hamishivi commited on
Commit
32a6e03
1 Parent(s): b807587

upload model

Browse files
config.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "T5ForConditionalGeneration"
4
+ ],
5
+ "d_ff": 5120,
6
+ "d_kv": 64,
7
+ "d_model": 2048,
8
+ "decoder_start_token_id": 0,
9
+ "dense_act_fn": "gelu_new",
10
+ "dropout_rate": 0.1,
11
+ "eos_token_id": 1,
12
+ "feed_forward_proj": "gated-gelu",
13
+ "gradient_checkpointing": false,
14
+ "initializer_factor": 1.0,
15
+ "is_encoder_decoder": true,
16
+ "is_gated_act": true,
17
+ "layer_norm_epsilon": 1e-06,
18
+ "model_type": "t5",
19
+ "num_decoder_layers": 24,
20
+ "num_heads": 32,
21
+ "num_layers": 24,
22
+ "output_past": true,
23
+ "pad_token_id": 0,
24
+ "relative_attention_max_distance": 128,
25
+ "relative_attention_num_buckets": 32,
26
+ "tie_word_embeddings": false,
27
+ "transformers_version": "4.21.1",
28
+ "use_cache": true,
29
+ "vocab_size": 32128
30
+ }
flax_model-00001-of-00002.msgpack ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7014608c1aacbc95774558c58195f488cd84c58efde98cc85e62f001455d9cbb
3
+ size 9975002822
flax_model-00002-of-00002.msgpack ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b499a8bd82509fa16c6bbd570706cdab499bd7eae3454bfb0803098de3b88fc3
3
+ size 1424050028
flax_model.msgpack.index.json ADDED
@@ -0,0 +1,565 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 11399028736
4
+ },
5
+ "weight_map": {
6
+ "decoder/block/0/layer/0/SelfAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
7
+ "decoder/block/0/layer/0/SelfAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
8
+ "decoder/block/0/layer/0/SelfAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
9
+ "decoder/block/0/layer/0/SelfAttention/relative_attention_bias/embedding": "flax_model-00001-of-00002.msgpack",
10
+ "decoder/block/0/layer/0/SelfAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
11
+ "decoder/block/0/layer/0/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
12
+ "decoder/block/0/layer/1/EncDecAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
13
+ "decoder/block/0/layer/1/EncDecAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
14
+ "decoder/block/0/layer/1/EncDecAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
15
+ "decoder/block/0/layer/1/EncDecAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
16
+ "decoder/block/0/layer/1/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
17
+ "decoder/block/0/layer/2/DenseReluDense/wi_0/kernel": "flax_model-00001-of-00002.msgpack",
18
+ "decoder/block/0/layer/2/DenseReluDense/wi_1/kernel": "flax_model-00001-of-00002.msgpack",
19
+ "decoder/block/0/layer/2/DenseReluDense/wo/kernel": "flax_model-00001-of-00002.msgpack",
20
+ "decoder/block/0/layer/2/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
21
+ "decoder/block/1/layer/0/SelfAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
22
+ "decoder/block/1/layer/0/SelfAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
23
+ "decoder/block/1/layer/0/SelfAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
24
+ "decoder/block/1/layer/0/SelfAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
25
+ "decoder/block/1/layer/0/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
26
+ "decoder/block/1/layer/1/EncDecAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
27
+ "decoder/block/1/layer/1/EncDecAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
28
+ "decoder/block/1/layer/1/EncDecAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
29
+ "decoder/block/1/layer/1/EncDecAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
30
+ "decoder/block/1/layer/1/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
31
+ "decoder/block/1/layer/2/DenseReluDense/wi_0/kernel": "flax_model-00001-of-00002.msgpack",
32
+ "decoder/block/1/layer/2/DenseReluDense/wi_1/kernel": "flax_model-00001-of-00002.msgpack",
33
+ "decoder/block/1/layer/2/DenseReluDense/wo/kernel": "flax_model-00001-of-00002.msgpack",
34
+ "decoder/block/1/layer/2/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
35
+ "decoder/block/10/layer/0/SelfAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
36
+ "decoder/block/10/layer/0/SelfAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
37
+ "decoder/block/10/layer/0/SelfAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
38
+ "decoder/block/10/layer/0/SelfAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
39
+ "decoder/block/10/layer/0/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
40
+ "decoder/block/10/layer/1/EncDecAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
41
+ "decoder/block/10/layer/1/EncDecAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
42
+ "decoder/block/10/layer/1/EncDecAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
43
+ "decoder/block/10/layer/1/EncDecAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
44
+ "decoder/block/10/layer/1/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
45
+ "decoder/block/10/layer/2/DenseReluDense/wi_0/kernel": "flax_model-00001-of-00002.msgpack",
46
+ "decoder/block/10/layer/2/DenseReluDense/wi_1/kernel": "flax_model-00001-of-00002.msgpack",
47
+ "decoder/block/10/layer/2/DenseReluDense/wo/kernel": "flax_model-00001-of-00002.msgpack",
48
+ "decoder/block/10/layer/2/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
49
+ "decoder/block/11/layer/0/SelfAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
50
+ "decoder/block/11/layer/0/SelfAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
51
+ "decoder/block/11/layer/0/SelfAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
52
+ "decoder/block/11/layer/0/SelfAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
53
+ "decoder/block/11/layer/0/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
54
+ "decoder/block/11/layer/1/EncDecAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
55
+ "decoder/block/11/layer/1/EncDecAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
56
+ "decoder/block/11/layer/1/EncDecAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
57
+ "decoder/block/11/layer/1/EncDecAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
58
+ "decoder/block/11/layer/1/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
59
+ "decoder/block/11/layer/2/DenseReluDense/wi_0/kernel": "flax_model-00001-of-00002.msgpack",
60
+ "decoder/block/11/layer/2/DenseReluDense/wi_1/kernel": "flax_model-00001-of-00002.msgpack",
61
+ "decoder/block/11/layer/2/DenseReluDense/wo/kernel": "flax_model-00001-of-00002.msgpack",
62
+ "decoder/block/11/layer/2/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
