jacobthebanana commited on
Commit
f8c4a0e
1 Parent(s): 7b3bcf1

Added JAX weights converted from facebook/galactica-6.7b.

Browse files
config.json ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "/content/standard",
3
+ "_remove_final_layer_norm": false,
4
+ "activation_dropout": 0.0,
5
+ "activation_function": "gelu",
6
+ "architectures": [
7
+ "OPTForCausalLM"
8
+ ],
9
+ "attention_dropout": 0.1,
10
+ "bos_token_id": 0,
11
+ "do_layer_norm_before": true,
12
+ "dropout": 0.1,
13
+ "enable_bias": true,
14
+ "eos_token_id": 2,
15
+ "ffn_dim": 16384,
16
+ "hidden_size": 4096,
17
+ "init_std": 0.02,
18
+ "layer_norm_elementwise_affine": true,
19
+ "layerdrop": 0.0,
20
+ "learned_embeddings": true,
21
+ "max_position_embeddings": 2048,
22
+ "model_type": "opt",
23
+ "num_attention_heads": 32,
24
+ "num_hidden_layers": 32,
25
+ "pad_token_id": 1,
26
+ "scale_embeddings": false,
27
+ "torch_dtype": "float16",
28
+ "transformers_version": "4.25.1",
29
+ "use_cache": true,
30
+ "vocab_size": 50000,
31
+ "word_embed_proj_dim": 4096
32
+ }
flax_model-00001-of-00002.msgpack ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:59bcf3698205c18ca0dcf95ef2b579fe1419465ca833583a2f9dd55f110c7419
3
+ size 9958413378
flax_model-00002-of-00002.msgpack ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:20efbc348c13b887950704ad04fa8a3aec05a571421d446e4f937091c172b50f
3
+ size 3356324346
flax_model.msgpack.index.json ADDED
@@ -0,0 +1,523 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 13314719744
4
+ },
5
+ "weight_map": {
6
+ "model/decoder/embed_positions/embedding": "flax_model-00001-of-00002.msgpack",
7
+ "model/decoder/embed_tokens/embedding": "flax_model-00001-of-00002.msgpack",
8
+ "model/decoder/final_layer_norm/bias": "flax_model-00001-of-00002.msgpack",
9
+ "model/decoder/final_layer_norm/scale": "flax_model-00001-of-00002.msgpack",
10
+ "model/decoder/layers/0/fc1/bias": "flax_model-00001-of-00002.msgpack",
11
+ "model/decoder/layers/0/fc1/kernel": "flax_model-00001-of-00002.msgpack",
12
+ "model/decoder/layers/0/fc2/bias": "flax_model-00001-of-00002.msgpack",
13
+ "model/decoder/layers/0/fc2/kernel": "flax_model-00001-of-00002.msgpack",
14
+ "model/decoder/layers/0/final_layer_norm/bias": "flax_model-00001-of-00002.msgpack",
15
+ "model/decoder/layers/0/final_layer_norm/scale": "flax_model-00001-of-00002.msgpack",
16
+ "model/decoder/layers/0/self_attn/k_proj/bias": "flax_model-00001-of-00002.msgpack",
17
+ "model/decoder/layers/0/self_attn/k_proj/kernel": "flax_model-00001-of-00002.msgpack",
18
+ "model/decoder/layers/0/self_attn/out_proj/bias": "flax_model-00001-of-00002.msgpack",
19
+ "model/decoder/layers/0/self_attn/out_proj/kernel": "flax_model-00001-of-00002.msgpack",
20
+ "model/decoder/layers/0/self_attn/q_proj/bias": "flax_model-00001-of-00002.msgpack",
21
+ "model/decoder/layers/0/self_attn/q_proj/kernel": "flax_model-00001-of-00002.msgpack",
22
+ "model/decoder/layers/0/self_attn/v_proj/bias": "flax_model-00001-of-00002.msgpack",
23
+ "model/decoder/layers/0/self_attn/v_proj/kernel": "flax_model-00001-of-00002.msgpack",
24
+ "model/decoder/layers/0/self_attn_layer_norm/bias": "flax_model-00001-of-00002.msgpack",
25
+ "model/decoder/layers/0/self_attn_layer_norm/scale": "flax_model-00001-of-00002.msgpack",
26
+ "model/decoder/layers/1/fc1/bias": "flax_model-00001-of-00002.msgpack",
27
+ "model/decoder/layers/1/fc1/kernel": "flax_model-00001-of-00002.msgpack",
28
+ "model/decoder/layers/1/fc2/bias": "flax_model-00001-of-00002.msgpack",
29
+ "model/decoder/layers/1/fc2/kernel": "flax_model-00001-of-00002.msgpack",
30
+ "model/decoder/layers/1/final_layer_norm/bias": "flax_model-00001-of-00002.msgpack",
31
+ "model/decoder/layers/1/final_layer_norm/scale": "flax_model-00001-of-00002.msgpack",
32
+ "model/decoder/layers/1/self_attn/k_proj/bias": "flax_model-00001-of-00002.msgpack",
33
+ "model/decoder/layers/1/self_attn/k_proj/kernel": "flax_model-00001-of-00002.msgpack",
34
+ "model/decoder/layers/1/self_attn/out_proj/bias": "flax_model-00001-of-00002.msgpack",
35
+ "model/decoder/layers/1/self_attn/out_proj/kernel": "flax_model-00001-of-00002.msgpack",
36
+ "model/decoder/layers/1/self_attn/q_proj/bias": "flax_model-00001-of-00002.msgpack",
37
+ "model/decoder/layers/1/self_attn/q_proj/kernel": "flax_model-00001-of-00002.msgpack",
38
+ "model/decoder/layers/1/self_attn/v_proj/bias": "flax_model-00001-of-00002.msgpack",
39
+ "model/decoder/layers/1/self_attn/v_proj/kernel": "flax_model-00001-of-00002.msgpack",
40
+ "model/decoder/layers/1/self_attn_layer_norm/bias": "flax_model-00001-of-00002.msgpack",
41
+ "model/decoder/layers/1/self_attn_layer_norm/scale": "flax_model-00001-of-00002.msgpack",
42
+ "model/decoder/layers/10/fc1/bias": "flax_model-00001-of-00002.msgpack",
43
+ "model/decoder/layers/10/fc1/kernel": "flax_model-00001-of-00002.msgpack",
44
+ "model/decoder/layers/10/fc2/bias": "flax_model-00001-of-00002.msgpack",
45
+ "model/decoder/layers/10/fc2/kernel": "flax_model-00001-of-00002.msgpack",
46
+ "model/decoder/layers/10/final_layer_norm/bias": "flax_model-00001-of-00002.msgpack",
47
+ "model/decoder/layers/10/final_layer_norm/scale": "flax_model-00001-of-00002.msgpack",
48
+ "model/decoder/layers/10/self_attn/k_proj/bias": "flax_model-00001-of-00002.msgpack",
49
+ "model/decoder/layers/10/self_attn/k_proj/kernel": "flax_model-00001-of-00002.msgpack",
50
+ "model/decoder/layers/10/self_attn/out_proj/bias": "flax_model-00001-of-00002.msgpack",
51
+ "model/decoder/layers/10/self_attn/out_proj/kernel": "flax_model-00001-of-00002.msgpack",
52
+ "model/decoder/layers/10/self_attn/q_proj/bias": "flax_model-00001-of-00002.msgpack",
53
+ "model/decoder/layers/10/self_attn/q_proj/kernel": "flax_model-00001-of-00002.msgpack",
54
+ "model/decoder/layers/10/self_attn/v_proj/bias": "flax_model-00001-of-00002.msgpack",
55
+ "model/decoder/layers/10/self_attn/v_proj/kernel": "flax_model-00001-of-00002.msgpack",
56
+ "model/decoder/layers/10/self_attn_layer_norm/bias": "flax_model-00001-of-00002.msgpack",
57
