Rocketknight1 HF staff commited on
Commit
06a50f2
1 Parent(s): d44f6ad

Add TF weights

Browse files

Model converted by the [`transformers`' `pt_to_tf` CLI](https://github.com/huggingface/transformers/blob/main/src/transformers/commands/pt_to_tf.py). All converted model outputs and hidden layers were validated against its Pytorch counterpart.

Maximum crossload output difference=2.566e-04; Maximum crossload hidden layer difference=2.197e-03;
Maximum conversion output difference=2.566e-04; Maximum conversion hidden layer difference=2.197e-03;


CAUTION: The maximum admissible error was manually increased to 0.003!

tf_model-00001-of-00002.h5 ADDED
@@ -0,0 +1,3 @@
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6e419ab27cca0914492525478957ec8e0952b5f8ee646a15bfdf0ab00a773240
3
+ size 9966769472
tf_model-00002-of-00002.h5 ADDED
@@ -0,0 +1,3 @@
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fb2d67f27f0343e37241bd9312db51150f530bed3a7929c7b0b1d8a64c65f9fc
3
+ size 1390425800
tf_model.h5.index.json ADDED
@@ -0,0 +1,628 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 11356359300
4
+ },
5
+ "weight_map": {
6
+ "tf_esm_for_masked_lm_1/esm/embeddings/word_embeddings/embeddings:0": "tf_model-00001-of-00002.h5",
7
+ "tf_esm_for_masked_lm_1/esm/encoder/emb_layer_norm_after/beta:0": "tf_model-00002-of-00002.h5",
8
+ "tf_esm_for_masked_lm_1/esm/encoder/emb_layer_norm_after/gamma:0": "tf_model-00002-of-00002.h5",
9
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._0/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
10
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._0/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
11
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._0/attention/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
12
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._0/attention/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
13
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._0/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
14
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._0/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
15
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._0/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
16
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._0/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
17
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._0/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
18
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._0/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
19
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._0/attention/self/rotary_embeddings/inv_freq:0": "tf_model-00001-of-00002.h5",
20
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._0/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
21
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._0/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
22
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._0/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
23
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._0/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
24
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._0/output/dense/bias:0": "tf_model-00001-of-00002.h5",
25
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._0/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
26
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._1/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
27
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._1/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
28
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._1/attention/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
29
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._1/attention/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
30
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._1/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
31
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._1/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
32
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._1/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
33
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._1/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
34
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._1/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
35
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._1/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
36
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._1/attention/self/rotary_embeddings/inv_freq:0": "tf_model-00001-of-00002.h5",
37
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._1/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
38
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._1/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
39
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._1/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
40
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._1/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
41
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._1/output/dense/bias:0": "tf_model-00001-of-00002.h5",
42
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._1/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
43
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._10/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
44
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._10/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
45
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._10/attention/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
46
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._10/attention/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
47
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._10/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
48
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._10/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
49
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._10/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
50
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._10/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
51
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._10/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
52
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._10/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
53
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._10/attention/self/rotary_embeddings/inv_freq:0": "tf_model-00001-of-00002.h5",
54
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._10/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
55
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._10/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
56
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._10/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
57
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._10/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
58
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._10/output/dense/bias:0": "tf_model-00001-of-00002.h5",
59
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._10/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
60
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._11/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
61
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._11/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
62
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._11/attention/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
63
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._11/attention/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
64
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._11/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
65
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._11/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
66
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._11/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
67
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._11/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
68
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._11/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
69
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._11/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
70
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._11/attention/self/rotary_embeddings/inv_freq:0": "tf_model-00001-of-00002.h5",
71
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._11/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
72
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._11/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
73
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._11/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
74
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._11/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
75
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._11/output/dense/bias:0": "tf_model-00001-of-00002.h5",
76
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._11/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
77
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._12/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
78
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._12/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
79
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._12/attention/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
80
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._12/attention/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
81
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._12/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
82
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._12/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
83
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._12/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
84
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._12/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
85
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._12/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
86
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._12/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
87
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._12/attention/self/rotary_embeddings/inv_freq:0": "tf_model-00001-of-00002.h5",
88
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._12/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
89
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._12/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
90
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._12/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
91
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._12/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
92
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._12/output/dense/bias:0": "tf_model-00001-of-00002.h5",
93
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._12/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
94
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._13/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
95
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._13/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
96
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._13/attention/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
97
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._13/attention/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
98
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._13/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
99
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._13/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
100
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._13/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
101
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._13/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
102
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._13/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