63
+ "decoder/block/12/layer/0/SelfAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
64
+ "decoder/block/12/layer/0/SelfAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
65
+ "decoder/block/12/layer/0/SelfAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
66
+ "decoder/block/12/layer/0/SelfAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
67
+ "decoder/block/12/layer/0/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
68
+ "decoder/block/12/layer/1/EncDecAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
69
+ "decoder/block/12/layer/1/EncDecAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
70
+ "decoder/block/12/layer/1/EncDecAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
71
+ "decoder/block/12/layer/1/EncDecAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
72
+ "decoder/block/12/layer/1/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
73
+ "decoder/block/12/layer/2/DenseReluDense/wi_0/kernel": "flax_model-00001-of-00002.msgpack",
74
+ "decoder/block/12/layer/2/DenseReluDense/wi_1/kernel": "flax_model-00001-of-00002.msgpack",
75
+ "decoder/block/12/layer/2/DenseReluDense/wo/kernel": "flax_model-00001-of-00002.msgpack",
76
+ "decoder/block/12/layer/2/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
77
+ "decoder/block/13/layer/0/SelfAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
78
+ "decoder/block/13/layer/0/SelfAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
79
+ "decoder/block/13/layer/0/SelfAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
80
+ "decoder/block/13/layer/0/SelfAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
81
+ "decoder/block/13/layer/0/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
82
+ "decoder/block/13/layer/1/EncDecAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
83
+ "decoder/block/13/layer/1/EncDecAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
84
+ "decoder/block/13/layer/1/EncDecAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
85
+ "decoder/block/13/layer/1/EncDecAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
86
+ "decoder/block/13/layer/1/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
87
+ "decoder/block/13/layer/2/DenseReluDense/wi_0/kernel": "flax_model-00001-of-00002.msgpack",
88
+ "decoder/block/13/layer/2/DenseReluDense/wi_1/kernel": "flax_model-00001-of-00002.msgpack",
89
+ "decoder/block/13/layer/2/DenseReluDense/wo/kernel": "flax_model-00001-of-00002.msgpack",
90
+ "decoder/block/13/layer/2/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
91
+ "decoder/block/14/layer/0/SelfAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
92
+ "decoder/block/14/layer/0/SelfAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
93
+ "decoder/block/14/layer/0/SelfAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
94
+ "decoder/block/14/layer/0/SelfAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
95
+ "decoder/block/14/layer/0/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
96
+ "decoder/block/14/layer/1/EncDecAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
97
+ "decoder/block/14/layer/1/EncDecAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
98
+ "decoder/block/14/layer/1/EncDecAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
99
+ "decoder/block/14/layer/1/EncDecAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
100
+ "decoder/block/14/layer/1/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
101
+ "decoder/block/14/layer/2/DenseReluDense/wi_0/kernel": "flax_model-00001-of-00002.msgpack",
102
+ "decoder/block/14/layer/2/DenseReluDense/wi_1/kernel": "flax_model-00001-of-00002.msgpack",
103
+ "decoder/block/14/layer/2/DenseReluDense/wo/kernel": "flax_model-00001-of-00002.msgpack",
104
+ "decoder/block/14/layer/2/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
105
+ "decoder/block/15/layer/0/SelfAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
106
+ "decoder/block/15/layer/0/SelfAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
107
+ "decoder/block/15/layer/0/SelfAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
108
+ "decoder/block/15/layer/0/SelfAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
109
+ "decoder/block/15/layer/0/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
110
+ "decoder/block/15/layer/1/EncDecAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
111
+ "decoder/block/15/layer/1/EncDecAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
112
+ "decoder/block/15/layer/1/EncDecAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
113
+ "decoder/block/15/layer/1/EncDecAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
114
+ "decoder/block/15/layer/1/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
115
+ "decoder/block/15/layer/2/DenseReluDense/wi_0/kernel": "flax_model-00001-of-00002.msgpack",
116
+ "decoder/block/15/layer/2/DenseReluDense/wi_1/kernel": "flax_model-00001-of-00002.msgpack",
117
+ "decoder/block/15/layer/2/DenseReluDense/wo/kernel": "flax_model-00001-of-00002.msgpack",
118
+ "decoder/block/15/layer/2/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
119
+ "decoder/block/16/layer/0/SelfAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
120
+ "decoder/block/16/layer/0/SelfAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
121
+ "decoder/block/16/layer/0/SelfAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
122
+ "decoder/block/16/layer/0/SelfAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
123
+ "decoder/block/16/layer/0/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
124
+ "decoder/block/16/layer/1/EncDecAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
125
+ "decoder/block/16/layer/1/EncDecAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
126
+ "decoder/block/16/layer/1/EncDecAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
127
+ "decoder/block/16/layer/1/EncDecAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
128
+ "decoder/block/16/layer/1/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
129
+ "decoder/block/16/layer/2/DenseReluDense/wi_0/kernel": "flax_model-00001-of-00002.msgpack",
130
+ "decoder/block/16/layer/2/DenseReluDense/wi_1/kernel": "flax_model-00001-of-00002.msgpack",
131
+ "decoder/block/16/layer/2/DenseReluDense/wo/kernel": "flax_model-00001-of-00002.msgpack",
132
+ "decoder/block/16/layer/2/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
133
+ "decoder/block/17/layer/0/SelfAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
134
