size 430634744
tf_model.h5.index.json ADDED
@@ -0,0 +1,398 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 1421438976
4
+ },
5
+ "weight_map": {
6
+ "roberta/embeddings/LayerNorm/beta:0": "tf_model-00002-of-00002.h5",
7
+ "roberta/embeddings/LayerNorm/gamma:0": "tf_model-00002-of-00002.h5",
8
+ "roberta/embeddings/position_embeddings/embeddings:0": "tf_model-00002-of-00002.h5",
9
+ "roberta/embeddings/token_type_embeddings/embeddings:0": "tf_model-00002-of-00002.h5",
10
+ "roberta/embeddings/word_embeddings/weight:0": "tf_model-00002-of-00002.h5",
11
+ "roberta/encoder/layer_._0/attention/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
12
+ "roberta/encoder/layer_._0/attention/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
13
+ "roberta/encoder/layer_._0/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
14
+ "roberta/encoder/layer_._0/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
15
+ "roberta/encoder/layer_._0/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
16
+ "roberta/encoder/layer_._0/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
17
+ "roberta/encoder/layer_._0/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
18
+ "roberta/encoder/layer_._0/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
19
+ "roberta/encoder/layer_._0/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
20
+ "roberta/encoder/layer_._0/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
21
+ "roberta/encoder/layer_._0/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
22
+ "roberta/encoder/layer_._0/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
23
+ "roberta/encoder/layer_._0/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
24
+ "roberta/encoder/layer_._0/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
25
+ "roberta/encoder/layer_._0/output/dense/bias:0": "tf_model-00001-of-00002.h5",
26
+ "roberta/encoder/layer_._0/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
27
+ "roberta/encoder/layer_._1/attention/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
28
+ "roberta/encoder/layer_._1/attention/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
29
+ "roberta/encoder/layer_._1/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
30
+ "roberta/encoder/layer_._1/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
31
+ "roberta/encoder/layer_._1/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
32
+ "roberta/encoder/layer_._1/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
33
+ "roberta/encoder/layer_._1/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
34
+ "roberta/encoder/layer_._1/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
35
+ "roberta/encoder/layer_._1/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
36
+ "roberta/encoder/layer_._1/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
37
+ "roberta/encoder/layer_._1/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
38
+ "roberta/encoder/layer_._1/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
39
+ "roberta/encoder/layer_._1/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
40
+ "roberta/encoder/layer_._1/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
41
+ "roberta/encoder/layer_._1/output/dense/bias:0": "tf_model-00001-of-00002.h5",
42
+ "roberta/encoder/layer_._1/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
43
+ "roberta/encoder/layer_._10/attention/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
44
+ "roberta/encoder/layer_._10/attention/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
45
+ "roberta/encoder/layer_._10/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
46
+ "roberta/encoder/layer_._10/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
47
+ "roberta/encoder/layer_._10/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
48
+ "roberta/encoder/layer_._10/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
49
+ "roberta/encoder/layer_._10/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
50
+ "roberta/encoder/layer_._10/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
51
+ "roberta/encoder/layer_._10/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
52
+ "roberta/encoder/layer_._10/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
53
+ "roberta/encoder/layer_._10/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
54
+ "roberta/encoder/layer_._10/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
55
+ "roberta/encoder/layer_._10/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
56
+ "roberta/encoder/layer_._10/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
57
+ "roberta/encoder/layer_._10/output/dense/bias:0": "tf_model-00001-of-00002.h5",
58
+ "roberta/encoder/layer_._10/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
59
+ "roberta/encoder/layer_._11/attention/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
60
+ "roberta/encoder/layer_._11/attention/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
61
+ "roberta/encoder/layer_._11/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