103
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._13/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
104
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._13/attention/self/rotary_embeddings/inv_freq:0": "tf_model-00001-of-00002.h5",
105
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._13/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
106
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._13/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
107
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._13/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
108
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._13/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
109
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._13/output/dense/bias:0": "tf_model-00001-of-00002.h5",
110
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._13/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
111
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._14/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
112
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._14/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
113
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._14/attention/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
114
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._14/attention/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
115
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._14/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
116
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._14/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
117
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._14/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
118
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._14/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
119
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._14/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
120
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._14/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
121
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._14/attention/self/rotary_embeddings/inv_freq:0": "tf_model-00001-of-00002.h5",
122
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._14/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
123
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._14/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
124
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._14/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
125
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._14/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
126
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._14/output/dense/bias:0": "tf_model-00001-of-00002.h5",
127
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._14/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
128
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._15/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
129
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._15/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
130
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._15/attention/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
131
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._15/attention/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
132
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._15/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
133
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._15/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
134
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._15/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
135
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._15/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
136
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._15/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
137
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._15/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
138
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._15/attention/self/rotary_embeddings/inv_freq:0": "tf_model-00001-of-00002.h5",
139
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._15/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
140
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._15/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
141
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._15/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
142
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._15/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
143
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._15/output/dense/bias:0": "tf_model-00001-of-00002.h5",
144
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._15/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
145
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._16/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
146
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._16/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
147
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._16/attention/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
148
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._16/attention/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
149
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._16/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
150
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._16/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
151
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._16/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
152
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._16/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
153
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._16/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
154
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._16/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
155
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._16/attention/self/rotary_embeddings/inv_freq:0": "tf_model-00001-of-00002.h5",
156
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._16/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
157
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._16/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
158
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._16/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
159
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._16/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
160
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._16/output/dense/bias:0": "tf_model-00001-of-00002.h5",
161
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._16/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
162
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._17/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
163
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._17/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
164
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._17/attention/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
165
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._17/attention/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
166
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._17/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
167
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._17/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
168
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._17/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
169
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._17/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
170
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._17/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
171
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._17/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
172
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._17/attention/self/rotary_embeddings/inv_freq:0": "tf_model-00001-of-00002.h5",
173
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._17/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
174
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._17/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
175
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._17/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
176
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._17/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
177
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._17/output/dense/bias:0": "tf_model-00001-of-00002.h5",
178
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._17/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
179
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._18/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
180
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._18/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
181
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._18/attention/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
182
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._18/attention/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
183
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._18/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
184
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._18/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
185
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._18/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
186
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._18/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
187
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._18/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
188
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._18/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
189
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._18/attention/self/rotary_embeddings/inv_freq:0": "tf_model-00001-of-00002.h5",
190
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._18/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
191
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._18/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
192
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._18/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
193
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._18/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
194
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._18/output/dense/bias:0": "tf_model-00001-of-00002.h5",
195
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._18/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
196
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._19/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
197
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._19/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
198
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._19/attention/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
199
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._19/attention/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
200
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._19/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
201
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._19/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
202
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._19/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
203
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._19/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
204
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._19/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
205
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._19/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
206
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._19/attention/self/rotary_embeddings/inv_freq:0": "tf_model-00001-of-00002.h5",
207