+ "decoder/block/17/layer/0/SelfAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
135
+ "decoder/block/17/layer/0/SelfAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
136
+ "decoder/block/17/layer/0/SelfAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
137
+ "decoder/block/17/layer/0/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
138
+ "decoder/block/17/layer/1/EncDecAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
139
+ "decoder/block/17/layer/1/EncDecAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
140
+ "decoder/block/17/layer/1/EncDecAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
141
+ "decoder/block/17/layer/1/EncDecAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
142
+ "decoder/block/17/layer/1/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
143
+ "decoder/block/17/layer/2/DenseReluDense/wi_0/kernel": "flax_model-00001-of-00002.msgpack",
144
+ "decoder/block/17/layer/2/DenseReluDense/wi_1/kernel": "flax_model-00001-of-00002.msgpack",
145
+ "decoder/block/17/layer/2/DenseReluDense/wo/kernel": "flax_model-00001-of-00002.msgpack",
146
+ "decoder/block/17/layer/2/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
147
+ "decoder/block/18/layer/0/SelfAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
148
+ "decoder/block/18/layer/0/SelfAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
149
+ "decoder/block/18/layer/0/SelfAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
150
+ "decoder/block/18/layer/0/SelfAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
151
+ "decoder/block/18/layer/0/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
152
+ "decoder/block/18/layer/1/EncDecAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
153
+ "decoder/block/18/layer/1/EncDecAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
154
+ "decoder/block/18/layer/1/EncDecAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
155
+ "decoder/block/18/layer/1/EncDecAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
156
+ "decoder/block/18/layer/1/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
157
+ "decoder/block/18/layer/2/DenseReluDense/wi_0/kernel": "flax_model-00001-of-00002.msgpack",
158
+ "decoder/block/18/layer/2/DenseReluDense/wi_1/kernel": "flax_model-00001-of-00002.msgpack",
159
+ "decoder/block/18/layer/2/DenseReluDense/wo/kernel": "flax_model-00001-of-00002.msgpack",
160
+ "decoder/block/18/layer/2/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
161
+ "decoder/block/19/layer/0/SelfAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
162
+ "decoder/block/19/layer/0/SelfAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
163
+ "decoder/block/19/layer/0/SelfAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
164
+ "decoder/block/19/layer/0/SelfAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
165
+ "decoder/block/19/layer/0/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
166
+ "decoder/block/19/layer/1/EncDecAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
167
+ "decoder/block/19/layer/1/EncDecAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
168
+ "decoder/block/19/layer/1/EncDecAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
169
+ "decoder/block/19/layer/1/EncDecAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
170
+ "decoder/block/19/layer/1/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
171
+ "decoder/block/19/layer/2/DenseReluDense/wi_0/kernel": "flax_model-00001-of-00002.msgpack",
172
+ "decoder/block/19/layer/2/DenseReluDense/wi_1/kernel": "flax_model-00001-of-00002.msgpack",
173
+ "decoder/block/19/layer/2/DenseReluDense/wo/kernel": "flax_model-00001-of-00002.msgpack",
174
+ "decoder/block/19/layer/2/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
175
+ "decoder/block/2/layer/0/SelfAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
176
+ "decoder/block/2/layer/0/SelfAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
177
+ "decoder/block/2/layer/0/SelfAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
178
+ "decoder/block/2/layer/0/SelfAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
179
+ "decoder/block/2/layer/0/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
180
+ "decoder/block/2/layer/1/EncDecAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
181
+ "decoder/block/2/layer/1/EncDecAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
182
+ "decoder/block/2/layer/1/EncDecAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
183
+ "decoder/block/2/layer/1/EncDecAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
184
+ "decoder/block/2/layer/1/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
185
+ "decoder/block/2/layer/2/DenseReluDense/wi_0/kernel": "flax_model-00001-of-00002.msgpack",
186
+ "decoder/block/2/layer/2/DenseReluDense/wi_1/kernel": "flax_model-00001-of-00002.msgpack",
187
+ "decoder/block/2/layer/2/DenseReluDense/wo/kernel": "flax_model-00001-of-00002.msgpack",
188
+ "decoder/block/2/layer/2/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
189
+ "decoder/block/20/layer/0/SelfAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
190
+ "decoder/block/20/layer/0/SelfAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
191
+ "decoder/block/20/layer/0/SelfAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
192
+ "decoder/block/20/layer/0/SelfAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
193
+ "decoder/block/20/layer/0/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
194
+ "decoder/block/20/layer/1/EncDecAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
195
+ "decoder/block/20/layer/1/EncDecAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
196
+ "decoder/block/20/layer/1/EncDecAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
197
+ "decoder/block/20/layer/1/EncDecAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
198
+ "decoder/block/20/layer/1/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
199
+ "decoder/block/20/layer/2/DenseReluDense/wi_0/kernel": "flax_model-00001-of-00002.msgpack",
200
+ "decoder/block/20/layer/2/DenseReluDense/wi_1/kernel": "flax_model-00001-of-00002.msgpack",
201
+ "decoder/block/20/layer/2/DenseReluDense/wo/kernel": "flax_model-00001-of-00002.msgpack",
202
+ "decoder/block/20/layer/2/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
203
+ "decoder/block/21/layer/0/SelfAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
204
+ "decoder/block/21/layer/0/SelfAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
205
+ "decoder/block/21/layer/0/SelfAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
