Hugging Face's logo

ariG23498
/
t5-v1-1-xxl-flax

Transformers
JAX
t5
Inference Endpoints
Model card Files Community
ariG23498 HF staff commited on
Commit
925e440
•
1 Parent(s): 10529a8

Upload model

Browse files
Files changed (4) hide show
  1. config.json +31 -0
  2. flax_model-00001-of-00002.msgpack +3 -0
  3. flax_model-00002-of-00002.msgpack +3 -0
  4. flax_model.msgpack.index.json +226 -0
config.json ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "/home/patrick/t5/t5-v1_1-xxl",
3
+ "architectures": [
4
+ "T5EncoderModel"
5
+ ],
6
+ "classifier_dropout": 0.0,
7
+ "d_ff": 10240,
8
+ "d_kv": 64,
9
+ "d_model": 4096,
10
+ "decoder_start_token_id": 0,
11
+ "dense_act_fn": "gelu_new",
12
+ "dropout_rate": 0.1,
13
+ "eos_token_id": 1,
14
+ "feed_forward_proj": "gated-gelu",
15
+ "initializer_factor": 1.0,
16
+ "is_encoder_decoder": true,
17
+ "is_gated_act": true,
18
+ "layer_norm_epsilon": 1e-06,
19
+ "model_type": "t5",
20
+ "num_decoder_layers": 24,
21
+ "num_heads": 64,
22
+ "num_layers": 24,
23
+ "output_past": true,
24
+ "pad_token_id": 0,
25
+ "relative_attention_max_distance": 128,
26
+ "relative_attention_num_buckets": 32,
27
+ "tie_word_embeddings": false,
28
+ "transformers_version": "4.45.2",
29
+ "use_cache": true,
30
+ "vocab_size": 32128
31
+ }
flax_model-00001-of-00002.msgpack ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ee027da5a91bc10d83209a748f5f2e9065371e8339415d9bc2aa826474f58911
3
+ size 4932715603
flax_model-00002-of-00002.msgpack ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:71707bbd21bd268bba4f6307c6c70880ec11936a815168a7ed672a00f61c5775
3
+ size 4591915533
flax_model.msgpack.index.json ADDED
@@ -0,0 +1,226 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 9524621312
4
+ },
5
+ "weight_map": {
6
+ "encoder/block/0/layer/0/SelfAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
7
+ "encoder/block/0/layer/0/SelfAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
8
+ "encoder/block/0/layer/0/SelfAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
9
+ "encoder/block/0/layer/0/SelfAttention/relative_attention_bias/embedding": "flax_model-00001-of-00002.msgpack",
10
+ "encoder/block/0/layer/0/SelfAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
11
+ "encoder/block/0/layer/0/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
12
+ "encoder/block/0/layer/1/DenseReluDense/wi_0/kernel": "flax_model-00001-of-00002.msgpack",
13
+ "encoder/block/0/layer/1/DenseReluDense/wi_1/kernel": "flax_model-00001-of-00002.msgpack",
14
+ "encoder/block/0/layer/1/DenseReluDense/wo/kernel": "flax_model-00001-of-00002.msgpack",
15
+ "encoder/block/0/layer/1/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
16
+ "encoder/block/1/layer/0/SelfAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
17
+ "encoder/block/1/layer/0/SelfAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
18
+ "encoder/block/1/layer/0/SelfAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
19
+ "encoder/block/1/layer/0/SelfAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
20
+ "encoder/block/1/layer/0/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
21
+ "encoder/block/1/layer/1/DenseReluDense/wi_0/kernel": "flax_model-00001-of-00002.msgpack",
22
+ "encoder/block/1/layer/1/DenseReluDense/wi_1/kernel": "flax_model-00001-of-00002.msgpack",
23
+ "encoder/block/1/layer/1/DenseReluDense/wo/kernel": "flax_model-00001-of-00002.msgpack",
24
+ "encoder/block/1/layer/1/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
25
+ "encoder/block/10/layer/0/SelfAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
26
+ "encoder/block/10/layer/0/SelfAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
