patrickvonplaten ArthurZ HF staff commited on
Commit
99cdfb7
1 Parent(s): a7fad6c

Add Flax weights (#5)

Browse files

- Add Flax weights (94f94ea6511f1401e61827769b5b9938a65b741f)


Co-authored-by: Arthur Zucker <ArthurZ@users.noreply.huggingface.co>

flax_model-00001-of-00007.msgpack ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c122ca9bf957999e4d70250c0664eb65080963dad0187d0fc70ae185f2a52cd9
3
+ size 9794416724
flax_model-00002-of-00007.msgpack ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f717b938c53a247530241a1c91e74ba8d8f1a3cd343bd3d4d2cfa1a9349ee044
3
+ size 9866498425
flax_model-00003-of-00007.msgpack ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2a4eaeff66638d9436e7ae330e6ecfad558293cc870ff3e3a8601a9776804619
3
+ size 9866498425
flax_model-00004-of-00007.msgpack ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5542c16c7bf87e6214d1b646091c1418cc6c2f46cf85bcae46b88754ea44c7ee
3
+ size 9866498425
flax_model-00005-of-00007.msgpack ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ab3c03aabbcdfb30af1e66aeb9257e30ef7430e9e8b709dda42eb291c274dec7
3
+ size 9866498425
flax_model-00006-of-00007.msgpack ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a69fd613a08e8f4de4deb999072153e46fb19a091b938261c5d649c2408d401a
3
+ size 9866498421
flax_model-00007-of-00007.msgpack ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b7b2fc207f423e70e10c210d558a6d2db904430561dd7459e80caa0db2ea6bac
3
+ size 822198762
flax_model.msgpack.index.json ADDED
@@ -0,0 +1,779 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 59949080576
4
+ },
5
+ "weight_map": {
6
+ "model/decoder/embed_positions/embedding": "flax_model-00001-of-00007.msgpack",
7
+ "model/decoder/embed_tokens/embedding": "flax_model-00001-of-00007.msgpack",
8
+ "model/decoder/final_layer_norm/bias": "flax_model-00001-of-00007.msgpack",
9
+ "model/decoder/final_layer_norm/scale": "flax_model-00001-of-00007.msgpack",
10
+ "model/decoder/layers/0/fc1/bias": "flax_model-00001-of-00007.msgpack",
11
+ "model/decoder/layers/0/fc1/kernel": "flax_model-00001-of-00007.msgpack",
12
+ "model/decoder/layers/0/fc2/bias": "flax_model-00001-of-00007.msgpack",
13
+ "model/decoder/layers/0/fc2/kernel": "flax_model-00001-of-00007.msgpack",
14
+ "model/decoder/layers/0/final_layer_norm/bias": "flax_model-00001-of-00007.msgpack",
15
+ "model/decoder/layers/0/final_layer_norm/scale": "flax_model-00001-of-00007.msgpack",
16
+ "model/decoder/layers/0/self_attn/k_proj/bias": "flax_model-00001-of-00007.msgpack",
17
+ "model/decoder/layers/0/self_attn/k_proj/kernel": "flax_model-00001-of-00007.msgpack",
18
+ "model/decoder/layers/0/self_attn/out_proj/bias": "flax_model-00001-of-00007.msgpack",
19
+ "model/decoder/layers/0/self_attn/out_proj/kernel": "flax_model-00001-of-00007.msgpack",
20
+ "model/decoder/layers/0/self_attn/q_proj/bias": "flax_model-00001-of-00007.msgpack",
21
+ "model/decoder/layers/0/self_attn/q_proj/kernel": "flax_model-00001-of-00007.msgpack",
22
+ "model/decoder/layers/0/self_attn/v_proj/bias": "flax_model-00001-of-00007.msgpack",
23
+ "model/decoder/layers/0/self_attn/v_proj/kernel": "flax_model-00001-of-00007.msgpack",
24
+ "model/decoder/layers/0/self_attn_layer_norm/bias": "flax_model-00001-of-00007.msgpack",
25
+ "model/decoder/layers/0/self_attn_layer_norm/scale": "flax_model-00001-of-00007.msgpack",
26
+ "model/decoder/layers/1/fc1/bias": "flax_model-00001-of-00007.msgpack",
27
+ "model/decoder/layers/1/fc1/kernel": "flax_model-00001-of-00007.msgpack",
28
+ "model/decoder/layers/1/fc2/bias": "flax_model-00001-of-00007.msgpack",
29
+ "model/decoder/layers/1/fc2/kernel": "flax_model-00001-of-00007.msgpack",
30
+ "model/decoder/layers/1/final_layer_norm/bias": "flax_model-00001-of-00007.msgpack",
31
+ "model/decoder/layers/1/final_layer_norm/scale": "flax_model-00001-of-00007.msgpack",
32
+ "model/decoder/layers/1/self_attn/k_proj/bias": "flax_model-00001-of-00007.msgpack",
33
+ "model/decoder/layers/1/self_attn/k_proj/kernel": "flax_model-00001-of-00007.msgpack",
34
+ "model/decoder/layers/1/self_attn/out_proj/bias": "flax_model-00001-of-00007.msgpack",
35
+ "model/decoder/layers/1/self_attn/out_proj/kernel": "flax_model-00001-of-00007.msgpack",
36
+ "model/decoder/layers/1/self_attn/q_proj/bias": "flax_model-00001-of-00007.msgpack",
37
+ "model/decoder/layers/1/self_attn/q_proj/kernel": "flax_model-00001-of-00007.msgpack",
38
+ "model/decoder/layers/1/self_attn/v_proj/bias": "flax_model-00001-of-00007.msgpack",
39
+ "model/decoder/layers/1/self_attn/v_proj/kernel": "flax_model-00001-of-00007.msgpack",
40
+ "model/decoder/layers/1/self_attn_layer_norm/bias": "flax_model-00001-of-00007.msgpack",
41
+ "model/decoder/layers/1/self_attn_layer_norm/scale": "flax_model-00001-of-00007.msgpack",
42
+ "model/decoder/layers/10/fc1/bias": "flax_model-00001-of-00007.msgpack",
43
+ "model/decoder/layers/10/fc1/kernel": "flax_model-00001-of-00007.msgpack",
44
+ "model/decoder/layers/10/fc2/bias": "flax_model-00001-of-00007.msgpack",
45
+ "model/decoder/layers/10/fc2/kernel": "flax_model-00001-of-00007.msgpack",
46
+ "model/decoder/layers/10/final_layer_norm/bias": "flax_model-00001-of-00007.msgpack",
47
+ "model/decoder/layers/10/final_layer_norm/scale": "flax_model-00001-of-00007.msgpack",
48
+ "model/decoder/layers/10/self_attn/k_proj/bias": "flax_model-00001-of-00007.msgpack",
49
+ "model/decoder/layers/10/self_attn/k_proj/kernel": "flax_model-00001-of-00007.msgpack",
50
+ "model/decoder/layers/10/self_attn/out_proj/bias": "flax_model-00001-of-00007.msgpack",
51
+ "model/decoder/layers/10/self_attn/out_proj/kernel": "flax_model-00001-of-00007.msgpack",
52
+ "model/decoder/layers/10/self_attn/q_proj/bias": "flax_model-00001-of-00007.msgpack",
53
+ "model/decoder/layers/10/self_attn/q_proj/kernel": "flax_model-00001-of-00007.msgpack",
54
+ "model/decoder/layers/10/self_attn/v_proj/bias": "flax_model-00001-of-00007.msgpack",
55
+ "model/decoder/layers/10/self_attn/v_proj/kernel": "flax_model-00001-of-00007.msgpack",
56
+ "model/decoder/layers/10/self_attn_layer_norm/bias": "flax_model-00001-of-00007.msgpack",
57
+ "model/decoder/layers/10/self_attn_layer_norm/scale": "flax_model-00001-of-00007.msgpack",
58
+ "model/decoder/layers/11/fc1/bias": "flax_model-00001-of-00007.msgpack",
59
+ "model/decoder/layers/11/fc1/kernel": "flax_model-00001-of-00007.msgpack",
60
+ "model/decoder/layers/11/fc2/bias": "flax_model-00001-of-00007.msgpack",
61
+ "model/decoder/layers/11/fc2/kernel": "flax_model-00001-of-00007.msgpack",
62
+ "model/decoder/layers/11/final_layer_norm/bias": "flax_model-00001-of-00007.msgpack",
63
+ "model/decoder/layers/11/final_layer_norm/scale": "flax_model-00001-of-00007.msgpack",
64
+ "model/decoder/layers/11/self_attn/k_proj/bias": "flax_model-00001-of-00007.msgpack",
65
+ "model/decoder/layers/11/self_attn/k_proj/kernel": "flax_model-00001-of-00007.msgpack",
66
+ "model/decoder/layers/11/self_attn/out_proj/bias": "flax_model-00001-of-00007.msgpack",
67
+ "model/decoder/layers/11/self_attn/out_proj/kernel": "flax_model-00001-of-00007.msgpack",
68
+ "model/decoder/layers/11/self_attn/q_proj/bias": "flax_model-00001-of-00007.msgpack",
69
+ "model/decoder/layers/11/self_attn/q_proj/kernel": "flax_model-00001-of-00007.msgpack",
70
+ "model/decoder/layers/11/self_attn/v_proj/bias": "flax_model-00001-of-00007.msgpack",
71
+ "model/decoder/layers/11/self_attn/v_proj/kernel": "flax_model-00001-of-00007.msgpack",
72
+ "model/decoder/layers/11/self_attn_layer_norm/bias": "flax_model-00001-of-00007.msgpack",
73
+ "model/decoder/layers/11/self_attn_layer_norm/scale": "flax_model-00001-of-00007.msgpack",
74
+ "model/decoder/layers/12/fc1/bias": "flax_model-00001-of-00007.msgpack",
75
+ "model/decoder/layers/12/fc1/kernel": "flax_model-00001-of-00007.msgpack",
76
+ "model/decoder/layers/12/fc2/bias": "flax_model-00001-of-00007.msgpack",
77
+ "model/decoder/layers/12/fc2/kernel": "flax_model-00001-of-00007.msgpack",
78
+ "model/decoder/layers/12/final_layer_norm/bias": "flax_model-00001-of-00007.msgpack",
79
+ "model/decoder/layers/12/final_layer_norm/scale": "flax_model-00001-of-00007.msgpack",
80
+ "model/decoder/layers/12/self_attn/k_proj/bias": "flax_model-00001-of-00007.msgpack",
81
+ "model/decoder/layers/12/self_attn/k_proj/kernel": "flax_model-00001-of-00007.msgpack",
82
+ "model/decoder/layers/12/self_attn/out_proj/bias": "flax_model-00001-of-00007.msgpack",
83
+ "model/decoder/layers/12/self_attn/out_proj/kernel": "flax_model-00001-of-00007.msgpack",
84
+ "model/decoder/layers/12/self_attn/q_proj/bias": "flax_model-00001-of-00007.msgpack",
85
+ "model/decoder/layers/12/self_attn/q_proj/kernel": "flax_model-00001-of-00007.msgpack",
86
+ "model/decoder/layers/12/self_attn/v_proj/bias": "flax_model-00001-of-00007.msgpack",
87
+ "model/decoder/layers/12/self_attn/v_proj/kernel": "flax_model-00001-of-00007.msgpack",
88