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._19/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
208
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._19/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
209
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._19/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
210
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._19/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
211
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._19/output/dense/bias:0": "tf_model-00001-of-00002.h5",
212
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._19/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
213
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._2/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
214
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._2/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
215
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._2/attention/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
216
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._2/attention/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
217
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._2/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
218
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._2/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
219
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._2/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
220
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._2/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
221
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._2/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
222
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._2/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
223
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._2/attention/self/rotary_embeddings/inv_freq:0": "tf_model-00001-of-00002.h5",
224
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._2/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
225
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._2/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
226
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._2/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
227
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._2/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
228
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._2/output/dense/bias:0": "tf_model-00001-of-00002.h5",
229
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._2/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
230
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._20/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
231
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._20/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
232
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._20/attention/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
233
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._20/attention/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
234
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._20/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
235
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._20/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
236
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._20/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
237
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._20/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
238
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._20/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
239
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._20/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
240
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._20/attention/self/rotary_embeddings/inv_freq:0": "tf_model-00001-of-00002.h5",
241
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._20/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
242
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._20/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
243
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._20/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
244
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._20/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
245
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._20/output/dense/bias:0": "tf_model-00001-of-00002.h5",
246
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._20/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
247
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._21/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
248
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._21/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
249
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._21/attention/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
250
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._21/attention/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
251
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._21/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
252
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._21/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
253
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._21/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
254
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._21/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
255
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._21/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
256
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._21/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
257
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._21/attention/self/rotary_embeddings/inv_freq:0": "tf_model-00001-of-00002.h5",
258
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._21/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
259
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._21/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
260
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._21/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
261
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._21/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
262
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._21/output/dense/bias:0": "tf_model-00001-of-00002.h5",
263
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._21/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
264
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._22/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
265
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._22/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
266
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._22/attention/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
267
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._22/attention/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
268
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._22/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
269
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._22/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
270
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._22/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
271
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._22/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
272
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._22/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
273
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._22/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
274
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._22/attention/self/rotary_embeddings/inv_freq:0": "tf_model-00001-of-00002.h5",
275
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._22/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
276
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._22/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
277
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._22/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
278
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._22/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
279
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._22/output/dense/bias:0": "tf_model-00001-of-00002.h5",
280
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._22/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
281
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._23/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
282
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._23/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
283
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._23/attention/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
284
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._23/attention/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
285
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._23/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
286
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._23/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
287
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._23/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
288
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._23/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
289
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._23/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
290
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._23/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
291
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._23/attention/self/rotary_embeddings/inv_freq:0": "tf_model-00001-of-00002.h5",
292
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._23/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
293
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._23/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
294
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._23/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
295
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._23/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
296
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._23/output/dense/bias:0": "tf_model-00001-of-00002.h5",
297
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._23/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
298
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._24/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
299
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._24/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
300
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._24/attention/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
301
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._24/attention/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
302
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._24/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
303
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._24/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
304
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._24/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
305
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._24/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
306
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._24/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
307
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._24/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
308
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._24/attention/self/rotary_embeddings/inv_freq:0": "tf_model-00001-of-00002.h5",
309
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._24/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
310
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._24/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
311
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._24/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
312