+ "model/decoder/layers/10/self_attn_layer_norm/scale": "flax_model-00001-of-00002.msgpack",
58
+ "model/decoder/layers/11/fc1/bias": "flax_model-00001-of-00002.msgpack",
59
+ "model/decoder/layers/11/fc1/kernel": "flax_model-00001-of-00002.msgpack",
60
+ "model/decoder/layers/11/fc2/bias": "flax_model-00001-of-00002.msgpack",
61
+ "model/decoder/layers/11/fc2/kernel": "flax_model-00001-of-00002.msgpack",
62
+ "model/decoder/layers/11/final_layer_norm/bias": "flax_model-00001-of-00002.msgpack",
63
+ "model/decoder/layers/11/final_layer_norm/scale": "flax_model-00001-of-00002.msgpack",
64
+ "model/decoder/layers/11/self_attn/k_proj/bias": "flax_model-00001-of-00002.msgpack",
65
+ "model/decoder/layers/11/self_attn/k_proj/kernel": "flax_model-00001-of-00002.msgpack",
66
+ "model/decoder/layers/11/self_attn/out_proj/bias": "flax_model-00001-of-00002.msgpack",
67
+ "model/decoder/layers/11/self_attn/out_proj/kernel": "flax_model-00001-of-00002.msgpack",
68
+ "model/decoder/layers/11/self_attn/q_proj/bias": "flax_model-00001-of-00002.msgpack",
69
+ "model/decoder/layers/11/self_attn/q_proj/kernel": "flax_model-00001-of-00002.msgpack",
70
+ "model/decoder/layers/11/self_attn/v_proj/bias": "flax_model-00001-of-00002.msgpack",
71
+ "model/decoder/layers/11/self_attn/v_proj/kernel": "flax_model-00001-of-00002.msgpack",
72
+ "model/decoder/layers/11/self_attn_layer_norm/bias": "flax_model-00001-of-00002.msgpack",
73
+ "model/decoder/layers/11/self_attn_layer_norm/scale": "flax_model-00001-of-00002.msgpack",
74
+ "model/decoder/layers/12/fc1/bias": "flax_model-00001-of-00002.msgpack",
75
+ "model/decoder/layers/12/fc1/kernel": "flax_model-00001-of-00002.msgpack",
76
+ "model/decoder/layers/12/fc2/bias": "flax_model-00001-of-00002.msgpack",
77
+ "model/decoder/layers/12/fc2/kernel": "flax_model-00001-of-00002.msgpack",
78
+ "model/decoder/layers/12/final_layer_norm/bias": "flax_model-00001-of-00002.msgpack",
79
+ "model/decoder/layers/12/final_layer_norm/scale": "flax_model-00001-of-00002.msgpack",
80
+ "model/decoder/layers/12/self_attn/k_proj/bias": "flax_model-00001-of-00002.msgpack",
81
+ "model/decoder/layers/12/self_attn/k_proj/kernel": "flax_model-00001-of-00002.msgpack",
82
+ "model/decoder/layers/12/self_attn/out_proj/bias": "flax_model-00001-of-00002.msgpack",
83
+ "model/decoder/layers/12/self_attn/out_proj/kernel": "flax_model-00001-of-00002.msgpack",
84
+ "model/decoder/layers/12/self_attn/q_proj/bias": "flax_model-00001-of-00002.msgpack",
85
+ "model/decoder/layers/12/self_attn/q_proj/kernel": "flax_model-00001-of-00002.msgpack",
86
+ "model/decoder/layers/12/self_attn/v_proj/bias": "flax_model-00001-of-00002.msgpack",
87
+ "model/decoder/layers/12/self_attn/v_proj/kernel": "flax_model-00001-of-00002.msgpack",
88
+ "model/decoder/layers/12/self_attn_layer_norm/bias": "flax_model-00001-of-00002.msgpack",
89
+ "model/decoder/layers/12/self_attn_layer_norm/scale": "flax_model-00001-of-00002.msgpack",
90
+ "model/decoder/layers/13/fc1/bias": "flax_model-00001-of-00002.msgpack",
91
+ "model/decoder/layers/13/fc1/kernel": "flax_model-00001-of-00002.msgpack",
92
+ "model/decoder/layers/13/fc2/bias": "flax_model-00001-of-00002.msgpack",
93
+ "model/decoder/layers/13/fc2/kernel": "flax_model-00001-of-00002.msgpack",
94
+ "model/decoder/layers/13/final_layer_norm/bias": "flax_model-00001-of-00002.msgpack",
95
+ "model/decoder/layers/13/final_layer_norm/scale": "flax_model-00001-of-00002.msgpack",
96
+ "model/decoder/layers/13/self_attn/k_proj/bias": "flax_model-00001-of-00002.msgpack",
97
+ "model/decoder/layers/13/self_attn/k_proj/kernel": "flax_model-00001-of-00002.msgpack",
98
+ "model/decoder/layers/13/self_attn/out_proj/bias": "flax_model-00001-of-00002.msgpack",
99
+ "model/decoder/layers/13/self_attn/out_proj/kernel": "flax_model-00001-of-00002.msgpack",
100
+ "model/decoder/layers/13/self_attn/q_proj/bias": "flax_model-00001-of-00002.msgpack",
101
+ "model/decoder/layers/13/self_attn/q_proj/kernel": "flax_model-00001-of-00002.msgpack",
102
+ "model/decoder/layers/13/self_attn/v_proj/bias": "flax_model-00001-of-00002.msgpack",
103
+ "model/decoder/layers/13/self_attn/v_proj/kernel": "flax_model-00001-of-00002.msgpack",
104
+ "model/decoder/layers/13/self_attn_layer_norm/bias": "flax_model-00001-of-00002.msgpack",
105
+ "model/decoder/layers/13/self_attn_layer_norm/scale": "flax_model-00001-of-00002.msgpack",
106
+ "model/decoder/layers/14/fc1/bias": "flax_model-00001-of-00002.msgpack",
107
+ "model/decoder/layers/14/fc1/kernel": "flax_model-00001-of-00002.msgpack",
108
+ "model/decoder/layers/14/fc2/bias": "flax_model-00001-of-00002.msgpack",
109
+ "model/decoder/layers/14/fc2/kernel": "flax_model-00001-of-00002.msgpack",
110
+ "model/decoder/layers/14/final_layer_norm/bias": "flax_model-00001-of-00002.msgpack",
111
+ "model/decoder/layers/14/final_layer_norm/scale": "flax_model-00001-of-00002.msgpack",
112
+ "model/decoder/layers/14/self_attn/k_proj/bias": "flax_model-00001-of-00002.msgpack",
113
+ "model/decoder/layers/14/self_attn/k_proj/kernel": "flax_model-00001-of-00002.msgpack",
114
+ "model/decoder/layers/14/self_attn/out_proj/bias": "flax_model-00001-of-00002.msgpack",
115
+ "model/decoder/layers/14/self_attn/out_proj/kernel": "flax_model-00001-of-00002.msgpack",
116
+ "model/decoder/layers/14/self_attn/q_proj/bias": "flax_model-00001-of-00002.msgpack",
117
+ "model/decoder/layers/14/self_attn/q_proj/kernel": "flax_model-00001-of-00002.msgpack",
118
+ "model/decoder/layers/14/self_attn/v_proj/bias": "flax_model-00001-of-00002.msgpack",
119
+ "model/decoder/layers/14/self_attn/v_proj/kernel": "flax_model-00001-of-00002.msgpack",
120
+ "model/decoder/layers/14/self_attn_layer_norm/bias": "flax_model-00001-of-00002.msgpack",
121
+ "model/decoder/layers/14/self_attn_layer_norm/scale": "flax_model-00001-of-00002.msgpack",
122
+ "model/decoder/layers/15/fc1/bias": "flax_model-00001-of-00002.msgpack",
123
+ "model/decoder/layers/15/fc1/kernel": "flax_model-00001-of-00002.msgpack",
124