+ "model/decoder/layers/12/self_attn_layer_norm/bias": "flax_model-00001-of-00007.msgpack",
89
+ "model/decoder/layers/12/self_attn_layer_norm/scale": "flax_model-00001-of-00007.msgpack",
90
+ "model/decoder/layers/13/fc1/bias": "flax_model-00001-of-00007.msgpack",
91
+ "model/decoder/layers/13/fc1/kernel": "flax_model-00001-of-00007.msgpack",
92
+ "model/decoder/layers/13/fc2/bias": "flax_model-00001-of-00007.msgpack",
93
+ "model/decoder/layers/13/fc2/kernel": "flax_model-00001-of-00007.msgpack",
94
+ "model/decoder/layers/13/final_layer_norm/bias": "flax_model-00001-of-00007.msgpack",
95
+ "model/decoder/layers/13/final_layer_norm/scale": "flax_model-00001-of-00007.msgpack",
96
+ "model/decoder/layers/13/self_attn/k_proj/bias": "flax_model-00001-of-00007.msgpack",
97
+ "model/decoder/layers/13/self_attn/k_proj/kernel": "flax_model-00001-of-00007.msgpack",
98
+ "model/decoder/layers/13/self_attn/out_proj/bias": "flax_model-00001-of-00007.msgpack",
99
+ "model/decoder/layers/13/self_attn/out_proj/kernel": "flax_model-00001-of-00007.msgpack",
100
+ "model/decoder/layers/13/self_attn/q_proj/bias": "flax_model-00001-of-00007.msgpack",
101
+ "model/decoder/layers/13/self_attn/q_proj/kernel": "flax_model-00001-of-00007.msgpack",
102
+ "model/decoder/layers/13/self_attn/v_proj/bias": "flax_model-00001-of-00007.msgpack",
103
+ "model/decoder/layers/13/self_attn/v_proj/kernel": "flax_model-00001-of-00007.msgpack",
104
+ "model/decoder/layers/13/self_attn_layer_norm/bias": "flax_model-00001-of-00007.msgpack",
105
+ "model/decoder/layers/13/self_attn_layer_norm/scale": "flax_model-00001-of-00007.msgpack",
106
+ "model/decoder/layers/14/fc1/bias": "flax_model-00001-of-00007.msgpack",
107
+ "model/decoder/layers/14/fc1/kernel": "flax_model-00001-of-00007.msgpack",
108
+ "model/decoder/layers/14/fc2/bias": "flax_model-00001-of-00007.msgpack",
109
+ "model/decoder/layers/14/fc2/kernel": "flax_model-00001-of-00007.msgpack",
110
+ "model/decoder/layers/14/final_layer_norm/bias": "flax_model-00001-of-00007.msgpack",
111
+ "model/decoder/layers/14/final_layer_norm/scale": "flax_model-00001-of-00007.msgpack",
112
+ "model/decoder/layers/14/self_attn/k_proj/bias": "flax_model-00001-of-00007.msgpack",
113
+ "model/decoder/layers/14/self_attn/k_proj/kernel": "flax_model-00001-of-00007.msgpack",
114
+ "model/decoder/layers/14/self_attn/out_proj/bias": "flax_model-00001-of-00007.msgpack",
115
+ "model/decoder/layers/14/self_attn/out_proj/kernel": "flax_model-00001-of-00007.msgpack",
116
+ "model/decoder/layers/14/self_attn/q_proj/bias": "flax_model-00001-of-00007.msgpack",
117
+ "model/decoder/layers/14/self_attn/q_proj/kernel": "flax_model-00001-of-00007.msgpack",
118
+ "model/decoder/layers/14/self_attn/v_proj/bias": "flax_model-00001-of-00007.msgpack",
119
+ "model/decoder/layers/14/self_attn/v_proj/kernel": "flax_model-00001-of-00007.msgpack",
120
+ "model/decoder/layers/14/self_attn_layer_norm/bias": "flax_model-00001-of-00007.msgpack",
121
+ "model/decoder/layers/14/self_attn_layer_norm/scale": "flax_model-00001-of-00007.msgpack",
122
+ "model/decoder/layers/15/fc1/bias": "flax_model-00001-of-00007.msgpack",
123
+ "model/decoder/layers/15/fc1/kernel": "flax_model-00001-of-00007.msgpack",
124
+ "model/decoder/layers/15/fc2/bias": "flax_model-00001-of-00007.msgpack",
125
+ "model/decoder/layers/15/fc2/kernel": "flax_model-00002-of-00007.msgpack",
126
+ "model/decoder/layers/15/final_layer_norm/bias": "flax_model-00002-of-00007.msgpack",
127
+ "model/decoder/layers/15/final_layer_norm/scale": "flax_model-00002-of-00007.msgpack",
128
+ "model/decoder/layers/15/self_attn/k_proj/bias": "flax_model-00002-of-00007.msgpack",
129
+ "model/decoder/layers/15/self_attn/k_proj/kernel": "flax_model-00002-of-00007.msgpack",
130
+ "model/decoder/layers/15/self_attn/out_proj/bias": "flax_model-00002-of-00007.msgpack",
131
+ "model/decoder/layers/15/self_attn/out_proj/kernel": "flax_model-00002-of-00007.msgpack",
132
+ "model/decoder/layers/15/self_attn/q_proj/bias": "flax_model-00002-of-00007.msgpack",
133
+ "model/decoder/layers/15/self_attn/q_proj/kernel": "flax_model-00002-of-00007.msgpack",
134
+ "model/decoder/layers/15/self_attn/v_proj/bias": "flax_model-00002-of-00007.msgpack",
135
+ "model/decoder/layers/15/self_attn/v_proj/kernel": "flax_model-00002-of-00007.msgpack",
136
+ "model/decoder/layers/15/self_attn_layer_norm/bias": "flax_model-00002-of-00007.msgpack",
137
+ "model/decoder/layers/15/self_attn_layer_norm/scale": "flax_model-00002-of-00007.msgpack",
138
+ "model/decoder/layers/16/fc1/bias": "flax_model-00002-of-00007.msgpack",
139
+ "model/decoder/layers/16/fc1/kernel": "flax_model-00002-of-00007.msgpack",
140
+ "model/decoder/layers/16/fc2/bias": "flax_model-00002-of-00007.msgpack",
141
+ "model/decoder/layers/16/fc2/kernel": "flax_model-00002-of-00007.msgpack",
142
+ "model/decoder/layers/16/final_layer_norm/bias": "flax_model-00002-of-00007.msgpack",
143
+ "model/decoder/layers/16/final_layer_norm/scale": "flax_model-00002-of-00007.msgpack",
144
+ "model/decoder/layers/16/self_attn/k_proj/bias": "flax_model-00002-of-00007.msgpack",
145
+ "model/decoder/layers/16/self_attn/k_proj/kernel": "flax_model-00002-of-00007.msgpack",
146
+ "model/decoder/layers/16/self_attn/out_proj/bias": "flax_model-00002-of-00007.msgpack",
147
+ "model/decoder/layers/16/self_attn/out_proj/kernel": "flax_model-00002-of-00007.msgpack",
148
+ "model/decoder/layers/16/self_attn/q_proj/bias": "flax_model-00002-of-00007.msgpack",
149
+ "model/decoder/layers/16/self_attn/q_proj/kernel": "flax_model-00002-of-00007.msgpack",
150
+ "model/decoder/layers/16/self_attn/v_proj/bias": "flax_model-00002-of-00007.msgpack",
151
+ "model/decoder/layers/16/self_attn/v_proj/kernel": "flax_model-00002-of-00007.msgpack",
152
+ "model/decoder/layers/16/self_attn_layer_norm/bias": "flax_model-00002-of-00007.msgpack",
153
+ "model/decoder/layers/16/self_attn_layer_norm/scale": "flax_model-00002-of-00007.msgpack",
154
+ "model/decoder/layers/17/fc1/bias": "flax_model-00002-of-00007.msgpack",
155
+ "model/decoder/layers/17/fc1/kernel": "flax_model-00002-of-00007.msgpack",
156
+ "model/decoder/layers/17/fc2/bias": "flax_model-00002-of-00007.msgpack",
157
+ "model/decoder/layers/17/fc2/kernel": "flax_model-00002-of-00007.msgpack",
158
+ "model/decoder/layers/17/final_layer_norm/bias": "flax_model-00002-of-00007.msgpack",
159
+ "model/decoder/layers/17/final_layer_norm/scale": "flax_model-00002-of-00007.msgpack",
160
+ "model/decoder/layers/17/self_attn/k_proj/bias": "flax_model-00002-of-00007.msgpack",
161
+ "model/decoder/layers/17/self_attn/k_proj/kernel": "flax_model-00002-of-00007.msgpack",
162
+ "model/decoder/layers/17/self_attn/out_proj/bias": "flax_model-00002-of-00007.msgpack",
163
+ "model/decoder/layers/17/self_attn/out_proj/kernel": "flax_model-00002-of-00007.msgpack",
164
+ "model/decoder/layers/17/self_attn/q_proj/bias": "flax_model-00002-of-00007.msgpack",
165
+ "model/decoder/layers/17/self_attn/q_proj/kernel": "flax_model-00002-of-00007.msgpack",
166
+ "model/decoder/layers/17/self_attn/v_proj/bias": "flax_model-00002-of-00007.msgpack",
167
+ "model/decoder/layers/17/self_attn/v_proj/kernel": "flax_model-00002-of-00007.msgpack",
168
+ "model/decoder/layers/17/self_attn_layer_norm/bias": "flax_model-00002-of-00007.msgpack",
169
+ "model/decoder/layers/17/self_attn_layer_norm/scale": "flax_model-00002-of-00007.msgpack",
170
+ "model/decoder/layers/18/fc1/bias": "flax_model-00002-of-00007.msgpack",
171
+ "model/decoder/layers/18/fc1/kernel": "flax_model-00002-of-00007.msgpack",
172
+ "model/decoder/layers/18/fc2/bias": "flax_model-00002-of-00007.msgpack",
173
+ "model/decoder/layers/18/fc2/kernel": "flax_model-00002-of-00007.msgpack",
174
+ "model/decoder/layers/18/final_layer_norm/bias": "flax_model-00002-of-00007.msgpack",
175
+ "model/decoder/layers/18/final_layer_norm/scale": "flax_model-00002-of-00007.msgpack",
176
+ "model/decoder/layers/18/self_attn/k_proj/bias": "flax_model-00002-of-00007.msgpack",
177
+ "model/decoder/layers/18/self_attn/k_proj/kernel": "flax_model-00002-of-00007.msgpack",
178
+ "model/decoder/layers/18/self_attn/out_proj/bias": "flax_model-00002-of-00007.msgpack",
179
+ "model/decoder/layers/18/self_attn/out_proj/kernel": "flax_model-00002-of-00007.msgpack",
180
+ "model/decoder/layers/18/self_attn/q_proj/bias": "flax_model-00002-of-00007.msgpack",
181
+ "model/decoder/layers/18/self_attn/q_proj/kernel": "flax_model-00002-of-00007.msgpack",
182
+ "model/decoder/layers/18/self_attn/v_proj/bias": "flax_model-00002-of-00007.msgpack",
183
+ "model/decoder/layers/18/self_attn/v_proj/kernel": "flax_model-00002-of-00007.msgpack",
184
+ "model/decoder/layers/18/self_attn_layer_norm/bias": "flax_model-00002-of-00007.msgpack",
185
+ "model/decoder/layers/18/self_attn_layer_norm/scale": "flax_model-00002-of-00007.msgpack",
186
+ "model/decoder/layers/19/fc1/bias": "flax_model-00002-of-00007.msgpack",
