dartpain commited on
Commit
f651bb7
1 Parent(s): ba1fc8f

Upload 15 files

Browse files
config.json ADDED
@@ -0,0 +1,33 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "tiiuae/falcon-40b",
3
+ "alibi": false,
4
+ "apply_residual_connection_post_layernorm": false,
5
+ "architectures": [
6
+ "RWForCausalLM"
7
+ ],
8
+ "attention_dropout": 0.0,
9
+ "auto_map": {
10
+ "AutoConfig": "tiiuae/falcon-40b--configuration_RW.RWConfig",
11
+ "AutoModel": "tiiuae/falcon-40b--modelling_RW.RWModel",
12
+ "AutoModelForCausalLM": "tiiuae/falcon-40b--modelling_RW.RWForCausalLM",
13
+ "AutoModelForQuestionAnswering": "tiiuae/falcon-40b--modelling_RW.RWForQuestionAnswering",
14
+ "AutoModelForSequenceClassification": "tiiuae/falcon-40b--modelling_RW.RWForSequenceClassification",
15
+ "AutoModelForTokenClassification": "tiiuae/falcon-40b--modelling_RW.RWForTokenClassification"
16
+ },
17
+ "bias": false,
18
+ "bos_token_id": 11,
19
+ "eos_token_id": 11,
20
+ "hidden_dropout": 0.0,
21
+ "hidden_size": 8192,
22
+ "initializer_range": 0.02,
23
+ "layer_norm_epsilon": 1e-05,
24
+ "model_type": "RefinedWeb",
25
+ "n_head": 128,
26
+ "n_head_kv": 8,
27
+ "n_layer": 60,
28
+ "parallel_attn": true,
29
+ "torch_dtype": "float16",
30
+ "transformers_version": "4.32.0.dev0",
31
+ "use_cache": true,
32
+ "vocab_size": 65024
33
+ }
generation_config.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 1,
4
+ "eos_token_id": 2,
5
+ "transformers_version": "4.32.0.dev0"
6
+ }
model-00001-of-00009.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6bd9bd22d64af8a3c126ed522e1ad10db0c771457280f393880bf2a7aa7bfdb3
3
+ size 9504757752
model-00002-of-00009.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2b949b3f6cf2290189df753d06cd9d761916d20d1a4dd3bba1703968ccf4bfee
3
+ size 9513146520
model-00003-of-00009.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5f66cf813f03d96092b016beafc5e356688573d2e00e7546f8c3127e637b7114
3
+ size 9513146584
model-00004-of-00009.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ec608bb411d7ff62ef0a0cd5866b2051269b6116962a282a1ba28290bee8caba
3
+ size 9513146584
model-00005-of-00009.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5bbeda5454232c80027fa68bd534fa4864e15dddc7aee26c425a4f8965936742
3
+ size 9513146584
model-00006-of-00009.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7111faa6eef847fbff9def182296bbed8e955dfb899b9be302ca3d2b055ac7ac
3
+ size 9513146584
model-00007-of-00009.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fa3054e78ea7b55cbcd51ac540080805e2be83c95649f04686581e67b0cff363
3
+ size 9513146584
model-00008-of-00009.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d9ce9fb5f732cd4a5a96fb75690a4860738951bb5fab7881ba6fffc0c42440a4
3
+ size 9513146584
model-00009-of-00009.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7889ecba4fdc0ebb0a27fb1aa2089093adbbf1cd015583d5b254853af36f6571
3
+ size 6509858800
model.safetensors.index.json ADDED
@@ -0,0 +1,490 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 82606587904
4
+ },
5
+ "weight_map": {
6
+ "transformer.h.0.ln_attn.bias": "model-00001-of-00009.safetensors",
7
+ "transformer.h.0.ln_attn.weight": "model-00001-of-00009.safetensors",
8
+ "transformer.h.0.ln_mlp.bias": "model-00001-of-00009.safetensors",
9
+ "transformer.h.0.ln_mlp.weight": "model-00001-of-00009.safetensors",
10
+ "transformer.h.0.mlp.dense_4h_to_h.weight": "model-00001-of-00009.safetensors",
11
+ "transformer.h.0.mlp.dense_h_to_4h.weight": "model-00001-of-00009.safetensors",
12
+ "transformer.h.0.self_attention.dense.weight": "model-00001-of-00009.safetensors",
13
+ "transformer.h.0.self_attention.query_key_value.weight": "model-00001-of-00009.safetensors",
14
+ "transformer.h.1.ln_attn.bias": "model-00001-of-00009.safetensors",
15
+ "transformer.h.1.ln_attn.weight": "model-00001-of-00009.safetensors",
16
+ "transformer.h.1.ln_mlp.bias": "model-00001-of-00009.safetensors",
17
+ "transformer.h.1.ln_mlp.weight": "model-00001-of-00009.safetensors",
18
+ "transformer.h.1.mlp.dense_4h_to_h.weight": "model-00001-of-00009.safetensors",
19
+ "transformer.h.1.mlp.dense_h_to_4h.weight": "model-00001-of-00009.safetensors",
20
+ "transformer.h.1.self_attention.dense.weight": "model-00001-of-00009.safetensors",
21
+ "transformer.h.1.self_attention.query_key_value.weight": "model-00001-of-00009.safetensors",
22
+ "transformer.h.10.ln_attn.bias": "model-00002-of-00009.safetensors",
23
+ "transformer.h.10.ln_attn.weight": "model-00002-of-00009.safetensors",
24
+ "transformer.h.10.ln_mlp.bias": "model-00002-of-00009.safetensors",
25
+ "transformer.h.10.ln_mlp.weight": "model-00002-of-00009.safetensors",
26
+ "transformer.h.10.mlp.dense_4h_to_h.weight": "model-00002-of-00009.safetensors",
27
+ "transformer.h.10.mlp.dense_h_to_4h.weight": "model-00002-of-00009.safetensors",
28
+ "transformer.h.10.self_attention.dense.weight": "model-00002-of-00009.safetensors",
29
+ "transformer.h.10.self_attention.query_key_value.weight": "model-00002-of-00009.safetensors",
30
+ "transformer.h.11.ln_attn.bias": "model-00002-of-00009.safetensors",
31
+ "transformer.h.11.ln_attn.weight": "model-00002-of-00009.safetensors",
32
+ "transformer.h.11.ln_mlp.bias": "model-00002-of-00009.safetensors",
33
+ "transformer.h.11.ln_mlp.weight": "model-00002-of-00009.safetensors",
34
+ "transformer.h.11.mlp.dense_4h_to_h.weight": "model-00002-of-00009.safetensors",
35
+ "transformer.h.11.mlp.dense_h_to_4h.weight": "model-00002-of-00009.safetensors",