27
+ "encoder/block/10/layer/0/SelfAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
28
+ "encoder/block/10/layer/0/SelfAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
29
+ "encoder/block/10/layer/0/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
30
+ "encoder/block/10/layer/1/DenseReluDense/wi_0/kernel": "flax_model-00001-of-00002.msgpack",
31
+ "encoder/block/10/layer/1/DenseReluDense/wi_1/kernel": "flax_model-00001-of-00002.msgpack",
32
+ "encoder/block/10/layer/1/DenseReluDense/wo/kernel": "flax_model-00001-of-00002.msgpack",
33
+ "encoder/block/10/layer/1/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
34
+ "encoder/block/11/layer/0/SelfAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
35
+ "encoder/block/11/layer/0/SelfAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
36
+ "encoder/block/11/layer/0/SelfAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
37
+ "encoder/block/11/layer/0/SelfAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
38
+ "encoder/block/11/layer/0/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
39
+ "encoder/block/11/layer/1/DenseReluDense/wi_0/kernel": "flax_model-00001-of-00002.msgpack",
40
+ "encoder/block/11/layer/1/DenseReluDense/wi_1/kernel": "flax_model-00001-of-00002.msgpack",
41
+ "encoder/block/11/layer/1/DenseReluDense/wo/kernel": "flax_model-00001-of-00002.msgpack",
42
+ "encoder/block/11/layer/1/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
43
+ "encoder/block/12/layer/0/SelfAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
44
+ "encoder/block/12/layer/0/SelfAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
45
+ "encoder/block/12/layer/0/SelfAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
46
+ "encoder/block/12/layer/0/SelfAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
47
+ "encoder/block/12/layer/0/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
48
+ "encoder/block/12/layer/1/DenseReluDense/wi_0/kernel": "flax_model-00001-of-00002.msgpack",
49
+ "encoder/block/12/layer/1/DenseReluDense/wi_1/kernel": "flax_model-00001-of-00002.msgpack",
50
+ "encoder/block/12/layer/1/DenseReluDense/wo/kernel": "flax_model-00001-of-00002.msgpack",
51
+ "encoder/block/12/layer/1/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
52
+ "encoder/block/13/layer/0/SelfAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
53
+ "encoder/block/13/layer/0/SelfAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
54
+ "encoder/block/13/layer/0/SelfAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
55
+ "encoder/block/13/layer/0/SelfAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
56
+ "encoder/block/13/layer/0/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
57
+ "encoder/block/13/layer/1/DenseReluDense/wi_0/kernel": "flax_model-00001-of-00002.msgpack",
58
+ "encoder/block/13/layer/1/DenseReluDense/wi_1/kernel": "flax_model-00001-of-00002.msgpack",
59
+ "encoder/block/13/layer/1/DenseReluDense/wo/kernel": "flax_model-00001-of-00002.msgpack",
60
+ "encoder/block/13/layer/1/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
61
+ "encoder/block/14/layer/0/SelfAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
62
+ "encoder/block/14/layer/0/SelfAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
63
+ "encoder/block/14/layer/0/SelfAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
64
+ "encoder/block/14/layer/0/SelfAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
65
+ "encoder/block/14/layer/0/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
66
+ "encoder/block/14/layer/1/DenseReluDense/wi_0/kernel": "flax_model-00001-of-00002.msgpack",