187
+ "model/decoder/layers/19/fc1/kernel": "flax_model-00002-of-00007.msgpack",
188
+ "model/decoder/layers/19/fc2/bias": "flax_model-00002-of-00007.msgpack",
189
+ "model/decoder/layers/19/fc2/kernel": "flax_model-00002-of-00007.msgpack",
190
+ "model/decoder/layers/19/final_layer_norm/bias": "flax_model-00002-of-00007.msgpack",
191
+ "model/decoder/layers/19/final_layer_norm/scale": "flax_model-00002-of-00007.msgpack",
192
+ "model/decoder/layers/19/self_attn/k_proj/bias": "flax_model-00002-of-00007.msgpack",
193
+ "model/decoder/layers/19/self_attn/k_proj/kernel": "flax_model-00002-of-00007.msgpack",
194
+ "model/decoder/layers/19/self_attn/out_proj/bias": "flax_model-00002-of-00007.msgpack",
195
+ "model/decoder/layers/19/self_attn/out_proj/kernel": "flax_model-00002-of-00007.msgpack",
196
+ "model/decoder/layers/19/self_attn/q_proj/bias": "flax_model-00002-of-00007.msgpack",
197
+ "model/decoder/layers/19/self_attn/q_proj/kernel": "flax_model-00002-of-00007.msgpack",
198
+ "model/decoder/layers/19/self_attn/v_proj/bias": "flax_model-00002-of-00007.msgpack",
199
+ "model/decoder/layers/19/self_attn/v_proj/kernel": "flax_model-00002-of-00007.msgpack",
200
+ "model/decoder/layers/19/self_attn_layer_norm/bias": "flax_model-00002-of-00007.msgpack",
201
+ "model/decoder/layers/19/self_attn_layer_norm/scale": "flax_model-00002-of-00007.msgpack",
202
+ "model/decoder/layers/2/fc1/bias": "flax_model-00002-of-00007.msgpack",
203
+ "model/decoder/layers/2/fc1/kernel": "flax_model-00002-of-00007.msgpack",
204
+ "model/decoder/layers/2/fc2/bias": "flax_model-00002-of-00007.msgpack",
205
+ "model/decoder/layers/2/fc2/kernel": "flax_model-00002-of-00007.msgpack",
206
+ "model/decoder/layers/2/final_layer_norm/bias": "flax_model-00002-of-00007.msgpack",
207
+ "model/decoder/layers/2/final_layer_norm/scale": "flax_model-00002-of-00007.msgpack",
208
+ "model/decoder/layers/2/self_attn/k_proj/bias": "flax_model-00002-of-00007.msgpack",
209
+ "model/decoder/layers/2/self_attn/k_proj/kernel": "flax_model-00002-of-00007.msgpack",
210
+ "model/decoder/layers/2/self_attn/out_proj/bias": "flax_model-00002-of-00007.msgpack",
211
+ "model/decoder/layers/2/self_attn/out_proj/kernel": "flax_model-00002-of-00007.msgpack",
212
+ "model/decoder/layers/2/self_attn/q_proj/bias": "flax_model-00002-of-00007.msgpack",
213
+ "model/decoder/layers/2/self_attn/q_proj/kernel": "flax_model-00002-of-00007.msgpack",
214
+ "model/decoder/layers/2/self_attn/v_proj/bias": "flax_model-00002-of-00007.msgpack",
215
+ "model/decoder/layers/2/self_attn/v_proj/kernel": "flax_model-00002-of-00007.msgpack",
216
+ "model/decoder/layers/2/self_attn_layer_norm/bias": "flax_model-00002-of-00007.msgpack",
217
+ "model/decoder/layers/2/self_attn_layer_norm/scale": "flax_model-00002-of-00007.msgpack",
218
+ "model/decoder/layers/20/fc1/bias": "flax_model-00002-of-00007.msgpack",
219
+ "model/decoder/layers/20/fc1/kernel": "flax_model-00002-of-00007.msgpack",
220
+ "model/decoder/layers/20/fc2/bias": "flax_model-00002-of-00007.msgpack",
221
+ "model/decoder/layers/20/fc2/kernel": "flax_model-00002-of-00007.msgpack",
222
+ "model/decoder/layers/20/final_layer_norm/bias": "flax_model-00002-of-00007.msgpack",
223
+ "model/decoder/layers/20/final_layer_norm/scale": "flax_model-00002-of-00007.msgpack",
224
+ "model/decoder/layers/20/self_attn/k_proj/bias": "flax_model-00002-of-00007.msgpack",
225
+ "model/decoder/layers/20/self_attn/k_proj/kernel": "flax_model-00002-of-00007.msgpack",
226
+ "model/decoder/layers/20/self_attn/out_proj/bias": "flax_model-00002-of-00007.msgpack",
227
+ "model/decoder/layers/20/self_attn/out_proj/kernel": "flax_model-00002-of-00007.msgpack",
228
+ "model/decoder/layers/20/self_attn/q_proj/bias": "flax_model-00002-of-00007.msgpack",
229
+ "model/decoder/layers/20/self_attn/q_proj/kernel": "flax_model-00002-of-00007.msgpack",
230
+ "model/decoder/layers/20/self_attn/v_proj/bias": "flax_model-00002-of-00007.msgpack",
231
+ "model/decoder/layers/20/self_attn/v_proj/kernel": "flax_model-00002-of-00007.msgpack",
232
+ "model/decoder/layers/20/self_attn_layer_norm/bias": "flax_model-00002-of-00007.msgpack",
233
+ "model/decoder/layers/20/self_attn_layer_norm/scale": "flax_model-00002-of-00007.msgpack",
234
+ "model/decoder/layers/21/fc1/bias": "flax_model-00002-of-00007.msgpack",
235
+ "model/decoder/layers/21/fc1/kernel": "flax_model-00002-of-00007.msgpack",
236
+ "model/decoder/layers/21/fc2/bias": "flax_model-00002-of-00007.msgpack",
237
+ "model/decoder/layers/21/fc2/kernel": "flax_model-00002-of-00007.msgpack",
238
+ "model/decoder/layers/21/final_layer_norm/bias": "flax_model-00002-of-00007.msgpack",
239
+ "model/decoder/layers/21/final_layer_norm/scale": "flax_model-00002-of-00007.msgpack",
240
+ "model/decoder/layers/21/self_attn/k_proj/bias": "flax_model-00002-of-00007.msgpack",
241
+ "model/decoder/layers/21/self_attn/k_proj/kernel": "flax_model-00002-of-00007.msgpack",
242
+ "model/decoder/layers/21/self_attn/out_proj/bias": "flax_model-00002-of-00007.msgpack",
243
+ "model/decoder/layers/21/self_attn/out_proj/kernel": "flax_model-00002-of-00007.msgpack",
244
+ "model/decoder/layers/21/self_attn/q_proj/bias": "flax_model-00002-of-00007.msgpack",
245
+ "model/decoder/layers/21/self_attn/q_proj/kernel": "flax_model-00002-of-00007.msgpack",
246
+ "model/decoder/layers/21/self_attn/v_proj/bias": "flax_model-00002-of-00007.msgpack",
247
+ "model/decoder/layers/21/self_attn/v_proj/kernel": "flax_model-00002-of-00007.msgpack",
248
+ "model/decoder/layers/21/self_attn_layer_norm/bias": "flax_model-00002-of-00007.msgpack",
249
+ "model/decoder/layers/21/self_attn_layer_norm/scale": "flax_model-00002-of-00007.msgpack",
250
+ "model/decoder/layers/22/fc1/bias": "flax_model-00002-of-00007.msgpack",
251
+ "model/decoder/layers/22/fc1/kernel": "flax_model-00002-of-00007.msgpack",
252
+ "model/decoder/layers/22/fc2/bias": "flax_model-00002-of-00007.msgpack",
253
+ "model/decoder/layers/22/fc2/kernel": "flax_model-00003-of-00007.msgpack",
254
+ "model/decoder/layers/22/final_layer_norm/bias": "flax_model-00003-of-00007.msgpack",
255
+ "model/decoder/layers/22/final_layer_norm/scale": "flax_model-00003-of-00007.msgpack",
256
+ "model/decoder/layers/22/self_attn/k_proj/bias": "flax_model-00003-of-00007.msgpack",
257
+ "model/decoder/layers/22/self_attn/k_proj/kernel": "flax_model-00003-of-00007.msgpack",
258
+ "model/decoder/layers/22/self_attn/out_proj/bias": "flax_model-00003-of-00007.msgpack",
259
+ "model/decoder/layers/22/self_attn/out_proj/kernel": "flax_model-00003-of-00007.msgpack",
260
+ "model/decoder/layers/22/self_attn/q_proj/bias": "flax_model-00003-of-00007.msgpack",
261
+ "model/decoder/layers/22/self_attn/q_proj/kernel": "flax_model-00003-of-00007.msgpack",
262
+ "model/decoder/layers/22/self_attn/v_proj/bias": "flax_model-00003-of-00007.msgpack",
263
+ "model/decoder/layers/22/self_attn/v_proj/kernel": "flax_model-00003-of-00007.msgpack",
264
+ "model/decoder/layers/22/self_attn_layer_norm/bias": "flax_model-00003-of-00007.msgpack",
265
+ "model/decoder/layers/22/self_attn_layer_norm/scale": "flax_model-00003-of-00007.msgpack",
266
+ "model/decoder/layers/23/fc1/bias": "flax_model-00003-of-00007.msgpack",
267
+ "model/decoder/layers/23/fc1/kernel": "flax_model-00003-of-00007.msgpack",
268
+ "model/decoder/layers/23/fc2/bias": "flax_model-00003-of-00007.msgpack",
269
+ "model/decoder/layers/23/fc2/kernel": "flax_model-00003-of-00007.msgpack",
270
+ "model/decoder/layers/23/final_layer_norm/bias": "flax_model-00003-of-00007.msgpack",
271
+ "model/decoder/layers/23/final_layer_norm/scale": "flax_model-00003-of-00007.msgpack",
272
+ "model/decoder/layers/23/self_attn/k_proj/bias": "flax_model-00003-of-00007.msgpack",
273
+ "model/decoder/layers/23/self_attn/k_proj/kernel": "flax_model-00003-of-00007.msgpack",
274
+ "model/decoder/layers/23/self_attn/out_proj/bias": "flax_model-00003-of-00007.msgpack",
275
+ "model/decoder/layers/23/self_attn/out_proj/kernel": "flax_model-00003-of-00007.msgpack",
276
+ "model/decoder/layers/23/self_attn/q_proj/bias": "flax_model-00003-of-00007.msgpack",
277
+ "model/decoder/layers/23/self_attn/q_proj/kernel": "flax_model-00003-of-00007.msgpack",
278
+ "model/decoder/layers/23/self_attn/v_proj/bias": "flax_model-00003-of-00007.msgpack",
279
+ "model/decoder/layers/23/self_attn/v_proj/kernel": "flax_model-00003-of-00007.msgpack",
280
+ "model/decoder/layers/23/self_attn_layer_norm/bias": "flax_model-00003-of-00007.msgpack",
281
+ "model/decoder/layers/23/self_attn_layer_norm/scale": "flax_model-00003-of-00007.msgpack",
282
+ "model/decoder/layers/24/fc1/bias": "flax_model-00003-of-00007.msgpack",
283
+ "model/decoder/layers/24/fc1/kernel": "flax_model-00003-of-00007.msgpack",
284
+ "model/decoder/layers/24/fc2/bias": "flax_model-00003-of-00007.msgpack",
285
+ "model/decoder/layers/24/fc2/kernel": "flax_model-00003-of-00007.msgpack",