36
+ "transformer.h.11.self_attention.dense.weight": "model-00002-of-00009.safetensors",
37
+ "transformer.h.11.self_attention.query_key_value.weight": "model-00002-of-00009.safetensors",
38
+ "transformer.h.12.ln_attn.bias": "model-00002-of-00009.safetensors",
39
+ "transformer.h.12.ln_attn.weight": "model-00002-of-00009.safetensors",
40
+ "transformer.h.12.ln_mlp.bias": "model-00002-of-00009.safetensors",
41
+ "transformer.h.12.ln_mlp.weight": "model-00002-of-00009.safetensors",
42
+ "transformer.h.12.mlp.dense_4h_to_h.weight": "model-00002-of-00009.safetensors",
43
+ "transformer.h.12.mlp.dense_h_to_4h.weight": "model-00002-of-00009.safetensors",
44
+ "transformer.h.12.self_attention.dense.weight": "model-00002-of-00009.safetensors",
45
+ "transformer.h.12.self_attention.query_key_value.weight": "model-00002-of-00009.safetensors",
46
+ "transformer.h.13.ln_attn.bias": "model-00002-of-00009.safetensors",
47
+ "transformer.h.13.ln_attn.weight": "model-00002-of-00009.safetensors",
48
+ "transformer.h.13.ln_mlp.bias": "model-00002-of-00009.safetensors",
49
+ "transformer.h.13.ln_mlp.weight": "model-00002-of-00009.safetensors",
50
+ "transformer.h.13.mlp.dense_4h_to_h.weight": "model-00003-of-00009.safetensors",
51
+ "transformer.h.13.mlp.dense_h_to_4h.weight": "model-00003-of-00009.safetensors",
52
+ "transformer.h.13.self_attention.dense.weight": "model-00002-of-00009.safetensors",
53
+ "transformer.h.13.self_attention.query_key_value.weight": "model-00002-of-00009.safetensors",
54
+ "transformer.h.14.ln_attn.bias": "model-00003-of-00009.safetensors",
55
+ "transformer.h.14.ln_attn.weight": "model-00003-of-00009.safetensors",
56
+ "transformer.h.14.ln_mlp.bias": "model-00003-of-00009.safetensors",
57
+ "transformer.h.14.ln_mlp.weight": "model-00003-of-00009.safetensors",
58
+ "transformer.h.14.mlp.dense_4h_to_h.weight": "model-00003-of-00009.safetensors",
59
+ "transformer.h.14.mlp.dense_h_to_4h.weight": "model-00003-of-00009.safetensors",
60
+ "transformer.h.14.self_attention.dense.weight": "model-00003-of-00009.safetensors",
61
+ "transformer.h.14.self_attention.query_key_value.weight": "model-00003-of-00009.safetensors",
62
+ "transformer.h.15.ln_attn.bias": "model-00003-of-00009.safetensors",
63
+ "transformer.h.15.ln_attn.weight": "model-00003-of-00009.safetensors",
64
+ "transformer.h.15.ln_mlp.bias": "model-00003-of-00009.safetensors",
65
+ "transformer.h.15.ln_mlp.weight": "model-00003-of-00009.safetensors",
66
+ "transformer.h.15.mlp.dense_4h_to_h.weight": "model-00003-of-00009.safetensors",
67
+ "transformer.h.15.mlp.dense_h_to_4h.weight": "model-00003-of-00009.safetensors",
68
+ "transformer.h.15.self_attention.dense.weight": "model-00003-of-00009.safetensors",
69
+ "transformer.h.15.self_attention.query_key_value.weight": "model-00003-of-00009.safetensors",
70
+ "transformer.h.16.ln_attn.bias": "model-00003-of-00009.safetensors",
71
+ "transformer.h.16.ln_attn.weight": "model-00003-of-00009.safetensors",
72
+ "transformer.h.16.ln_mlp.bias": "model-00003-of-00009.safetensors",
73
+ "transformer.h.16.ln_mlp.weight": "model-00003-of-00009.safetensors",
74
+ "transformer.h.16.mlp.dense_4h_to_h.weight": "model-00003-of-00009.safetensors",
75
+ "transformer.h.16.mlp.dense_h_to_4h.weight": "model-00003-of-00009.safetensors",
76
+ "transformer.h.16.self_attention.dense.weight": "model-00003-of-00009.safetensors",
77
+ "transformer.h.16.self_attention.query_key_value.weight": "model-00003-of-00009.safetensors",
78
+ "transformer.h.17.ln_attn.bias": "model-00003-of-00009.safetensors",
79
+ "transformer.h.17.ln_attn.weight": "model-00003-of-00009.safetensors",
80
+ "transformer.h.17.ln_mlp.bias": "model-00003-of-00009.safetensors",
81
+ "transformer.h.17.ln_mlp.weight": "model-00003-of-00009.safetensors",
82
+ "transformer.h.17.mlp.dense_4h_to_h.weight": "model-00003-of-00009.safetensors",
83
+ "transformer.h.17.mlp.dense_h_to_4h.weight": "model-00003-of-00009.safetensors",
84
+ "transformer.h.17.self_attention.dense.weight": "model-00003-of-00009.safetensors",
85
+ "transformer.h.17.self_attention.query_key_value.weight": "model-00003-of-00009.safetensors",
86
+ "transformer.h.18.ln_attn.bias": "model-00003-of-00009.safetensors",
87
+ "transformer.h.18.ln_attn.weight": "model-00003-of-00009.safetensors",
88
+ "transformer.h.18.ln_mlp.bias": "model-00003-of-00009.safetensors",
89
+ "transformer.h.18.ln_mlp.weight": "model-00003-of-00009.safetensors",
90
+ "transformer.h.18.mlp.dense_4h_to_h.weight": "model-00003-of-00009.safetensors",
91
+ "transformer.h.18.mlp.dense_h_to_4h.weight": "model-00003-of-00009.safetensors",
92
+ "transformer.h.18.self_attention.dense.weight": "model-00003-of-00009.safetensors",
93
+ "transformer.h.18.self_attention.query_key_value.weight": "model-00003-of-00009.safetensors",
94
+ "transformer.h.19.ln_attn.bias": "model-00003-of-00009.safetensors",
95
+ "transformer.h.19.ln_attn.weight": "model-00003-of-00009.safetensors",
96
+ "transformer.h.19.ln_mlp.bias": "model-00003-of-00009.safetensors",
97
+ "transformer.h.19.ln_mlp.weight": "model-00003-of-00009.safetensors",
98
+ "transformer.h.19.mlp.dense_4h_to_h.weight": "model-00003-of-00009.safetensors",
99
+ "transformer.h.19.mlp.dense_h_to_4h.weight": "model-00003-of-00009.safetensors",
100
+ "transformer.h.19.self_attention.dense.weight": "model-00003-of-00009.safetensors",
101
+ "transformer.h.19.self_attention.query_key_value.weight": "model-00003-of-00009.safetensors",
