Hugging Face's logo

Tele-AI
/
TeleChat2.5-115B

Safetensors
telechat
custom_code
Model card Files
xet
Community
jianfbsdc commited on
Commit
3ceb32e
·
verified ·
1 Parent(s): 2648018

Upload folder using huggingface_hub

Browse files
Files changed (5) hide show
  1. config.json +49 -0
  2. generation_config.json +15 -0
  3. model.safetensors.index.json +970 -0
  4. special_tokens_map.json +35 -0
  5. tokenizer_config.json +114 -0
config.json ADDED
@@ -0,0 +1,49 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "apply_residual_connection_post_layernorm": false,
3
+ "architectures": [
4
+ "TeleChat2ForCausalLM"
5
+ ],
6
+ "attention_dropout": 0.0,
7
+ "attention_softmax_in_fp32": true,
8
+ "auto_map": {
9
+ "AutoConfig": "configuration_telechat2.Telechat2Config",
10
+ "AutoModelForCausalLM": "modeling_telechat2.Telechat2ForCausalLM"
11
+ },
12
+ "base_seqlen": 32768,
13
+ "bias_dropout_fusion": true,
14
+ "bos_token_id": 1,
15
+ "embed_layernorm": false,
16
+ "eos_token_id": 2,
17
+ "ffn_hidden_size": 40960,
18
+ "flash_attn": true,
19
+ "hidden_dropout": 0.0,
20
+ "hidden_size": 8192,
21
+ "initializer_range": 0.02,
22
+ "layer_norm_epsilon": 1e-05,
23
+ "logn": true,
24
+ "masked_softmax_fusion": true,
25
+ "max_position_embeddings": 32768,
26
+ "model_type": "telechat",
27
+ "n_head": 64,
28
+ "n_inner": null,
29
+ "n_layer": 96,
30
+ "num_key_value_heads": 8,
31
+ "pad_token_id": 3,
32
+ "pretraining_tp": 2,
33
+ "rope_scaling": {
34
+ "factor": 1.0,
35
+ "rope_type": "dynamic"
36
+ },
37
+ "rope_theta": 1000000,
38
+ "seq_length": 32768,
39
+ "skip_bias_add": false,
40
+ "skip_bias_add_qkv": false,
41
+ "slow_but_exact": false,
42
+ "tie_word_embeddings": false,
43
+ "torch_dtype": "bfloat16",
44
+ "training_seqlen": 32768,
45
+ "transformers_version": "4.46.0",
46
+ "unk_token_id": 0,
47
+ "use_cache": true,
48
+ "vocab_size": 131072
49
+ }
generation_config.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 1,
3
+ "bot_token_id": 5,
4
+ "do_sample": true,
5
+ "eos_token_id": 2,
6
+ "max_new_tokens": 1000,
7
+ "pad_token_id": 3,
8
+ "repetition_penalty": 1.03,
9
+ "start_token_id": 1,
10
+ "temperature": 0.3,
11
+ "top_k": 5,
12
+ "top_p": 0.85,
13
+ "transformers_version": "4.46.0",
14
+ "user_token_id": 4
15
+ }
model.safetensors.index.json ADDED
@@ -0,0 +1,970 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 226565832704
4
+ },
5
+ "weight_map": {
6
+ "lm_head.weight": "model-00059-of-00059.safetensors",
7
+ "transformer.h.0.input_layernorm.weight": "model-00001-of-00059.safetensors",
8
+ "transformer.h.0.mlp.down_proj.bias": "model-00002-of-00059.safetensors",
9
+ "transformer.h.0.mlp.down_proj.weight": "model-00002-of-00059.safetensors",
10
+ "transformer.h.0.mlp.gate_proj.weight": "model-00001-of-00059.safetensors",
11
+ "transformer.h.0.mlp.up_proj.weight": "model-00001-of-00059.safetensors",
12
+ "transformer.h.0.post_attention_layernorm.weight": "model-00001-of-00059.safetensors",
13
+ "transformer.h.0.self_attention.dense.bias": "model-00001-of-00059.safetensors",
14
+ "transformer.h.0.self_attention.dense.weight": "model-00001-of-00059.safetensors",
15
+ "transformer.h.0.self_attention.key_value.weight": "model-00001-of-00059.safetensors",
16
+ "transformer.h.0.self_attention.query.weight": "model-00001-of-00059.safetensors",
17
+ "transformer.h.1.input_layernorm.weight": "model-00002-of-00059.safetensors",
18
+ "transformer.h.1.mlp.down_proj.bias": "model-00002-of-00059.safetensors",
19
+ "transformer.h.1.mlp.down_proj.weight": "model-00002-of-00059.safetensors",
20
+ "transformer.h.1.mlp.gate_proj.weight": "model-00002-of-00059.safetensors",
21
+ "transformer.h.1.mlp.up_proj.weight": "model-00002-of-00059.safetensors",
22
+ "transformer.h.1.post_attention_layernorm.weight": "model-00002-of-00059.safetensors",
23
+ "transformer.h.1.self_attention.dense.bias": "model-00002-of-00059.safetensors",
24