+ "model/decoder/layers/15/fc2/bias": "flax_model-00001-of-00002.msgpack",
125
+ "model/decoder/layers/15/fc2/kernel": "flax_model-00001-of-00002.msgpack",
126
+ "model/decoder/layers/15/final_layer_norm/bias": "flax_model-00001-of-00002.msgpack",
127
+ "model/decoder/layers/15/final_layer_norm/scale": "flax_model-00001-of-00002.msgpack",
128
+ "model/decoder/layers/15/self_attn/k_proj/bias": "flax_model-00001-of-00002.msgpack",
129
+ "model/decoder/layers/15/self_attn/k_proj/kernel": "flax_model-00001-of-00002.msgpack",
130
+ "model/decoder/layers/15/self_attn/out_proj/bias": "flax_model-00001-of-00002.msgpack",
131
+ "model/decoder/layers/15/self_attn/out_proj/kernel": "flax_model-00001-of-00002.msgpack",
132
+ "model/decoder/layers/15/self_attn/q_proj/bias": "flax_model-00001-of-00002.msgpack",
133
+ "model/decoder/layers/15/self_attn/q_proj/kernel": "flax_model-00001-of-00002.msgpack",
134
+ "model/decoder/layers/15/self_attn/v_proj/bias": "flax_model-00001-of-00002.msgpack",
135
+ "model/decoder/layers/15/self_attn/v_proj/kernel": "flax_model-00001-of-00002.msgpack",
136
+ "model/decoder/layers/15/self_attn_layer_norm/bias": "flax_model-00001-of-00002.msgpack",
137
+ "model/decoder/layers/15/self_attn_layer_norm/scale": "flax_model-00001-of-00002.msgpack",
138
+ "model/decoder/layers/16/fc1/bias": "flax_model-00001-of-00002.msgpack",
139
+ "model/decoder/layers/16/fc1/kernel": "flax_model-00001-of-00002.msgpack",
140
+ "model/decoder/layers/16/fc2/bias": "flax_model-00001-of-00002.msgpack",
141
+ "model/decoder/layers/16/fc2/kernel": "flax_model-00001-of-00002.msgpack",
142
+ "model/decoder/layers/16/final_layer_norm/bias": "flax_model-00001-of-00002.msgpack",
143
+ "model/decoder/layers/16/final_layer_norm/scale": "flax_model-00001-of-00002.msgpack",
144
+ "model/decoder/layers/16/self_attn/k_proj/bias": "flax_model-00001-of-00002.msgpack",
145
+ "model/decoder/layers/16/self_attn/k_proj/kernel": "flax_model-00001-of-00002.msgpack",
146
+ "model/decoder/layers/16/self_attn/out_proj/bias": "flax_model-00001-of-00002.msgpack",
147
+ "model/decoder/layers/16/self_attn/out_proj/kernel": "flax_model-00001-of-00002.msgpack",
148
+ "model/decoder/layers/16/self_attn/q_proj/bias": "flax_model-00001-of-00002.msgpack",
149
+ "model/decoder/layers/16/self_attn/q_proj/kernel": "flax_model-00001-of-00002.msgpack",
150
+ "model/decoder/layers/16/self_attn/v_proj/bias": "flax_model-00001-of-00002.msgpack",
151
+ "model/decoder/layers/16/self_attn/v_proj/kernel": "flax_model-00001-of-00002.msgpack",
152
+ "model/decoder/layers/16/self_attn_layer_norm/bias": "flax_model-00001-of-00002.msgpack",
153
+ "model/decoder/layers/16/self_attn_layer_norm/scale": "flax_model-00001-of-00002.msgpack",
154
+ "model/decoder/layers/17/fc1/bias": "flax_model-00001-of-00002.msgpack",
155
+ "model/decoder/layers/17/fc1/kernel": "flax_model-00001-of-00002.msgpack",
156
+ "model/decoder/layers/17/fc2/bias": "flax_model-00001-of-00002.msgpack",
157
+ "model/decoder/layers/17/fc2/kernel": "flax_model-00001-of-00002.msgpack",
158
+ "model/decoder/layers/17/final_layer_norm/bias": "flax_model-00001-of-00002.msgpack",
159
+ "model/decoder/layers/17/final_layer_norm/scale": "flax_model-00001-of-00002.msgpack",
160
+ "model/decoder/layers/17/self_attn/k_proj/bias": "flax_model-00001-of-00002.msgpack",
161
+ "model/decoder/layers/17/self_attn/k_proj/kernel": "flax_model-00001-of-00002.msgpack",
162
+ "model/decoder/layers/17/self_attn/out_proj/bias": "flax_model-00001-of-00002.msgpack",
163
+ "model/decoder/layers/17/self_attn/out_proj/kernel": "flax_model-00001-of-00002.msgpack",
164
+ "model/decoder/layers/17/self_attn/q_proj/bias": "flax_model-00001-of-00002.msgpack",
165
+ "model/decoder/layers/17/self_attn/q_proj/kernel": "flax_model-00001-of-00002.msgpack",
166
+ "model/decoder/layers/17/self_attn/v_proj/bias": "flax_model-00001-of-00002.msgpack",
167
+ "model/decoder/layers/17/self_attn/v_proj/kernel": "flax_model-00001-of-00002.msgpack",
168
+ "model/decoder/layers/17/self_attn_layer_norm/bias": "flax_model-00001-of-00002.msgpack",
169
+ "model/decoder/layers/17/self_attn_layer_norm/scale": "flax_model-00001-of-00002.msgpack",
170
+ "model/decoder/layers/18/fc1/bias": "flax_model-00001-of-00002.msgpack",
171
+ "model/decoder/layers/18/fc1/kernel": "flax_model-00001-of-00002.msgpack",
172
+ "model/decoder/layers/18/fc2/bias": "flax_model-00001-of-00002.msgpack",
173
+ "model/decoder/layers/18/fc2/kernel": "flax_model-00001-of-00002.msgpack",
174
+ "model/decoder/layers/18/final_layer_norm/bias": "flax_model-00001-of-00002.msgpack",
175
+ "model/decoder/layers/18/final_layer_norm/scale": "flax_model-00001-of-00002.msgpack",
176
+ "model/decoder/layers/18/self_attn/k_proj/bias": "flax_model-00001-of-00002.msgpack",
177
+ "model/decoder/layers/18/self_attn/k_proj/kernel": "flax_model-00001-of-00002.msgpack",
178
+ "model/decoder/layers/18/self_attn/out_proj/bias": "flax_model-00001-of-00002.msgpack",
179
+ "model/decoder/layers/18/self_attn/out_proj/kernel": "flax_model-00001-of-00002.msgpack",
180
+ "model/decoder/layers/18/self_attn/q_proj/bias": "flax_model-00001-of-00002.msgpack",
181
+ "model/decoder/layers/18/self_attn/q_proj/kernel": "flax_model-00001-of-00002.msgpack",
182
+ "model/decoder/layers/18/self_attn/v_proj/bias": "flax_model-00001-of-00002.msgpack",
183
+ "model/decoder/layers/18/self_attn/v_proj/kernel": "flax_model-00001-of-00002.msgpack",
184
+ "model/decoder/layers/18/self_attn_layer_norm/bias": "flax_model-00001-of-00002.msgpack",
185
+ "model/decoder/layers/18/self_attn_layer_norm/scale": "flax_model-00001-of-00002.msgpack",
186
+ "model/decoder/layers/19/fc1/bias": "flax_model-00001-of-00002.msgpack",
187
+ "model/decoder/layers/19/fc1/kernel": "flax_model-00001-of-00002.msgpack",
188
+ "model/decoder/layers/19/fc2/bias": "flax_model-00001-of-00002.msgpack",
189
+ "model/decoder/layers/19/fc2/kernel": "flax_model-00001-of-00002.msgpack",
190
+ "model/decoder/layers/19/final_layer_norm/bias": "flax_model-00001-of-00002.msgpack",
191