206
+ "decoder/block/21/layer/0/SelfAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
207
+ "decoder/block/21/layer/0/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
208
+ "decoder/block/21/layer/1/EncDecAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
209
+ "decoder/block/21/layer/1/EncDecAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
210
+ "decoder/block/21/layer/1/EncDecAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
211
+ "decoder/block/21/layer/1/EncDecAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
212
+ "decoder/block/21/layer/1/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
213
+ "decoder/block/21/layer/2/DenseReluDense/wi_0/kernel": "flax_model-00001-of-00002.msgpack",
214
+ "decoder/block/21/layer/2/DenseReluDense/wi_1/kernel": "flax_model-00001-of-00002.msgpack",
215
+ "decoder/block/21/layer/2/DenseReluDense/wo/kernel": "flax_model-00001-of-00002.msgpack",
216
+ "decoder/block/21/layer/2/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
217
+ "decoder/block/22/layer/0/SelfAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
218
+ "decoder/block/22/layer/0/SelfAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
219
+ "decoder/block/22/layer/0/SelfAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
220
+ "decoder/block/22/layer/0/SelfAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
221
+ "decoder/block/22/layer/0/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
222
+ "decoder/block/22/layer/1/EncDecAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
223
+ "decoder/block/22/layer/1/EncDecAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
224
+ "decoder/block/22/layer/1/EncDecAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
225
+ "decoder/block/22/layer/1/EncDecAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
226
+ "decoder/block/22/layer/1/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
227
+ "decoder/block/22/layer/2/DenseReluDense/wi_0/kernel": "flax_model-00001-of-00002.msgpack",
228
+ "decoder/block/22/layer/2/DenseReluDense/wi_1/kernel": "flax_model-00001-of-00002.msgpack",
229
+ "decoder/block/22/layer/2/DenseReluDense/wo/kernel": "flax_model-00001-of-00002.msgpack",
230
+ "decoder/block/22/layer/2/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
231
+ "decoder/block/23/layer/0/SelfAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
232
+ "decoder/block/23/layer/0/SelfAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
233
+ "decoder/block/23/layer/0/SelfAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
234
+ "decoder/block/23/layer/0/SelfAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
235
+ "decoder/block/23/layer/0/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
236
+ "decoder/block/23/layer/1/EncDecAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
237
+ "decoder/block/23/layer/1/EncDecAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
238
+ "decoder/block/23/layer/1/EncDecAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
239
+ "decoder/block/23/layer/1/EncDecAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
240
+ "decoder/block/23/layer/1/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
241
+ "decoder/block/23/layer/2/DenseReluDense/wi_0/kernel": "flax_model-00001-of-00002.msgpack",
242
+ "decoder/block/23/layer/2/DenseReluDense/wi_1/kernel": "flax_model-00001-of-00002.msgpack",
243
+ "decoder/block/23/layer/2/DenseReluDense/wo/kernel": "flax_model-00001-of-00002.msgpack",
244
+ "decoder/block/23/layer/2/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
245
+ "decoder/block/3/layer/0/SelfAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
246
+ "decoder/block/3/layer/0/SelfAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
247
+ "decoder/block/3/layer/0/SelfAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
248
+ "decoder/block/3/layer/0/SelfAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
249
+ "decoder/block/3/layer/0/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
250
+ "decoder/block/3/layer/1/EncDecAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
251
+ "decoder/block/3/layer/1/EncDecAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
252
+ "decoder/block/3/layer/1/EncDecAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
253
+ "decoder/block/3/layer/1/EncDecAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
254
+ "decoder/block/3/layer/1/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
255
+ "decoder/block/3/layer/2/DenseReluDense/wi_0/kernel": "flax_model-00001-of-00002.msgpack",
256
+ "decoder/block/3/layer/2/DenseReluDense/wi_1/kernel": "flax_model-00001-of-00002.msgpack",
257
+ "decoder/block/3/layer/2/DenseReluDense/wo/kernel": "flax_model-00001-of-00002.msgpack",
258
+ "decoder/block/3/layer/2/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
259
+ "decoder/block/4/layer/0/SelfAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
260
+ "decoder/block/4/layer/0/SelfAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
261
+ "decoder/block/4/layer/0/SelfAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
262
+ "decoder/block/4/layer/0/SelfAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
263
+ "decoder/block/4/layer/0/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
264
+ "decoder/block/4/layer/1/EncDecAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
265
+ "decoder/block/4/layer/1/EncDecAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
266
+ "decoder/block/4/layer/1/EncDecAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
267
+ "decoder/block/4/layer/1/EncDecAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
268
+ "decoder/block/4/layer/1/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
269
+ "decoder/block/4/layer/2/DenseReluDense/wi_0/kernel": "flax_model-00001-of-00002.msgpack",
270
+ "decoder/block/4/layer/2/DenseReluDense/wi_1/kernel": "flax_model-00001-of-00002.msgpack",
271
+ "decoder/block/4/layer/2/DenseReluDense/wo/kernel": "flax_model-00001-of-00002.msgpack",
272
+ "decoder/block/4/layer/2/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
273
+ "decoder/block/5/layer/0/SelfAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
274
+ "decoder/block/5/layer/0/SelfAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
275
+ "decoder/block/5/layer/0/SelfAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
276
+ "decoder/block/5/layer/0/SelfAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
277
+ "decoder/block/5/layer/0/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