286
+ "model/decoder/layers/24/final_layer_norm/bias": "flax_model-00003-of-00007.msgpack",
287
+ "model/decoder/layers/24/final_layer_norm/scale": "flax_model-00003-of-00007.msgpack",
288
+ "model/decoder/layers/24/self_attn/k_proj/bias": "flax_model-00003-of-00007.msgpack",
289
+ "model/decoder/layers/24/self_attn/k_proj/kernel": "flax_model-00003-of-00007.msgpack",
290
+ "model/decoder/layers/24/self_attn/out_proj/bias": "flax_model-00003-of-00007.msgpack",
291
+ "model/decoder/layers/24/self_attn/out_proj/kernel": "flax_model-00003-of-00007.msgpack",
292
+ "model/decoder/layers/24/self_attn/q_proj/bias": "flax_model-00003-of-00007.msgpack",
293
+ "model/decoder/layers/24/self_attn/q_proj/kernel": "flax_model-00003-of-00007.msgpack",
294
+ "model/decoder/layers/24/self_attn/v_proj/bias": "flax_model-00003-of-00007.msgpack",
295
+ "model/decoder/layers/24/self_attn/v_proj/kernel": "flax_model-00003-of-00007.msgpack",
296
+ "model/decoder/layers/24/self_attn_layer_norm/bias": "flax_model-00003-of-00007.msgpack",
297
+ "model/decoder/layers/24/self_attn_layer_norm/scale": "flax_model-00003-of-00007.msgpack",
298
+ "model/decoder/layers/25/fc1/bias": "flax_model-00003-of-00007.msgpack",
299
+ "model/decoder/layers/25/fc1/kernel": "flax_model-00003-of-00007.msgpack",
300
+ "model/decoder/layers/25/fc2/bias": "flax_model-00003-of-00007.msgpack",
301
+ "model/decoder/layers/25/fc2/kernel": "flax_model-00003-of-00007.msgpack",
302
+ "model/decoder/layers/25/final_layer_norm/bias": "flax_model-00003-of-00007.msgpack",
303
+ "model/decoder/layers/25/final_layer_norm/scale": "flax_model-00003-of-00007.msgpack",
304
+ "model/decoder/layers/25/self_attn/k_proj/bias": "flax_model-00003-of-00007.msgpack",
305
+ "model/decoder/layers/25/self_attn/k_proj/kernel": "flax_model-00003-of-00007.msgpack",
306
+ "model/decoder/layers/25/self_attn/out_proj/bias": "flax_model-00003-of-00007.msgpack",
307
+ "model/decoder/layers/25/self_attn/out_proj/kernel": "flax_model-00003-of-00007.msgpack",
308
+ "model/decoder/layers/25/self_attn/q_proj/bias": "flax_model-00003-of-00007.msgpack",
309
+ "model/decoder/layers/25/self_attn/q_proj/kernel": "flax_model-00003-of-00007.msgpack",
310
+ "model/decoder/layers/25/self_attn/v_proj/bias": "flax_model-00003-of-00007.msgpack",
311
+ "model/decoder/layers/25/self_attn/v_proj/kernel": "flax_model-00003-of-00007.msgpack",
312
+ "model/decoder/layers/25/self_attn_layer_norm/bias": "flax_model-00003-of-00007.msgpack",
313
+ "model/decoder/layers/25/self_attn_layer_norm/scale": "flax_model-00003-of-00007.msgpack",
314
+ "model/decoder/layers/26/fc1/bias": "flax_model-00003-of-00007.msgpack",
315
+ "model/decoder/layers/26/fc1/kernel": "flax_model-00003-of-00007.msgpack",
316
+ "model/decoder/layers/26/fc2/bias": "flax_model-00003-of-00007.msgpack",
317
+ "model/decoder/layers/26/fc2/kernel": "flax_model-00003-of-00007.msgpack",
318
+ "model/decoder/layers/26/final_layer_norm/bias": "flax_model-00003-of-00007.msgpack",
319
+ "model/decoder/layers/26/final_layer_norm/scale": "flax_model-00003-of-00007.msgpack",
320
+ "model/decoder/layers/26/self_attn/k_proj/bias": "flax_model-00003-of-00007.msgpack",
321
+ "model/decoder/layers/26/self_attn/k_proj/kernel": "flax_model-00003-of-00007.msgpack",
322
+ "model/decoder/layers/26/self_attn/out_proj/bias": "flax_model-00003-of-00007.msgpack",
323
+ "model/decoder/layers/26/self_attn/out_proj/kernel": "flax_model-00003-of-00007.msgpack",
324
+ "model/decoder/layers/26/self_attn/q_proj/bias": "flax_model-00003-of-00007.msgpack",
325
+ "model/decoder/layers/26/self_attn/q_proj/kernel": "flax_model-00003-of-00007.msgpack",
326
+ "model/decoder/layers/26/self_attn/v_proj/bias": "flax_model-00003-of-00007.msgpack",
327
+ "model/decoder/layers/26/self_attn/v_proj/kernel": "flax_model-00003-of-00007.msgpack",
328
+ "model/decoder/layers/26/self_attn_layer_norm/bias": "flax_model-00003-of-00007.msgpack",
329
+ "model/decoder/layers/26/self_attn_layer_norm/scale": "flax_model-00003-of-00007.msgpack",
330
+ "model/decoder/layers/27/fc1/bias": "flax_model-00003-of-00007.msgpack",
331
+ "model/decoder/layers/27/fc1/kernel": "flax_model-00003-of-00007.msgpack",
332
+ "model/decoder/layers/27/fc2/bias": "flax_model-00003-of-00007.msgpack",
333
+ "model/decoder/layers/27/fc2/kernel": "flax_model-00003-of-00007.msgpack",
334
+ "model/decoder/layers/27/final_layer_norm/bias": "flax_model-00003-of-00007.msgpack",
335
+ "model/decoder/layers/27/final_layer_norm/scale": "flax_model-00003-of-00007.msgpack",
336
+ "model/decoder/layers/27/self_attn/k_proj/bias": "flax_model-00003-of-00007.msgpack",
337
+ "model/decoder/layers/27/self_attn/k_proj/kernel": "flax_model-00003-of-00007.msgpack",
338
+ "model/decoder/layers/27/self_attn/out_proj/bias": "flax_model-00003-of-00007.msgpack",
339
+ "model/decoder/layers/27/self_attn/out_proj/kernel": "flax_model-00003-of-00007.msgpack",
340
+ "model/decoder/layers/27/self_attn/q_proj/bias": "flax_model-00003-of-00007.msgpack",
341
+ "model/decoder/layers/27/self_attn/q_proj/kernel": "flax_model-00003-of-00007.msgpack",
342
+ "model/decoder/layers/27/self_attn/v_proj/bias": "flax_model-00003-of-00007.msgpack",
343
+ "model/decoder/layers/27/self_attn/v_proj/kernel": "flax_model-00003-of-00007.msgpack",
344
+ "model/decoder/layers/27/self_attn_layer_norm/bias": "flax_model-00003-of-00007.msgpack",
345
+ "model/decoder/layers/27/self_attn_layer_norm/scale": "flax_model-00003-of-00007.msgpack",
346
+ "model/decoder/layers/28/fc1/bias": "flax_model-00003-of-00007.msgpack",
347
+ "model/decoder/layers/28/fc1/kernel": "flax_model-00003-of-00007.msgpack",
348
+ "model/decoder/layers/28/fc2/bias": "flax_model-00003-of-00007.msgpack",
349
+ "model/decoder/layers/28/fc2/kernel": "flax_model-00003-of-00007.msgpack",
350
+ "model/decoder/layers/28/final_layer_norm/bias": "flax_model-00003-of-00007.msgpack",
351
+ "model/decoder/layers/28/final_layer_norm/scale": "flax_model-00003-of-00007.msgpack",
352
+ "model/decoder/layers/28/self_attn/k_proj/bias": "flax_model-00003-of-00007.msgpack",
353
+ "model/decoder/layers/28/self_attn/k_proj/kernel": "flax_model-00003-of-00007.msgpack",
354
+ "model/decoder/layers/28/self_attn/out_proj/bias": "flax_model-00003-of-00007.msgpack",
355
+ "model/decoder/layers/28/self_attn/out_proj/kernel": "flax_model-00003-of-00007.msgpack",
356
+ "model/decoder/layers/28/self_attn/q_proj/bias": "flax_model-00003-of-00007.msgpack",
357
+ "model/decoder/layers/28/self_attn/q_proj/kernel": "flax_model-00003-of-00007.msgpack",
358
+ "model/decoder/layers/28/self_attn/v_proj/bias": "flax_model-00003-of-00007.msgpack",
359
+ "model/decoder/layers/28/self_attn/v_proj/kernel": "flax_model-00003-of-00007.msgpack",
360
+ "model/decoder/layers/28/self_attn_layer_norm/bias": "flax_model-00003-of-00007.msgpack",
361
+ "model/decoder/layers/28/self_attn_layer_norm/scale": "flax_model-00003-of-00007.msgpack",
362
+ "model/decoder/layers/29/fc1/bias": "flax_model-00003-of-00007.msgpack",
363
+ "model/decoder/layers/29/fc1/kernel": "flax_model-00003-of-00007.msgpack",
364
+ "model/decoder/layers/29/fc2/bias": "flax_model-00003-of-00007.msgpack",
365
+ "model/decoder/layers/29/fc2/kernel": "flax_model-00003-of-00007.msgpack",
366
+ "model/decoder/layers/29/final_layer_norm/bias": "flax_model-00003-of-00007.msgpack",
367
+ "model/decoder/layers/29/final_layer_norm/scale": "flax_model-00003-of-00007.msgpack",
368
+ "model/decoder/layers/29/self_attn/k_proj/bias": "flax_model-00003-of-00007.msgpack",
369
+ "model/decoder/layers/29/self_attn/k_proj/kernel": "flax_model-00003-of-00007.msgpack",
370
+ "model/decoder/layers/29/self_attn/out_proj/bias": "flax_model-00003-of-00007.msgpack",
371
+ "model/decoder/layers/29/self_attn/out_proj/kernel": "flax_model-00003-of-00007.msgpack",
372
+ "model/decoder/layers/29/self_attn/q_proj/bias": "flax_model-00003-of-00007.msgpack",
373
+ "model/decoder/layers/29/self_attn/q_proj/kernel": "flax_model-00003-of-00007.msgpack",
374
+ "model/decoder/layers/29/self_attn/v_proj/bias": "flax_model-00003-of-00007.msgpack",
375
+ "model/decoder/layers/29/self_attn/v_proj/kernel": "flax_model-00003-of-00007.msgpack",
376
+ "model/decoder/layers/29/self_attn_layer_norm/bias": "flax_model-00003-of-00007.msgpack",
377
+ "model/decoder/layers/29/self_attn_layer_norm/scale": "flax_model-00003-of-00007.msgpack",
378
+ "model/decoder/layers/3/fc1/bias": "flax_model-00003-of-00007.msgpack",
379
+ "model/decoder/layers/3/fc1/kernel": "flax_model-00003-of-00007.msgpack",
380
+ "model/decoder/layers/3/fc2/bias": "flax_model-00003-of-00007.msgpack",
381
+ "model/decoder/layers/3/fc2/kernel": "flax_model-00004-of-00007.msgpack",
382
+ "model/decoder/layers/3/final_layer_norm/bias": "flax_model-00004-of-00007.msgpack",
383
+ "model/decoder/layers/3/final_layer_norm/scale": "flax_model-00004-of-00007.msgpack",
384