+ "transformer.h.1.self_attention.dense.weight": "model-00002-of-00059.safetensors",
25
+ "transformer.h.1.self_attention.key_value.weight": "model-00002-of-00059.safetensors",
26
+ "transformer.h.1.self_attention.query.weight": "model-00002-of-00059.safetensors",
27
+ "transformer.h.10.input_layernorm.weight": "model-00007-of-00059.safetensors",
28
+ "transformer.h.10.mlp.down_proj.bias": "model-00008-of-00059.safetensors",
29
+ "transformer.h.10.mlp.down_proj.weight": "model-00008-of-00059.safetensors",
30
+ "transformer.h.10.mlp.gate_proj.weight": "model-00007-of-00059.safetensors",
31
+ "transformer.h.10.mlp.up_proj.weight": "model-00007-of-00059.safetensors",
32
+ "transformer.h.10.post_attention_layernorm.weight": "model-00007-of-00059.safetensors",
33
+ "transformer.h.10.self_attention.dense.bias": "model-00007-of-00059.safetensors",
34
+ "transformer.h.10.self_attention.dense.weight": "model-00007-of-00059.safetensors",
35
+ "transformer.h.10.self_attention.key_value.weight": "model-00007-of-00059.safetensors",
36
+ "transformer.h.10.self_attention.query.weight": "model-00007-of-00059.safetensors",
37
+ "transformer.h.11.input_layernorm.weight": "model-00008-of-00059.safetensors",
38
+ "transformer.h.11.mlp.down_proj.bias": "model-00008-of-00059.safetensors",
39
+ "transformer.h.11.mlp.down_proj.weight": "model-00008-of-00059.safetensors",
40
+ "transformer.h.11.mlp.gate_proj.weight": "model-00008-of-00059.safetensors",
41
+ "transformer.h.11.mlp.up_proj.weight": "model-00008-of-00059.safetensors",
42
+ "transformer.h.11.post_attention_layernorm.weight": "model-00008-of-00059.safetensors",
43
+ "transformer.h.11.self_attention.dense.bias": "model-00008-of-00059.safetensors",
44
+ "transformer.h.11.self_attention.dense.weight": "model-00008-of-00059.safetensors",
45
+ "transformer.h.11.self_attention.key_value.weight": "model-00008-of-00059.safetensors",
46
+ "transformer.h.11.self_attention.query.weight": "model-00008-of-00059.safetensors",
47
+ "transformer.h.12.input_layernorm.weight": "model-00008-of-00059.safetensors",
48
+ "transformer.h.12.mlp.down_proj.bias": "model-00009-of-00059.safetensors",
49
+ "transformer.h.12.mlp.down_proj.weight": "model-00009-of-00059.safetensors",
50
+ "transformer.h.12.mlp.gate_proj.weight": "model-00008-of-00059.safetensors",
51
+ "transformer.h.12.mlp.up_proj.weight": "model-00009-of-00059.safetensors",
52
+ "transformer.h.12.post_attention_layernorm.weight": "model-00008-of-00059.safetensors",
53
+ "transformer.h.12.self_attention.dense.bias": "model-00008-of-00059.safetensors",
54
+ "transformer.h.12.self_attention.dense.weight": "model-00008-of-00059.safetensors",
55
+ "transformer.h.12.self_attention.key_value.weight": "model-00008-of-00059.safetensors",
56
+ "transformer.h.12.self_attention.query.weight": "model-00008-of-00059.safetensors",
57
+ "transformer.h.13.input_layernorm.weight": "model-00009-of-00059.safetensors",
58
+ "transformer.h.13.mlp.down_proj.bias": "model-00009-of-00059.safetensors",
59
+ "transformer.h.13.mlp.down_proj.weight": "model-00009-of-00059.safetensors",
60
+ "transformer.h.13.mlp.gate_proj.weight": "model-00009-of-00059.safetensors",
61
+ "transformer.h.13.mlp.up_proj.weight": "model-00009-of-00059.safetensors",
62
+ "transformer.h.13.post_attention_layernorm.weight": "model-00009-of-00059.safetensors",
63
+ "transformer.h.13.self_attention.dense.bias": "model-00009-of-00059.safetensors",
64
+ "transformer.h.13.self_attention.dense.weight": "model-00009-of-00059.safetensors",
65
+ "transformer.h.13.self_attention.key_value.weight": "model-00009-of-00059.safetensors",
66
+ "transformer.h.13.self_attention.query.weight": "model-00009-of-00059.safetensors",
67
+ "transformer.h.14.input_layernorm.weight": "model-00009-of-00059.safetensors",
68
+ "transformer.h.14.mlp.down_proj.bias": "model-00010-of-00059.safetensors",
69
+ "transformer.h.14.mlp.down_proj.weight": "model-00010-of-00059.safetensors",