278
+ "decoder/block/5/layer/1/EncDecAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
279
+ "decoder/block/5/layer/1/EncDecAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
280
+ "decoder/block/5/layer/1/EncDecAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
281
+ "decoder/block/5/layer/1/EncDecAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
282
+ "decoder/block/5/layer/1/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
283
+ "decoder/block/5/layer/2/DenseReluDense/wi_0/kernel": "flax_model-00001-of-00002.msgpack",
284
+ "decoder/block/5/layer/2/DenseReluDense/wi_1/kernel": "flax_model-00001-of-00002.msgpack",
285
+ "decoder/block/5/layer/2/DenseReluDense/wo/kernel": "flax_model-00001-of-00002.msgpack",
286
+ "decoder/block/5/layer/2/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
287
+ "decoder/block/6/layer/0/SelfAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
288
+ "decoder/block/6/layer/0/SelfAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
289
+ "decoder/block/6/layer/0/SelfAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
290
+ "decoder/block/6/layer/0/SelfAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
291
+ "decoder/block/6/layer/0/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
292
+ "decoder/block/6/layer/1/EncDecAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
293
+ "decoder/block/6/layer/1/EncDecAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
294
+ "decoder/block/6/layer/1/EncDecAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
295
+ "decoder/block/6/layer/1/EncDecAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
296
+ "decoder/block/6/layer/1/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
297
+ "decoder/block/6/layer/2/DenseReluDense/wi_0/kernel": "flax_model-00001-of-00002.msgpack",
298
+ "decoder/block/6/layer/2/DenseReluDense/wi_1/kernel": "flax_model-00001-of-00002.msgpack",
299
+ "decoder/block/6/layer/2/DenseReluDense/wo/kernel": "flax_model-00001-of-00002.msgpack",
300
+ "decoder/block/6/layer/2/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
301
+ "decoder/block/7/layer/0/SelfAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
302
+ "decoder/block/7/layer/0/SelfAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
303
+ "decoder/block/7/layer/0/SelfAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
304
+ "decoder/block/7/layer/0/SelfAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
305
+ "decoder/block/7/layer/0/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
306
+ "decoder/block/7/layer/1/EncDecAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
307
+ "decoder/block/7/layer/1/EncDecAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
308
+ "decoder/block/7/layer/1/EncDecAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
309
+ "decoder/block/7/layer/1/EncDecAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
310
+ "decoder/block/7/layer/1/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
311
+ "decoder/block/7/layer/2/DenseReluDense/wi_0/kernel": "flax_model-00001-of-00002.msgpack",
312
+ "decoder/block/7/layer/2/DenseReluDense/wi_1/kernel": "flax_model-00001-of-00002.msgpack",
313
+ "decoder/block/7/layer/2/DenseReluDense/wo/kernel": "flax_model-00001-of-00002.msgpack",
314
+ "decoder/block/7/layer/2/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
315
+ "decoder/block/8/layer/0/SelfAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
316
+ "decoder/block/8/layer/0/SelfAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
317
+ "decoder/block/8/layer/0/SelfAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
318
+ "decoder/block/8/layer/0/SelfAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
319
+ "decoder/block/8/layer/0/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
320
+ "decoder/block/8/layer/1/EncDecAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
321
+ "decoder/block/8/layer/1/EncDecAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
322
+ "decoder/block/8/layer/1/EncDecAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
323
+ "decoder/block/8/layer/1/EncDecAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
324
+ "decoder/block/8/layer/1/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
325
+ "decoder/block/8/layer/2/DenseReluDense/wi_0/kernel": "flax_model-00001-of-00002.msgpack",
326
+ "decoder/block/8/layer/2/DenseReluDense/wi_1/kernel": "flax_model-00001-of-00002.msgpack",
327
+ "decoder/block/8/layer/2/DenseReluDense/wo/kernel": "flax_model-00001-of-00002.msgpack",
328
+ "decoder/block/8/layer/2/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
329
+ "decoder/block/9/layer/0/SelfAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
330
+ "decoder/block/9/layer/0/SelfAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
331
+ "decoder/block/9/layer/0/SelfAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
332
+ "decoder/block/9/layer/0/SelfAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
333
+ "decoder/block/9/layer/0/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
334
+ "decoder/block/9/layer/1/EncDecAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
335
+ "decoder/block/9/layer/1/EncDecAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
336
+ "decoder/block/9/layer/1/EncDecAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
337
+ "decoder/block/9/layer/1/EncDecAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
338
+ "decoder/block/9/layer/1/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
339
+ "decoder/block/9/layer/2/DenseReluDense/wi_0/kernel": "flax_model-00001-of-00002.msgpack",
340
+ "decoder/block/9/layer/2/DenseReluDense/wi_1/kernel": "flax_model-00001-of-00002.msgpack",
341
+ "decoder/block/9/layer/2/DenseReluDense/wo/kernel": "flax_model-00001-of-00002.msgpack",
342
+ "decoder/block/9/layer/2/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
343
+ "decoder/final_layer_norm/weight": "flax_model-00001-of-00002.msgpack",
344
+ "encoder/block/0/layer/0/SelfAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
345
+ "encoder/block/0/layer/0/SelfAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
346
+ "encoder/block/0/layer/0/SelfAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
347
+ "encoder/block/0/layer/0/SelfAttention/relative_attention_bias/embedding": "flax_model-00001-of-00002.msgpack",
348
+ "encoder/block/0/layer/0/SelfAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
349
+ "encoder/block/0/layer/0/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