62
+ "roberta/encoder/layer_._11/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
63
+ "roberta/encoder/layer_._11/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
64
+ "roberta/encoder/layer_._11/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
65
+ "roberta/encoder/layer_._11/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
66
+ "roberta/encoder/layer_._11/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
67
+ "roberta/encoder/layer_._11/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
68
+ "roberta/encoder/layer_._11/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
69
+ "roberta/encoder/layer_._11/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
70
+ "roberta/encoder/layer_._11/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
71
+ "roberta/encoder/layer_._11/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
72
+ "roberta/encoder/layer_._11/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
73
+ "roberta/encoder/layer_._11/output/dense/bias:0": "tf_model-00001-of-00002.h5",
74
+ "roberta/encoder/layer_._11/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
75
+ "roberta/encoder/layer_._12/attention/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
76
+ "roberta/encoder/layer_._12/attention/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
77
+ "roberta/encoder/layer_._12/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
78
+ "roberta/encoder/layer_._12/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
79
+ "roberta/encoder/layer_._12/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
80
+ "roberta/encoder/layer_._12/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
81
+ "roberta/encoder/layer_._12/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
82
+ "roberta/encoder/layer_._12/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
83
+ "roberta/encoder/layer_._12/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
84
+ "roberta/encoder/layer_._12/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
85
+ "roberta/encoder/layer_._12/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
86
+ "roberta/encoder/layer_._12/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
87
+ "roberta/encoder/layer_._12/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
88
+ "roberta/encoder/layer_._12/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
89
+ "roberta/encoder/layer_._12/output/dense/bias:0": "tf_model-00001-of-00002.h5",
90
+ "roberta/encoder/layer_._12/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
91
+ "roberta/encoder/layer_._13/attention/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
92
+ "roberta/encoder/layer_._13/attention/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
93
+ "roberta/encoder/layer_._13/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
94
+ "roberta/encoder/layer_._13/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
95
+ "roberta/encoder/layer_._13/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
96
+ "roberta/encoder/layer_._13/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
97
+ "roberta/encoder/layer_._13/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
98
+ "roberta/encoder/layer_._13/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
99
+ "roberta/encoder/layer_._13/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
100
+ "roberta/encoder/layer_._13/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
101
+ "roberta/encoder/layer_._13/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
102
+ "roberta/encoder/layer_._13/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
103
+ "roberta/encoder/layer_._13/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
104
+ "roberta/encoder/layer_._13/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
105
+ "roberta/encoder/layer_._13/output/dense/bias:0": "tf_model-00001-of-00002.h5",
106
+ "roberta/encoder/layer_._13/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
107
+ "roberta/encoder/layer_._14/attention/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
108
+ "roberta/encoder/layer_._14/attention/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
109
+ "roberta/encoder/layer_._14/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
110
+ "roberta/encoder/layer_._14/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
111
+ "roberta/encoder/layer_._14/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
112
+ "roberta/encoder/layer_._14/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
113
+ "roberta/encoder/layer_._14/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
114
+ "roberta/encoder/layer_._14/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
115
+ "roberta/encoder/layer_._14/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
116
+ "roberta/encoder/layer_._14/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
117
+ "roberta/encoder/layer_._14/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
118