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._24/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
313
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._24/output/dense/bias:0": "tf_model-00001-of-00002.h5",
314
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._24/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
315
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._25/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
316
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._25/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
317
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._25/attention/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
318
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._25/attention/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
319
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._25/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
320
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._25/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
321
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._25/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
322
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._25/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
323
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._25/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
324
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._25/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
325
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._25/attention/self/rotary_embeddings/inv_freq:0": "tf_model-00001-of-00002.h5",
326
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._25/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
327
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._25/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
328
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._25/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
329
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._25/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
330
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._25/output/dense/bias:0": "tf_model-00001-of-00002.h5",
331
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._25/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
332
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._26/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
333
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._26/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
334
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._26/attention/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
335
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._26/attention/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
336
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._26/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
337
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._26/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
338
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._26/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
339
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._26/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
340
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._26/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
341
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._26/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
342
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._26/attention/self/rotary_embeddings/inv_freq:0": "tf_model-00001-of-00002.h5",
343
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._26/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
344
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._26/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
345
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._26/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
346
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._26/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
347
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._26/output/dense/bias:0": "tf_model-00001-of-00002.h5",
348
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._26/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
349
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._27/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
350
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._27/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
351
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._27/attention/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
352
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._27/attention/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
353
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._27/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
354
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._27/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
355
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._27/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
356
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._27/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
357
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._27/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
358
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._27/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
359
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._27/attention/self/rotary_embeddings/inv_freq:0": "tf_model-00001-of-00002.h5",
360
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._27/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
361
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._27/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
362
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._27/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
363
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._27/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
364
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._27/output/dense/bias:0": "tf_model-00001-of-00002.h5",
365
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._27/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
366
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._28/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
367
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._28/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
368
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._28/attention/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
369
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._28/attention/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
370
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._28/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
371
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._28/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
372
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._28/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
373
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._28/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
374
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._28/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
375
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._28/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
376
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._28/attention/self/rotary_embeddings/inv_freq:0": "tf_model-00001-of-00002.h5",
377
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._28/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
378
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._28/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
379
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._28/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
380
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._28/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
381
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._28/output/dense/bias:0": "tf_model-00001-of-00002.h5",
382
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._28/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
383
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._29/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
384
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._29/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
385
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._29/attention/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
386
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._29/attention/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
387
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._29/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
388
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._29/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
389
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._29/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
390
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._29/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
391
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._29/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
392
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._29/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
393
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._29/attention/self/rotary_embeddings/inv_freq:0": "tf_model-00001-of-00002.h5",
394
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._29/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
395
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._29/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
396
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._29/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
397
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._29/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
398
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._29/output/dense/bias:0": "tf_model-00001-of-00002.h5",
399
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._29/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
400
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._3/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
401
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._3/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
402
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._3/attention/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
403
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._3/attention/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
404
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._3/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
405
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._3/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
406
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._3/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
407
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._3/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
408
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._3/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
409
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._3/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
410
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._3/attention/self/rotary_embeddings/inv_freq:0": "tf_model-00001-of-00002.h5",
411
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._3/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
412
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._3/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
413
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._3/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
414
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._3/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
415
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._3/output/dense/bias:0": "tf_model-00001-of-00002.h5",
416
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._3/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
417