70
+ "transformer.h.14.mlp.gate_proj.weight": "model-00010-of-00059.safetensors",
71
+ "transformer.h.14.mlp.up_proj.weight": "model-00010-of-00059.safetensors",
72
+ "transformer.h.14.post_attention_layernorm.weight": "model-00009-of-00059.safetensors",
73
+ "transformer.h.14.self_attention.dense.bias": "model-00009-of-00059.safetensors",
74
+ "transformer.h.14.self_attention.dense.weight": "model-00009-of-00059.safetensors",
75
+ "transformer.h.14.self_attention.key_value.weight": "model-00009-of-00059.safetensors",
76
+ "transformer.h.14.self_attention.query.weight": "model-00009-of-00059.safetensors",
77
+ "transformer.h.15.input_layernorm.weight": "model-00010-of-00059.safetensors",
78
+ "transformer.h.15.mlp.down_proj.bias": "model-00011-of-00059.safetensors",
79
+ "transformer.h.15.mlp.down_proj.weight": "model-00011-of-00059.safetensors",
80
+ "transformer.h.15.mlp.gate_proj.weight": "model-00010-of-00059.safetensors",
81
+ "transformer.h.15.mlp.up_proj.weight": "model-00010-of-00059.safetensors",
82
+ "transformer.h.15.post_attention_layernorm.weight": "model-00010-of-00059.safetensors",
83
+ "transformer.h.15.self_attention.dense.bias": "model-00010-of-00059.safetensors",
84
+ "transformer.h.15.self_attention.dense.weight": "model-00010-of-00059.safetensors",
85
+ "transformer.h.15.self_attention.key_value.weight": "model-00010-of-00059.safetensors",
86
+ "transformer.h.15.self_attention.query.weight": "model-00010-of-00059.safetensors",
87
+ "transformer.h.16.input_layernorm.weight": "model-00011-of-00059.safetensors",
88
+ "transformer.h.16.mlp.down_proj.bias": "model-00011-of-00059.safetensors",
89
+ "transformer.h.16.mlp.down_proj.weight": "model-00011-of-00059.safetensors",
90
+ "transformer.h.16.mlp.gate_proj.weight": "model-00011-of-00059.safetensors",
91
+ "transformer.h.16.mlp.up_proj.weight": "model-00011-of-00059.safetensors",
92
+ "transformer.h.16.post_attention_layernorm.weight": "model-00011-of-00059.safetensors",
93
+ "transformer.h.16.self_attention.dense.bias": "model-00011-of-00059.safetensors",
94
+ "transformer.h.16.self_attention.dense.weight": "model-00011-of-00059.safetensors",
95
+ "transformer.h.16.self_attention.key_value.weight": "model-00011-of-00059.safetensors",
96
+ "transformer.h.16.self_attention.query.weight": "model-00011-of-00059.safetensors",
97
+ "transformer.h.17.input_layernorm.weight": "model-00011-of-00059.safetensors",
98
+ "transformer.h.17.mlp.down_proj.bias": "model-00012-of-00059.safetensors",
99
+ "transformer.h.17.mlp.down_proj.weight": "model-00012-of-00059.safetensors",
100
+ "transformer.h.17.mlp.gate_proj.weight": "model-00011-of-00059.safetensors",
101
+ "transformer.h.17.mlp.up_proj.weight": "model-00012-of-00059.safetensors",
102
+ "transformer.h.17.post_attention_layernorm.weight": "model-00011-of-00059.safetensors",
103
+ "transformer.h.17.self_attention.dense.bias": "model-00011-of-00059.safetensors",
104
+ "transformer.h.17.self_attention.dense.weight": "model-00011-of-00059.safetensors",
105
+ "transformer.h.17.self_attention.key_value.weight": "model-00011-of-00059.safetensors",
106
+ "transformer.h.17.self_attention.query.weight": "model-00011-of-00059.safetensors",
107
+ "transformer.h.18.input_layernorm.weight": "model-00012-of-00059.safetensors",
108
+ "transformer.h.18.mlp.down_proj.bias": "model-00012-of-00059.safetensors",
109
+ "transformer.h.18.mlp.down_proj.weight": "model-00012-of-00059.safetensors",
110
+ "transformer.h.18.mlp.gate_proj.weight": "model-00012-of-00059.safetensors",
111
+ "transformer.h.18.mlp.up_proj.weight": "model-00012-of-00059.safetensors",
112
+ "transformer.h.18.post_attention_layernorm.weight": "model-00012-of-00059.safetensors",
113
+ "transformer.h.18.self_attention.dense.bias": "model-00012-of-00059.safetensors",
114
+ "transformer.h.18.self_attention.dense.weight": "model-00012-of-00059.safetensors",
115