+ "roberta/encoder/layer_._14/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
119
+ "roberta/encoder/layer_._14/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
120
+ "roberta/encoder/layer_._14/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
121
+ "roberta/encoder/layer_._14/output/dense/bias:0": "tf_model-00001-of-00002.h5",
122
+ "roberta/encoder/layer_._14/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
123
+ "roberta/encoder/layer_._15/attention/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
124
+ "roberta/encoder/layer_._15/attention/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
125
+ "roberta/encoder/layer_._15/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
126
+ "roberta/encoder/layer_._15/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
127
+ "roberta/encoder/layer_._15/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
128
+ "roberta/encoder/layer_._15/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
129
+ "roberta/encoder/layer_._15/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
130
+ "roberta/encoder/layer_._15/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
131
+ "roberta/encoder/layer_._15/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
132
+ "roberta/encoder/layer_._15/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
133
+ "roberta/encoder/layer_._15/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
134
+ "roberta/encoder/layer_._15/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
135
+ "roberta/encoder/layer_._15/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
136
+ "roberta/encoder/layer_._15/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
137
+ "roberta/encoder/layer_._15/output/dense/bias:0": "tf_model-00001-of-00002.h5",
138
+ "roberta/encoder/layer_._15/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
139
+ "roberta/encoder/layer_._16/attention/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
140
+ "roberta/encoder/layer_._16/attention/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
141
+ "roberta/encoder/layer_._16/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
142
+ "roberta/encoder/layer_._16/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
143
+ "roberta/encoder/layer_._16/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
144
+ "roberta/encoder/layer_._16/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
145
+ "roberta/encoder/layer_._16/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
146
+ "roberta/encoder/layer_._16/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
147
+ "roberta/encoder/layer_._16/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
148
+ "roberta/encoder/layer_._16/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
149
+ "roberta/encoder/layer_._16/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
150
+ "roberta/encoder/layer_._16/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
151
+ "roberta/encoder/layer_._16/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
152
+ "roberta/encoder/layer_._16/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
153
+ "roberta/encoder/layer_._16/output/dense/bias:0": "tf_model-00001-of-00002.h5",
154
+ "roberta/encoder/layer_._16/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
155
+ "roberta/encoder/layer_._17/attention/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
156
+ "roberta/encoder/layer_._17/attention/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
157
+ "roberta/encoder/layer_._17/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
158
+ "roberta/encoder/layer_._17/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
159
+ "roberta/encoder/layer_._17/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
160
+ "roberta/encoder/layer_._17/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
161
+ "roberta/encoder/layer_._17/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
162
+ "roberta/encoder/layer_._17/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
163
+ "roberta/encoder/layer_._17/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
164
+ "roberta/encoder/layer_._17/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
165
+ "roberta/encoder/layer_._17/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
166
+ "roberta/encoder/layer_._17/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
167
+ "roberta/encoder/layer_._17/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
168
+ "roberta/encoder/layer_._17/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
169
+ "roberta/encoder/layer_._17/output/dense/bias:0": "tf_model-00001-of-00002.h5",
170
+ "roberta/encoder/layer_._17/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
171
+ "roberta/encoder/layer_._18/attention/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
172
+ "roberta/encoder/layer_._18/attention/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
173
+ "roberta/encoder/layer_._18/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
174