+ "model/decoder/layers/19/final_layer_norm/scale": "flax_model-00001-of-00002.msgpack",
192
+ "model/decoder/layers/19/self_attn/k_proj/bias": "flax_model-00001-of-00002.msgpack",
193
+ "model/decoder/layers/19/self_attn/k_proj/kernel": "flax_model-00001-of-00002.msgpack",
194
+ "model/decoder/layers/19/self_attn/out_proj/bias": "flax_model-00001-of-00002.msgpack",
195
+ "model/decoder/layers/19/self_attn/out_proj/kernel": "flax_model-00001-of-00002.msgpack",
196
+ "model/decoder/layers/19/self_attn/q_proj/bias": "flax_model-00001-of-00002.msgpack",
197
+ "model/decoder/layers/19/self_attn/q_proj/kernel": "flax_model-00001-of-00002.msgpack",
198
+ "model/decoder/layers/19/self_attn/v_proj/bias": "flax_model-00001-of-00002.msgpack",
199
+ "model/decoder/layers/19/self_attn/v_proj/kernel": "flax_model-00001-of-00002.msgpack",
200
+ "model/decoder/layers/19/self_attn_layer_norm/bias": "flax_model-00001-of-00002.msgpack",
201
+ "model/decoder/layers/19/self_attn_layer_norm/scale": "flax_model-00001-of-00002.msgpack",
202
+ "model/decoder/layers/2/fc1/bias": "flax_model-00001-of-00002.msgpack",
203
+ "model/decoder/layers/2/fc1/kernel": "flax_model-00001-of-00002.msgpack",
204
+ "model/decoder/layers/2/fc2/bias": "flax_model-00001-of-00002.msgpack",
205
+ "model/decoder/layers/2/fc2/kernel": "flax_model-00001-of-00002.msgpack",
206
+ "model/decoder/layers/2/final_layer_norm/bias": "flax_model-00001-of-00002.msgpack",
207
+ "model/decoder/layers/2/final_layer_norm/scale": "flax_model-00001-of-00002.msgpack",
208
+ "model/decoder/layers/2/self_attn/k_proj/bias": "flax_model-00001-of-00002.msgpack",
209
+ "model/decoder/layers/2/self_attn/k_proj/kernel": "flax_model-00001-of-00002.msgpack",
210
+ "model/decoder/layers/2/self_attn/out_proj/bias": "flax_model-00001-of-00002.msgpack",
211
+ "model/decoder/layers/2/self_attn/out_proj/kernel": "flax_model-00001-of-00002.msgpack",
212
+ "model/decoder/layers/2/self_attn/q_proj/bias": "flax_model-00001-of-00002.msgpack",
213
+ "model/decoder/layers/2/self_attn/q_proj/kernel": "flax_model-00001-of-00002.msgpack",
214
+ "model/decoder/layers/2/self_attn/v_proj/bias": "flax_model-00001-of-00002.msgpack",
215
+ "model/decoder/layers/2/self_attn/v_proj/kernel": "flax_model-00001-of-00002.msgpack",
216
+ "model/decoder/layers/2/self_attn_layer_norm/bias": "flax_model-00001-of-00002.msgpack",
217
+ "model/decoder/layers/2/self_attn_layer_norm/scale": "flax_model-00001-of-00002.msgpack",
218
+ "model/decoder/layers/20/fc1/bias": "flax_model-00001-of-00002.msgpack",
219
+ "model/decoder/layers/20/fc1/kernel": "flax_model-00001-of-00002.msgpack",
220
+ "model/decoder/layers/20/fc2/bias": "flax_model-00001-of-00002.msgpack",
221
+ "model/decoder/layers/20/fc2/kernel": "flax_model-00001-of-00002.msgpack",
222
+ "model/decoder/layers/20/final_layer_norm/bias": "flax_model-00001-of-00002.msgpack",
223
+ "model/decoder/layers/20/final_layer_norm/scale": "flax_model-00001-of-00002.msgpack",
224
+ "model/decoder/layers/20/self_attn/k_proj/bias": "flax_model-00001-of-00002.msgpack",
225
+ "model/decoder/layers/20/self_attn/k_proj/kernel": "flax_model-00001-of-00002.msgpack",
226
+ "model/decoder/layers/20/self_attn/out_proj/bias": "flax_model-00001-of-00002.msgpack",
227
+ "model/decoder/layers/20/self_attn/out_proj/kernel": "flax_model-00001-of-00002.msgpack",
228
+ "model/decoder/layers/20/self_attn/q_proj/bias": "flax_model-00001-of-00002.msgpack",
229
+ "model/decoder/layers/20/self_attn/q_proj/kernel": "flax_model-00001-of-00002.msgpack",
230
+ "model/decoder/layers/20/self_attn/v_proj/bias": "flax_model-00001-of-00002.msgpack",
231
+ "model/decoder/layers/20/self_attn/v_proj/kernel": "flax_model-00001-of-00002.msgpack",
232
+ "model/decoder/layers/20/self_attn_layer_norm/bias": "flax_model-00001-of-00002.msgpack",
233
+ "model/decoder/layers/20/self_attn_layer_norm/scale": "flax_model-00001-of-00002.msgpack",
234
+ "model/decoder/layers/21/fc1/bias": "flax_model-00001-of-00002.msgpack",
235
+ "model/decoder/layers/21/fc1/kernel": "flax_model-00001-of-00002.msgpack",
236
+ "model/decoder/layers/21/fc2/bias": "flax_model-00001-of-00002.msgpack",
237
+ "model/decoder/layers/21/fc2/kernel": "flax_model-00001-of-00002.msgpack",
238
+ "model/decoder/layers/21/final_layer_norm/bias": "flax_model-00001-of-00002.msgpack",
239
+ "model/decoder/layers/21/final_layer_norm/scale": "flax_model-00001-of-00002.msgpack",
240
+ "model/decoder/layers/21/self_attn/k_proj/bias": "flax_model-00001-of-00002.msgpack",
241
+ "model/decoder/layers/21/self_attn/k_proj/kernel": "flax_model-00001-of-00002.msgpack",
242
+ "model/decoder/layers/21/self_attn/out_proj/bias": "flax_model-00001-of-00002.msgpack",
243
+ "model/decoder/layers/21/self_attn/out_proj/kernel": "flax_model-00001-of-00002.msgpack",
244
+ "model/decoder/layers/21/self_attn/q_proj/bias": "flax_model-00001-of-00002.msgpack",
245
+ "model/decoder/layers/21/self_attn/q_proj/kernel": "flax_model-00001-of-00002.msgpack",
246
+ "model/decoder/layers/21/self_attn/v_proj/bias": "flax_model-00001-of-00002.msgpack",
247
+ "model/decoder/layers/21/self_attn/v_proj/kernel": "flax_model-00001-of-00002.msgpack",
248
+ "model/decoder/layers/21/self_attn_layer_norm/bias": "flax_model-00001-of-00002.msgpack",
249
+ "model/decoder/layers/21/self_attn_layer_norm/scale": "flax_model-00001-of-00002.msgpack",
250
+ "model/decoder/layers/22/fc1/bias": "flax_model-00001-of-00002.msgpack",
251
+ "model/decoder/layers/22/fc1/kernel": "flax_model-00001-of-00002.msgpack",
252
+ "model/decoder/layers/22/fc2/bias": "flax_model-00001-of-00002.msgpack",
253
+ "model/decoder/layers/22/fc2/kernel": "flax_model-00001-of-00002.msgpack",
254
+ "model/decoder/layers/22/final_layer_norm/bias": "flax_model-00001-of-00002.msgpack",
255
+ "model/decoder/layers/22/final_layer_norm/scale": "flax_model-00001-of-00002.msgpack",
256
+ "model/decoder/layers/22/self_attn/k_proj/bias": "flax_model-00001-of-00002.msgpack",
257
+ "model/decoder/layers/22/self_attn/k_proj/kernel": "flax_model-00001-of-00002.msgpack",
258