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._30/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
418
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._30/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
419
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._30/attention/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
420
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._30/attention/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
421
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._30/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
422
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._30/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
423
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._30/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
424
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._30/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
425
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._30/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
426
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._30/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
427
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._30/attention/self/rotary_embeddings/inv_freq:0": "tf_model-00001-of-00002.h5",
428
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._30/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
429
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._30/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
430
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._30/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
431
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._30/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
432
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._30/output/dense/bias:0": "tf_model-00001-of-00002.h5",
433
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._30/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
434
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._31/LayerNorm/beta:0": "tf_model-00002-of-00002.h5",
435
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._31/LayerNorm/gamma:0": "tf_model-00002-of-00002.h5",
436
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._31/attention/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
437
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._31/attention/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
438
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._31/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
439
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._31/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
440
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._31/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
441
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._31/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
442
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._31/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
443
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._31/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
444
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._31/attention/self/rotary_embeddings/inv_freq:0": "tf_model-00001-of-00002.h5",
445
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._31/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
446
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._31/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
447
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._31/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
448
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._31/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
449
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._31/output/dense/bias:0": "tf_model-00002-of-00002.h5",
450
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._31/output/dense/kernel:0": "tf_model-00002-of-00002.h5",
451
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._32/LayerNorm/beta:0": "tf_model-00002-of-00002.h5",
452
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._32/LayerNorm/gamma:0": "tf_model-00002-of-00002.h5",
453
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._32/attention/LayerNorm/beta:0": "tf_model-00002-of-00002.h5",
454
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._32/attention/LayerNorm/gamma:0": "tf_model-00002-of-00002.h5",
455
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._32/attention/output/dense/bias:0": "tf_model-00002-of-00002.h5",
456
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._32/attention/output/dense/kernel:0": "tf_model-00002-of-00002.h5",
457
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._32/attention/self/key/bias:0": "tf_model-00002-of-00002.h5",
458
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._32/attention/self/key/kernel:0": "tf_model-00002-of-00002.h5",
459
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._32/attention/self/query/bias:0": "tf_model-00002-of-00002.h5",
460
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._32/attention/self/query/kernel:0": "tf_model-00002-of-00002.h5",
461
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._32/attention/self/rotary_embeddings/inv_freq:0": "tf_model-00002-of-00002.h5",
462
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._32/attention/self/value/bias:0": "tf_model-00002-of-00002.h5",
463
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._32/attention/self/value/kernel:0": "tf_model-00002-of-00002.h5",
464
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._32/intermediate/dense/bias:0": "tf_model-00002-of-00002.h5",
465
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._32/intermediate/dense/kernel:0": "tf_model-00002-of-00002.h5",
466
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._32/output/dense/bias:0": "tf_model-00002-of-00002.h5",
467
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._32/output/dense/kernel:0": "tf_model-00002-of-00002.h5",
468
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._33/LayerNorm/beta:0": "tf_model-00002-of-00002.h5",
469
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._33/LayerNorm/gamma:0": "tf_model-00002-of-00002.h5",
470
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._33/attention/LayerNorm/beta:0": "tf_model-00002-of-00002.h5",
471
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._33/attention/LayerNorm/gamma:0": "tf_model-00002-of-00002.h5",
472
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._33/attention/output/dense/bias:0": "tf_model-00002-of-00002.h5",
473
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._33/attention/output/dense/kernel:0": "tf_model-00002-of-00002.h5",
474
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._33/attention/self/key/bias:0": "tf_model-00002-of-00002.h5",
475
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._33/attention/self/key/kernel:0": "tf_model-00002-of-00002.h5",
476
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._33/attention/self/query/bias:0": "tf_model-00002-of-00002.h5",
477
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._33/attention/self/query/kernel:0": "tf_model-00002-of-00002.h5",
478
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._33/attention/self/rotary_embeddings/inv_freq:0": "tf_model-00002-of-00002.h5",
479
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._33/attention/self/value/bias:0": "tf_model-00002-of-00002.h5",
480
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._33/attention/self/value/kernel:0": "tf_model-00002-of-00002.h5",
481
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._33/intermediate/dense/bias:0": "tf_model-00002-of-00002.h5",
482
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._33/intermediate/dense/kernel:0": "tf_model-00002-of-00002.h5",
483
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._33/output/dense/bias:0": "tf_model-00002-of-00002.h5",
484
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._33/output/dense/kernel:0": "tf_model-00002-of-00002.h5",
485
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._34/LayerNorm/beta:0": "tf_model-00002-of-00002.h5",
486
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._34/LayerNorm/gamma:0": "tf_model-00002-of-00002.h5",
487
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._34/attention/LayerNorm/beta:0": "tf_model-00002-of-00002.h5",
488
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._34/attention/LayerNorm/gamma:0": "tf_model-00002-of-00002.h5",
489
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._34/attention/output/dense/bias:0": "tf_model-00002-of-00002.h5",
490
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._34/attention/output/dense/kernel:0": "tf_model-00002-of-00002.h5",
491
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._34/attention/self/key/bias:0": "tf_model-00002-of-00002.h5",
492
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._34/attention/self/key/kernel:0": "tf_model-00002-of-00002.h5",
493
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._34/attention/self/query/bias:0": "tf_model-00002-of-00002.h5",
494
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._34/attention/self/query/kernel:0": "tf_model-00002-of-00002.h5",
495
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._34/attention/self/rotary_embeddings/inv_freq:0": "tf_model-00002-of-00002.h5",
496
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._34/attention/self/value/bias:0": "tf_model-00002-of-00002.h5",
497
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._34/attention/self/value/kernel:0": "tf_model-00002-of-00002.h5",
498
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._34/intermediate/dense/bias:0": "tf_model-00002-of-00002.h5",
499
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._34/intermediate/dense/kernel:0": "tf_model-00002-of-00002.h5",
500
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._34/output/dense/bias:0": "tf_model-00002-of-00002.h5",
501
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._34/output/dense/kernel:0": "tf_model-00002-of-00002.h5",
502
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._35/LayerNorm/beta:0": "tf_model-00002-of-00002.h5",
503
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._35/LayerNorm/gamma:0": "tf_model-00002-of-00002.h5",
504
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._35/attention/LayerNorm/beta:0": "tf_model-00002-of-00002.h5",
505
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._35/attention/LayerNorm/gamma:0": "tf_model-00002-of-00002.h5",
506
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._35/attention/output/dense/bias:0": "tf_model-00002-of-00002.h5",
507
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._35/attention/output/dense/kernel:0": "tf_model-00002-of-00002.h5",
508
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._35/attention/self/key/bias:0": "tf_model-00002-of-00002.h5",
509
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._35/attention/self/key/kernel:0": "tf_model-00002-of-00002.h5",
510
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._35/attention/self/query/bias:0": "tf_model-00002-of-00002.h5",
511
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._35/attention/self/query/kernel:0": "tf_model-00002-of-00002.h5",
512
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._35/attention/self/rotary_embeddings/inv_freq:0": "tf_model-00002-of-00002.h5",
513
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._35/attention/self/value/bias:0": "tf_model-00002-of-00002.h5",
514
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._35/attention/self/value/kernel:0": "tf_model-00002-of-00002.h5",
515
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._35/intermediate/dense/bias:0": "tf_model-00002-of-00002.h5",
516
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._35/intermediate/dense/kernel:0": "tf_model-00002-of-00002.h5",
517
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._35/output/dense/bias:0": "tf_model-00002-of-00002.h5",
518
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._35/output/dense/kernel:0": "tf_model-00002-of-00002.h5",
519
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._4/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
520