+ "model/decoder/layers/3/self_attn/k_proj/bias": "flax_model-00004-of-00007.msgpack",
385
+ "model/decoder/layers/3/self_attn/k_proj/kernel": "flax_model-00004-of-00007.msgpack",
386
+ "model/decoder/layers/3/self_attn/out_proj/bias": "flax_model-00004-of-00007.msgpack",
387
+ "model/decoder/layers/3/self_attn/out_proj/kernel": "flax_model-00004-of-00007.msgpack",
388
+ "model/decoder/layers/3/self_attn/q_proj/bias": "flax_model-00004-of-00007.msgpack",
389
+ "model/decoder/layers/3/self_attn/q_proj/kernel": "flax_model-00004-of-00007.msgpack",
390
+ "model/decoder/layers/3/self_attn/v_proj/bias": "flax_model-00004-of-00007.msgpack",
391
+ "model/decoder/layers/3/self_attn/v_proj/kernel": "flax_model-00004-of-00007.msgpack",
392
+ "model/decoder/layers/3/self_attn_layer_norm/bias": "flax_model-00004-of-00007.msgpack",
393
+ "model/decoder/layers/3/self_attn_layer_norm/scale": "flax_model-00004-of-00007.msgpack",
394
+ "model/decoder/layers/30/fc1/bias": "flax_model-00004-of-00007.msgpack",
395
+ "model/decoder/layers/30/fc1/kernel": "flax_model-00004-of-00007.msgpack",
396
+ "model/decoder/layers/30/fc2/bias": "flax_model-00004-of-00007.msgpack",
397
+ "model/decoder/layers/30/fc2/kernel": "flax_model-00004-of-00007.msgpack",
398
+ "model/decoder/layers/30/final_layer_norm/bias": "flax_model-00004-of-00007.msgpack",
399
+ "model/decoder/layers/30/final_layer_norm/scale": "flax_model-00004-of-00007.msgpack",
400
+ "model/decoder/layers/30/self_attn/k_proj/bias": "flax_model-00004-of-00007.msgpack",
401
+ "model/decoder/layers/30/self_attn/k_proj/kernel": "flax_model-00004-of-00007.msgpack",
402
+ "model/decoder/layers/30/self_attn/out_proj/bias": "flax_model-00004-of-00007.msgpack",
403
+ "model/decoder/layers/30/self_attn/out_proj/kernel": "flax_model-00004-of-00007.msgpack",
404
+ "model/decoder/layers/30/self_attn/q_proj/bias": "flax_model-00004-of-00007.msgpack",
405
+ "model/decoder/layers/30/self_attn/q_proj/kernel": "flax_model-00004-of-00007.msgpack",
406
+ "model/decoder/layers/30/self_attn/v_proj/bias": "flax_model-00004-of-00007.msgpack",
407
+ "model/decoder/layers/30/self_attn/v_proj/kernel": "flax_model-00004-of-00007.msgpack",
408
+ "model/decoder/layers/30/self_attn_layer_norm/bias": "flax_model-00004-of-00007.msgpack",
409
+ "model/decoder/layers/30/self_attn_layer_norm/scale": "flax_model-00004-of-00007.msgpack",
410
+ "model/decoder/layers/31/fc1/bias": "flax_model-00004-of-00007.msgpack",
411
+ "model/decoder/layers/31/fc1/kernel": "flax_model-00004-of-00007.msgpack",
412
+ "model/decoder/layers/31/fc2/bias": "flax_model-00004-of-00007.msgpack",
413
+ "model/decoder/layers/31/fc2/kernel": "flax_model-00004-of-00007.msgpack",
414
+ "model/decoder/layers/31/final_layer_norm/bias": "flax_model-00004-of-00007.msgpack",
415
+ "model/decoder/layers/31/final_layer_norm/scale": "flax_model-00004-of-00007.msgpack",
416
+ "model/decoder/layers/31/self_attn/k_proj/bias": "flax_model-00004-of-00007.msgpack",
417
+ "model/decoder/layers/31/self_attn/k_proj/kernel": "flax_model-00004-of-00007.msgpack",
418
+ "model/decoder/layers/31/self_attn/out_proj/bias": "flax_model-00004-of-00007.msgpack",
419
+ "model/decoder/layers/31/self_attn/out_proj/kernel": "flax_model-00004-of-00007.msgpack",
420
+ "model/decoder/layers/31/self_attn/q_proj/bias": "flax_model-00004-of-00007.msgpack",
421
+ "model/decoder/layers/31/self_attn/q_proj/kernel": "flax_model-00004-of-00007.msgpack",
422
+ "model/decoder/layers/31/self_attn/v_proj/bias": "flax_model-00004-of-00007.msgpack",
423
+ "model/decoder/layers/31/self_attn/v_proj/kernel": "flax_model-00004-of-00007.msgpack",
424
+ "model/decoder/layers/31/self_attn_layer_norm/bias": "flax_model-00004-of-00007.msgpack",
425
+ "model/decoder/layers/31/self_attn_layer_norm/scale": "flax_model-00004-of-00007.msgpack",
426
+ "model/decoder/layers/32/fc1/bias": "flax_model-00004-of-00007.msgpack",
427
+ "model/decoder/layers/32/fc1/kernel": "flax_model-00004-of-00007.msgpack",
428
+ "model/decoder/layers/32/fc2/bias": "flax_model-00004-of-00007.msgpack",
429
+ "model/decoder/layers/32/fc2/kernel": "flax_model-00004-of-00007.msgpack",
430
+ "model/decoder/layers/32/final_layer_norm/bias": "flax_model-00004-of-00007.msgpack",
431
+ "model/decoder/layers/32/final_layer_norm/scale": "flax_model-00004-of-00007.msgpack",
432
+ "model/decoder/layers/32/self_attn/k_proj/bias": "flax_model-00004-of-00007.msgpack",
433
+ "model/decoder/layers/32/self_attn/k_proj/kernel": "flax_model-00004-of-00007.msgpack",
434
+ "model/decoder/layers/32/self_attn/out_proj/bias": "flax_model-00004-of-00007.msgpack",
435
+ "model/decoder/layers/32/self_attn/out_proj/kernel": "flax_model-00004-of-00007.msgpack",
436
+ "model/decoder/layers/32/self_attn/q_proj/bias": "flax_model-00004-of-00007.msgpack",
437
+ "model/decoder/layers/32/self_attn/q_proj/kernel": "flax_model-00004-of-00007.msgpack",
438
+ "model/decoder/layers/32/self_attn/v_proj/bias": "flax_model-00004-of-00007.msgpack",
439
+ "model/decoder/layers/32/self_attn/v_proj/kernel": "flax_model-00004-of-00007.msgpack",
440
+ "model/decoder/layers/32/self_attn_layer_norm/bias": "flax_model-00004-of-00007.msgpack",
441
+ "model/decoder/layers/32/self_attn_layer_norm/scale": "flax_model-00004-of-00007.msgpack",
442
+ "model/decoder/layers/33/fc1/bias": "flax_model-00004-of-00007.msgpack",
443
+ "model/decoder/layers/33/fc1/kernel": "flax_model-00004-of-00007.msgpack",
444
+ "model/decoder/layers/33/fc2/bias": "flax_model-00004-of-00007.msgpack",
445
+ "model/decoder/layers/33/fc2/kernel": "flax_model-00004-of-00007.msgpack",
446
+ "model/decoder/layers/33/final_layer_norm/bias": "flax_model-00004-of-00007.msgpack",
447
+ "model/decoder/layers/33/final_layer_norm/scale": "flax_model-00004-of-00007.msgpack",
448
+ "model/decoder/layers/33/self_attn/k_proj/bias": "flax_model-00004-of-00007.msgpack",
449
+ "model/decoder/layers/33/self_attn/k_proj/kernel": "flax_model-00004-of-00007.msgpack",
450
+ "model/decoder/layers/33/self_attn/out_proj/bias": "flax_model-00004-of-00007.msgpack",
451
+ "model/decoder/layers/33/self_attn/out_proj/kernel": "flax_model-00004-of-00007.msgpack",
452
+ "model/decoder/layers/33/self_attn/q_proj/bias": "flax_model-00004-of-00007.msgpack",
453
+ "model/decoder/layers/33/self_attn/q_proj/kernel": "flax_model-00004-of-00007.msgpack",
454
+ "model/decoder/layers/33/self_attn/v_proj/bias": "flax_model-00004-of-00007.msgpack",
455
+ "model/decoder/layers/33/self_attn/v_proj/kernel": "flax_model-00004-of-00007.msgpack",
456
+ "model/decoder/layers/33/self_attn_layer_norm/bias": "flax_model-00004-of-00007.msgpack",
457
+ "model/decoder/layers/33/self_attn_layer_norm/scale": "flax_model-00004-of-00007.msgpack",
458
+ "model/decoder/layers/34/fc1/bias": "flax_model-00004-of-00007.msgpack",
459
+ "model/decoder/layers/34/fc1/kernel": "flax_model-00004-of-00007.msgpack",
460
+ "model/decoder/layers/34/fc2/bias": "flax_model-00004-of-00007.msgpack",
461
+ "model/decoder/layers/34/fc2/kernel": "flax_model-00004-of-00007.msgpack",
462
+ "model/decoder/layers/34/final_layer_norm/bias": "flax_model-00004-of-00007.msgpack",
463
+ "model/decoder/layers/34/final_layer_norm/scale": "flax_model-00004-of-00007.msgpack",
464
+ "model/decoder/layers/34/self_attn/k_proj/bias": "flax_model-00004-of-00007.msgpack",
465
+ "model/decoder/layers/34/self_attn/k_proj/kernel": "flax_model-00004-of-00007.msgpack",
466
+ "model/decoder/layers/34/self_attn/out_proj/bias": "flax_model-00004-of-00007.msgpack",
467
+ "model/decoder/layers/34/self_attn/out_proj/kernel": "flax_model-00004-of-00007.msgpack",
468
+ "model/decoder/layers/34/self_attn/q_proj/bias": "flax_model-00004-of-00007.msgpack",
469
+ "model/decoder/layers/34/self_attn/q_proj/kernel": "flax_model-00004-of-00007.msgpack",
470
+ "model/decoder/layers/34/self_attn/v_proj/bias": "flax_model-00004-of-00007.msgpack",
471
+ "model/decoder/layers/34/self_attn/v_proj/kernel": "flax_model-00004-of-00007.msgpack",
472
+ "model/decoder/layers/34/self_attn_layer_norm/bias": "flax_model-00004-of-00007.msgpack",
473
+ "model/decoder/layers/34/self_attn_layer_norm/scale": "flax_model-00004-of-00007.msgpack",
474
+ "model/decoder/layers/35/fc1/bias": "flax_model-00004-of-00007.msgpack",
475
+ "model/decoder/layers/35/fc1/kernel": "flax_model-00004-of-00007.msgpack",
476
+ "model/decoder/layers/35/fc2/bias": "flax_model-00004-of-00007.msgpack",
477
+ "model/decoder/layers/35/fc2/kernel": "flax_model-00004-of-00007.msgpack",
478
+ "model/decoder/layers/35/final_layer_norm/bias": "flax_model-00004-of-00007.msgpack",
479
+ "model/decoder/layers/35/final_layer_norm/scale": "flax_model-00004-of-00007.msgpack",
480
+ "model/decoder/layers/35/self_attn/k_proj/bias": "flax_model-00004-of-00007.msgpack",
481
+ "model/decoder/layers/35/self_attn/k_proj/kernel": "flax_model-00004-of-00007.msgpack",
482