+ "model/decoder/layers/22/self_attn/out_proj/bias": "flax_model-00001-of-00002.msgpack",
259
+ "model/decoder/layers/22/self_attn/out_proj/kernel": "flax_model-00001-of-00002.msgpack",
260
+ "model/decoder/layers/22/self_attn/q_proj/bias": "flax_model-00001-of-00002.msgpack",
261
+ "model/decoder/layers/22/self_attn/q_proj/kernel": "flax_model-00001-of-00002.msgpack",
262
+ "model/decoder/layers/22/self_attn/v_proj/bias": "flax_model-00001-of-00002.msgpack",
263
+ "model/decoder/layers/22/self_attn/v_proj/kernel": "flax_model-00001-of-00002.msgpack",
264
+ "model/decoder/layers/22/self_attn_layer_norm/bias": "flax_model-00001-of-00002.msgpack",
265
+ "model/decoder/layers/22/self_attn_layer_norm/scale": "flax_model-00001-of-00002.msgpack",
266
+ "model/decoder/layers/23/fc1/bias": "flax_model-00001-of-00002.msgpack",
267
+ "model/decoder/layers/23/fc1/kernel": "flax_model-00001-of-00002.msgpack",
268
+ "model/decoder/layers/23/fc2/bias": "flax_model-00002-of-00002.msgpack",
269
+ "model/decoder/layers/23/fc2/kernel": "flax_model-00002-of-00002.msgpack",
270
+ "model/decoder/layers/23/final_layer_norm/bias": "flax_model-00002-of-00002.msgpack",
271
+ "model/decoder/layers/23/final_layer_norm/scale": "flax_model-00002-of-00002.msgpack",
272
+ "model/decoder/layers/23/self_attn/k_proj/bias": "flax_model-00001-of-00002.msgpack",
273
+ "model/decoder/layers/23/self_attn/k_proj/kernel": "flax_model-00001-of-00002.msgpack",
274
+ "model/decoder/layers/23/self_attn/out_proj/bias": "flax_model-00001-of-00002.msgpack",
275
+ "model/decoder/layers/23/self_attn/out_proj/kernel": "flax_model-00001-of-00002.msgpack",
276
+ "model/decoder/layers/23/self_attn/q_proj/bias": "flax_model-00001-of-00002.msgpack",
277
+ "model/decoder/layers/23/self_attn/q_proj/kernel": "flax_model-00001-of-00002.msgpack",
278
+ "model/decoder/layers/23/self_attn/v_proj/bias": "flax_model-00001-of-00002.msgpack",
279
+ "model/decoder/layers/23/self_attn/v_proj/kernel": "flax_model-00001-of-00002.msgpack",
280
+ "model/decoder/layers/23/self_attn_layer_norm/bias": "flax_model-00001-of-00002.msgpack",
281
+ "model/decoder/layers/23/self_attn_layer_norm/scale": "flax_model-00001-of-00002.msgpack",
282
+ "model/decoder/layers/24/fc1/bias": "flax_model-00002-of-00002.msgpack",
283
+ "model/decoder/layers/24/fc1/kernel": "flax_model-00002-of-00002.msgpack",
284
+ "model/decoder/layers/24/fc2/bias": "flax_model-00002-of-00002.msgpack",
285
+ "model/decoder/layers/24/fc2/kernel": "flax_model-00002-of-00002.msgpack",
286
+ "model/decoder/layers/24/final_layer_norm/bias": "flax_model-00002-of-00002.msgpack",
287
+ "model/decoder/layers/24/final_layer_norm/scale": "flax_model-00002-of-00002.msgpack",
288
+ "model/decoder/layers/24/self_attn/k_proj/bias": "flax_model-00002-of-00002.msgpack",
289
+ "model/decoder/layers/24/self_attn/k_proj/kernel": "flax_model-00002-of-00002.msgpack",
290
+ "model/decoder/layers/24/self_attn/out_proj/bias": "flax_model-00002-of-00002.msgpack",
291
+ "model/decoder/layers/24/self_attn/out_proj/kernel": "flax_model-00002-of-00002.msgpack",
292
+ "model/decoder/layers/24/self_attn/q_proj/bias": "flax_model-00002-of-00002.msgpack",
293
+ "model/decoder/layers/24/self_attn/q_proj/kernel": "flax_model-00002-of-00002.msgpack",
294
+ "model/decoder/layers/24/self_attn/v_proj/bias": "flax_model-00002-of-00002.msgpack",
295
+ "model/decoder/layers/24/self_attn/v_proj/kernel": "flax_model-00002-of-00002.msgpack",
296
+ "model/decoder/layers/24/self_attn_layer_norm/bias": "flax_model-00002-of-00002.msgpack",
297
+ "model/decoder/layers/24/self_attn_layer_norm/scale": "flax_model-00002-of-00002.msgpack",
298
+ "model/decoder/layers/25/fc1/bias": "flax_model-00002-of-00002.msgpack",
299
+ "model/decoder/layers/25/fc1/kernel": "flax_model-00002-of-00002.msgpack",
300
+ "model/decoder/layers/25/fc2/bias": "flax_model-00002-of-00002.msgpack",
301
+ "model/decoder/layers/25/fc2/kernel": "flax_model-00002-of-00002.msgpack",
302
+ "model/decoder/layers/25/final_layer_norm/bias": "flax_model-00002-of-00002.msgpack",
303
+ "model/decoder/layers/25/final_layer_norm/scale": "flax_model-00002-of-00002.msgpack",
304
+ "model/decoder/layers/25/self_attn/k_proj/bias": "flax_model-00002-of-00002.msgpack",
305
+ "model/decoder/layers/25/self_attn/k_proj/kernel": "flax_model-00002-of-00002.msgpack",
306
+ "model/decoder/layers/25/self_attn/out_proj/bias": "flax_model-00002-of-00002.msgpack",
307
+ "model/decoder/layers/25/self_attn/out_proj/kernel": "flax_model-00002-of-00002.msgpack",
308
+ "model/decoder/layers/25/self_attn/q_proj/bias": "flax_model-00002-of-00002.msgpack",
309
+ "model/decoder/layers/25/self_attn/q_proj/kernel": "flax_model-00002-of-00002.msgpack",
310
+ "model/decoder/layers/25/self_attn/v_proj/bias": "flax_model-00002-of-00002.msgpack",
311
+ "model/decoder/layers/25/self_attn/v_proj/kernel": "flax_model-00002-of-00002.msgpack",
312
+ "model/decoder/layers/25/self_attn_layer_norm/bias": "flax_model-00002-of-00002.msgpack",
313
+ "model/decoder/layers/25/self_attn_layer_norm/scale": "flax_model-00002-of-00002.msgpack",
314
+ "model/decoder/layers/26/fc1/bias": "flax_model-00002-of-00002.msgpack",
315
+ "model/decoder/layers/26/fc1/kernel": "flax_model-00002-of-00002.msgpack",
316
+ "model/decoder/layers/26/fc2/bias": "flax_model-00002-of-00002.msgpack",
317
+ "model/decoder/layers/26/fc2/kernel": "flax_model-00002-of-00002.msgpack",
318
+ "model/decoder/layers/26/final_layer_norm/bias": "flax_model-00002-of-00002.msgpack",
319
+ "model/decoder/layers/26/final_layer_norm/scale": "flax_model-00002-of-00002.msgpack",
320
+ "model/decoder/layers/26/self_attn/k_proj/bias": "flax_model-00002-of-00002.msgpack",
321
+ "model/decoder/layers/26/self_attn/k_proj/kernel": "flax_model-00002-of-00002.msgpack",
322
+ "model/decoder/layers/26/self_attn/out_proj/bias": "flax_model-00002-of-00002.msgpack",
323
+ "model/decoder/layers/26/self_attn/out_proj/kernel": "flax_model-00002-of-00002.msgpack",
324
+ "model/decoder/layers/26/self_attn/q_proj/bias": "flax_model-00002-of-00002.msgpack",