350
+ "encoder/block/0/layer/1/DenseReluDense/wi_0/kernel": "flax_model-00001-of-00002.msgpack",
351
+ "encoder/block/0/layer/1/DenseReluDense/wi_1/kernel": "flax_model-00001-of-00002.msgpack",
352
+ "encoder/block/0/layer/1/DenseReluDense/wo/kernel": "flax_model-00001-of-00002.msgpack",
353
+ "encoder/block/0/layer/1/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
354
+ "encoder/block/1/layer/0/SelfAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
355
+ "encoder/block/1/layer/0/SelfAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
356
+ "encoder/block/1/layer/0/SelfAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
357
+ "encoder/block/1/layer/0/SelfAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
358
+ "encoder/block/1/layer/0/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
359
+ "encoder/block/1/layer/1/DenseReluDense/wi_0/kernel": "flax_model-00001-of-00002.msgpack",
360
+ "encoder/block/1/layer/1/DenseReluDense/wi_1/kernel": "flax_model-00001-of-00002.msgpack",
361
+ "encoder/block/1/layer/1/DenseReluDense/wo/kernel": "flax_model-00001-of-00002.msgpack",
362
+ "encoder/block/1/layer/1/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
363
+ "encoder/block/10/layer/0/SelfAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
364
+ "encoder/block/10/layer/0/SelfAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
365
+ "encoder/block/10/layer/0/SelfAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
366
+ "encoder/block/10/layer/0/SelfAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
367
+ "encoder/block/10/layer/0/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
368
+ "encoder/block/10/layer/1/DenseReluDense/wi_0/kernel": "flax_model-00001-of-00002.msgpack",
369
+ "encoder/block/10/layer/1/DenseReluDense/wi_1/kernel": "flax_model-00001-of-00002.msgpack",
370
+ "encoder/block/10/layer/1/DenseReluDense/wo/kernel": "flax_model-00001-of-00002.msgpack",
371
+ "encoder/block/10/layer/1/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
372
+ "encoder/block/11/layer/0/SelfAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
373
+ "encoder/block/11/layer/0/SelfAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
374
+ "encoder/block/11/layer/0/SelfAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
375
+ "encoder/block/11/layer/0/SelfAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
376
+ "encoder/block/11/layer/0/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
377
+ "encoder/block/11/layer/1/DenseReluDense/wi_0/kernel": "flax_model-00001-of-00002.msgpack",
378
+ "encoder/block/11/layer/1/DenseReluDense/wi_1/kernel": "flax_model-00001-of-00002.msgpack",
379
+ "encoder/block/11/layer/1/DenseReluDense/wo/kernel": "flax_model-00001-of-00002.msgpack",
380
+ "encoder/block/11/layer/1/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
381
+ "encoder/block/12/layer/0/SelfAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
382
+ "encoder/block/12/layer/0/SelfAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
383
+ "encoder/block/12/layer/0/SelfAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
384
+ "encoder/block/12/layer/0/SelfAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
385
+ "encoder/block/12/layer/0/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
386
+ "encoder/block/12/layer/1/DenseReluDense/wi_0/kernel": "flax_model-00001-of-00002.msgpack",
387
+ "encoder/block/12/layer/1/DenseReluDense/wi_1/kernel": "flax_model-00001-of-00002.msgpack",
388
+ "encoder/block/12/layer/1/DenseReluDense/wo/kernel": "flax_model-00001-of-00002.msgpack",
389
+ "encoder/block/12/layer/1/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
390
+ "encoder/block/13/layer/0/SelfAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
391
+ "encoder/block/13/layer/0/SelfAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
392
+ "encoder/block/13/layer/0/SelfAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
393
+ "encoder/block/13/layer/0/SelfAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
394
+ "encoder/block/13/layer/0/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
395
+ "encoder/block/13/layer/1/DenseReluDense/wi_0/kernel": "flax_model-00001-of-00002.msgpack",
396
+ "encoder/block/13/layer/1/DenseReluDense/wi_1/kernel": "flax_model-00001-of-00002.msgpack",
397
+ "encoder/block/13/layer/1/DenseReluDense/wo/kernel": "flax_model-00001-of-00002.msgpack",
398
+ "encoder/block/13/layer/1/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
399
+ "encoder/block/14/layer/0/SelfAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
400
+ "encoder/block/14/layer/0/SelfAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
401
+ "encoder/block/14/layer/0/SelfAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
402
+ "encoder/block/14/layer/0/SelfAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
403
+ "encoder/block/14/layer/0/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
404
+ "encoder/block/14/layer/1/DenseReluDense/wi_0/kernel": "flax_model-00001-of-00002.msgpack",
405
+ "encoder/block/14/layer/1/DenseReluDense/wi_1/kernel": "flax_model-00001-of-00002.msgpack",
406
+ "encoder/block/14/layer/1/DenseReluDense/wo/kernel": "flax_model-00001-of-00002.msgpack",
407
+ "encoder/block/14/layer/1/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
408
+ "encoder/block/15/layer/0/SelfAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
409
+ "encoder/block/15/layer/0/SelfAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
410
+ "encoder/block/15/layer/0/SelfAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
411
+ "encoder/block/15/layer/0/SelfAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
412
+ "encoder/block/15/layer/0/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
413
+ "encoder/block/15/layer/1/DenseReluDense/wi_0/kernel": "flax_model-00001-of-00002.msgpack",
414
+ "encoder/block/15/layer/1/DenseReluDense/wi_1/kernel": "flax_model-00001-of-00002.msgpack",
415
+ "encoder/block/15/layer/1/DenseReluDense/wo/kernel": "flax_model-00001-of-00002.msgpack",
416
+ "encoder/block/15/layer/1/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
417
+ "encoder/block/16/layer/0/SelfAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
418
+ "encoder/block/16/layer/0/SelfAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
419
+ "encoder/block/16/layer/0/SelfAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
420
+ "encoder/block/16/layer/0/SelfAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
421