102
+ "transformer.h.2.ln_attn.bias": "model-00001-of-00009.safetensors",
103
+ "transformer.h.2.ln_attn.weight": "model-00001-of-00009.safetensors",
104
+ "transformer.h.2.ln_mlp.bias": "model-00001-of-00009.safetensors",
105
+ "transformer.h.2.ln_mlp.weight": "model-00001-of-00009.safetensors",
106
+ "transformer.h.2.mlp.dense_4h_to_h.weight": "model-00001-of-00009.safetensors",
107
+ "transformer.h.2.mlp.dense_h_to_4h.weight": "model-00001-of-00009.safetensors",
108
+ "transformer.h.2.self_attention.dense.weight": "model-00001-of-00009.safetensors",
109
+ "transformer.h.2.self_attention.query_key_value.weight": "model-00001-of-00009.safetensors",
110
+ "transformer.h.20.ln_attn.bias": "model-00003-of-00009.safetensors",
111
+ "transformer.h.20.ln_attn.weight": "model-00003-of-00009.safetensors",
112
+ "transformer.h.20.ln_mlp.bias": "model-00003-of-00009.safetensors",
113
+ "transformer.h.20.ln_mlp.weight": "model-00003-of-00009.safetensors",
114
+ "transformer.h.20.mlp.dense_4h_to_h.weight": "model-00004-of-00009.safetensors",
115
+ "transformer.h.20.mlp.dense_h_to_4h.weight": "model-00004-of-00009.safetensors",
116
+ "transformer.h.20.self_attention.dense.weight": "model-00003-of-00009.safetensors",
117
+ "transformer.h.20.self_attention.query_key_value.weight": "model-00003-of-00009.safetensors",
118
+ "transformer.h.21.ln_attn.bias": "model-00004-of-00009.safetensors",
119
+ "transformer.h.21.ln_attn.weight": "model-00004-of-00009.safetensors",
120
+ "transformer.h.21.ln_mlp.bias": "model-00004-of-00009.safetensors",
121
+ "transformer.h.21.ln_mlp.weight": "model-00004-of-00009.safetensors",
122
+ "transformer.h.21.mlp.dense_4h_to_h.weight": "model-00004-of-00009.safetensors",
123
+ "transformer.h.21.mlp.dense_h_to_4h.weight": "model-00004-of-00009.safetensors",
124
+ "transformer.h.21.self_attention.dense.weight": "model-00004-of-00009.safetensors",
125
+ "transformer.h.21.self_attention.query_key_value.weight": "model-00004-of-00009.safetensors",
126
+ "transformer.h.22.ln_attn.bias": "model-00004-of-00009.safetensors",
127
+ "transformer.h.22.ln_attn.weight": "model-00004-of-00009.safetensors",
128
+ "transformer.h.22.ln_mlp.bias": "model-00004-of-00009.safetensors",
129
+ "transformer.h.22.ln_mlp.weight": "model-00004-of-00009.safetensors",
130
+ "transformer.h.22.mlp.dense_4h_to_h.weight": "model-00004-of-00009.safetensors",
131
+ "transformer.h.22.mlp.dense_h_to_4h.weight": "model-00004-of-00009.safetensors",
132
+ "transformer.h.22.self_attention.dense.weight": "model-00004-of-00009.safetensors",
133
+ "transformer.h.22.self_attention.query_key_value.weight": "model-00004-of-00009.safetensors",
134
+ "transformer.h.23.ln_attn.bias": "model-00004-of-00009.safetensors",
135
+ "transformer.h.23.ln_attn.weight": "model-00004-of-00009.safetensors",
136
+ "transformer.h.23.ln_mlp.bias": "model-00004-of-00009.safetensors",
137
+ "transformer.h.23.ln_mlp.weight": "model-00004-of-00009.safetensors",
138
+ "transformer.h.23.mlp.dense_4h_to_h.weight": "model-00004-of-00009.safetensors",
139
+ "transformer.h.23.mlp.dense_h_to_4h.weight": "model-00004-of-00009.safetensors",
140
+ "transformer.h.23.self_attention.dense.weight": "model-00004-of-00009.safetensors",
141
+ "transformer.h.23.self_attention.query_key_value.weight": "model-00004-of-00009.safetensors",
142
+ "transformer.h.24.ln_attn.bias": "model-00004-of-00009.safetensors",
143
+ "transformer.h.24.ln_attn.weight": "model-00004-of-00009.safetensors",
144
+ "transformer.h.24.ln_mlp.bias": "model-00004-of-00009.safetensors",
145
+ "transformer.h.24.ln_mlp.weight": "model-00004-of-00009.safetensors",
146
+ "transformer.h.24.mlp.dense_4h_to_h.weight": "model-00004-of-00009.safetensors",
147
+ "transformer.h.24.mlp.dense_h_to_4h.weight": "model-00004-of-00009.safetensors",
148
+ "transformer.h.24.self_attention.dense.weight": "model-00004-of-00009.safetensors",
149
+ "transformer.h.24.self_attention.query_key_value.weight": "model-00004-of-00009.safetensors",
150
+ "transformer.h.25.ln_attn.bias": "model-00004-of-00009.safetensors",
151
+ "transformer.h.25.ln_attn.weight": "model-00004-of-00009.safetensors",
152
+ "transformer.h.25.ln_mlp.bias": "model-00004-of-00009.safetensors",
153
+ "transformer.h.25.ln_mlp.weight": "model-00004-of-00009.safetensors",
154
+ "transformer.h.25.mlp.dense_4h_to_h.weight": "model-00004-of-00009.safetensors",
155
+ "transformer.h.25.mlp.dense_h_to_4h.weight": "model-00004-of-00009.safetensors",
156
+ "transformer.h.25.self_attention.dense.weight": "model-00004-of-00009.safetensors",
157
+ "transformer.h.25.self_attention.query_key_value.weight": "model-00004-of-00009.safetensors",
158
+ "transformer.h.26.ln_attn.bias": "model-00004-of-00009.safetensors",
159
+ "transformer.h.26.ln_attn.weight": "model-00004-of-00009.safetensors",
160
+ "transformer.h.26.ln_mlp.bias": "model-00004-of-00009.safetensors",
161
+ "transformer.h.26.ln_mlp.weight": "model-00004-of-00009.safetensors",
162
+ "transformer.h.26.mlp.dense_4h_to_h.weight": "model-00004-of-00009.safetensors",
163
+ "transformer.h.26.mlp.dense_h_to_4h.weight": "model-00004-of-00009.safetensors",
164
+ "transformer.h.26.self_attention.dense.weight": "model-00004-of-00009.safetensors",
165
+ "transformer.h.26.self_attention.query_key_value.weight": "model-00004-of-00009.safetensors",
166
+ "transformer.h.27.ln_attn.bias": "model-00004-of-00009.safetensors",
167