67
+ "encoder/block/14/layer/1/DenseReluDense/wi_1/kernel": "flax_model-00001-of-00002.msgpack",
68
+ "encoder/block/14/layer/1/DenseReluDense/wo/kernel": "flax_model-00001-of-00002.msgpack",
69
+ "encoder/block/14/layer/1/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
70
+ "encoder/block/15/layer/0/SelfAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
71
+ "encoder/block/15/layer/0/SelfAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
72
+ "encoder/block/15/layer/0/SelfAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
73
+ "encoder/block/15/layer/0/SelfAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
74
+ "encoder/block/15/layer/0/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
75
+ "encoder/block/15/layer/1/DenseReluDense/wi_0/kernel": "flax_model-00001-of-00002.msgpack",
76
+ "encoder/block/15/layer/1/DenseReluDense/wi_1/kernel": "flax_model-00001-of-00002.msgpack",
77
+ "encoder/block/15/layer/1/DenseReluDense/wo/kernel": "flax_model-00001-of-00002.msgpack",
78
+ "encoder/block/15/layer/1/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
79
+ "encoder/block/16/layer/0/SelfAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
80
+ "encoder/block/16/layer/0/SelfAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
81
+ "encoder/block/16/layer/0/SelfAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
82
+ "encoder/block/16/layer/0/SelfAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
83
+ "encoder/block/16/layer/0/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
84
+ "encoder/block/16/layer/1/DenseReluDense/wi_0/kernel": "flax_model-00001-of-00002.msgpack",
85
+ "encoder/block/16/layer/1/DenseReluDense/wi_1/kernel": "flax_model-00001-of-00002.msgpack",
86
+ "encoder/block/16/layer/1/DenseReluDense/wo/kernel": "flax_model-00001-of-00002.msgpack",
87
+ "encoder/block/16/layer/1/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
88
+ "encoder/block/17/layer/0/SelfAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
89
+ "encoder/block/17/layer/0/SelfAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
90
+ "encoder/block/17/layer/0/SelfAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
91
+ "encoder/block/17/layer/0/SelfAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
92
+ "encoder/block/17/layer/0/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
93
+ "encoder/block/17/layer/1/DenseReluDense/wi_0/kernel": "flax_model-00001-of-00002.msgpack",
94
+ "encoder/block/17/layer/1/DenseReluDense/wi_1/kernel": "flax_model-00001-of-00002.msgpack",
95
+ "encoder/block/17/layer/1/DenseReluDense/wo/kernel": "flax_model-00001-of-00002.msgpack",
96
+ "encoder/block/17/layer/1/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
97
+ "encoder/block/18/layer/0/SelfAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
98
+ "encoder/block/18/layer/0/SelfAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
99
+ "encoder/block/18/layer/0/SelfAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
100
+ "encoder/block/18/layer/0/SelfAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
101
+ "encoder/block/18/layer/0/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
102
+ "encoder/block/18/layer/1/DenseReluDense/wi_0/kernel": "flax_model-00001-of-00002.msgpack",
103
+ "encoder/block/18/layer/1/DenseReluDense/wi_1/kernel": "flax_model-00001-of-00002.msgpack",
104
+ "encoder/block/18/layer/1/DenseReluDense/wo/kernel": "flax_model-00001-of-00002.msgpack",
105
+ "encoder/block/18/layer/1/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
106
+ "encoder/block/19/layer/0/SelfAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