+ "roberta/encoder/layer_._18/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
175
+ "roberta/encoder/layer_._18/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
176
+ "roberta/encoder/layer_._18/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
177
+ "roberta/encoder/layer_._18/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
178
+ "roberta/encoder/layer_._18/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
179
+ "roberta/encoder/layer_._18/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
180
+ "roberta/encoder/layer_._18/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
181
+ "roberta/encoder/layer_._18/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
182
+ "roberta/encoder/layer_._18/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
183
+ "roberta/encoder/layer_._18/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
184
+ "roberta/encoder/layer_._18/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
185
+ "roberta/encoder/layer_._18/output/dense/bias:0": "tf_model-00001-of-00002.h5",
186
+ "roberta/encoder/layer_._18/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
187
+ "roberta/encoder/layer_._19/attention/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
188
+ "roberta/encoder/layer_._19/attention/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
189
+ "roberta/encoder/layer_._19/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
190
+ "roberta/encoder/layer_._19/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
191
+ "roberta/encoder/layer_._19/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
192
+ "roberta/encoder/layer_._19/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
193
+ "roberta/encoder/layer_._19/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
194
+ "roberta/encoder/layer_._19/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
195
+ "roberta/encoder/layer_._19/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
196
+ "roberta/encoder/layer_._19/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
197
+ "roberta/encoder/layer_._19/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
198
+ "roberta/encoder/layer_._19/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
199
+ "roberta/encoder/layer_._19/output/LayerNorm/beta:0": "tf_model-00002-of-00002.h5",
200
+ "roberta/encoder/layer_._19/output/LayerNorm/gamma:0": "tf_model-00002-of-00002.h5",
201
+ "roberta/encoder/layer_._19/output/dense/bias:0": "tf_model-00002-of-00002.h5",
202
+ "roberta/encoder/layer_._19/output/dense/kernel:0": "tf_model-00002-of-00002.h5",
203
+ "roberta/encoder/layer_._2/attention/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
204
+ "roberta/encoder/layer_._2/attention/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
205
+ "roberta/encoder/layer_._2/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
206
+ "roberta/encoder/layer_._2/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
207
+ "roberta/encoder/layer_._2/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
208
+ "roberta/encoder/layer_._2/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
209
+ "roberta/encoder/layer_._2/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
210
+ "roberta/encoder/layer_._2/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
211
+ "roberta/encoder/layer_._2/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
212
+ "roberta/encoder/layer_._2/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
213
+ "roberta/encoder/layer_._2/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
214
+ "roberta/encoder/layer_._2/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
215
+ "roberta/encoder/layer_._2/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
216
+ "roberta/encoder/layer_._2/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
217
+ "roberta/encoder/layer_._2/output/dense/bias:0": "tf_model-00001-of-00002.h5",
218
+ "roberta/encoder/layer_._2/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
219
+ "roberta/encoder/layer_._20/attention/output/LayerNorm/beta:0": "tf_model-00002-of-00002.h5",
220
+ "roberta/encoder/layer_._20/attention/output/LayerNorm/gamma:0": "tf_model-00002-of-00002.h5",
221
+ "roberta/encoder/layer_._20/attention/output/dense/bias:0": "tf_model-00002-of-00002.h5",
222
+ "roberta/encoder/layer_._20/attention/output/dense/kernel:0": "tf_model-00002-of-00002.h5",
223
+ "roberta/encoder/layer_._20/attention/self/key/bias:0": "tf_model-00002-of-00002.h5",
224
+ "roberta/encoder/layer_._20/attention/self/key/kernel:0": "tf_model-00002-of-00002.h5",
225
+ "roberta/encoder/layer_._20/attention/self/query/bias:0": "tf_model-00002-of-00002.h5",
226
+ "roberta/encoder/layer_._20/attention/self/query/kernel:0": "tf_model-00002-of-00002.h5",
227
+ "roberta/encoder/layer_._20/attention/self/value/bias:0": "tf_model-00002-of-00002.h5",
228
+ "roberta/encoder/layer_._20/attention/self/value/kernel:0": "tf_model-00002-of-00002.h5",
229
+ "roberta/encoder/layer_._20/intermediate/dense/bias:0": "tf_model-00002-of-00002.h5",