+ "transformer.h.27.ln_attn.weight": "model-00004-of-00009.safetensors",
168
+ "transformer.h.27.ln_mlp.bias": "model-00004-of-00009.safetensors",
169
+ "transformer.h.27.ln_mlp.weight": "model-00004-of-00009.safetensors",
170
+ "transformer.h.27.mlp.dense_4h_to_h.weight": "model-00005-of-00009.safetensors",
171
+ "transformer.h.27.mlp.dense_h_to_4h.weight": "model-00005-of-00009.safetensors",
172
+ "transformer.h.27.self_attention.dense.weight": "model-00004-of-00009.safetensors",
173
+ "transformer.h.27.self_attention.query_key_value.weight": "model-00004-of-00009.safetensors",
174
+ "transformer.h.28.ln_attn.bias": "model-00005-of-00009.safetensors",
175
+ "transformer.h.28.ln_attn.weight": "model-00005-of-00009.safetensors",
176
+ "transformer.h.28.ln_mlp.bias": "model-00005-of-00009.safetensors",
177
+ "transformer.h.28.ln_mlp.weight": "model-00005-of-00009.safetensors",
178
+ "transformer.h.28.mlp.dense_4h_to_h.weight": "model-00005-of-00009.safetensors",
179
+ "transformer.h.28.mlp.dense_h_to_4h.weight": "model-00005-of-00009.safetensors",
180
+ "transformer.h.28.self_attention.dense.weight": "model-00005-of-00009.safetensors",
181
+ "transformer.h.28.self_attention.query_key_value.weight": "model-00005-of-00009.safetensors",
182
+ "transformer.h.29.ln_attn.bias": "model-00005-of-00009.safetensors",
183
+ "transformer.h.29.ln_attn.weight": "model-00005-of-00009.safetensors",
184
+ "transformer.h.29.ln_mlp.bias": "model-00005-of-00009.safetensors",
185
+ "transformer.h.29.ln_mlp.weight": "model-00005-of-00009.safetensors",
186
+ "transformer.h.29.mlp.dense_4h_to_h.weight": "model-00005-of-00009.safetensors",
187
+ "transformer.h.29.mlp.dense_h_to_4h.weight": "model-00005-of-00009.safetensors",
188
+ "transformer.h.29.self_attention.dense.weight": "model-00005-of-00009.safetensors",
189
+ "transformer.h.29.self_attention.query_key_value.weight": "model-00005-of-00009.safetensors",
190
+ "transformer.h.3.ln_attn.bias": "model-00001-of-00009.safetensors",
191
+ "transformer.h.3.ln_attn.weight": "model-00001-of-00009.safetensors",
192
+ "transformer.h.3.ln_mlp.bias": "model-00001-of-00009.safetensors",
193
+ "transformer.h.3.ln_mlp.weight": "model-00001-of-00009.safetensors",
194
+ "transformer.h.3.mlp.dense_4h_to_h.weight": "model-00001-of-00009.safetensors",
195
+ "transformer.h.3.mlp.dense_h_to_4h.weight": "model-00001-of-00009.safetensors",
196
+ "transformer.h.3.self_attention.dense.weight": "model-00001-of-00009.safetensors",
197
+ "transformer.h.3.self_attention.query_key_value.weight": "model-00001-of-00009.safetensors",
198
+ "transformer.h.30.ln_attn.bias": "model-00005-of-00009.safetensors",
199
+ "transformer.h.30.ln_attn.weight": "model-00005-of-00009.safetensors",
200
+ "transformer.h.30.ln_mlp.bias": "model-00005-of-00009.safetensors",
201
+ "transformer.h.30.ln_mlp.weight": "model-00005-of-00009.safetensors",
202
+ "transformer.h.30.mlp.dense_4h_to_h.weight": "model-00005-of-00009.safetensors",
203
+ "transformer.h.30.mlp.dense_h_to_4h.weight": "model-00005-of-00009.safetensors",
204
+ "transformer.h.30.self_attention.dense.weight": "model-00005-of-00009.safetensors",
205
+ "transformer.h.30.self_attention.query_key_value.weight": "model-00005-of-00009.safetensors",
206
+ "transformer.h.31.ln_attn.bias": "model-00005-of-00009.safetensors",
207
+ "transformer.h.31.ln_attn.weight": "model-00005-of-00009.safetensors",
208
+ "transformer.h.31.ln_mlp.bias": "model-00005-of-00009.safetensors",
209
+ "transformer.h.31.ln_mlp.weight": "model-00005-of-00009.safetensors",
210
+ "transformer.h.31.mlp.dense_4h_to_h.weight": "model-00005-of-00009.safetensors",
211
+ "transformer.h.31.mlp.dense_h_to_4h.weight": "model-00005-of-00009.safetensors",
212
+ "transformer.h.31.self_attention.dense.weight": "model-00005-of-00009.safetensors",
213
+ "transformer.h.31.self_attention.query_key_value.weight": "model-00005-of-00009.safetensors",
214
+ "transformer.h.32.ln_attn.bias": "model-00005-of-00009.safetensors",
215
+ "transformer.h.32.ln_attn.weight": "model-00005-of-00009.safetensors",
216
+ "transformer.h.32.ln_mlp.bias": "model-00005-of-00009.safetensors",
217
+ "transformer.h.32.ln_mlp.weight": "model-00005-of-00009.safetensors",
218
+ "transformer.h.32.mlp.dense_4h_to_h.weight": "model-00005-of-00009.safetensors",
219
+ "transformer.h.32.mlp.dense_h_to_4h.weight": "model-00005-of-00009.safetensors",
220
+ "transformer.h.32.self_attention.dense.weight": "model-00005-of-00009.safetensors",
221
+ "transformer.h.32.self_attention.query_key_value.weight": "model-00005-of-00009.safetensors",
222
+ "transformer.h.33.ln_attn.bias": "model-00005-of-00009.safetensors",
223
+ "transformer.h.33.ln_attn.weight": "model-00005-of-00009.safetensors",
224
+ "transformer.h.33.ln_mlp.bias": "model-00005-of-00009.safetensors",
225
+ "transformer.h.33.ln_mlp.weight": "model-00005-of-00009.safetensors",
226
+ "transformer.h.33.mlp.dense_4h_to_h.weight": "model-00005-of-00009.safetensors",
227
+ "transformer.h.33.mlp.dense_h_to_4h.weight": "model-00005-of-00009.safetensors",
228
+ "transformer.h.33.self_attention.dense.weight": "model-00005-of-00009.safetensors",
229
+ "transformer.h.33.self_attention.query_key_value.weight": "model-00005-of-00009.safetensors",
230
+ "transformer.h.34.ln_attn.bias": "model-00005-of-00009.safetensors",
231
+ "transformer.h.34.ln_attn.weight": "model-00005-of-00009.safetensors",
232
+ "transformer.h.34.ln_mlp.bias": "model-00005-of-00009.safetensors",
233