107
+ "encoder/block/19/layer/0/SelfAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
108
+ "encoder/block/19/layer/0/SelfAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
109
+ "encoder/block/19/layer/0/SelfAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
110
+ "encoder/block/19/layer/0/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
111
+ "encoder/block/19/layer/1/DenseReluDense/wi_0/kernel": "flax_model-00001-of-00002.msgpack",
112
+ "encoder/block/19/layer/1/DenseReluDense/wi_1/kernel": "flax_model-00001-of-00002.msgpack",
113
+ "encoder/block/19/layer/1/DenseReluDense/wo/kernel": "flax_model-00001-of-00002.msgpack",
114
+ "encoder/block/19/layer/1/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
115
+ "encoder/block/2/layer/0/SelfAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
116
+ "encoder/block/2/layer/0/SelfAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
117
+ "encoder/block/2/layer/0/SelfAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
118
+ "encoder/block/2/layer/0/SelfAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
119
+ "encoder/block/2/layer/0/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
120
+ "encoder/block/2/layer/1/DenseReluDense/wi_0/kernel": "flax_model-00001-of-00002.msgpack",
121
+ "encoder/block/2/layer/1/DenseReluDense/wi_1/kernel": "flax_model-00001-of-00002.msgpack",
122
+ "encoder/block/2/layer/1/DenseReluDense/wo/kernel": "flax_model-00002-of-00002.msgpack",
123
+ "encoder/block/2/layer/1/layer_norm/weight": "flax_model-00002-of-00002.msgpack",
124
+ "encoder/block/20/layer/0/SelfAttention/k/kernel": "flax_model-00002-of-00002.msgpack",
125
+ "encoder/block/20/layer/0/SelfAttention/o/kernel": "flax_model-00002-of-00002.msgpack",
126
+ "encoder/block/20/layer/0/SelfAttention/q/kernel": "flax_model-00002-of-00002.msgpack",
127
+ "encoder/block/20/layer/0/SelfAttention/v/kernel": "flax_model-00002-of-00002.msgpack",
128
+ "encoder/block/20/layer/0/layer_norm/weight": "flax_model-00002-of-00002.msgpack",
129
+ "encoder/block/20/layer/1/DenseReluDense/wi_0/kernel": "flax_model-00002-of-00002.msgpack",
130
+ "encoder/block/20/layer/1/DenseReluDense/wi_1/kernel": "flax_model-00002-of-00002.msgpack",
131
+ "encoder/block/20/layer/1/DenseReluDense/wo/kernel": "flax_model-00002-of-00002.msgpack",
132
+ "encoder/block/20/layer/1/layer_norm/weight": "flax_model-00002-of-00002.msgpack",
133
+ "encoder/block/21/layer/0/SelfAttention/k/kernel": "flax_model-00002-of-00002.msgpack",
134
+ "encoder/block/21/layer/0/SelfAttention/o/kernel": "flax_model-00002-of-00002.msgpack",
135
+ "encoder/block/21/layer/0/SelfAttention/q/kernel": "flax_model-00002-of-00002.msgpack",
136
+ "encoder/block/21/layer/0/SelfAttention/v/kernel": "flax_model-00002-of-00002.msgpack",
137
+ "encoder/block/21/layer/0/layer_norm/weight": "flax_model-00002-of-00002.msgpack",
138
+ "encoder/block/21/layer/1/DenseReluDense/wi_0/kernel": "flax_model-00002-of-00002.msgpack",
139
+ "encoder/block/21/layer/1/DenseReluDense/wi_1/kernel": "flax_model-00002-of-00002.msgpack",
140
+ "encoder/block/21/layer/1/DenseReluDense/wo/kernel": "flax_model-00002-of-00002.msgpack",
141
+ "encoder/block/21/layer/1/layer_norm/weight": "flax_model-00002-of-00002.msgpack",
142
+ "encoder/block/22/layer/0/SelfAttention/k/kernel": "flax_model-00002-of-00002.msgpack",
143
+ "encoder/block/22/layer/0/SelfAttention/o/kernel": "flax_model-00002-of-00002.msgpack",
144
+ "encoder/block/22/layer/0/SelfAttention/q/kernel": "flax_model-00002-of-00002.msgpack",
145
+ "encoder/block/22/layer/0/SelfAttention/v/kernel": "flax_model-00002-of-00002.msgpack",
146