+ "model/decoder/layers/35/self_attn/out_proj/bias": "flax_model-00004-of-00007.msgpack",
483
+ "model/decoder/layers/35/self_attn/out_proj/kernel": "flax_model-00004-of-00007.msgpack",
484
+ "model/decoder/layers/35/self_attn/q_proj/bias": "flax_model-00004-of-00007.msgpack",
485
+ "model/decoder/layers/35/self_attn/q_proj/kernel": "flax_model-00004-of-00007.msgpack",
486
+ "model/decoder/layers/35/self_attn/v_proj/bias": "flax_model-00004-of-00007.msgpack",
487
+ "model/decoder/layers/35/self_attn/v_proj/kernel": "flax_model-00004-of-00007.msgpack",
488
+ "model/decoder/layers/35/self_attn_layer_norm/bias": "flax_model-00004-of-00007.msgpack",
489
+ "model/decoder/layers/35/self_attn_layer_norm/scale": "flax_model-00004-of-00007.msgpack",
490
+ "model/decoder/layers/36/fc1/bias": "flax_model-00004-of-00007.msgpack",
491
+ "model/decoder/layers/36/fc1/kernel": "flax_model-00004-of-00007.msgpack",
492
+ "model/decoder/layers/36/fc2/bias": "flax_model-00004-of-00007.msgpack",
493
+ "model/decoder/layers/36/fc2/kernel": "flax_model-00004-of-00007.msgpack",
494
+ "model/decoder/layers/36/final_layer_norm/bias": "flax_model-00004-of-00007.msgpack",
495
+ "model/decoder/layers/36/final_layer_norm/scale": "flax_model-00004-of-00007.msgpack",
496
+ "model/decoder/layers/36/self_attn/k_proj/bias": "flax_model-00004-of-00007.msgpack",
497
+ "model/decoder/layers/36/self_attn/k_proj/kernel": "flax_model-00004-of-00007.msgpack",
498
+ "model/decoder/layers/36/self_attn/out_proj/bias": "flax_model-00004-of-00007.msgpack",
499
+ "model/decoder/layers/36/self_attn/out_proj/kernel": "flax_model-00004-of-00007.msgpack",
500
+ "model/decoder/layers/36/self_attn/q_proj/bias": "flax_model-00004-of-00007.msgpack",
501
+ "model/decoder/layers/36/self_attn/q_proj/kernel": "flax_model-00004-of-00007.msgpack",
502
+ "model/decoder/layers/36/self_attn/v_proj/bias": "flax_model-00004-of-00007.msgpack",
503
+ "model/decoder/layers/36/self_attn/v_proj/kernel": "flax_model-00004-of-00007.msgpack",
504
+ "model/decoder/layers/36/self_attn_layer_norm/bias": "flax_model-00004-of-00007.msgpack",
505
+ "model/decoder/layers/36/self_attn_layer_norm/scale": "flax_model-00004-of-00007.msgpack",
506
+ "model/decoder/layers/37/fc1/bias": "flax_model-00004-of-00007.msgpack",
507
+ "model/decoder/layers/37/fc1/kernel": "flax_model-00004-of-00007.msgpack",
508
+ "model/decoder/layers/37/fc2/bias": "flax_model-00004-of-00007.msgpack",
509
+ "model/decoder/layers/37/fc2/kernel": "flax_model-00005-of-00007.msgpack",
510
+ "model/decoder/layers/37/final_layer_norm/bias": "flax_model-00005-of-00007.msgpack",
511
+ "model/decoder/layers/37/final_layer_norm/scale": "flax_model-00005-of-00007.msgpack",
512
+ "model/decoder/layers/37/self_attn/k_proj/bias": "flax_model-00005-of-00007.msgpack",
513
+ "model/decoder/layers/37/self_attn/k_proj/kernel": "flax_model-00005-of-00007.msgpack",
514
+ "model/decoder/layers/37/self_attn/out_proj/bias": "flax_model-00005-of-00007.msgpack",
515
+ "model/decoder/layers/37/self_attn/out_proj/kernel": "flax_model-00005-of-00007.msgpack",
516
+ "model/decoder/layers/37/self_attn/q_proj/bias": "flax_model-00005-of-00007.msgpack",
517
+ "model/decoder/layers/37/self_attn/q_proj/kernel": "flax_model-00005-of-00007.msgpack",
518
+ "model/decoder/layers/37/self_attn/v_proj/bias": "flax_model-00005-of-00007.msgpack",
519
+ "model/decoder/layers/37/self_attn/v_proj/kernel": "flax_model-00005-of-00007.msgpack",
520
+ "model/decoder/layers/37/self_attn_layer_norm/bias": "flax_model-00005-of-00007.msgpack",
521
+ "model/decoder/layers/37/self_attn_layer_norm/scale": "flax_model-00005-of-00007.msgpack",
522
+ "model/decoder/layers/38/fc1/bias": "flax_model-00005-of-00007.msgpack",
523
+ "model/decoder/layers/38/fc1/kernel": "flax_model-00005-of-00007.msgpack",
524
+ "model/decoder/layers/38/fc2/bias": "flax_model-00005-of-00007.msgpack",
525
+ "model/decoder/layers/38/fc2/kernel": "flax_model-00005-of-00007.msgpack",
526
+ "model/decoder/layers/38/final_layer_norm/bias": "flax_model-00005-of-00007.msgpack",
527
+ "model/decoder/layers/38/final_layer_norm/scale": "flax_model-00005-of-00007.msgpack",
528
+ "model/decoder/layers/38/self_attn/k_proj/bias": "flax_model-00005-of-00007.msgpack",
529
+ "model/decoder/layers/38/self_attn/k_proj/kernel": "flax_model-00005-of-00007.msgpack",
530
+ "model/decoder/layers/38/self_attn/out_proj/bias": "flax_model-00005-of-00007.msgpack",
531
+ "model/decoder/layers/38/self_attn/out_proj/kernel": "flax_model-00005-of-00007.msgpack",
532
+ "model/decoder/layers/38/self_attn/q_proj/bias": "flax_model-00005-of-00007.msgpack",
533
+ "model/decoder/layers/38/self_attn/q_proj/kernel": "flax_model-00005-of-00007.msgpack",
534
+ "model/decoder/layers/38/self_attn/v_proj/bias": "flax_model-00005-of-00007.msgpack",
535
+ "model/decoder/layers/38/self_attn/v_proj/kernel": "flax_model-00005-of-00007.msgpack",
536
+ "model/decoder/layers/38/self_attn_layer_norm/bias": "flax_model-00005-of-00007.msgpack",
537
+ "model/decoder/layers/38/self_attn_layer_norm/scale": "flax_model-00005-of-00007.msgpack",
538
+ "model/decoder/layers/39/fc1/bias": "flax_model-00005-of-00007.msgpack",
539
+ "model/decoder/layers/39/fc1/kernel": "flax_model-00005-of-00007.msgpack",
540
+ "model/decoder/layers/39/fc2/bias": "flax_model-00005-of-00007.msgpack",
541
+ "model/decoder/layers/39/fc2/kernel": "flax_model-00005-of-00007.msgpack",
542
+ "model/decoder/layers/39/final_layer_norm/bias": "flax_model-00005-of-00007.msgpack",
543
+ "model/decoder/layers/39/final_layer_norm/scale": "flax_model-00005-of-00007.msgpack",
544
+ "model/decoder/layers/39/self_attn/k_proj/bias": "flax_model-00005-of-00007.msgpack",
545
+ "model/decoder/layers/39/self_attn/k_proj/kernel": "flax_model-00005-of-00007.msgpack",
546
+ "model/decoder/layers/39/self_attn/out_proj/bias": "flax_model-00005-of-00007.msgpack",
547
+ "model/decoder/layers/39/self_attn/out_proj/kernel": "flax_model-00005-of-00007.msgpack",
548
+ "model/decoder/layers/39/self_attn/q_proj/bias": "flax_model-00005-of-00007.msgpack",
549
+ "model/decoder/layers/39/self_attn/q_proj/kernel": "flax_model-00005-of-00007.msgpack",
550
+ "model/decoder/layers/39/self_attn/v_proj/bias": "flax_model-00005-of-00007.msgpack",
551
+ "model/decoder/layers/39/self_attn/v_proj/kernel": "flax_model-00005-of-00007.msgpack",
552
+ "model/decoder/layers/39/self_attn_layer_norm/bias": "flax_model-00005-of-00007.msgpack",
553
+ "model/decoder/layers/39/self_attn_layer_norm/scale": "flax_model-00005-of-00007.msgpack",
554
+ "model/decoder/layers/4/fc1/bias": "flax_model-00005-of-00007.msgpack",
555
+ "model/decoder/layers/4/fc1/kernel": "flax_model-00005-of-00007.msgpack",
556
+ "model/decoder/layers/4/fc2/bias": "flax_model-00005-of-00007.msgpack",
557
+ "model/decoder/layers/4/fc2/kernel": "flax_model-00005-of-00007.msgpack",
558
+ "model/decoder/layers/4/final_layer_norm/bias": "flax_model-00005-of-00007.msgpack",
559
+ "model/decoder/layers/4/final_layer_norm/scale": "flax_model-00005-of-00007.msgpack",
560
+ "model/decoder/layers/4/self_attn/k_proj/bias": "flax_model-00005-of-00007.msgpack",
561
+ "model/decoder/layers/4/self_attn/k_proj/kernel": "flax_model-00005-of-00007.msgpack",
562
+ "model/decoder/layers/4/self_attn/out_proj/bias": "flax_model-00005-of-00007.msgpack",
563
+ "model/decoder/layers/4/self_attn/out_proj/kernel": "flax_model-00005-of-00007.msgpack",
564
+ "model/decoder/layers/4/self_attn/q_proj/bias": "flax_model-00005-of-00007.msgpack",
565
+ "model/decoder/layers/4/self_attn/q_proj/kernel": "flax_model-00005-of-00007.msgpack",
566
+ "model/decoder/layers/4/self_attn/v_proj/bias": "flax_model-00005-of-00007.msgpack",
567
+ "model/decoder/layers/4/self_attn/v_proj/kernel": "flax_model-00005-of-00007.msgpack",
568
+ "model/decoder/layers/4/self_attn_layer_norm/bias": "flax_model-00005-of-00007.msgpack",
569
+ "model/decoder/layers/4/self_attn_layer_norm/scale": "flax_model-00005-of-00007.msgpack",
570
+ "model/decoder/layers/40/fc1/bias": "flax_model-00005-of-00007.msgpack",
571
+ "model/decoder/layers/40/fc1/kernel": "flax_model-00005-of-00007.msgpack",
572
+ "model/decoder/layers/40/fc2/bias": "flax_model-00005-of-00007.msgpack",
573
+ "model/decoder/layers/40/fc2/kernel": "flax_model-00005-of-00007.msgpack",
574
+ "model/decoder/layers/40/final_layer_norm/bias": "flax_model-00005-of-00007.msgpack",
575
+ "model/decoder/layers/40/final_layer_norm/scale": "flax_model-00005-of-00007.msgpack",
576
+ "model/decoder/layers/40/self_attn/k_proj/bias": "flax_model-00005-of-00007.msgpack",
577
+ "model/decoder/layers/40/self_attn/k_proj/kernel": "flax_model-00005-of-00007.msgpack",
578
+ "model/decoder/layers/40/self_attn/out_proj/bias": "flax_model-00005-of-00007.msgpack",
579
+ "model/decoder/layers/40/self_attn/out_proj/kernel": "flax_model-00005-of-00007.msgpack",
580
+ "model/decoder/layers/40/self_attn/q_proj/bias": "flax_model-00005-of-00007.msgpack",