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._4/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
521
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._4/attention/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
522
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._4/attention/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
523
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._4/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
524
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._4/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
525
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._4/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
526
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._4/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
527
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._4/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
528
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._4/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
529
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._4/attention/self/rotary_embeddings/inv_freq:0": "tf_model-00001-of-00002.h5",
530
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._4/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
531
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._4/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
532
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._4/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
533
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._4/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
534
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._4/output/dense/bias:0": "tf_model-00001-of-00002.h5",
535
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._4/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
536
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._5/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
537
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._5/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
538
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._5/attention/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
539
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._5/attention/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
540
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._5/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
541
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._5/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
542
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._5/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
543
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._5/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
544
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._5/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
545
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._5/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
546
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._5/attention/self/rotary_embeddings/inv_freq:0": "tf_model-00001-of-00002.h5",
547
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._5/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
548
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._5/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
549
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._5/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
550
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._5/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
551
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._5/output/dense/bias:0": "tf_model-00001-of-00002.h5",
552
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._5/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
553
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._6/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
554
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._6/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
555
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._6/attention/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
556
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._6/attention/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
557
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._6/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
558
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._6/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
559
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._6/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
560
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._6/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
561
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._6/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
562
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._6/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
563
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._6/attention/self/rotary_embeddings/inv_freq:0": "tf_model-00001-of-00002.h5",
564
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._6/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
565
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._6/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
566
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._6/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
567
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._6/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
568
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._6/output/dense/bias:0": "tf_model-00001-of-00002.h5",
569
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._6/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
570
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._7/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
571
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._7/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
572
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._7/attention/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
573
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._7/attention/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
574
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._7/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
575
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._7/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
576
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._7/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
577
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._7/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
578
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._7/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
579
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._7/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
580
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._7/attention/self/rotary_embeddings/inv_freq:0": "tf_model-00001-of-00002.h5",
581
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._7/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
582
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._7/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
583
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._7/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
584
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._7/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
585
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._7/output/dense/bias:0": "tf_model-00001-of-00002.h5",
586
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._7/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
587
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._8/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
588
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._8/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
589
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._8/attention/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
590
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._8/attention/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
591
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._8/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
592
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._8/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
593
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._8/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
594
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._8/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
595
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._8/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
596
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._8/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
597
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._8/attention/self/rotary_embeddings/inv_freq:0": "tf_model-00001-of-00002.h5",
598
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._8/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
599
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._8/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
600
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._8/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
601
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._8/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
602
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._8/output/dense/bias:0": "tf_model-00001-of-00002.h5",
603
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._8/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
604
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._9/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
605
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._9/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
606
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._9/attention/LayerNorm/beta:0": "tf_model-00001-of-00002.h5",
607
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._9/attention/LayerNorm/gamma:0": "tf_model-00001-of-00002.h5",
608
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._9/attention/output/dense/bias:0": "tf_model-00001-of-00002.h5",
609
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._9/attention/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
610
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._9/attention/self/key/bias:0": "tf_model-00001-of-00002.h5",
611
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._9/attention/self/key/kernel:0": "tf_model-00001-of-00002.h5",
612
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._9/attention/self/query/bias:0": "tf_model-00001-of-00002.h5",
613
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._9/attention/self/query/kernel:0": "tf_model-00001-of-00002.h5",
614
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._9/attention/self/rotary_embeddings/inv_freq:0": "tf_model-00001-of-00002.h5",
615
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._9/attention/self/value/bias:0": "tf_model-00001-of-00002.h5",
616
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._9/attention/self/value/kernel:0": "tf_model-00001-of-00002.h5",
617
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._9/intermediate/dense/bias:0": "tf_model-00001-of-00002.h5",
618
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._9/intermediate/dense/kernel:0": "tf_model-00001-of-00002.h5",
619
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._9/output/dense/bias:0": "tf_model-00001-of-00002.h5",
620
+ "tf_esm_for_masked_lm_1/esm/encoder/layer_._9/output/dense/kernel:0": "tf_model-00001-of-00002.h5",
621
+ "tf_esm_for_masked_lm_1/lm_head/bias:0": "tf_model-00002-of-00002.h5",
622
+ "tf_esm_for_masked_lm_1/lm_head/decoder/kernel:0": "tf_model-00002-of-00002.h5",
623
+ "tf_esm_for_masked_lm_1/lm_head/dense/bias:0": "tf_model-00002-of-00002.h5",
624
+ "tf_esm_for_masked_lm_1/lm_head/dense/kernel:0": "tf_model-00002-of-00002.h5",
625
+ "tf_esm_for_masked_lm_1/lm_head/layer_norm/beta:0": "tf_model-00002-of-00002.h5",
626
+ "tf_esm_for_masked_lm_1/lm_head/layer_norm/gamma:0": "tf_model-00002-of-00002.h5"
627
+ }
628
+ }