+ "transformer.h.34.ln_mlp.weight": "model-00005-of-00009.safetensors",
234
+ "transformer.h.34.mlp.dense_4h_to_h.weight": "model-00006-of-00009.safetensors",
235
+ "transformer.h.34.mlp.dense_h_to_4h.weight": "model-00006-of-00009.safetensors",
236
+ "transformer.h.34.self_attention.dense.weight": "model-00005-of-00009.safetensors",
237
+ "transformer.h.34.self_attention.query_key_value.weight": "model-00005-of-00009.safetensors",
238
+ "transformer.h.35.ln_attn.bias": "model-00006-of-00009.safetensors",
239
+ "transformer.h.35.ln_attn.weight": "model-00006-of-00009.safetensors",
240
+ "transformer.h.35.ln_mlp.bias": "model-00006-of-00009.safetensors",
241
+ "transformer.h.35.ln_mlp.weight": "model-00006-of-00009.safetensors",
242
+ "transformer.h.35.mlp.dense_4h_to_h.weight": "model-00006-of-00009.safetensors",
243
+ "transformer.h.35.mlp.dense_h_to_4h.weight": "model-00006-of-00009.safetensors",
244
+ "transformer.h.35.self_attention.dense.weight": "model-00006-of-00009.safetensors",
245
+ "transformer.h.35.self_attention.query_key_value.weight": "model-00006-of-00009.safetensors",
246
+ "transformer.h.36.ln_attn.bias": "model-00006-of-00009.safetensors",
247
+ "transformer.h.36.ln_attn.weight": "model-00006-of-00009.safetensors",
248
+ "transformer.h.36.ln_mlp.bias": "model-00006-of-00009.safetensors",
249
+ "transformer.h.36.ln_mlp.weight": "model-00006-of-00009.safetensors",
250
+ "transformer.h.36.mlp.dense_4h_to_h.weight": "model-00006-of-00009.safetensors",
251
+ "transformer.h.36.mlp.dense_h_to_4h.weight": "model-00006-of-00009.safetensors",
252
+ "transformer.h.36.self_attention.dense.weight": "model-00006-of-00009.safetensors",
253
+ "transformer.h.36.self_attention.query_key_value.weight": "model-00006-of-00009.safetensors",
254
+ "transformer.h.37.ln_attn.bias": "model-00006-of-00009.safetensors",
255
+ "transformer.h.37.ln_attn.weight": "model-00006-of-00009.safetensors",
256
+ "transformer.h.37.ln_mlp.bias": "model-00006-of-00009.safetensors",
257
+ "transformer.h.37.ln_mlp.weight": "model-00006-of-00009.safetensors",
258
+ "transformer.h.37.mlp.dense_4h_to_h.weight": "model-00006-of-00009.safetensors",
259
+ "transformer.h.37.mlp.dense_h_to_4h.weight": "model-00006-of-00009.safetensors",
260
+ "transformer.h.37.self_attention.dense.weight": "model-00006-of-00009.safetensors",
261
+ "transformer.h.37.self_attention.query_key_value.weight": "model-00006-of-00009.safetensors",
262
+ "transformer.h.38.ln_attn.bias": "model-00006-of-00009.safetensors",
263
+ "transformer.h.38.ln_attn.weight": "model-00006-of-00009.safetensors",
264
+ "transformer.h.38.ln_mlp.bias": "model-00006-of-00009.safetensors",
265
+ "transformer.h.38.ln_mlp.weight": "model-00006-of-00009.safetensors",
266
+ "transformer.h.38.mlp.dense_4h_to_h.weight": "model-00006-of-00009.safetensors",
267
+ "transformer.h.38.mlp.dense_h_to_4h.weight": "model-00006-of-00009.safetensors",
268
+ "transformer.h.38.self_attention.dense.weight": "model-00006-of-00009.safetensors",
269
+ "transformer.h.38.self_attention.query_key_value.weight": "model-00006-of-00009.safetensors",
270
+ "transformer.h.39.ln_attn.bias": "model-00006-of-00009.safetensors",
271
+ "transformer.h.39.ln_attn.weight": "model-00006-of-00009.safetensors",
272
+ "transformer.h.39.ln_mlp.bias": "model-00006-of-00009.safetensors",
273
+ "transformer.h.39.ln_mlp.weight": "model-00006-of-00009.safetensors",
274
+ "transformer.h.39.mlp.dense_4h_to_h.weight": "model-00006-of-00009.safetensors",
275
+ "transformer.h.39.mlp.dense_h_to_4h.weight": "model-00006-of-00009.safetensors",
276
+ "transformer.h.39.self_attention.dense.weight": "model-00006-of-00009.safetensors",
277
+ "transformer.h.39.self_attention.query_key_value.weight": "model-00006-of-00009.safetensors",
278
+ "transformer.h.4.ln_attn.bias": "model-00001-of-00009.safetensors",
279
+ "transformer.h.4.ln_attn.weight": "model-00001-of-00009.safetensors",
280
+ "transformer.h.4.ln_mlp.bias": "model-00001-of-00009.safetensors",
281
+ "transformer.h.4.ln_mlp.weight": "model-00001-of-00009.safetensors",
282
+ "transformer.h.4.mlp.dense_4h_to_h.weight": "model-00001-of-00009.safetensors",
283
+ "transformer.h.4.mlp.dense_h_to_4h.weight": "model-00001-of-00009.safetensors",
284
+ "transformer.h.4.self_attention.dense.weight": "model-00001-of-00009.safetensors",
285
+ "transformer.h.4.self_attention.query_key_value.weight": "model-00001-of-00009.safetensors",
286
+ "transformer.h.40.ln_attn.bias": "model-00006-of-00009.safetensors",
287
+ "transformer.h.40.ln_attn.weight": "model-00006-of-00009.safetensors",
288
+ "transformer.h.40.ln_mlp.bias": "model-00006-of-00009.safetensors",
289
+ "transformer.h.40.ln_mlp.weight": "model-00006-of-00009.safetensors",
290
+ "transformer.h.40.mlp.dense_4h_to_h.weight": "model-00006-of-00009.safetensors",
291
+ "transformer.h.40.mlp.dense_h_to_4h.weight": "model-00006-of-00009.safetensors",
292
+ "transformer.h.40.self_attention.dense.weight": "model-00006-of-00009.safetensors",
293
+ "transformer.h.40.self_attention.query_key_value.weight": "model-00006-of-00009.safetensors",
294
+ "transformer.h.41.ln_attn.bias": "model-00006-of-00009.safetensors",
295
+ "transformer.h.41.ln_attn.weight": "model-00006-of-00009.safetensors",
296
+ "transformer.h.41.ln_mlp.bias": "model-00006-of-00009.safetensors",
297
+ "transformer.h.41.ln_mlp.weight": "model-00006-of-00009.safetensors",
298
+ "transformer.h.41.mlp.dense_4h_to_h.weight": "model-00007-of-00009.safetensors",