+ "encoder/block/16/layer/0/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
422
+ "encoder/block/16/layer/1/DenseReluDense/wi_0/kernel": "flax_model-00001-of-00002.msgpack",
423
+ "encoder/block/16/layer/1/DenseReluDense/wi_1/kernel": "flax_model-00001-of-00002.msgpack",
424
+ "encoder/block/16/layer/1/DenseReluDense/wo/kernel": "flax_model-00001-of-00002.msgpack",
425
+ "encoder/block/16/layer/1/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
426
+ "encoder/block/17/layer/0/SelfAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
427
+ "encoder/block/17/layer/0/SelfAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
428
+ "encoder/block/17/layer/0/SelfAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
429
+ "encoder/block/17/layer/0/SelfAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
430
+ "encoder/block/17/layer/0/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
431
+ "encoder/block/17/layer/1/DenseReluDense/wi_0/kernel": "flax_model-00001-of-00002.msgpack",
432
+ "encoder/block/17/layer/1/DenseReluDense/wi_1/kernel": "flax_model-00001-of-00002.msgpack",
433
+ "encoder/block/17/layer/1/DenseReluDense/wo/kernel": "flax_model-00001-of-00002.msgpack",
434
+ "encoder/block/17/layer/1/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
435
+ "encoder/block/18/layer/0/SelfAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
436
+ "encoder/block/18/layer/0/SelfAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
437
+ "encoder/block/18/layer/0/SelfAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
438
+ "encoder/block/18/layer/0/SelfAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
439
+ "encoder/block/18/layer/0/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
440
+ "encoder/block/18/layer/1/DenseReluDense/wi_0/kernel": "flax_model-00001-of-00002.msgpack",
441
+ "encoder/block/18/layer/1/DenseReluDense/wi_1/kernel": "flax_model-00001-of-00002.msgpack",
442
+ "encoder/block/18/layer/1/DenseReluDense/wo/kernel": "flax_model-00001-of-00002.msgpack",
443
+ "encoder/block/18/layer/1/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
444
+ "encoder/block/19/layer/0/SelfAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
445
+ "encoder/block/19/layer/0/SelfAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
446
+ "encoder/block/19/layer/0/SelfAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
447
+ "encoder/block/19/layer/0/SelfAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
448
+ "encoder/block/19/layer/0/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
449
+ "encoder/block/19/layer/1/DenseReluDense/wi_0/kernel": "flax_model-00001-of-00002.msgpack",
450
+ "encoder/block/19/layer/1/DenseReluDense/wi_1/kernel": "flax_model-00001-of-00002.msgpack",
451
+ "encoder/block/19/layer/1/DenseReluDense/wo/kernel": "flax_model-00001-of-00002.msgpack",
452
+ "encoder/block/19/layer/1/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
453
+ "encoder/block/2/layer/0/SelfAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
454
+ "encoder/block/2/layer/0/SelfAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
455
+ "encoder/block/2/layer/0/SelfAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
456
+ "encoder/block/2/layer/0/SelfAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
457
+ "encoder/block/2/layer/0/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
458
+ "encoder/block/2/layer/1/DenseReluDense/wi_0/kernel": "flax_model-00001-of-00002.msgpack",
459
+ "encoder/block/2/layer/1/DenseReluDense/wi_1/kernel": "flax_model-00001-of-00002.msgpack",
460
+ "encoder/block/2/layer/1/DenseReluDense/wo/kernel": "flax_model-00001-of-00002.msgpack",
461
+ "encoder/block/2/layer/1/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
462
+ "encoder/block/20/layer/0/SelfAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
463
+ "encoder/block/20/layer/0/SelfAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
464
+ "encoder/block/20/layer/0/SelfAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
465
+ "encoder/block/20/layer/0/SelfAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
466
+ "encoder/block/20/layer/0/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
467
+ "encoder/block/20/layer/1/DenseReluDense/wi_0/kernel": "flax_model-00001-of-00002.msgpack",
468
+ "encoder/block/20/layer/1/DenseReluDense/wi_1/kernel": "flax_model-00001-of-00002.msgpack",
469
+ "encoder/block/20/layer/1/DenseReluDense/wo/kernel": "flax_model-00001-of-00002.msgpack",
470
+ "encoder/block/20/layer/1/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
471
+ "encoder/block/21/layer/0/SelfAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
472
+ "encoder/block/21/layer/0/SelfAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
473
+ "encoder/block/21/layer/0/SelfAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
474
+ "encoder/block/21/layer/0/SelfAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
475
+ "encoder/block/21/layer/0/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
476
+ "encoder/block/21/layer/1/DenseReluDense/wi_0/kernel": "flax_model-00001-of-00002.msgpack",
477
+ "encoder/block/21/layer/1/DenseReluDense/wi_1/kernel": "flax_model-00001-of-00002.msgpack",
478
+ "encoder/block/21/layer/1/DenseReluDense/wo/kernel": "flax_model-00001-of-00002.msgpack",
479
+ "encoder/block/21/layer/1/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
480
+ "encoder/block/22/layer/0/SelfAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
481
+ "encoder/block/22/layer/0/SelfAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
482
+ "encoder/block/22/layer/0/SelfAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
483
+ "encoder/block/22/layer/0/SelfAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
484
+ "encoder/block/22/layer/0/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
485
+ "encoder/block/22/layer/1/DenseReluDense/wi_0/kernel": "flax_model-00001-of-00002.msgpack",
486
+ "encoder/block/22/layer/1/DenseReluDense/wi_1/kernel": "flax_model-00001-of-00002.msgpack",
487
+ "encoder/block/22/layer/1/DenseReluDense/wo/kernel": "flax_model-00001-of-00002.msgpack",
488
+ "encoder/block/22/layer/1/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
489
+ "encoder/block/23/layer/0/SelfAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
490
+ "encoder/block/23/layer/0/SelfAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
491
+ "encoder/block/23/layer/0/SelfAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
492