230
+ "roberta/encoder/layer_._20/intermediate/dense/kernel:0": "tf_model-00002-of-00002.h5",
231
+ "roberta/encoder/layer_._20/output/LayerNorm/beta:0": "tf_model-00002-of-00002.h5",
232
+ "roberta/encoder/layer_._20/output/LayerNorm/gamma:0": "tf_model-00002-of-00002.h5",
233
+ "roberta/encoder/layer_._20/output/dense/bias:0": "tf_model-00002-of-00002.h5",
234
+ "roberta/encoder/layer_._20/output/dense/kernel:0": "tf_model-00002-of-00002.h5",
235
+ "roberta/encoder/layer_._21/attention/output/LayerNorm/beta:0": "tf_model-00002-of-00002.h5",
236
+ "roberta/encoder/layer_._21/attention/output/LayerNorm/gamma:0": "tf_model-00002-of-00002.h5",
237
+ "roberta/encoder/layer_._21/attention/output/dense/bias:0": "tf_model-00002-of-00002.h5",
238
+ "roberta/encoder/layer_._21/attention/output/dense/kernel:0": "tf_model-00002-of-00002.h5",
239
+ "roberta/encoder/layer_._21/attention/self/key/bias:0": "tf_model-00002-of-00002.h5",
240
+ "roberta/encoder/layer_._21/attention/self/key/kernel:0": "tf_model-00002-of-00002.h5",
241
+ "roberta/encoder/layer_._21/attention/self/query/bias:0": "tf_model-00002-of-00002.h5",
242
+ "roberta/encoder/layer_._21/attention/self/query/kernel:0": "tf_model-00002-of-00002.h5",
243
+ "roberta/encoder/layer_._21/attention/self/value/bias:0": "tf_model-00002-of-00002.h5",
244
+ "roberta/encoder/layer_._21/attention/self/value/kernel:0": "tf_model-00002-of-00002.h5",
245
+ "roberta/encoder/layer_._21/intermediate/dense/bias:0": "tf_model-00002-of-00002.h5",
246
+ "roberta/encoder/layer_._21/intermediate/dense/kernel:0": "tf_model-00002-of-00002.h5",
247
+ "roberta/encoder/layer_._21/output/LayerNorm/beta:0": "tf_model-00002-of-00002.h5",
248
+ "roberta/encoder/layer_._21/output/LayerNorm/gamma:0": "tf_model-00002-of-00002.h5",
249
+ "roberta/encoder/layer_._21/output/dense/bias:0": "tf_model-00002-of-00002.h5",
250
+ "roberta/encoder/layer_._21/output/dense/kernel:0": "tf_model-00002-of-00002.h5",
251
+ "roberta/encoder/layer_._22/attention/output/LayerNorm/beta:0": "tf_model-00002-of-00002.h5",
252
+ "roberta/encoder/layer_._22/attention/output/LayerNorm/gamma:0": "tf_model-00002-of-00002.h5",
253
+ "roberta/encoder/layer_._22/attention/output/dense/bias:0": "tf_model-00002-of-00002.h5",
254
+ "roberta/encoder/layer_._22/attention/output/dense/kernel:0": "tf_model-00002-of-00002.h5",
255
+ "roberta/encoder/layer_._22/attention/self/key/bias:0": "tf_model-00002-of-00002.h5",
256
+ "roberta/encoder/layer_._22/attention/self/key/kernel:0": "tf_model-00002-of-00002.h5",
257
+ "roberta/encoder/layer_._22/attention/self/query/bias:0": "tf_model-00002-of-00002.h5",
258
+ "roberta/encoder/layer_._22/attention/self/query/kernel:0": "tf_model-00002-of-00002.h5",
259
+ "roberta/encoder/layer_._22/attention/self/value/bias:0": "tf_model-00002-of-00002.h5",
260
+ "roberta/encoder/layer_._22/attention/self/value/kernel:0": "tf_model-00002-of-00002.h5",
261
+ "roberta/encoder/layer_._22/intermediate/dense/bias:0": "tf_model-00002-of-00002.h5",
262
+ "roberta/encoder/layer_._22/intermediate/dense/kernel:0": "tf_model-00002-of-00002.h5",
263
+ "roberta/encoder/layer_._22/output/LayerNorm/beta:0": "tf_model-00002-of-00002.h5",
264
+ "roberta/encoder/layer_._22/output/LayerNorm/gamma:0": "tf_model-00002-of-00002.h5",
265
+ "roberta/encoder/layer_._22/output/dense/bias:0": "tf_model-00002-of-00002.h5",
266
+ "roberta/encoder/layer_._22/output/dense/kernel:0": "tf_model-00002-of-00002.h5",
267
+ "roberta/encoder/layer_._23/attention/output/LayerNorm/beta:0": "tf_model-00002-of-00002.h5",
268
+ "roberta/encoder/layer_._23/attention/output/LayerNorm/gamma:0": "tf_model-00002-of-00002.h5",
269
+ "roberta/encoder/layer_._23/attention/output/dense/bias:0": "tf_model-00002-of-00002.h5",
270
+ "roberta/encoder/layer_._23/attention/output/dense/kernel:0": "tf_model-00002-of-00002.h5",
271
+ "roberta/encoder/layer_._23/attention/self/key/bias:0": "tf_model-00002-of-00002.h5",
272
+ "roberta/encoder/layer_._23/attention/self/key/kernel:0": "tf_model-00002-of-00002.h5",
273
+ "roberta/encoder/layer_._23/attention/self/query/bias:0": "tf_model-00002-of-00002.h5",
274
+ "roberta/encoder/layer_._23/attention/self/query/kernel:0": "tf_model-00002-of-00002.h5",
275
+ "roberta/encoder/layer_._23/attention/self/value/bias:0": "tf_model-00002-of-00002.h5",
276
+ "roberta/encoder/layer_._23/attention/self/value/kernel:0": "tf_model-00002-of-00002.h5",
277
+ "roberta/encoder/layer_._23/intermediate/dense/bias:0": "tf_model-00002-of-00002.h5",
278
+ "roberta/encoder/layer_._23/intermediate/dense/kernel:0": "tf_model-00002-of-00002.h5",
279
+ "roberta/encoder/layer_._23/output/LayerNorm/beta:0": "tf_model-00002-of-00002.h5",
280
+ "roberta/encoder/layer_._23/output/LayerNorm/gamma:0": "tf_model-00002-of-00002.h5",
281
+ "roberta/encoder/layer_._23/output/dense/bias:0": "tf_model-00002-of-00002.h5",
282
+ "roberta/encoder/layer_._23/output/dense/kernel:0": "tf_model-00002-of-00002.h5",
283