581
+ "model/decoder/layers/40/self_attn/q_proj/kernel": "flax_model-00005-of-00007.msgpack",
582
+ "model/decoder/layers/40/self_attn/v_proj/bias": "flax_model-00005-of-00007.msgpack",
583
+ "model/decoder/layers/40/self_attn/v_proj/kernel": "flax_model-00005-of-00007.msgpack",
584
+ "model/decoder/layers/40/self_attn_layer_norm/bias": "flax_model-00005-of-00007.msgpack",
585
+ "model/decoder/layers/40/self_attn_layer_norm/scale": "flax_model-00005-of-00007.msgpack",
586
+ "model/decoder/layers/41/fc1/bias": "flax_model-00005-of-00007.msgpack",
587
+ "model/decoder/layers/41/fc1/kernel": "flax_model-00005-of-00007.msgpack",
588
+ "model/decoder/layers/41/fc2/bias": "flax_model-00005-of-00007.msgpack",
589
+ "model/decoder/layers/41/fc2/kernel": "flax_model-00005-of-00007.msgpack",
590
+ "model/decoder/layers/41/final_layer_norm/bias": "flax_model-00005-of-00007.msgpack",
591
+ "model/decoder/layers/41/final_layer_norm/scale": "flax_model-00005-of-00007.msgpack",
592
+ "model/decoder/layers/41/self_attn/k_proj/bias": "flax_model-00005-of-00007.msgpack",
593
+ "model/decoder/layers/41/self_attn/k_proj/kernel": "flax_model-00005-of-00007.msgpack",
594
+ "model/decoder/layers/41/self_attn/out_proj/bias": "flax_model-00005-of-00007.msgpack",
595
+ "model/decoder/layers/41/self_attn/out_proj/kernel": "flax_model-00005-of-00007.msgpack",
596
+ "model/decoder/layers/41/self_attn/q_proj/bias": "flax_model-00005-of-00007.msgpack",
597
+ "model/decoder/layers/41/self_attn/q_proj/kernel": "flax_model-00005-of-00007.msgpack",
598
+ "model/decoder/layers/41/self_attn/v_proj/bias": "flax_model-00005-of-00007.msgpack",
599
+ "model/decoder/layers/41/self_attn/v_proj/kernel": "flax_model-00005-of-00007.msgpack",
600
+ "model/decoder/layers/41/self_attn_layer_norm/bias": "flax_model-00005-of-00007.msgpack",
601
+ "model/decoder/layers/41/self_attn_layer_norm/scale": "flax_model-00005-of-00007.msgpack",
602
+ "model/decoder/layers/42/fc1/bias": "flax_model-00005-of-00007.msgpack",
603
+ "model/decoder/layers/42/fc1/kernel": "flax_model-00005-of-00007.msgpack",
604
+ "model/decoder/layers/42/fc2/bias": "flax_model-00005-of-00007.msgpack",
605
+ "model/decoder/layers/42/fc2/kernel": "flax_model-00005-of-00007.msgpack",
606
+ "model/decoder/layers/42/final_layer_norm/bias": "flax_model-00005-of-00007.msgpack",
607
+ "model/decoder/layers/42/final_layer_norm/scale": "flax_model-00005-of-00007.msgpack",
608
+ "model/decoder/layers/42/self_attn/k_proj/bias": "flax_model-00005-of-00007.msgpack",
609
+ "model/decoder/layers/42/self_attn/k_proj/kernel": "flax_model-00005-of-00007.msgpack",
610
+ "model/decoder/layers/42/self_attn/out_proj/bias": "flax_model-00005-of-00007.msgpack",
611
+ "model/decoder/layers/42/self_attn/out_proj/kernel": "flax_model-00005-of-00007.msgpack",
612
+ "model/decoder/layers/42/self_attn/q_proj/bias": "flax_model-00005-of-00007.msgpack",
613
+ "model/decoder/layers/42/self_attn/q_proj/kernel": "flax_model-00005-of-00007.msgpack",
614
+ "model/decoder/layers/42/self_attn/v_proj/bias": "flax_model-00005-of-00007.msgpack",
615
+ "model/decoder/layers/42/self_attn/v_proj/kernel": "flax_model-00005-of-00007.msgpack",
616
+ "model/decoder/layers/42/self_attn_layer_norm/bias": "flax_model-00005-of-00007.msgpack",
617
+ "model/decoder/layers/42/self_attn_layer_norm/scale": "flax_model-00005-of-00007.msgpack",
618
+ "model/decoder/layers/43/fc1/bias": "flax_model-00005-of-00007.msgpack",
619
+ "model/decoder/layers/43/fc1/kernel": "flax_model-00005-of-00007.msgpack",
620
+ "model/decoder/layers/43/fc2/bias": "flax_model-00005-of-00007.msgpack",
621
+ "model/decoder/layers/43/fc2/kernel": "flax_model-00005-of-00007.msgpack",
622
+ "model/decoder/layers/43/final_layer_norm/bias": "flax_model-00005-of-00007.msgpack",
623
+ "model/decoder/layers/43/final_layer_norm/scale": "flax_model-00005-of-00007.msgpack",
624
+ "model/decoder/layers/43/self_attn/k_proj/bias": "flax_model-00005-of-00007.msgpack",
625
+ "model/decoder/layers/43/self_attn/k_proj/kernel": "flax_model-00005-of-00007.msgpack",
626
+ "model/decoder/layers/43/self_attn/out_proj/bias": "flax_model-00005-of-00007.msgpack",
627
+ "model/decoder/layers/43/self_attn/out_proj/kernel": "flax_model-00005-of-00007.msgpack",
628
+ "model/decoder/layers/43/self_attn/q_proj/bias": "flax_model-00005-of-00007.msgpack",
629
+ "model/decoder/layers/43/self_attn/q_proj/kernel": "flax_model-00005-of-00007.msgpack",
630
+ "model/decoder/layers/43/self_attn/v_proj/bias": "flax_model-00005-of-00007.msgpack",
631
+ "model/decoder/layers/43/self_attn/v_proj/kernel": "flax_model-00005-of-00007.msgpack",
632
+ "model/decoder/layers/43/self_attn_layer_norm/bias": "flax_model-00005-of-00007.msgpack",
633
+ "model/decoder/layers/43/self_attn_layer_norm/scale": "flax_model-00005-of-00007.msgpack",
634
+ "model/decoder/layers/44/fc1/bias": "flax_model-00005-of-00007.msgpack",
635
+ "model/decoder/layers/44/fc1/kernel": "flax_model-00005-of-00007.msgpack",
636
+ "model/decoder/layers/44/fc2/bias": "flax_model-00005-of-00007.msgpack",
637
+ "model/decoder/layers/44/fc2/kernel": "flax_model-00006-of-00007.msgpack",
638
+ "model/decoder/layers/44/final_layer_norm/bias": "flax_model-00006-of-00007.msgpack",
639
+ "model/decoder/layers/44/final_layer_norm/scale": "flax_model-00006-of-00007.msgpack",
640
+ "model/decoder/layers/44/self_attn/k_proj/bias": "flax_model-00006-of-00007.msgpack",
641
+ "model/decoder/layers/44/self_attn/k_proj/kernel": "flax_model-00006-of-00007.msgpack",
642
+ "model/decoder/layers/44/self_attn/out_proj/bias": "flax_model-00006-of-00007.msgpack",
643
+ "model/decoder/layers/44/self_attn/out_proj/kernel": "flax_model-00006-of-00007.msgpack",
644
+ "model/decoder/layers/44/self_attn/q_proj/bias": "flax_model-00006-of-00007.msgpack",
645
+ "model/decoder/layers/44/self_attn/q_proj/kernel": "flax_model-00006-of-00007.msgpack",
646
+ "model/decoder/layers/44/self_attn/v_proj/bias": "flax_model-00006-of-00007.msgpack",
647
+ "model/decoder/layers/44/self_attn/v_proj/kernel": "flax_model-00006-of-00007.msgpack",
648
+ "model/decoder/layers/44/self_attn_layer_norm/bias": "flax_model-00006-of-00007.msgpack",
649
+ "model/decoder/layers/44/self_attn_layer_norm/scale": "flax_model-00006-of-00007.msgpack",
650
+ "model/decoder/layers/45/fc1/bias": "flax_model-00006-of-00007.msgpack",
651
+ "model/decoder/layers/45/fc1/kernel": "flax_model-00006-of-00007.msgpack",
652
+ "model/decoder/layers/45/fc2/bias": "flax_model-00006-of-00007.msgpack",
653
+ "model/decoder/layers/45/fc2/kernel": "flax_model-00006-of-00007.msgpack",
654
+ "model/decoder/layers/45/final_layer_norm/bias": "flax_model-00006-of-00007.msgpack",
655
+ "model/decoder/layers/45/final_layer_norm/scale": "flax_model-00006-of-00007.msgpack",
656
+ "model/decoder/layers/45/self_attn/k_proj/bias": "flax_model-00006-of-00007.msgpack",
657
+ "model/decoder/layers/45/self_attn/k_proj/kernel": "flax_model-00006-of-00007.msgpack",
658
+ "model/decoder/layers/45/self_attn/out_proj/bias": "flax_model-00006-of-00007.msgpack",
659
+ "model/decoder/layers/45/self_attn/out_proj/kernel": "flax_model-00006-of-00007.msgpack",
660
+ "model/decoder/layers/45/self_attn/q_proj/bias": "flax_model-00006-of-00007.msgpack",
661
+ "model/decoder/layers/45/self_attn/q_proj/kernel": "flax_model-00006-of-00007.msgpack",
662
+ "model/decoder/layers/45/self_attn/v_proj/bias": "flax_model-00006-of-00007.msgpack",
663
+ "model/decoder/layers/45/self_attn/v_proj/kernel": "flax_model-00006-of-00007.msgpack",
664
+ "model/decoder/layers/45/self_attn_layer_norm/bias": "flax_model-00006-of-00007.msgpack",
665
+ "model/decoder/layers/45/self_attn_layer_norm/scale": "flax_model-00006-of-00007.msgpack",
666
+ "model/decoder/layers/46/fc1/bias": "flax_model-00006-of-00007.msgpack",
667
+ "model/decoder/layers/46/fc1/kernel": "flax_model-00006-of-00007.msgpack",
668
+ "model/decoder/layers/46/fc2/bias": "flax_model-00006-of-00007.msgpack",
669
+ "model/decoder/layers/46/fc2/kernel": "flax_model-00006-of-00007.msgpack",
670
+ "model/decoder/layers/46/final_layer_norm/bias": "flax_model-00006-of-00007.msgpack",
671
+ "model/decoder/layers/46/final_layer_norm/scale": "flax_model-00006-of-00007.msgpack",
672
+ "model/decoder/layers/46/self_attn/k_proj/bias": "flax_model-00006-of-00007.msgpack",
673
+ "model/decoder/layers/46/self_attn/k_proj/kernel": "flax_model-00006-of-00007.msgpack",
674
+ "model/decoder/layers/46/self_attn/out_proj/bias": "flax_model-00006-of-00007.msgpack",
675
+ "model/decoder/layers/46/self_attn/out_proj/kernel": "flax_model-00006-of-00007.msgpack",
676
+ "model/decoder/layers/46/self_attn/q_proj/bias": "flax_model-00006-of-00007.msgpack",
677
+ "model/decoder/layers/46/self_attn/q_proj/kernel": "flax_model-00006-of-00007.msgpack",
678
+ "model/decoder/layers/46/self_attn/v_proj/bias": "flax_model-00006-of-00007.msgpack",
679