299
+ "transformer.h.41.mlp.dense_h_to_4h.weight": "model-00007-of-00009.safetensors",
300
+ "transformer.h.41.self_attention.dense.weight": "model-00006-of-00009.safetensors",
301
+ "transformer.h.41.self_attention.query_key_value.weight": "model-00006-of-00009.safetensors",
302
+ "transformer.h.42.ln_attn.bias": "model-00007-of-00009.safetensors",
303
+ "transformer.h.42.ln_attn.weight": "model-00007-of-00009.safetensors",
304
+ "transformer.h.42.ln_mlp.bias": "model-00007-of-00009.safetensors",
305
+ "transformer.h.42.ln_mlp.weight": "model-00007-of-00009.safetensors",
306
+ "transformer.h.42.mlp.dense_4h_to_h.weight": "model-00007-of-00009.safetensors",
307
+ "transformer.h.42.mlp.dense_h_to_4h.weight": "model-00007-of-00009.safetensors",
308
+ "transformer.h.42.self_attention.dense.weight": "model-00007-of-00009.safetensors",
309
+ "transformer.h.42.self_attention.query_key_value.weight": "model-00007-of-00009.safetensors",
310
+ "transformer.h.43.ln_attn.bias": "model-00007-of-00009.safetensors",
311
+ "transformer.h.43.ln_attn.weight": "model-00007-of-00009.safetensors",
312
+ "transformer.h.43.ln_mlp.bias": "model-00007-of-00009.safetensors",
313
+ "transformer.h.43.ln_mlp.weight": "model-00007-of-00009.safetensors",
314
+ "transformer.h.43.mlp.dense_4h_to_h.weight": "model-00007-of-00009.safetensors",
315
+ "transformer.h.43.mlp.dense_h_to_4h.weight": "model-00007-of-00009.safetensors",
316
+ "transformer.h.43.self_attention.dense.weight": "model-00007-of-00009.safetensors",
317
+ "transformer.h.43.self_attention.query_key_value.weight": "model-00007-of-00009.safetensors",
318
+ "transformer.h.44.ln_attn.bias": "model-00007-of-00009.safetensors",
319
+ "transformer.h.44.ln_attn.weight": "model-00007-of-00009.safetensors",
320
+ "transformer.h.44.ln_mlp.bias": "model-00007-of-00009.safetensors",
321
+ "transformer.h.44.ln_mlp.weight": "model-00007-of-00009.safetensors",
322
+ "transformer.h.44.mlp.dense_4h_to_h.weight": "model-00007-of-00009.safetensors",
323
+ "transformer.h.44.mlp.dense_h_to_4h.weight": "model-00007-of-00009.safetensors",
324
+ "transformer.h.44.self_attention.dense.weight": "model-00007-of-00009.safetensors",
325
+ "transformer.h.44.self_attention.query_key_value.weight": "model-00007-of-00009.safetensors",
326
+ "transformer.h.45.ln_attn.bias": "model-00007-of-00009.safetensors",
327
+ "transformer.h.45.ln_attn.weight": "model-00007-of-00009.safetensors",
328
+ "transformer.h.45.ln_mlp.bias": "model-00007-of-00009.safetensors",
329
+ "transformer.h.45.ln_mlp.weight": "model-00007-of-00009.safetensors",
330
+ "transformer.h.45.mlp.dense_4h_to_h.weight": "model-00007-of-00009.safetensors",
331
+ "transformer.h.45.mlp.dense_h_to_4h.weight": "model-00007-of-00009.safetensors",
332
+ "transformer.h.45.self_attention.dense.weight": "model-00007-of-00009.safetensors",
333
+ "transformer.h.45.self_attention.query_key_value.weight": "model-00007-of-00009.safetensors",
334
+ "transformer.h.46.ln_attn.bias": "model-00007-of-00009.safetensors",
335
+ "transformer.h.46.ln_attn.weight": "model-00007-of-00009.safetensors",
336
+ "transformer.h.46.ln_mlp.bias": "model-00007-of-00009.safetensors",
337
+ "transformer.h.46.ln_mlp.weight": "model-00007-of-00009.safetensors",
338
+ "transformer.h.46.mlp.dense_4h_to_h.weight": "model-00007-of-00009.safetensors",
339
+ "transformer.h.46.mlp.dense_h_to_4h.weight": "model-00007-of-00009.safetensors",
340
+ "transformer.h.46.self_attention.dense.weight": "model-00007-of-00009.safetensors",
341
+ "transformer.h.46.self_attention.query_key_value.weight": "model-00007-of-00009.safetensors",
342
+ "transformer.h.47.ln_attn.bias": "model-00007-of-00009.safetensors",
343
+ "transformer.h.47.ln_attn.weight": "model-00007-of-00009.safetensors",
344
+ "transformer.h.47.ln_mlp.bias": "model-00007-of-00009.safetensors",
345
+ "transformer.h.47.ln_mlp.weight": "model-00007-of-00009.safetensors",
346
+ "transformer.h.47.mlp.dense_4h_to_h.weight": "model-00007-of-00009.safetensors",
347
+ "transformer.h.47.mlp.dense_h_to_4h.weight": "model-00007-of-00009.safetensors",
348
+ "transformer.h.47.self_attention.dense.weight": "model-00007-of-00009.safetensors",
349
+ "transformer.h.47.self_attention.query_key_value.weight": "model-00007-of-00009.safetensors",
350
+ "transformer.h.48.ln_attn.bias": "model-00007-of-00009.safetensors",
351
+ "transformer.h.48.ln_attn.weight": "model-00007-of-00009.safetensors",
352
+ "transformer.h.48.ln_mlp.bias": "model-00007-of-00009.safetensors",
353
+ "transformer.h.48.ln_mlp.weight": "model-00007-of-00009.safetensors",
354
+ "transformer.h.48.mlp.dense_4h_to_h.weight": "model-00008-of-00009.safetensors",
355
+ "transformer.h.48.mlp.dense_h_to_4h.weight": "model-00008-of-00009.safetensors",
356
+ "transformer.h.48.self_attention.dense.weight": "model-00007-of-00009.safetensors",
357
+ "transformer.h.48.self_attention.query_key_value.weight": "model-00007-of-00009.safetensors",
358
+ "transformer.h.49.ln_attn.bias": "model-00008-of-00009.safetensors",
359
+ "transformer.h.49.ln_attn.weight": "model-00008-of-00009.safetensors",
360
+ "transformer.h.49.ln_mlp.bias": "model-00008-of-00009.safetensors",
361
+ "transformer.h.49.ln_mlp.weight": "model-00008-of-00009.safetensors",
362
+ "transformer.h.49.mlp.dense_4h_to_h.weight": "model-00008-of-00009.safetensors",
363
+ "transformer.h.49.mlp.dense_h_to_4h.weight": "model-00008-of-00009.safetensors",
364