325
+ "model/decoder/layers/26/self_attn/q_proj/kernel": "flax_model-00002-of-00002.msgpack",
326
+ "model/decoder/layers/26/self_attn/v_proj/bias": "flax_model-00002-of-00002.msgpack",
327
+ "model/decoder/layers/26/self_attn/v_proj/kernel": "flax_model-00002-of-00002.msgpack",
328
+ "model/decoder/layers/26/self_attn_layer_norm/bias": "flax_model-00002-of-00002.msgpack",
329
+ "model/decoder/layers/26/self_attn_layer_norm/scale": "flax_model-00002-of-00002.msgpack",
330
+ "model/decoder/layers/27/fc1/bias": "flax_model-00002-of-00002.msgpack",
331
+ "model/decoder/layers/27/fc1/kernel": "flax_model-00002-of-00002.msgpack",
332
+ "model/decoder/layers/27/fc2/bias": "flax_model-00002-of-00002.msgpack",
333
+ "model/decoder/layers/27/fc2/kernel": "flax_model-00002-of-00002.msgpack",
334
+ "model/decoder/layers/27/final_layer_norm/bias": "flax_model-00002-of-00002.msgpack",
335
+ "model/decoder/layers/27/final_layer_norm/scale": "flax_model-00002-of-00002.msgpack",
336
+ "model/decoder/layers/27/self_attn/k_proj/bias": "flax_model-00002-of-00002.msgpack",
337
+ "model/decoder/layers/27/self_attn/k_proj/kernel": "flax_model-00002-of-00002.msgpack",
338
+ "model/decoder/layers/27/self_attn/out_proj/bias": "flax_model-00002-of-00002.msgpack",
339
+ "model/decoder/layers/27/self_attn/out_proj/kernel": "flax_model-00002-of-00002.msgpack",
340
+ "model/decoder/layers/27/self_attn/q_proj/bias": "flax_model-00002-of-00002.msgpack",
341
+ "model/decoder/layers/27/self_attn/q_proj/kernel": "flax_model-00002-of-00002.msgpack",
342
+ "model/decoder/layers/27/self_attn/v_proj/bias": "flax_model-00002-of-00002.msgpack",
343
+ "model/decoder/layers/27/self_attn/v_proj/kernel": "flax_model-00002-of-00002.msgpack",
344
+ "model/decoder/layers/27/self_attn_layer_norm/bias": "flax_model-00002-of-00002.msgpack",
345
+ "model/decoder/layers/27/self_attn_layer_norm/scale": "flax_model-00002-of-00002.msgpack",
346
+ "model/decoder/layers/28/fc1/bias": "flax_model-00002-of-00002.msgpack",
347
+ "model/decoder/layers/28/fc1/kernel": "flax_model-00002-of-00002.msgpack",
348
+ "model/decoder/layers/28/fc2/bias": "flax_model-00002-of-00002.msgpack",
349
+ "model/decoder/layers/28/fc2/kernel": "flax_model-00002-of-00002.msgpack",
350
+ "model/decoder/layers/28/final_layer_norm/bias": "flax_model-00002-of-00002.msgpack",
351
+ "model/decoder/layers/28/final_layer_norm/scale": "flax_model-00002-of-00002.msgpack",
352
+ "model/decoder/layers/28/self_attn/k_proj/bias": "flax_model-00002-of-00002.msgpack",
353
+ "model/decoder/layers/28/self_attn/k_proj/kernel": "flax_model-00002-of-00002.msgpack",
354
+ "model/decoder/layers/28/self_attn/out_proj/bias": "flax_model-00002-of-00002.msgpack",
355
+ "model/decoder/layers/28/self_attn/out_proj/kernel": "flax_model-00002-of-00002.msgpack",
356
+ "model/decoder/layers/28/self_attn/q_proj/bias": "flax_model-00002-of-00002.msgpack",
357
+ "model/decoder/layers/28/self_attn/q_proj/kernel": "flax_model-00002-of-00002.msgpack",
358
+ "model/decoder/layers/28/self_attn/v_proj/bias": "flax_model-00002-of-00002.msgpack",
359
+ "model/decoder/layers/28/self_attn/v_proj/kernel": "flax_model-00002-of-00002.msgpack",
360
+ "model/decoder/layers/28/self_attn_layer_norm/bias": "flax_model-00002-of-00002.msgpack",
361
+ "model/decoder/layers/28/self_attn_layer_norm/scale": "flax_model-00002-of-00002.msgpack",
362
+ "model/decoder/layers/29/fc1/bias": "flax_model-00002-of-00002.msgpack",
363
+ "model/decoder/layers/29/fc1/kernel": "flax_model-00002-of-00002.msgpack",
364
+ "model/decoder/layers/29/fc2/bias": "flax_model-00002-of-00002.msgpack",
365
+ "model/decoder/layers/29/fc2/kernel": "flax_model-00002-of-00002.msgpack",
366
+ "model/decoder/layers/29/final_layer_norm/bias": "flax_model-00002-of-00002.msgpack",
367
+ "model/decoder/layers/29/final_layer_norm/scale": "flax_model-00002-of-00002.msgpack",
368
+ "model/decoder/layers/29/self_attn/k_proj/bias": "flax_model-00002-of-00002.msgpack",
369
+ "model/decoder/layers/29/self_attn/k_proj/kernel": "flax_model-00002-of-00002.msgpack",
370
+ "model/decoder/layers/29/self_attn/out_proj/bias": "flax_model-00002-of-00002.msgpack",
371
+ "model/decoder/layers/29/self_attn/out_proj/kernel": "flax_model-00002-of-00002.msgpack",
372
+ "model/decoder/layers/29/self_attn/q_proj/bias": "flax_model-00002-of-00002.msgpack",
373
+ "model/decoder/layers/29/self_attn/q_proj/kernel": "flax_model-00002-of-00002.msgpack",
374
+ "model/decoder/layers/29/self_attn/v_proj/bias": "flax_model-00002-of-00002.msgpack",
375
+ "model/decoder/layers/29/self_attn/v_proj/kernel": "flax_model-00002-of-00002.msgpack",
376
+ "model/decoder/layers/29/self_attn_layer_norm/bias": "flax_model-00002-of-00002.msgpack",
377
+ "model/decoder/layers/29/self_attn_layer_norm/scale": "flax_model-00002-of-00002.msgpack",
378
+ "model/decoder/layers/3/fc1/bias": "flax_model-00001-of-00002.msgpack",
379
+ "model/decoder/layers/3/fc1/kernel": "flax_model-00001-of-00002.msgpack",
380
+ "model/decoder/layers/3/fc2/bias": "flax_model-00001-of-00002.msgpack",
381
+ "model/decoder/layers/3/fc2/kernel": "flax_model-00001-of-00002.msgpack",
382
+ "model/decoder/layers/3/final_layer_norm/bias": "flax_model-00001-of-00002.msgpack",
383
+ "model/decoder/layers/3/final_layer_norm/scale": "flax_model-00001-of-00002.msgpack",
384
+ "model/decoder/layers/3/self_attn/k_proj/bias": "flax_model-00001-of-00002.msgpack",
385
+ "model/decoder/layers/3/self_attn/k_proj/kernel": "flax_model-00001-of-00002.msgpack",
386
+ "model/decoder/layers/3/self_attn/out_proj/bias": "flax_model-00001-of-00002.msgpack",
387
+ "model/decoder/layers/3/self_attn/out_proj/kernel": "flax_model-00001-of-00002.msgpack",
388
+ "model/decoder/layers/3/self_attn/q_proj/bias": "flax_model-00001-of-00002.msgpack",
389
+ "model/decoder/layers/3/self_attn/q_proj/kernel": "flax_model-00001-of-00002.msgpack",
390
+ "model/decoder/layers/3/self_attn/v_proj/bias": "flax_model-00001-of-00002.msgpack",
391
+ "model/decoder/layers/3/self_attn/v_proj/kernel": "flax_model-00001-of-00002.msgpack",