+ "encoder/block/22/layer/0/layer_norm/weight": "flax_model-00002-of-00002.msgpack",
147
+ "encoder/block/22/layer/1/DenseReluDense/wi_0/kernel": "flax_model-00002-of-00002.msgpack",
148
+ "encoder/block/22/layer/1/DenseReluDense/wi_1/kernel": "flax_model-00002-of-00002.msgpack",
149
+ "encoder/block/22/layer/1/DenseReluDense/wo/kernel": "flax_model-00002-of-00002.msgpack",
150
+ "encoder/block/22/layer/1/layer_norm/weight": "flax_model-00002-of-00002.msgpack",
151
+ "encoder/block/23/layer/0/SelfAttention/k/kernel": "flax_model-00002-of-00002.msgpack",
152
+ "encoder/block/23/layer/0/SelfAttention/o/kernel": "flax_model-00002-of-00002.msgpack",
153
+ "encoder/block/23/layer/0/SelfAttention/q/kernel": "flax_model-00002-of-00002.msgpack",
154
+ "encoder/block/23/layer/0/SelfAttention/v/kernel": "flax_model-00002-of-00002.msgpack",
155
+ "encoder/block/23/layer/0/layer_norm/weight": "flax_model-00002-of-00002.msgpack",
156
+ "encoder/block/23/layer/1/DenseReluDense/wi_0/kernel": "flax_model-00002-of-00002.msgpack",
157
+ "encoder/block/23/layer/1/DenseReluDense/wi_1/kernel": "flax_model-00002-of-00002.msgpack",
158
+ "encoder/block/23/layer/1/DenseReluDense/wo/kernel": "flax_model-00002-of-00002.msgpack",
159
+ "encoder/block/23/layer/1/layer_norm/weight": "flax_model-00002-of-00002.msgpack",
160
+ "encoder/block/3/layer/0/SelfAttention/k/kernel": "flax_model-00002-of-00002.msgpack",
161
+ "encoder/block/3/layer/0/SelfAttention/o/kernel": "flax_model-00002-of-00002.msgpack",
162
+ "encoder/block/3/layer/0/SelfAttention/q/kernel": "flax_model-00002-of-00002.msgpack",
163
+ "encoder/block/3/layer/0/SelfAttention/v/kernel": "flax_model-00002-of-00002.msgpack",
164
+ "encoder/block/3/layer/0/layer_norm/weight": "flax_model-00002-of-00002.msgpack",
165
+ "encoder/block/3/layer/1/DenseReluDense/wi_0/kernel": "flax_model-00002-of-00002.msgpack",
166
+ "encoder/block/3/layer/1/DenseReluDense/wi_1/kernel": "flax_model-00002-of-00002.msgpack",
167
+ "encoder/block/3/layer/1/DenseReluDense/wo/kernel": "flax_model-00002-of-00002.msgpack",
168
+ "encoder/block/3/layer/1/layer_norm/weight": "flax_model-00002-of-00002.msgpack",
169
+ "encoder/block/4/layer/0/SelfAttention/k/kernel": "flax_model-00002-of-00002.msgpack",
170
+ "encoder/block/4/layer/0/SelfAttention/o/kernel": "flax_model-00002-of-00002.msgpack",
171
+ "encoder/block/4/layer/0/SelfAttention/q/kernel": "flax_model-00002-of-00002.msgpack",
172
+ "encoder/block/4/layer/0/SelfAttention/v/kernel": "flax_model-00002-of-00002.msgpack",
173
+ "encoder/block/4/layer/0/layer_norm/weight": "flax_model-00002-of-00002.msgpack",
174
+ "encoder/block/4/layer/1/DenseReluDense/wi_0/kernel": "flax_model-00002-of-00002.msgpack",
175
+ "encoder/block/4/layer/1/DenseReluDense/wi_1/kernel": "flax_model-00002-of-00002.msgpack",
176
+ "encoder/block/4/layer/1/DenseReluDense/wo/kernel": "flax_model-00002-of-00002.msgpack",
177
+ "encoder/block/4/layer/1/layer_norm/weight": "flax_model-00002-of-00002.msgpack",
178
+ "encoder/block/5/layer/0/SelfAttention/k/kernel": "flax_model-00002-of-00002.msgpack",
179
+ "encoder/block/5/layer/0/SelfAttention/o/kernel": "flax_model-00002-of-00002.msgpack",
180
+ "encoder/block/5/layer/0/SelfAttention/q/kernel": "flax_model-00002-of-00002.msgpack",
181
+ "encoder/block/5/layer/0/SelfAttention/v/kernel": "flax_model-00002-of-00002.msgpack",
182
+ "encoder/block/5/layer/0/layer_norm/weight": "flax_model-00002-of-00002.msgpack",
183
+ "encoder/block/5/layer/1/DenseReluDense/wi_0/kernel": "flax_model-00002-of-00002.msgpack",
184
+ "encoder/block/5/layer/1/DenseReluDense/wi_1/kernel": "flax_model-00002-of-00002.msgpack",
185