+ "roberta/encoder/layer_._3/attention/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
284
+ "roberta/encoder/layer_._3/attention/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
285
+ "roberta/encoder/layer_._3/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
286
+ "roberta/encoder/layer_._3/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
287
+ "roberta/encoder/layer_._3/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
288
+ "roberta/encoder/layer_._3/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
289
+ "roberta/encoder/layer_._3/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
290
+ "roberta/encoder/layer_._3/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
291
+ "roberta/encoder/layer_._3/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
292
+ "roberta/encoder/layer_._3/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
293
+ "roberta/encoder/layer_._3/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
294
+ "roberta/encoder/layer_._3/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
295
+ "roberta/encoder/layer_._3/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
296
+ "roberta/encoder/layer_._3/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
297
+ "roberta/encoder/layer_._3/output/dense/bias:0": "tf_model-00001-of-00002.h5",
298
+ "roberta/encoder/layer_._3/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
299
+ "roberta/encoder/layer_._4/attention/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
300
+ "roberta/encoder/layer_._4/attention/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
301
+ "roberta/encoder/layer_._4/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
302
+ "roberta/encoder/layer_._4/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
303
+ "roberta/encoder/layer_._4/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
304
+ "roberta/encoder/layer_._4/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
305
+ "roberta/encoder/layer_._4/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
306
+ "roberta/encoder/layer_._4/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
307
+ "roberta/encoder/layer_._4/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
308
+ "roberta/encoder/layer_._4/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
309
+ "roberta/encoder/layer_._4/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
310
+ "roberta/encoder/layer_._4/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
311
+ "roberta/encoder/layer_._4/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
312
+ "roberta/encoder/layer_._4/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
313
+ "roberta/encoder/layer_._4/output/dense/bias:0": "tf_model-00001-of-00002.h5",
314
+ "roberta/encoder/layer_._4/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
315
+ "roberta/encoder/layer_._5/attention/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
316
+ "roberta/encoder/layer_._5/attention/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
317
+ "roberta/encoder/layer_._5/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
318
+ "roberta/encoder/layer_._5/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
319
+ "roberta/encoder/layer_._5/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
320
+ "roberta/encoder/layer_._5/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
321
+ "roberta/encoder/layer_._5/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
322
+ "roberta/encoder/layer_._5/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
323
+ "roberta/encoder/layer_._5/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
324
+ "roberta/encoder/layer_._5/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
325
+ "roberta/encoder/layer_._5/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
326
+ "roberta/encoder/layer_._5/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
327
+ "roberta/encoder/layer_._5/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
328
+ "roberta/encoder/layer_._5/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
329
+ "roberta/encoder/layer_._5/output/dense/bias:0": "tf_model-00001-of-00002.h5",
330
+ "roberta/encoder/layer_._5/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
331
+ "roberta/encoder/layer_._6/attention/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
332
+ "roberta/encoder/layer_._6/attention/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
333
+ "roberta/encoder/layer_._6/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
334
+ "roberta/encoder/layer_._6/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
335
+ "roberta/encoder/layer_._6/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
336
+ "roberta/encoder/layer_._6/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
337
+ "roberta/encoder/layer_._6/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
338
+ "roberta/encoder/layer_._6/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
339
+ "roberta/encoder/layer_._6/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