+ "transformer.h.18.self_attention.key_value.weight": "model-00012-of-00059.safetensors",
116
+ "transformer.h.18.self_attention.query.weight": "model-00012-of-00059.safetensors",
117
+ "transformer.h.19.input_layernorm.weight": "model-00012-of-00059.safetensors",
118
+ "transformer.h.19.mlp.down_proj.bias": "model-00013-of-00059.safetensors",
119
+ "transformer.h.19.mlp.down_proj.weight": "model-00013-of-00059.safetensors",
120
+ "transformer.h.19.mlp.gate_proj.weight": "model-00013-of-00059.safetensors",
121
+ "transformer.h.19.mlp.up_proj.weight": "model-00013-of-00059.safetensors",
122
+ "transformer.h.19.post_attention_layernorm.weight": "model-00012-of-00059.safetensors",
123
+ "transformer.h.19.self_attention.dense.bias": "model-00012-of-00059.safetensors",
124
+ "transformer.h.19.self_attention.dense.weight": "model-00012-of-00059.safetensors",
125
+ "transformer.h.19.self_attention.key_value.weight": "model-00012-of-00059.safetensors",
126
+ "transformer.h.19.self_attention.query.weight": "model-00012-of-00059.safetensors",
127
+ "transformer.h.2.input_layernorm.weight": "model-00002-of-00059.safetensors",
128
+ "transformer.h.2.mlp.down_proj.bias": "model-00003-of-00059.safetensors",
129
+ "transformer.h.2.mlp.down_proj.weight": "model-00003-of-00059.safetensors",
130
+ "transformer.h.2.mlp.gate_proj.weight": "model-00002-of-00059.safetensors",
131
+ "transformer.h.2.mlp.up_proj.weight": "model-00003-of-00059.safetensors",
132
+ "transformer.h.2.post_attention_layernorm.weight": "model-00002-of-00059.safetensors",
133
+ "transformer.h.2.self_attention.dense.bias": "model-00002-of-00059.safetensors",
134
+ "transformer.h.2.self_attention.dense.weight": "model-00002-of-00059.safetensors",
135
+ "transformer.h.2.self_attention.key_value.weight": "model-00002-of-00059.safetensors",
136
+ "transformer.h.2.self_attention.query.weight": "model-00002-of-00059.safetensors",
137
+ "transformer.h.20.input_layernorm.weight": "model-00013-of-00059.safetensors",
138
+ "transformer.h.20.mlp.down_proj.bias": "model-00014-of-00059.safetensors",
139
+ "transformer.h.20.mlp.down_proj.weight": "model-00014-of-00059.safetensors",
140
+ "transformer.h.20.mlp.gate_proj.weight": "model-00013-of-00059.safetensors",
141
+ "transformer.h.20.mlp.up_proj.weight": "model-00013-of-00059.safetensors",
142
+ "transformer.h.20.post_attention_layernorm.weight": "model-00013-of-00059.safetensors",
143
+ "transformer.h.20.self_attention.dense.bias": "model-00013-of-00059.safetensors",
144
+ "transformer.h.20.self_attention.dense.weight": "model-00013-of-00059.safetensors",
145
+ "transformer.h.20.self_attention.key_value.weight": "model-00013-of-00059.safetensors",
146
+ "transformer.h.20.self_attention.query.weight": "model-00013-of-00059.safetensors",
147
+ "transformer.h.21.input_layernorm.weight": "model-00014-of-00059.safetensors",
148
+ "transformer.h.21.mlp.down_proj.bias": "model-00014-of-00059.safetensors",
149
+ "transformer.h.21.mlp.down_proj.weight": "model-00014-of-00059.safetensors",
150
+ "transformer.h.21.mlp.gate_proj.weight": "model-00014-of-00059.safetensors",
151
+ "transformer.h.21.mlp.up_proj.weight": "model-00014-of-00059.safetensors",
152
+ "transformer.h.21.post_attention_layernorm.weight": "model-00014-of-00059.safetensors",
153
+ "transformer.h.21.self_attention.dense.bias": "model-00014-of-00059.safetensors",
154
+ "transformer.h.21.self_attention.dense.weight": "model-00014-of-00059.safetensors",
155
+ "transformer.h.21.self_attention.key_value.weight": "model-00014-of-00059.safetensors",
156
+ "transformer.h.21.self_attention.query.weight": "model-00014-of-00059.safetensors",
157
+ "transformer.h.22.input_layernorm.weight": "model-00014-of-00059.safetensors",
158
+ "transformer.h.22.mlp.down_proj.bias": "model-00015-of-00059.safetensors",
159
+ "transformer.h.22.mlp.down_proj.weight": "model-00015-of-00059.safetensors",
160
+ "transformer.h.22.mlp.gate_proj.weight": "model-00014-of-00059.safetensors",