+ "model/decoder/layers/46/self_attn/v_proj/kernel": "flax_model-00006-of-00007.msgpack",
680
+ "model/decoder/layers/46/self_attn_layer_norm/bias": "flax_model-00006-of-00007.msgpack",
681
+ "model/decoder/layers/46/self_attn_layer_norm/scale": "flax_model-00006-of-00007.msgpack",
682
+ "model/decoder/layers/47/fc1/bias": "flax_model-00006-of-00007.msgpack",
683
+ "model/decoder/layers/47/fc1/kernel": "flax_model-00006-of-00007.msgpack",
684
+ "model/decoder/layers/47/fc2/bias": "flax_model-00006-of-00007.msgpack",
685
+ "model/decoder/layers/47/fc2/kernel": "flax_model-00006-of-00007.msgpack",
686
+ "model/decoder/layers/47/final_layer_norm/bias": "flax_model-00006-of-00007.msgpack",
687
+ "model/decoder/layers/47/final_layer_norm/scale": "flax_model-00006-of-00007.msgpack",
688
+ "model/decoder/layers/47/self_attn/k_proj/bias": "flax_model-00006-of-00007.msgpack",
689
+ "model/decoder/layers/47/self_attn/k_proj/kernel": "flax_model-00006-of-00007.msgpack",
690
+ "model/decoder/layers/47/self_attn/out_proj/bias": "flax_model-00006-of-00007.msgpack",
691
+ "model/decoder/layers/47/self_attn/out_proj/kernel": "flax_model-00006-of-00007.msgpack",
692
+ "model/decoder/layers/47/self_attn/q_proj/bias": "flax_model-00006-of-00007.msgpack",
693
+ "model/decoder/layers/47/self_attn/q_proj/kernel": "flax_model-00006-of-00007.msgpack",
694
+ "model/decoder/layers/47/self_attn/v_proj/bias": "flax_model-00006-of-00007.msgpack",
695
+ "model/decoder/layers/47/self_attn/v_proj/kernel": "flax_model-00006-of-00007.msgpack",
696
+ "model/decoder/layers/47/self_attn_layer_norm/bias": "flax_model-00006-of-00007.msgpack",
697
+ "model/decoder/layers/47/self_attn_layer_norm/scale": "flax_model-00006-of-00007.msgpack",
698
+ "model/decoder/layers/5/fc1/bias": "flax_model-00006-of-00007.msgpack",
699
+ "model/decoder/layers/5/fc1/kernel": "flax_model-00006-of-00007.msgpack",
700
+ "model/decoder/layers/5/fc2/bias": "flax_model-00006-of-00007.msgpack",
701
+ "model/decoder/layers/5/fc2/kernel": "flax_model-00006-of-00007.msgpack",
702
+ "model/decoder/layers/5/final_layer_norm/bias": "flax_model-00006-of-00007.msgpack",
703
+ "model/decoder/layers/5/final_layer_norm/scale": "flax_model-00006-of-00007.msgpack",
704
+ "model/decoder/layers/5/self_attn/k_proj/bias": "flax_model-00006-of-00007.msgpack",
705
+ "model/decoder/layers/5/self_attn/k_proj/kernel": "flax_model-00006-of-00007.msgpack",
706
+ "model/decoder/layers/5/self_attn/out_proj/bias": "flax_model-00006-of-00007.msgpack",
707
+ "model/decoder/layers/5/self_attn/out_proj/kernel": "flax_model-00006-of-00007.msgpack",
708
+ "model/decoder/layers/5/self_attn/q_proj/bias": "flax_model-00006-of-00007.msgpack",
709
+ "model/decoder/layers/5/self_attn/q_proj/kernel": "flax_model-00006-of-00007.msgpack",
710
+ "model/decoder/layers/5/self_attn/v_proj/bias": "flax_model-00006-of-00007.msgpack",
711
+ "model/decoder/layers/5/self_attn/v_proj/kernel": "flax_model-00006-of-00007.msgpack",
712
+ "model/decoder/layers/5/self_attn_layer_norm/bias": "flax_model-00006-of-00007.msgpack",
713
+ "model/decoder/layers/5/self_attn_layer_norm/scale": "flax_model-00006-of-00007.msgpack",
714
+ "model/decoder/layers/6/fc1/bias": "flax_model-00006-of-00007.msgpack",
715
+ "model/decoder/layers/6/fc1/kernel": "flax_model-00006-of-00007.msgpack",
716
+ "model/decoder/layers/6/fc2/bias": "flax_model-00006-of-00007.msgpack",
717
+ "model/decoder/layers/6/fc2/kernel": "flax_model-00006-of-00007.msgpack",
718
+ "model/decoder/layers/6/final_layer_norm/bias": "flax_model-00006-of-00007.msgpack",
719
+ "model/decoder/layers/6/final_layer_norm/scale": "flax_model-00006-of-00007.msgpack",
720
+ "model/decoder/layers/6/self_attn/k_proj/bias": "flax_model-00006-of-00007.msgpack",
721
+ "model/decoder/layers/6/self_attn/k_proj/kernel": "flax_model-00006-of-00007.msgpack",
722
+ "model/decoder/layers/6/self_attn/out_proj/bias": "flax_model-00006-of-00007.msgpack",
723
+ "model/decoder/layers/6/self_attn/out_proj/kernel": "flax_model-00006-of-00007.msgpack",
724
+ "model/decoder/layers/6/self_attn/q_proj/bias": "flax_model-00006-of-00007.msgpack",
725
+ "model/decoder/layers/6/self_attn/q_proj/kernel": "flax_model-00006-of-00007.msgpack",
726
+ "model/decoder/layers/6/self_attn/v_proj/bias": "flax_model-00006-of-00007.msgpack",
727
+ "model/decoder/layers/6/self_attn/v_proj/kernel": "flax_model-00006-of-00007.msgpack",
728
+ "model/decoder/layers/6/self_attn_layer_norm/bias": "flax_model-00006-of-00007.msgpack",
729
+ "model/decoder/layers/6/self_attn_layer_norm/scale": "flax_model-00006-of-00007.msgpack",
730
+ "model/decoder/layers/7/fc1/bias": "flax_model-00006-of-00007.msgpack",
731
+ "model/decoder/layers/7/fc1/kernel": "flax_model-00006-of-00007.msgpack",
732
+ "model/decoder/layers/7/fc2/bias": "flax_model-00006-of-00007.msgpack",
733
+ "model/decoder/layers/7/fc2/kernel": "flax_model-00006-of-00007.msgpack",
734
+ "model/decoder/layers/7/final_layer_norm/bias": "flax_model-00006-of-00007.msgpack",
735
+ "model/decoder/layers/7/final_layer_norm/scale": "flax_model-00006-of-00007.msgpack",
736
+ "model/decoder/layers/7/self_attn/k_proj/bias": "flax_model-00006-of-00007.msgpack",
737
+ "model/decoder/layers/7/self_attn/k_proj/kernel": "flax_model-00006-of-00007.msgpack",
738
+ "model/decoder/layers/7/self_attn/out_proj/bias": "flax_model-00006-of-00007.msgpack",
739
+ "model/decoder/layers/7/self_attn/out_proj/kernel": "flax_model-00006-of-00007.msgpack",
740
+ "model/decoder/layers/7/self_attn/q_proj/bias": "flax_model-00006-of-00007.msgpack",
741
+ "model/decoder/layers/7/self_attn/q_proj/kernel": "flax_model-00006-of-00007.msgpack",
742
+ "model/decoder/layers/7/self_attn/v_proj/bias": "flax_model-00006-of-00007.msgpack",
743
+ "model/decoder/layers/7/self_attn/v_proj/kernel": "flax_model-00006-of-00007.msgpack",
744
+ "model/decoder/layers/7/self_attn_layer_norm/bias": "flax_model-00006-of-00007.msgpack",
745
+ "model/decoder/layers/7/self_attn_layer_norm/scale": "flax_model-00006-of-00007.msgpack",
746
+ "model/decoder/layers/8/fc1/bias": "flax_model-00006-of-00007.msgpack",
747
+ "model/decoder/layers/8/fc1/kernel": "flax_model-00006-of-00007.msgpack",
748
+ "model/decoder/layers/8/fc2/bias": "flax_model-00006-of-00007.msgpack",
749
+ "model/decoder/layers/8/fc2/kernel": "flax_model-00006-of-00007.msgpack",
750
+ "model/decoder/layers/8/final_layer_norm/bias": "flax_model-00006-of-00007.msgpack",
751
+ "model/decoder/layers/8/final_layer_norm/scale": "flax_model-00006-of-00007.msgpack",
752
+ "model/decoder/layers/8/self_attn/k_proj/bias": "flax_model-00006-of-00007.msgpack",
753
+ "model/decoder/layers/8/self_attn/k_proj/kernel": "flax_model-00006-of-00007.msgpack",
754
+ "model/decoder/layers/8/self_attn/out_proj/bias": "flax_model-00006-of-00007.msgpack",
755
+ "model/decoder/layers/8/self_attn/out_proj/kernel": "flax_model-00006-of-00007.msgpack",
756
+ "model/decoder/layers/8/self_attn/q_proj/bias": "flax_model-00006-of-00007.msgpack",
757
+ "model/decoder/layers/8/self_attn/q_proj/kernel": "flax_model-00006-of-00007.msgpack",
758
+ "model/decoder/layers/8/self_attn/v_proj/bias": "flax_model-00006-of-00007.msgpack",
759
+ "model/decoder/layers/8/self_attn/v_proj/kernel": "flax_model-00006-of-00007.msgpack",
760
+ "model/decoder/layers/8/self_attn_layer_norm/bias": "flax_model-00006-of-00007.msgpack",
761
+ "model/decoder/layers/8/self_attn_layer_norm/scale": "flax_model-00006-of-00007.msgpack",
762
+ "model/decoder/layers/9/fc1/bias": "flax_model-00006-of-00007.msgpack",
763
+ "model/decoder/layers/9/fc1/kernel": "flax_model-00006-of-00007.msgpack",
764
+ "model/decoder/layers/9/fc2/bias": "flax_model-00006-of-00007.msgpack",
765
+ "model/decoder/layers/9/fc2/kernel": "flax_model-00007-of-00007.msgpack",
766
+ "model/decoder/layers/9/final_layer_norm/bias": "flax_model-00007-of-00007.msgpack",
767
+ "model/decoder/layers/9/final_layer_norm/scale": "flax_model-00007-of-00007.msgpack",
768
+ "model/decoder/layers/9/self_attn/k_proj/bias": "flax_model-00007-of-00007.msgpack",
769
+ "model/decoder/layers/9/self_attn/k_proj/kernel": "flax_model-00007-of-00007.msgpack",
770
+ "model/decoder/layers/9/self_attn/out_proj/bias": "flax_model-00007-of-00007.msgpack",
771
+ "model/decoder/layers/9/self_attn/out_proj/kernel": "flax_model-00007-of-00007.msgpack",
772
+ "model/decoder/layers/9/self_attn/q_proj/bias": "flax_model-00007-of-00007.msgpack",
773
+ "model/decoder/layers/9/self_attn/q_proj/kernel": "flax_model-00007-of-00007.msgpack",
774
+ "model/decoder/layers/9/self_attn/v_proj/bias": "flax_model-00007-of-00007.msgpack",
775
+ "model/decoder/layers/9/self_attn/v_proj/kernel": "flax_model-00007-of-00007.msgpack",
776
+ "model/decoder/layers/9/self_attn_layer_norm/bias": "flax_model-00007-of-00007.msgpack",
777
+ "model/decoder/layers/9/self_attn_layer_norm/scale": "flax_model-00007-of-00007.msgpack"
778
+ }
779
+ }