340
+ "roberta/encoder/layer_._6/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
341
+ "roberta/encoder/layer_._6/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
342
+ "roberta/encoder/layer_._6/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
343
+ "roberta/encoder/layer_._6/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
344
+ "roberta/encoder/layer_._6/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
345
+ "roberta/encoder/layer_._6/output/dense/bias:0": "tf_model-00001-of-00002.h5",
346
+ "roberta/encoder/layer_._6/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
347
+ "roberta/encoder/layer_._7/attention/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
348
+ "roberta/encoder/layer_._7/attention/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
349
+ "roberta/encoder/layer_._7/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
350
+ "roberta/encoder/layer_._7/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
351
+ "roberta/encoder/layer_._7/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
352
+ "roberta/encoder/layer_._7/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
353
+ "roberta/encoder/layer_._7/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
354
+ "roberta/encoder/layer_._7/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
355
+ "roberta/encoder/layer_._7/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
356
+ "roberta/encoder/layer_._7/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
357
+ "roberta/encoder/layer_._7/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
358
+ "roberta/encoder/layer_._7/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
359
+ "roberta/encoder/layer_._7/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
360
+ "roberta/encoder/layer_._7/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
361
+ "roberta/encoder/layer_._7/output/dense/bias:0": "tf_model-00001-of-00002.h5",
362
+ "roberta/encoder/layer_._7/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
363
+ "roberta/encoder/layer_._8/attention/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
364
+ "roberta/encoder/layer_._8/attention/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
365
+ "roberta/encoder/layer_._8/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
366
+ "roberta/encoder/layer_._8/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
367
+ "roberta/encoder/layer_._8/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
368
+ "roberta/encoder/layer_._8/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
369
+ "roberta/encoder/layer_._8/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
370
+ "roberta/encoder/layer_._8/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
371
+ "roberta/encoder/layer_._8/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
372
+ "roberta/encoder/layer_._8/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
373
+ "roberta/encoder/layer_._8/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
374
+ "roberta/encoder/layer_._8/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
375
+ "roberta/encoder/layer_._8/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
376
+ "roberta/encoder/layer_._8/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
377
+ "roberta/encoder/layer_._8/output/dense/bias:0": "tf_model-00001-of-00002.h5",
378
+ "roberta/encoder/layer_._8/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
379
+ "roberta/encoder/layer_._9/attention/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
380
+ "roberta/encoder/layer_._9/attention/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
381
+ "roberta/encoder/layer_._9/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
382
+ "roberta/encoder/layer_._9/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
383
+ "roberta/encoder/layer_._9/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
384
+ "roberta/encoder/layer_._9/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
385
+ "roberta/encoder/layer_._9/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
386
+ "roberta/encoder/layer_._9/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
387
+ "roberta/encoder/layer_._9/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
388
+ "roberta/encoder/layer_._9/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
389
+ "roberta/encoder/layer_._9/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
390
+ "roberta/encoder/layer_._9/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
391
+ "roberta/encoder/layer_._9/output/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
392
+ "roberta/encoder/layer_._9/output/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
393
+ "roberta/encoder/layer_._9/output/dense/bias:0": "tf_model-00001-of-00002.h5",
394
+ "roberta/encoder/layer_._9/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
395
+ "roberta/pooler/dense/bias:0": "tf_model-00002-of-00002.h5",
396
+ "roberta/pooler/dense/kernel:0": "tf_model-00002-of-00002.h5"
397
+ }
398
+ }