161
+ "transformer.h.22.mlp.up_proj.weight": "model-00015-of-00059.safetensors",
162
+ "transformer.h.22.post_attention_layernorm.weight": "model-00014-of-00059.safetensors",
163
+ "transformer.h.22.self_attention.dense.bias": "model-00014-of-00059.safetensors",
164
+ "transformer.h.22.self_attention.dense.weight": "model-00014-of-00059.safetensors",
165
+ "transformer.h.22.self_attention.key_value.weight": "model-00014-of-00059.safetensors",
166
+ "transformer.h.22.self_attention.query.weight": "model-00014-of-00059.safetensors",
167
+ "transformer.h.23.input_layernorm.weight": "model-00015-of-00059.safetensors",
168
+ "transformer.h.23.mlp.down_proj.bias": "model-00015-of-00059.safetensors",
169
+ "transformer.h.23.mlp.down_proj.weight": "model-00015-of-00059.safetensors",
170
+ "transformer.h.23.mlp.gate_proj.weight": "model-00015-of-00059.safetensors",
171
+ "transformer.h.23.mlp.up_proj.weight": "model-00015-of-00059.safetensors",
172
+ "transformer.h.23.post_attention_layernorm.weight": "model-00015-of-00059.safetensors",
173
+ "transformer.h.23.self_attention.dense.bias": "model-00015-of-00059.safetensors",
174
+ "transformer.h.23.self_attention.dense.weight": "model-00015-of-00059.safetensors",
175
+ "transformer.h.23.self_attention.key_value.weight": "model-00015-of-00059.safetensors",
176
+ "transformer.h.23.self_attention.query.weight": "model-00015-of-00059.safetensors",
177
+ "transformer.h.24.input_layernorm.weight": "model-00015-of-00059.safetensors",
178
+ "transformer.h.24.mlp.down_proj.bias": "model-00016-of-00059.safetensors",
179
+ "transformer.h.24.mlp.down_proj.weight": "model-00016-of-00059.safetensors",
180
+ "transformer.h.24.mlp.gate_proj.weight": "model-00016-of-00059.safetensors",
181
+ "transformer.h.24.mlp.up_proj.weight": "model-00016-of-00059.safetensors",
182
+ "transformer.h.24.post_attention_layernorm.weight": "model-00015-of-00059.safetensors",
183
+ "transformer.h.24.self_attention.dense.bias": "model-00015-of-00059.safetensors",
184
+ "transformer.h.24.self_attention.dense.weight": "model-00015-of-00059.safetensors",
185
+ "transformer.h.24.self_attention.key_value.weight": "model-00015-of-00059.safetensors",
186
+ "transformer.h.24.self_attention.query.weight": "model-00015-of-00059.safetensors",
187
+ "transformer.h.25.input_layernorm.weight": "model-00016-of-00059.safetensors",
188
+ "transformer.h.25.mlp.down_proj.bias": "model-00017-of-00059.safetensors",
189
+ "transformer.h.25.mlp.down_proj.weight": "model-00017-of-00059.safetensors",
190
+ "transformer.h.25.mlp.gate_proj.weight": "model-00016-of-00059.safetensors",
191
+ "transformer.h.25.mlp.up_proj.weight": "model-00016-of-00059.safetensors",
192
+ "transformer.h.25.post_attention_layernorm.weight": "model-00016-of-00059.safetensors",
193
+ "transformer.h.25.self_attention.dense.bias": "model-00016-of-00059.safetensors",
194
+ "transformer.h.25.self_attention.dense.weight": "model-00016-of-00059.safetensors",
195
+ "transformer.h.25.self_attention.key_value.weight": "model-00016-of-00059.safetensors",
196
+ "transformer.h.25.self_attention.query.weight": "model-00016-of-00059.safetensors",
197
+ "transformer.h.26.input_layernorm.weight": "model-00017-of-00059.safetensors",
198
+ "transformer.h.26.mlp.down_proj.bias": "model-00017-of-00059.safetensors",
199
+ "transformer.h.26.mlp.down_proj.weight": "model-00017-of-00059.safetensors",
200
+ "transformer.h.26.mlp.gate_proj.weight": "model-00017-of-00059.safetensors",
201
+ "transformer.h.26.mlp.up_proj.weight": "model-00017-of-00059.safetensors",
202
+ "transformer.h.26.post_attention_layernorm.weight": "model-00017-of-00059.safetensors",
203
+ "transformer.h.26.self_attention.dense.bias": "model-00017-of-00059.safetensors",
204
+ "transformer.h.26.self_attention.dense.weight": "model-00017-of-00059.safetensors",
205
+ "transformer.h.26.self_attention.key_value.weight": "model-00017-of-00059.safetensors",