+ "encoder/block/5/layer/1/DenseReluDense/wo/kernel": "flax_model-00002-of-00002.msgpack",
186
+ "encoder/block/5/layer/1/layer_norm/weight": "flax_model-00002-of-00002.msgpack",
187
+ "encoder/block/6/layer/0/SelfAttention/k/kernel": "flax_model-00002-of-00002.msgpack",
188
+ "encoder/block/6/layer/0/SelfAttention/o/kernel": "flax_model-00002-of-00002.msgpack",
189
+ "encoder/block/6/layer/0/SelfAttention/q/kernel": "flax_model-00002-of-00002.msgpack",
190
+ "encoder/block/6/layer/0/SelfAttention/v/kernel": "flax_model-00002-of-00002.msgpack",
191
+ "encoder/block/6/layer/0/layer_norm/weight": "flax_model-00002-of-00002.msgpack",
192
+ "encoder/block/6/layer/1/DenseReluDense/wi_0/kernel": "flax_model-00002-of-00002.msgpack",
193
+ "encoder/block/6/layer/1/DenseReluDense/wi_1/kernel": "flax_model-00002-of-00002.msgpack",
194
+ "encoder/block/6/layer/1/DenseReluDense/wo/kernel": "flax_model-00002-of-00002.msgpack",
195
+ "encoder/block/6/layer/1/layer_norm/weight": "flax_model-00002-of-00002.msgpack",
196
+ "encoder/block/7/layer/0/SelfAttention/k/kernel": "flax_model-00002-of-00002.msgpack",
197
+ "encoder/block/7/layer/0/SelfAttention/o/kernel": "flax_model-00002-of-00002.msgpack",
198
+ "encoder/block/7/layer/0/SelfAttention/q/kernel": "flax_model-00002-of-00002.msgpack",
199
+ "encoder/block/7/layer/0/SelfAttention/v/kernel": "flax_model-00002-of-00002.msgpack",
200
+ "encoder/block/7/layer/0/layer_norm/weight": "flax_model-00002-of-00002.msgpack",
201
+ "encoder/block/7/layer/1/DenseReluDense/wi_0/kernel": "flax_model-00002-of-00002.msgpack",
202
+ "encoder/block/7/layer/1/DenseReluDense/wi_1/kernel": "flax_model-00002-of-00002.msgpack",
203
+ "encoder/block/7/layer/1/DenseReluDense/wo/kernel": "flax_model-00002-of-00002.msgpack",
204
+ "encoder/block/7/layer/1/layer_norm/weight": "flax_model-00002-of-00002.msgpack",
205
+ "encoder/block/8/layer/0/SelfAttention/k/kernel": "flax_model-00002-of-00002.msgpack",
206
+ "encoder/block/8/layer/0/SelfAttention/o/kernel": "flax_model-00002-of-00002.msgpack",
207
+ "encoder/block/8/layer/0/SelfAttention/q/kernel": "flax_model-00002-of-00002.msgpack",
208
+ "encoder/block/8/layer/0/SelfAttention/v/kernel": "flax_model-00002-of-00002.msgpack",
209
+ "encoder/block/8/layer/0/layer_norm/weight": "flax_model-00002-of-00002.msgpack",
210
+ "encoder/block/8/layer/1/DenseReluDense/wi_0/kernel": "flax_model-00002-of-00002.msgpack",
211
+ "encoder/block/8/layer/1/DenseReluDense/wi_1/kernel": "flax_model-00002-of-00002.msgpack",
212
+ "encoder/block/8/layer/1/DenseReluDense/wo/kernel": "flax_model-00002-of-00002.msgpack",
213
+ "encoder/block/8/layer/1/layer_norm/weight": "flax_model-00002-of-00002.msgpack",
214
+ "encoder/block/9/layer/0/SelfAttention/k/kernel": "flax_model-00002-of-00002.msgpack",
215
+ "encoder/block/9/layer/0/SelfAttention/o/kernel": "flax_model-00002-of-00002.msgpack",
216
+ "encoder/block/9/layer/0/SelfAttention/q/kernel": "flax_model-00002-of-00002.msgpack",
217
+ "encoder/block/9/layer/0/SelfAttention/v/kernel": "flax_model-00002-of-00002.msgpack",
218
+ "encoder/block/9/layer/0/layer_norm/weight": "flax_model-00002-of-00002.msgpack",
219
+ "encoder/block/9/layer/1/DenseReluDense/wi_0/kernel": "flax_model-00002-of-00002.msgpack",
220
+ "encoder/block/9/layer/1/DenseReluDense/wi_1/kernel": "flax_model-00002-of-00002.msgpack",
221
+ "encoder/block/9/layer/1/DenseReluDense/wo/kernel": "flax_model-00002-of-00002.msgpack",
222
+ "encoder/block/9/layer/1/layer_norm/weight": "flax_model-00002-of-00002.msgpack",
223
+ "encoder/final_layer_norm/weight": "flax_model-00002-of-00002.msgpack",
224
+ "shared/embedding": "flax_model-00002-of-00002.msgpack"
225
+ }
226
+ }