392
+ "model/decoder/layers/3/self_attn_layer_norm/bias": "flax_model-00001-of-00002.msgpack",
393
+ "model/decoder/layers/3/self_attn_layer_norm/scale": "flax_model-00001-of-00002.msgpack",
394
+ "model/decoder/layers/30/fc1/bias": "flax_model-00002-of-00002.msgpack",
395
+ "model/decoder/layers/30/fc1/kernel": "flax_model-00002-of-00002.msgpack",
396
+ "model/decoder/layers/30/fc2/bias": "flax_model-00002-of-00002.msgpack",
397
+ "model/decoder/layers/30/fc2/kernel": "flax_model-00002-of-00002.msgpack",
398
+ "model/decoder/layers/30/final_layer_norm/bias": "flax_model-00002-of-00002.msgpack",
399
+ "model/decoder/layers/30/final_layer_norm/scale": "flax_model-00002-of-00002.msgpack",
400
+ "model/decoder/layers/30/self_attn/k_proj/bias": "flax_model-00002-of-00002.msgpack",
401
+ "model/decoder/layers/30/self_attn/k_proj/kernel": "flax_model-00002-of-00002.msgpack",
402
+ "model/decoder/layers/30/self_attn/out_proj/bias": "flax_model-00002-of-00002.msgpack",
403
+ "model/decoder/layers/30/self_attn/out_proj/kernel": "flax_model-00002-of-00002.msgpack",
404
+ "model/decoder/layers/30/self_attn/q_proj/bias": "flax_model-00002-of-00002.msgpack",
405
+ "model/decoder/layers/30/self_attn/q_proj/kernel": "flax_model-00002-of-00002.msgpack",
406
+ "model/decoder/layers/30/self_attn/v_proj/bias": "flax_model-00002-of-00002.msgpack",
407
+ "model/decoder/layers/30/self_attn/v_proj/kernel": "flax_model-00002-of-00002.msgpack",
408
+ "model/decoder/layers/30/self_attn_layer_norm/bias": "flax_model-00002-of-00002.msgpack",
409
+ "model/decoder/layers/30/self_attn_layer_norm/scale": "flax_model-00002-of-00002.msgpack",
410
+ "model/decoder/layers/31/fc1/bias": "flax_model-00002-of-00002.msgpack",
411
+ "model/decoder/layers/31/fc1/kernel": "flax_model-00002-of-00002.msgpack",
412
+ "model/decoder/layers/31/fc2/bias": "flax_model-00002-of-00002.msgpack",
413
+ "model/decoder/layers/31/fc2/kernel": "flax_model-00002-of-00002.msgpack",
414
+ "model/decoder/layers/31/final_layer_norm/bias": "flax_model-00002-of-00002.msgpack",
415
+ "model/decoder/layers/31/final_layer_norm/scale": "flax_model-00002-of-00002.msgpack",
416
+ "model/decoder/layers/31/self_attn/k_proj/bias": "flax_model-00002-of-00002.msgpack",
417
+ "model/decoder/layers/31/self_attn/k_proj/kernel": "flax_model-00002-of-00002.msgpack",
418
+ "model/decoder/layers/31/self_attn/out_proj/bias": "flax_model-00002-of-00002.msgpack",
419
+ "model/decoder/layers/31/self_attn/out_proj/kernel": "flax_model-00002-of-00002.msgpack",
420
+ "model/decoder/layers/31/self_attn/q_proj/bias": "flax_model-00002-of-00002.msgpack",
421
+ "model/decoder/layers/31/self_attn/q_proj/kernel": "flax_model-00002-of-00002.msgpack",
422
+ "model/decoder/layers/31/self_attn/v_proj/bias": "flax_model-00002-of-00002.msgpack",
423
+ "model/decoder/layers/31/self_attn/v_proj/kernel": "flax_model-00002-of-00002.msgpack",
424
+ "model/decoder/layers/31/self_attn_layer_norm/bias": "flax_model-00002-of-00002.msgpack",
425
+ "model/decoder/layers/31/self_attn_layer_norm/scale": "flax_model-00002-of-00002.msgpack",
426
+ "model/decoder/layers/4/fc1/bias": "flax_model-00001-of-00002.msgpack",
427
+ "model/decoder/layers/4/fc1/kernel": "flax_model-00001-of-00002.msgpack",
428
+ "model/decoder/layers/4/fc2/bias": "flax_model-00001-of-00002.msgpack",
429
+ "model/decoder/layers/4/fc2/kernel": "flax_model-00001-of-00002.msgpack",
430
+ "model/decoder/layers/4/final_layer_norm/bias": "flax_model-00001-of-00002.msgpack",
431
+ "model/decoder/layers/4/final_layer_norm/scale": "flax_model-00001-of-00002.msgpack",
432
+ "model/decoder/layers/4/self_attn/k_proj/bias": "flax_model-00001-of-00002.msgpack",
433
+ "model/decoder/layers/4/self_attn/k_proj/kernel": "flax_model-00001-of-00002.msgpack",
434
+ "model/decoder/layers/4/self_attn/out_proj/bias": "flax_model-00001-of-00002.msgpack",
435
+ "model/decoder/layers/4/self_attn/out_proj/kernel": "flax_model-00001-of-00002.msgpack",
436
+ "model/decoder/layers/4/self_attn/q_proj/bias": "flax_model-00001-of-00002.msgpack",
437
+ "model/decoder/layers/4/self_attn/q_proj/kernel": "flax_model-00001-of-00002.msgpack",
438
+ "model/decoder/layers/4/self_attn/v_proj/bias": "flax_model-00001-of-00002.msgpack",
439
+ "model/decoder/layers/4/self_attn/v_proj/kernel": "flax_model-00001-of-00002.msgpack",
440
+ "model/decoder/layers/4/self_attn_layer_norm/bias": "flax_model-00001-of-00002.msgpack",
441
+ "model/decoder/layers/4/self_attn_layer_norm/scale": "flax_model-00001-of-00002.msgpack",
442
+ "model/decoder/layers/5/fc1/bias": "flax_model-00001-of-00002.msgpack",
443
+ "model/decoder/layers/5/fc1/kernel": "flax_model-00001-of-00002.msgpack",
444
+ "model/decoder/layers/5/fc2/bias": "flax_model-00001-of-00002.msgpack",
445
+ "model/decoder/layers/5/fc2/kernel": "flax_model-00001-of-00002.msgpack",
446
+ "model/decoder/layers/5/final_layer_norm/bias": "flax_model-00001-of-00002.msgpack",
447
+ "model/decoder/layers/5/final_layer_norm/scale": "flax_model-00001-of-00002.msgpack",
448
+ "model/decoder/layers/5/self_attn/k_proj/bias": "flax_model-00001-of-00002.msgpack",
449
+ "model/decoder/layers/5/self_attn/k_proj/kernel": "flax_model-00001-of-00002.msgpack",
450
+ "model/decoder/layers/5/self_attn/out_proj/bias": "flax_model-00001-of-00002.msgpack",
451
+ "model/decoder/layers/5/self_attn/out_proj/kernel": "flax_model-00001-of-00002.msgpack",
452
+ "model/decoder/layers/5/self_attn/q_proj/bias": "flax_model-00001-of-00002.msgpack",
453
+ "model/decoder/layers/5/self_attn/q_proj/kernel": "flax_model-00001-of-00002.msgpack",
454
+ "model/decoder/layers/5/self_attn/v_proj/bias": "flax_model-00001-of-00002.msgpack",
455
+ "model/decoder/layers/5/self_attn/v_proj/kernel": "flax_model-00001-of-00002.msgpack",
456
+ "model/decoder/layers/5/self_attn_layer_norm/bias": "flax_model-00001-of-00002.msgpack",
457
+ "model/decoder/layers/5/self_attn_layer_norm/scale": "flax_model-00001-of-00002.msgpack",
458
+ "model/decoder/layers/6/fc1/bias": "flax_model-00001-of-00002.msgpack",