206
+ "transformer.h.26.self_attention.query.weight": "model-00017-of-00059.safetensors",
207
+ "transformer.h.27.input_layernorm.weight": "model-00017-of-00059.safetensors",
208
+ "transformer.h.27.mlp.down_proj.bias": "model-00018-of-00059.safetensors",
209
+ "transformer.h.27.mlp.down_proj.weight": "model-00018-of-00059.safetensors",
210
+ "transformer.h.27.mlp.gate_proj.weight": "model-00017-of-00059.safetensors",
211
+ "transformer.h.27.mlp.up_proj.weight": "model-00018-of-00059.safetensors",
212
+ "transformer.h.27.post_attention_layernorm.weight": "model-00017-of-00059.safetensors",
213
+ "transformer.h.27.self_attention.dense.bias": "model-00017-of-00059.safetensors",
214
+ "transformer.h.27.self_attention.dense.weight": "model-00017-of-00059.safetensors",
215
+ "transformer.h.27.self_attention.key_value.weight": "model-00017-of-00059.safetensors",
216
+ "transformer.h.27.self_attention.query.weight": "model-00017-of-00059.safetensors",
217
+ "transformer.h.28.input_layernorm.weight": "model-00018-of-00059.safetensors",
218
+ "transformer.h.28.mlp.down_proj.bias": "model-00018-of-00059.safetensors",
219
+ "transformer.h.28.mlp.down_proj.weight": "model-00018-of-00059.safetensors",
220
+ "transformer.h.28.mlp.gate_proj.weight": "model-00018-of-00059.safetensors",
221
+ "transformer.h.28.mlp.up_proj.weight": "model-00018-of-00059.safetensors",
222
+ "transformer.h.28.post_attention_layernorm.weight": "model-00018-of-00059.safetensors",
223
+ "transformer.h.28.self_attention.dense.bias": "model-00018-of-00059.safetensors",
224
+ "transformer.h.28.self_attention.dense.weight": "model-00018-of-00059.safetensors",
225
+ "transformer.h.28.self_attention.key_value.weight": "model-00018-of-00059.safetensors",
226
+ "transformer.h.28.self_attention.query.weight": "model-00018-of-00059.safetensors",
227
+ "transformer.h.29.input_layernorm.weight": "model-00018-of-00059.safetensors",
228
+ "transformer.h.29.mlp.down_proj.bias": "model-00019-of-00059.safetensors",
229
+ "transformer.h.29.mlp.down_proj.weight": "model-00019-of-00059.safetensors",
230
+ "transformer.h.29.mlp.gate_proj.weight": "model-00019-of-00059.safetensors",
231
+ "transformer.h.29.mlp.up_proj.weight": "model-00019-of-00059.safetensors",
232
+ "transformer.h.29.post_attention_layernorm.weight": "model-00018-of-00059.safetensors",
233
+ "transformer.h.29.self_attention.dense.bias": "model-00018-of-00059.safetensors",
234
+ "transformer.h.29.self_attention.dense.weight": "model-00018-of-00059.safetensors",
235
+ "transformer.h.29.self_attention.key_value.weight": "model-00018-of-00059.safetensors",
236
+ "transformer.h.29.self_attention.query.weight": "model-00018-of-00059.safetensors",
237
+ "transformer.h.3.input_layernorm.weight": "model-00003-of-00059.safetensors",
238
+ "transformer.h.3.mlp.down_proj.bias": "model-00003-of-00059.safetensors",
239
+ "transformer.h.3.mlp.down_proj.weight": "model-00003-of-00059.safetensors",
240
+ "transformer.h.3.mlp.gate_proj.weight": "model-00003-of-00059.safetensors",
241
+ "transformer.h.3.mlp.up_proj.weight": "model-00003-of-00059.safetensors",
242
+ "transformer.h.3.post_attention_layernorm.weight": "model-00003-of-00059.safetensors",
243
+ "transformer.h.3.self_attention.dense.bias": "model-00003-of-00059.safetensors",
244
+ "transformer.h.3.self_attention.dense.weight": "model-00003-of-00059.safetensors",
245
+ "transformer.h.3.self_attention.key_value.weight": "model-00003-of-00059.safetensors",
246
+ "transformer.h.3.self_attention.query.weight": "model-00003-of-00059.safetensors",
247
+ "transformer.h.30.input_layernorm.weight": "model-00019-of-00059.safetensors",
248
+ "transformer.h.30.mlp.down_proj.bias": "model-00020-of-00059.safetensors",
249
+ "transformer.h.30.mlp.down_proj.weight": "model-00020-of-00059.safetensors",
250
+ "transformer.h.30.mlp.gate_proj.weight": "model-00019-of-00059.safetensors",
251