+ "transformer.h.49.self_attention.dense.weight": "model-00008-of-00009.safetensors",
365
+ "transformer.h.49.self_attention.query_key_value.weight": "model-00008-of-00009.safetensors",
366
+ "transformer.h.5.ln_attn.bias": "model-00001-of-00009.safetensors",
367
+ "transformer.h.5.ln_attn.weight": "model-00001-of-00009.safetensors",
368
+ "transformer.h.5.ln_mlp.bias": "model-00001-of-00009.safetensors",
369
+ "transformer.h.5.ln_mlp.weight": "model-00001-of-00009.safetensors",
370
+ "transformer.h.5.mlp.dense_4h_to_h.weight": "model-00001-of-00009.safetensors",
371
+ "transformer.h.5.mlp.dense_h_to_4h.weight": "model-00001-of-00009.safetensors",
372
+ "transformer.h.5.self_attention.dense.weight": "model-00001-of-00009.safetensors",
373
+ "transformer.h.5.self_attention.query_key_value.weight": "model-00001-of-00009.safetensors",
374
+ "transformer.h.50.ln_attn.bias": "model-00008-of-00009.safetensors",
375
+ "transformer.h.50.ln_attn.weight": "model-00008-of-00009.safetensors",
376
+ "transformer.h.50.ln_mlp.bias": "model-00008-of-00009.safetensors",
377
+ "transformer.h.50.ln_mlp.weight": "model-00008-of-00009.safetensors",
378
+ "transformer.h.50.mlp.dense_4h_to_h.weight": "model-00008-of-00009.safetensors",
379
+ "transformer.h.50.mlp.dense_h_to_4h.weight": "model-00008-of-00009.safetensors",
380
+ "transformer.h.50.self_attention.dense.weight": "model-00008-of-00009.safetensors",
381
+ "transformer.h.50.self_attention.query_key_value.weight": "model-00008-of-00009.safetensors",
382
+ "transformer.h.51.ln_attn.bias": "model-00008-of-00009.safetensors",
383
+ "transformer.h.51.ln_attn.weight": "model-00008-of-00009.safetensors",
384
+ "transformer.h.51.ln_mlp.bias": "model-00008-of-00009.safetensors",
385
+ "transformer.h.51.ln_mlp.weight": "model-00008-of-00009.safetensors",
386
+ "transformer.h.51.mlp.dense_4h_to_h.weight": "model-00008-of-00009.safetensors",
387
+ "transformer.h.51.mlp.dense_h_to_4h.weight": "model-00008-of-00009.safetensors",
388
+ "transformer.h.51.self_attention.dense.weight": "model-00008-of-00009.safetensors",
389
+ "transformer.h.51.self_attention.query_key_value.weight": "model-00008-of-00009.safetensors",
390
+ "transformer.h.52.ln_attn.bias": "model-00008-of-00009.safetensors",
391
+ "transformer.h.52.ln_attn.weight": "model-00008-of-00009.safetensors",
392
+ "transformer.h.52.ln_mlp.bias": "model-00008-of-00009.safetensors",
393
+ "transformer.h.52.ln_mlp.weight": "model-00008-of-00009.safetensors",
394
+ "transformer.h.52.mlp.dense_4h_to_h.weight": "model-00008-of-00009.safetensors",
395
+ "transformer.h.52.mlp.dense_h_to_4h.weight": "model-00008-of-00009.safetensors",
396
+ "transformer.h.52.self_attention.dense.weight": "model-00008-of-00009.safetensors",
397
+ "transformer.h.52.self_attention.query_key_value.weight": "model-00008-of-00009.safetensors",
398
+ "transformer.h.53.ln_attn.bias": "model-00008-of-00009.safetensors",
399
+ "transformer.h.53.ln_attn.weight": "model-00008-of-00009.safetensors",
400
+ "transformer.h.53.ln_mlp.bias": "model-00008-of-00009.safetensors",
401
+ "transformer.h.53.ln_mlp.weight": "model-00008-of-00009.safetensors",
402
+ "transformer.h.53.mlp.dense_4h_to_h.weight": "model-00008-of-00009.safetensors",
403
+ "transformer.h.53.mlp.dense_h_to_4h.weight": "model-00008-of-00009.safetensors",
404
+ "transformer.h.53.self_attention.dense.weight": "model-00008-of-00009.safetensors",
405
+ "transformer.h.53.self_attention.query_key_value.weight": "model-00008-of-00009.safetensors",
406
+ "transformer.h.54.ln_attn.bias": "model-00008-of-00009.safetensors",
407
+ "transformer.h.54.ln_attn.weight": "model-00008-of-00009.safetensors",
408
+ "transformer.h.54.ln_mlp.bias": "model-00008-of-00009.safetensors",
409
+ "transformer.h.54.ln_mlp.weight": "model-00008-of-00009.safetensors",
410
+ "transformer.h.54.mlp.dense_4h_to_h.weight": "model-00008-of-00009.safetensors",
411
+ "transformer.h.54.mlp.dense_h_to_4h.weight": "model-00008-of-00009.safetensors",
412
+ "transformer.h.54.self_attention.dense.weight": "model-00008-of-00009.safetensors",
413
+ "transformer.h.54.self_attention.query_key_value.weight": "model-00008-of-00009.safetensors",
414
+ "transformer.h.55.ln_attn.bias": "model-00008-of-00009.safetensors",
415
+ "transformer.h.55.ln_attn.weight": "model-00008-of-00009.safetensors",
416
+ "transformer.h.55.ln_mlp.bias": "model-00008-of-00009.safetensors",
417
+ "transformer.h.55.ln_mlp.weight": "model-00008-of-00009.safetensors",
418
+ "transformer.h.55.mlp.dense_4h_to_h.weight": "model-00009-of-00009.safetensors",
419
+ "transformer.h.55.mlp.dense_h_to_4h.weight": "model-00009-of-00009.safetensors",
420
+ "transformer.h.55.self_attention.dense.weight": "model-00008-of-00009.safetensors",
421
+ "transformer.h.55.self_attention.query_key_value.weight": "model-00008-of-00009.safetensors",
422
+ "transformer.h.56.ln_attn.bias": "model-00009-of-00009.safetensors",
423
+ "transformer.h.56.ln_attn.weight": "model-00009-of-00009.safetensors",
424
+ "transformer.h.56.ln_mlp.bias": "model-00009-of-00009.safetensors",
425
+ "transformer.h.56.ln_mlp.weight": "model-00009-of-00009.safetensors",
426
+ "transformer.h.56.mlp.dense_4h_to_h.weight": "model-00009-of-00009.safetensors",
427
+ "transformer.h.56.mlp.dense_h_to_4h.weight": "model-00009-of-00009.safetensors",
428
+ "transformer.h.56.self_attention.dense.weight": "model-00009-of-00009.safetensors",
429
+ "transformer.h.56.self_attention.query_key_value.weight": "model-00009-of-00009.safetensors",