459
+ "model/decoder/layers/6/fc1/kernel": "flax_model-00001-of-00002.msgpack",
460
+ "model/decoder/layers/6/fc2/bias": "flax_model-00001-of-00002.msgpack",
461
+ "model/decoder/layers/6/fc2/kernel": "flax_model-00001-of-00002.msgpack",
462
+ "model/decoder/layers/6/final_layer_norm/bias": "flax_model-00001-of-00002.msgpack",
463
+ "model/decoder/layers/6/final_layer_norm/scale": "flax_model-00001-of-00002.msgpack",
464
+ "model/decoder/layers/6/self_attn/k_proj/bias": "flax_model-00001-of-00002.msgpack",
465
+ "model/decoder/layers/6/self_attn/k_proj/kernel": "flax_model-00001-of-00002.msgpack",
466
+ "model/decoder/layers/6/self_attn/out_proj/bias": "flax_model-00001-of-00002.msgpack",
467
+ "model/decoder/layers/6/self_attn/out_proj/kernel": "flax_model-00001-of-00002.msgpack",
468
+ "model/decoder/layers/6/self_attn/q_proj/bias": "flax_model-00001-of-00002.msgpack",
469
+ "model/decoder/layers/6/self_attn/q_proj/kernel": "flax_model-00001-of-00002.msgpack",
470
+ "model/decoder/layers/6/self_attn/v_proj/bias": "flax_model-00001-of-00002.msgpack",
471
+ "model/decoder/layers/6/self_attn/v_proj/kernel": "flax_model-00001-of-00002.msgpack",
472
+ "model/decoder/layers/6/self_attn_layer_norm/bias": "flax_model-00001-of-00002.msgpack",
473
+ "model/decoder/layers/6/self_attn_layer_norm/scale": "flax_model-00001-of-00002.msgpack",
474
+ "model/decoder/layers/7/fc1/bias": "flax_model-00001-of-00002.msgpack",
475
+ "model/decoder/layers/7/fc1/kernel": "flax_model-00001-of-00002.msgpack",
476
+ "model/decoder/layers/7/fc2/bias": "flax_model-00001-of-00002.msgpack",
477
+ "model/decoder/layers/7/fc2/kernel": "flax_model-00001-of-00002.msgpack",
478
+ "model/decoder/layers/7/final_layer_norm/bias": "flax_model-00001-of-00002.msgpack",
479
+ "model/decoder/layers/7/final_layer_norm/scale": "flax_model-00001-of-00002.msgpack",
480
+ "model/decoder/layers/7/self_attn/k_proj/bias": "flax_model-00001-of-00002.msgpack",
481
+ "model/decoder/layers/7/self_attn/k_proj/kernel": "flax_model-00001-of-00002.msgpack",
482
+ "model/decoder/layers/7/self_attn/out_proj/bias": "flax_model-00001-of-00002.msgpack",
483
+ "model/decoder/layers/7/self_attn/out_proj/kernel": "flax_model-00001-of-00002.msgpack",
484
+ "model/decoder/layers/7/self_attn/q_proj/bias": "flax_model-00001-of-00002.msgpack",
485
+ "model/decoder/layers/7/self_attn/q_proj/kernel": "flax_model-00001-of-00002.msgpack",
486
+ "model/decoder/layers/7/self_attn/v_proj/bias": "flax_model-00001-of-00002.msgpack",
487
+ "model/decoder/layers/7/self_attn/v_proj/kernel": "flax_model-00001-of-00002.msgpack",
488
+ "model/decoder/layers/7/self_attn_layer_norm/bias": "flax_model-00001-of-00002.msgpack",
489
+ "model/decoder/layers/7/self_attn_layer_norm/scale": "flax_model-00001-of-00002.msgpack",
490
+ "model/decoder/layers/8/fc1/bias": "flax_model-00001-of-00002.msgpack",
491
+ "model/decoder/layers/8/fc1/kernel": "flax_model-00001-of-00002.msgpack",
492
+ "model/decoder/layers/8/fc2/bias": "flax_model-00001-of-00002.msgpack",
493
+ "model/decoder/layers/8/fc2/kernel": "flax_model-00001-of-00002.msgpack",
494
+ "model/decoder/layers/8/final_layer_norm/bias": "flax_model-00001-of-00002.msgpack",
495
+ "model/decoder/layers/8/final_layer_norm/scale": "flax_model-00001-of-00002.msgpack",
496
+ "model/decoder/layers/8/self_attn/k_proj/bias": "flax_model-00001-of-00002.msgpack",
497
+ "model/decoder/layers/8/self_attn/k_proj/kernel": "flax_model-00001-of-00002.msgpack",
498
+ "model/decoder/layers/8/self_attn/out_proj/bias": "flax_model-00001-of-00002.msgpack",
499
+ "model/decoder/layers/8/self_attn/out_proj/kernel": "flax_model-00001-of-00002.msgpack",
500
+ "model/decoder/layers/8/self_attn/q_proj/bias": "flax_model-00001-of-00002.msgpack",
501
+ "model/decoder/layers/8/self_attn/q_proj/kernel": "flax_model-00001-of-00002.msgpack",
502
+ "model/decoder/layers/8/self_attn/v_proj/bias": "flax_model-00001-of-00002.msgpack",
503
+ "model/decoder/layers/8/self_attn/v_proj/kernel": "flax_model-00001-of-00002.msgpack",
504
+ "model/decoder/layers/8/self_attn_layer_norm/bias": "flax_model-00001-of-00002.msgpack",
505
+ "model/decoder/layers/8/self_attn_layer_norm/scale": "flax_model-00001-of-00002.msgpack",
506
+ "model/decoder/layers/9/fc1/bias": "flax_model-00001-of-00002.msgpack",
507
+ "model/decoder/layers/9/fc1/kernel": "flax_model-00001-of-00002.msgpack",
508
+ "model/decoder/layers/9/fc2/bias": "flax_model-00001-of-00002.msgpack",
509
+ "model/decoder/layers/9/fc2/kernel": "flax_model-00001-of-00002.msgpack",
510
+ "model/decoder/layers/9/final_layer_norm/bias": "flax_model-00001-of-00002.msgpack",
511
+ "model/decoder/layers/9/final_layer_norm/scale": "flax_model-00001-of-00002.msgpack",
512
+ "model/decoder/layers/9/self_attn/k_proj/bias": "flax_model-00001-of-00002.msgpack",
513
+ "model/decoder/layers/9/self_attn/k_proj/kernel": "flax_model-00001-of-00002.msgpack",
514
+ "model/decoder/layers/9/self_attn/out_proj/bias": "flax_model-00001-of-00002.msgpack",
515
+ "model/decoder/layers/9/self_attn/out_proj/kernel": "flax_model-00001-of-00002.msgpack",
516
+ "model/decoder/layers/9/self_attn/q_proj/bias": "flax_model-00001-of-00002.msgpack",
517
+ "model/decoder/layers/9/self_attn/q_proj/kernel": "flax_model-00001-of-00002.msgpack",
518
+ "model/decoder/layers/9/self_attn/v_proj/bias": "flax_model-00001-of-00002.msgpack",
519
+ "model/decoder/layers/9/self_attn/v_proj/kernel": "flax_model-00001-of-00002.msgpack",
520
+ "model/decoder/layers/9/self_attn_layer_norm/bias": "flax_model-00001-of-00002.msgpack",
521
+ "model/decoder/layers/9/self_attn_layer_norm/scale": "flax_model-00001-of-00002.msgpack"
522
+ }
523
+ }
special_tokens_map.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {}
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "model_max_length": 1000000000000000019884624838656,
3
+ "name_or_path": "facebook/galactica-6.7b",
4
+ "special_tokens_map_file": "/content/tokenizer/special_tokens_map.json",
5
+ "tokenizer_class": "PreTrainedTokenizerFast"
6
+ }