+ "transformer.h.30.mlp.up_proj.weight": "model-00019-of-00059.safetensors",
252
+ "transformer.h.30.post_attention_layernorm.weight": "model-00019-of-00059.safetensors",
253
+ "transformer.h.30.self_attention.dense.bias": "model-00019-of-00059.safetensors",
254
+ "transformer.h.30.self_attention.dense.weight": "model-00019-of-00059.safetensors",
255
+ "transformer.h.30.self_attention.key_value.weight": "model-00019-of-00059.safetensors",
256
+ "transformer.h.30.self_attention.query.weight": "model-00019-of-00059.safetensors",
257
+ "transformer.h.31.input_layernorm.weight": "model-00020-of-00059.safetensors",
258
+ "transformer.h.31.mlp.down_proj.bias": "model-00020-of-00059.safetensors",
259
+ "transformer.h.31.mlp.down_proj.weight": "model-00020-of-00059.safetensors",
260
+ "transformer.h.31.mlp.gate_proj.weight": "model-00020-of-00059.safetensors",
261
+ "transformer.h.31.mlp.up_proj.weight": "model-00020-of-00059.safetensors",
262
+ "transformer.h.31.post_attention_layernorm.weight": "model-00020-of-00059.safetensors",
263
+ "transformer.h.31.self_attention.dense.bias": "model-00020-of-00059.safetensors",
264
+ "transformer.h.31.self_attention.dense.weight": "model-00020-of-00059.safetensors",
265
+ "transformer.h.31.self_attention.key_value.weight": "model-00020-of-00059.safetensors",
266
+ "transformer.h.31.self_attention.query.weight": "model-00020-of-00059.safetensors",
267
+ "transformer.h.32.input_layernorm.weight": "model-00020-of-00059.safetensors",
268
+ "transformer.h.32.mlp.down_proj.bias": "model-00021-of-00059.safetensors",
269
+ "transformer.h.32.mlp.down_proj.weight": "model-00021-of-00059.safetensors",
270
+ "transformer.h.32.mlp.gate_proj.weight": "model-00020-of-00059.safetensors",
271
+ "transformer.h.32.mlp.up_proj.weight": "model-00021-of-00059.safetensors",
272
+ "transformer.h.32.post_attention_layernorm.weight": "model-00020-of-00059.safetensors",
273
+ "transformer.h.32.self_attention.dense.bias": "model-00020-of-00059.safetensors",
274
+ "transformer.h.32.self_attention.dense.weight": "model-00020-of-00059.safetensors",
275
+ "transformer.h.32.self_attention.key_value.weight": "model-00020-of-00059.safetensors",
276
+ "transformer.h.32.self_attention.query.weight": "model-00020-of-00059.safetensors",
277
+ "transformer.h.33.input_layernorm.weight": "model-00021-of-00059.safetensors",
278
+ "transformer.h.33.mlp.down_proj.bias": "model-00021-of-00059.safetensors",
279
+ "transformer.h.33.mlp.down_proj.weight": "model-00021-of-00059.safetensors",
280
+ "transformer.h.33.mlp.gate_proj.weight": "model-00021-of-00059.safetensors",
281
+ "transformer.h.33.mlp.up_proj.weight": "model-00021-of-00059.safetensors",
282
+ "transformer.h.33.post_attention_layernorm.weight": "model-00021-of-00059.safetensors",
283
+ "transformer.h.33.self_attention.dense.bias": "model-00021-of-00059.safetensors",
284
+ "transformer.h.33.self_attention.dense.weight": "model-00021-of-00059.safetensors",
285
+ "transformer.h.33.self_attention.key_value.weight": "model-00021-of-00059.safetensors",
286
+ "transformer.h.33.self_attention.query.weight": "model-00021-of-00059.safetensors",
287
+ "transformer.h.34.input_layernorm.weight": "model-00021-of-00059.safetensors",
288
+ "transformer.h.34.mlp.down_proj.bias": "model-00022-of-00059.safetensors",
289
+ "transformer.h.34.mlp.down_proj.weight": "model-00022-of-00059.safetensors",
290
+ "transformer.h.34.mlp.gate_proj.weight": "model-00022-of-00059.safetensors",
291
+ "transformer.h.34.mlp.up_proj.weight": "model-00022-of-00059.safetensors",
292
+ "transformer.h.34.post_attention_layernorm.weight": "model-00021-of-00059.safetensors",
293
+ "transformer.h.34.self_attention.dense.bias": "model-00021-of-00059.safetensors",
294
+ "transformer.h.34.self_attention.dense.weight": "model-00021-of-00059.safetensors",
295
+ "transformer.h.34.self_attention.key_value.weight": "model-00021-of-00059.safetensors",