+ "encoder/block/23/layer/0/SelfAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
493
+ "encoder/block/23/layer/0/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
494
+ "encoder/block/23/layer/1/DenseReluDense/wi_0/kernel": "flax_model-00001-of-00002.msgpack",
495
+ "encoder/block/23/layer/1/DenseReluDense/wi_1/kernel": "flax_model-00001-of-00002.msgpack",
496
+ "encoder/block/23/layer/1/DenseReluDense/wo/kernel": "flax_model-00001-of-00002.msgpack",
497
+ "encoder/block/23/layer/1/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
498
+ "encoder/block/3/layer/0/SelfAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
499
+ "encoder/block/3/layer/0/SelfAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
500
+ "encoder/block/3/layer/0/SelfAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
501
+ "encoder/block/3/layer/0/SelfAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
502
+ "encoder/block/3/layer/0/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
503
+ "encoder/block/3/layer/1/DenseReluDense/wi_0/kernel": "flax_model-00001-of-00002.msgpack",
504
+ "encoder/block/3/layer/1/DenseReluDense/wi_1/kernel": "flax_model-00001-of-00002.msgpack",
505
+ "encoder/block/3/layer/1/DenseReluDense/wo/kernel": "flax_model-00001-of-00002.msgpack",
506
+ "encoder/block/3/layer/1/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
507
+ "encoder/block/4/layer/0/SelfAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
508
+ "encoder/block/4/layer/0/SelfAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
509
+ "encoder/block/4/layer/0/SelfAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
510
+ "encoder/block/4/layer/0/SelfAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
511
+ "encoder/block/4/layer/0/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
512
+ "encoder/block/4/layer/1/DenseReluDense/wi_0/kernel": "flax_model-00001-of-00002.msgpack",
513
+ "encoder/block/4/layer/1/DenseReluDense/wi_1/kernel": "flax_model-00001-of-00002.msgpack",
514
+ "encoder/block/4/layer/1/DenseReluDense/wo/kernel": "flax_model-00001-of-00002.msgpack",
515
+ "encoder/block/4/layer/1/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
516
+ "encoder/block/5/layer/0/SelfAttention/k/kernel": "flax_model-00001-of-00002.msgpack",
517
+ "encoder/block/5/layer/0/SelfAttention/o/kernel": "flax_model-00001-of-00002.msgpack",
518
+ "encoder/block/5/layer/0/SelfAttention/q/kernel": "flax_model-00001-of-00002.msgpack",
519
+ "encoder/block/5/layer/0/SelfAttention/v/kernel": "flax_model-00001-of-00002.msgpack",
520
+ "encoder/block/5/layer/0/layer_norm/weight": "flax_model-00001-of-00002.msgpack",
521
+ "encoder/block/5/layer/1/DenseReluDense/wi_0/kernel": "flax_model-00002-of-00002.msgpack",
522
+ "encoder/block/5/layer/1/DenseReluDense/wi_1/kernel": "flax_model-00002-of-00002.msgpack",
523
+ "encoder/block/5/layer/1/DenseReluDense/wo/kernel": "flax_model-00002-of-00002.msgpack",
524
+ "encoder/block/5/layer/1/layer_norm/weight": "flax_model-00002-of-00002.msgpack",
525
+ "encoder/block/6/layer/0/SelfAttention/k/kernel": "flax_model-00002-of-00002.msgpack",
526
+ "encoder/block/6/layer/0/SelfAttention/o/kernel": "flax_model-00002-of-00002.msgpack",
527
+ "encoder/block/6/layer/0/SelfAttention/q/kernel": "flax_model-00002-of-00002.msgpack",
528
+ "encoder/block/6/layer/0/SelfAttention/v/kernel": "flax_model-00002-of-00002.msgpack",
529
+ "encoder/block/6/layer/0/layer_norm/weight": "flax_model-00002-of-00002.msgpack",
530
+ "encoder/block/6/layer/1/DenseReluDense/wi_0/kernel": "flax_model-00002-of-00002.msgpack",
531
+ "encoder/block/6/layer/1/DenseReluDense/wi_1/kernel": "flax_model-00002-of-00002.msgpack",
532
+ "encoder/block/6/layer/1/DenseReluDense/wo/kernel": "flax_model-00002-of-00002.msgpack",
533
+ "encoder/block/6/layer/1/layer_norm/weight": "flax_model-00002-of-00002.msgpack",
534
+ "encoder/block/7/layer/0/SelfAttention/k/kernel": "flax_model-00002-of-00002.msgpack",
535
+ "encoder/block/7/layer/0/SelfAttention/o/kernel": "flax_model-00002-of-00002.msgpack",
536
+ "encoder/block/7/layer/0/SelfAttention/q/kernel": "flax_model-00002-of-00002.msgpack",
537
+ "encoder/block/7/layer/0/SelfAttention/v/kernel": "flax_model-00002-of-00002.msgpack",
538
+ "encoder/block/7/layer/0/layer_norm/weight": "flax_model-00002-of-00002.msgpack",
539
+ "encoder/block/7/layer/1/DenseReluDense/wi_0/kernel": "flax_model-00002-of-00002.msgpack",
540
+ "encoder/block/7/layer/1/DenseReluDense/wi_1/kernel": "flax_model-00002-of-00002.msgpack",
541
+ "encoder/block/7/layer/1/DenseReluDense/wo/kernel": "flax_model-00002-of-00002.msgpack",
542
+ "encoder/block/7/layer/1/layer_norm/weight": "flax_model-00002-of-00002.msgpack",
543
+ "encoder/block/8/layer/0/SelfAttention/k/kernel": "flax_model-00002-of-00002.msgpack",
544
+ "encoder/block/8/layer/0/SelfAttention/o/kernel": "flax_model-00002-of-00002.msgpack",
545
+ "encoder/block/8/layer/0/SelfAttention/q/kernel": "flax_model-00002-of-00002.msgpack",
546
+ "encoder/block/8/layer/0/SelfAttention/v/kernel": "flax_model-00002-of-00002.msgpack",
547
+ "encoder/block/8/layer/0/layer_norm/weight": "flax_model-00002-of-00002.msgpack",
548
+ "encoder/block/8/layer/1/DenseReluDense/wi_0/kernel": "flax_model-00002-of-00002.msgpack",
549
+ "encoder/block/8/layer/1/DenseReluDense/wi_1/kernel": "flax_model-00002-of-00002.msgpack",
550
+ "encoder/block/8/layer/1/DenseReluDense/wo/kernel": "flax_model-00002-of-00002.msgpack",
551
+ "encoder/block/8/layer/1/layer_norm/weight": "flax_model-00002-of-00002.msgpack",
552
+ "encoder/block/9/layer/0/SelfAttention/k/kernel": "flax_model-00002-of-00002.msgpack",
553
+ "encoder/block/9/layer/0/SelfAttention/o/kernel": "flax_model-00002-of-00002.msgpack",
554
+ "encoder/block/9/layer/0/SelfAttention/q/kernel": "flax_model-00002-of-00002.msgpack",
555
+ "encoder/block/9/layer/0/SelfAttention/v/kernel": "flax_model-00002-of-00002.msgpack",
556
+ "encoder/block/9/layer/0/layer_norm/weight": "flax_model-00002-of-00002.msgpack",
557
+ "encoder/block/9/layer/1/DenseReluDense/wi_0/kernel": "flax_model-00002-of-00002.msgpack",
558
+ "encoder/block/9/layer/1/DenseReluDense/wi_1/kernel": "flax_model-00002-of-00002.msgpack",
559
+ "encoder/block/9/layer/1/DenseReluDense/wo/kernel": "flax_model-00002-of-00002.msgpack",
560
+ "encoder/block/9/layer/1/layer_norm/weight": "flax_model-00002-of-00002.msgpack",
561
+ "encoder/final_layer_norm/weight": "flax_model-00002-of-00002.msgpack",
562
+ "lm_head/kernel": "flax_model-00002-of-00002.msgpack",
563
+ "shared/embedding": "flax_model-00002-of-00002.msgpack"
564
+ }
565
+ }