430
+ "transformer.h.57.ln_attn.bias": "model-00009-of-00009.safetensors",
431
+ "transformer.h.57.ln_attn.weight": "model-00009-of-00009.safetensors",
432
+ "transformer.h.57.ln_mlp.bias": "model-00009-of-00009.safetensors",
433
+ "transformer.h.57.ln_mlp.weight": "model-00009-of-00009.safetensors",
434
+ "transformer.h.57.mlp.dense_4h_to_h.weight": "model-00009-of-00009.safetensors",
435
+ "transformer.h.57.mlp.dense_h_to_4h.weight": "model-00009-of-00009.safetensors",
436
+ "transformer.h.57.self_attention.dense.weight": "model-00009-of-00009.safetensors",
437
+ "transformer.h.57.self_attention.query_key_value.weight": "model-00009-of-00009.safetensors",
438
+ "transformer.h.58.ln_attn.bias": "model-00009-of-00009.safetensors",
439
+ "transformer.h.58.ln_attn.weight": "model-00009-of-00009.safetensors",
440
+ "transformer.h.58.ln_mlp.bias": "model-00009-of-00009.safetensors",
441
+ "transformer.h.58.ln_mlp.weight": "model-00009-of-00009.safetensors",
442
+ "transformer.h.58.mlp.dense_4h_to_h.weight": "model-00009-of-00009.safetensors",
443
+ "transformer.h.58.mlp.dense_h_to_4h.weight": "model-00009-of-00009.safetensors",
444
+ "transformer.h.58.self_attention.dense.weight": "model-00009-of-00009.safetensors",
445
+ "transformer.h.58.self_attention.query_key_value.weight": "model-00009-of-00009.safetensors",
446
+ "transformer.h.59.ln_attn.bias": "model-00009-of-00009.safetensors",
447
+ "transformer.h.59.ln_attn.weight": "model-00009-of-00009.safetensors",
448
+ "transformer.h.59.ln_mlp.bias": "model-00009-of-00009.safetensors",
449
+ "transformer.h.59.ln_mlp.weight": "model-00009-of-00009.safetensors",
450
+ "transformer.h.59.mlp.dense_4h_to_h.weight": "model-00009-of-00009.safetensors",
451
+ "transformer.h.59.mlp.dense_h_to_4h.weight": "model-00009-of-00009.safetensors",
452
+ "transformer.h.59.self_attention.dense.weight": "model-00009-of-00009.safetensors",
453
+ "transformer.h.59.self_attention.query_key_value.weight": "model-00009-of-00009.safetensors",
454
+ "transformer.h.6.ln_attn.bias": "model-00001-of-00009.safetensors",
455
+ "transformer.h.6.ln_attn.weight": "model-00001-of-00009.safetensors",
456
+ "transformer.h.6.ln_mlp.bias": "model-00001-of-00009.safetensors",
457
+ "transformer.h.6.ln_mlp.weight": "model-00001-of-00009.safetensors",
458
+ "transformer.h.6.mlp.dense_4h_to_h.weight": "model-00002-of-00009.safetensors",
459
+ "transformer.h.6.mlp.dense_h_to_4h.weight": "model-00002-of-00009.safetensors",
460
+ "transformer.h.6.self_attention.dense.weight": "model-00001-of-00009.safetensors",
461
+ "transformer.h.6.self_attention.query_key_value.weight": "model-00001-of-00009.safetensors",
462
+ "transformer.h.7.ln_attn.bias": "model-00002-of-00009.safetensors",
463
+ "transformer.h.7.ln_attn.weight": "model-00002-of-00009.safetensors",
464
+ "transformer.h.7.ln_mlp.bias": "model-00002-of-00009.safetensors",
465
+ "transformer.h.7.ln_mlp.weight": "model-00002-of-00009.safetensors",
466
+ "transformer.h.7.mlp.dense_4h_to_h.weight": "model-00002-of-00009.safetensors",
467
+ "transformer.h.7.mlp.dense_h_to_4h.weight": "model-00002-of-00009.safetensors",
468
+ "transformer.h.7.self_attention.dense.weight": "model-00002-of-00009.safetensors",
469
+ "transformer.h.7.self_attention.query_key_value.weight": "model-00002-of-00009.safetensors",
470
+ "transformer.h.8.ln_attn.bias": "model-00002-of-00009.safetensors",
471
+ "transformer.h.8.ln_attn.weight": "model-00002-of-00009.safetensors",
472
+ "transformer.h.8.ln_mlp.bias": "model-00002-of-00009.safetensors",
473
+ "transformer.h.8.ln_mlp.weight": "model-00002-of-00009.safetensors",
474
+ "transformer.h.8.mlp.dense_4h_to_h.weight": "model-00002-of-00009.safetensors",
475
+ "transformer.h.8.mlp.dense_h_to_4h.weight": "model-00002-of-00009.safetensors",
476
+ "transformer.h.8.self_attention.dense.weight": "model-00002-of-00009.safetensors",
477
+ "transformer.h.8.self_attention.query_key_value.weight": "model-00002-of-00009.safetensors",
478
+ "transformer.h.9.ln_attn.bias": "model-00002-of-00009.safetensors",
479
+ "transformer.h.9.ln_attn.weight": "model-00002-of-00009.safetensors",
480
+ "transformer.h.9.ln_mlp.bias": "model-00002-of-00009.safetensors",
481
+ "transformer.h.9.ln_mlp.weight": "model-00002-of-00009.safetensors",
482
+ "transformer.h.9.mlp.dense_4h_to_h.weight": "model-00002-of-00009.safetensors",
483
+ "transformer.h.9.mlp.dense_h_to_4h.weight": "model-00002-of-00009.safetensors",
484
+ "transformer.h.9.self_attention.dense.weight": "model-00002-of-00009.safetensors",
485
+ "transformer.h.9.self_attention.query_key_value.weight": "model-00002-of-00009.safetensors",
486
+ "transformer.ln_f.bias": "model-00009-of-00009.safetensors",
487
+ "transformer.ln_f.weight": "model-00009-of-00009.safetensors",
488
+ "transformer.word_embeddings.weight": "model-00001-of-00009.safetensors"
489
+ }
490
+ }
special_tokens_map.json ADDED
@@ -0,0 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ ">>TITLE<<",
4
+ ">>ABSTRACT<<",
5
+ ">>INTRODUCTION<<",
6
+ ">>SUMMARY<<",
7
+ ">>COMMENT<<",
8
+ ">>ANSWER<<",
9
+ ">>QUESTION<<",
10
+ ">>DOMAIN<<",
11
+ ">>PREFIX<<",
12
+ ">>SUFFIX<<",
13
+ ">>MIDDLE<<"
14
+ ],
15
+ "eos_token": "<|endoftext|>"
16
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "clean_up_tokenization_spaces": true,
4
+ "eos_token": "<|endoftext|>",
5
+ "model_max_length": 2048,
6
+ "tokenizer_class": "PreTrainedTokenizerFast"
7
+ }