296
+ "transformer.h.34.self_attention.query.weight": "model-00021-of-00059.safetensors",
297
+ "transformer.h.35.input_layernorm.weight": "model-00022-of-00059.safetensors",
298
+ "transformer.h.35.mlp.down_proj.bias": "model-00023-of-00059.safetensors",
299
+ "transformer.h.35.mlp.down_proj.weight": "model-00023-of-00059.safetensors",
300
+ "transformer.h.35.mlp.gate_proj.weight": "model-00022-of-00059.safetensors",
301
+ "transformer.h.35.mlp.up_proj.weight": "model-00022-of-00059.safetensors",
302
+ "transformer.h.35.post_attention_layernorm.weight": "model-00022-of-00059.safetensors",
303
+ "transformer.h.35.self_attention.dense.bias": "model-00022-of-00059.safetensors",
304
+ "transformer.h.35.self_attention.dense.weight": "model-00022-of-00059.safetensors",
305
+ "transformer.h.35.self_attention.key_value.weight": "model-00022-of-00059.safetensors",
306
+ "transformer.h.35.self_attention.query.weight": "model-00022-of-00059.safetensors",
307
+ "transformer.h.36.input_layernorm.weight": "model-00023-of-00059.safetensors",
308
+ "transformer.h.36.mlp.down_proj.bias": "model-00023-of-00059.safetensors",
309
+ "transformer.h.36.mlp.down_proj.weight": "model-00023-of-00059.safetensors",
310
+ "transformer.h.36.mlp.gate_proj.weight": "model-00023-of-00059.safetensors",
311
+ "transformer.h.36.mlp.up_proj.weight": "model-00023-of-00059.safetensors",
312
+ "transformer.h.36.post_attention_layernorm.weight": "model-00023-of-00059.safetensors",
313
+ "transformer.h.36.self_attention.dense.bias": "model-00023-of-00059.safetensors",
314
+ "transformer.h.36.self_attention.dense.weight": "model-00023-of-00059.safetensors",
315
+ "transformer.h.36.self_attention.key_value.weight": "model-00023-of-00059.safetensors",
316
+ "transformer.h.36.self_attention.query.weight": "model-00023-of-00059.safetensors",
317
+ "transformer.h.37.input_layernorm.weight": "model-00023-of-00059.safetensors",
318
+ "transformer.h.37.mlp.down_proj.bias": "model-00024-of-00059.safetensors",
319
+ "transformer.h.37.mlp.down_proj.weight": "model-00024-of-00059.safetensors",
320
+ "transformer.h.37.mlp.gate_proj.weight": "model-00023-of-00059.safetensors",
321
+ "transformer.h.37.mlp.up_proj.weight": "model-00024-of-00059.safetensors",
322
+ "transformer.h.37.post_attention_layernorm.weight": "model-00023-of-00059.safetensors",
323
+ "transformer.h.37.self_attention.dense.bias": "model-00023-of-00059.safetensors",
324
+ "transformer.h.37.self_attention.dense.weight": "model-00023-of-00059.safetensors",
325
+ "transformer.h.37.self_attention.key_value.weight": "model-00023-of-00059.safetensors",
326
+ "transformer.h.37.self_attention.query.weight": "model-00023-of-00059.safetensors",
327
+ "transformer.h.38.input_layernorm.weight": "model-00024-of-00059.safetensors",
328
+ "transformer.h.38.mlp.down_proj.bias": "model-00024-of-00059.safetensors",
329
+ "transformer.h.38.mlp.down_proj.weight": "model-00024-of-00059.safetensors",
330
+ "transformer.h.38.mlp.gate_proj.weight": "model-00024-of-00059.safetensors",
331
+ "transformer.h.38.mlp.up_proj.weight": "model-00024-of-00059.safetensors",
332
+ "transformer.h.38.post_attention_layernorm.weight": "model-00024-of-00059.safetensors",
333
+ "transformer.h.38.self_attention.dense.bias": "model-00024-of-00059.safetensors",
334
+ "transformer.h.38.self_attention.dense.weight": "model-00024-of-00059.safetensors",
335
+ "transformer.h.38.self_attention.key_value.weight": "model-00024-of-00059.safetensors",
336
+ "transformer.h.38.self_attention.query.weight": "model-00024-of-00059.safetensors",
337
+ "transformer.h.39.input_layernorm.weight": "model-00024-of-00059.safetensors",
338
+ "transformer.h.39.mlp.down_proj.bias": "model-00025-of-00059.safetensors",
339
+ "transformer.h.39.mlp.down_proj.weight": "model-00025-of-00059.safetensors",
340
+ "transformer.h.39.mlp.gate_proj.weight": "model-00025-of-00059.safetensors",
341
+ "transformer.h.39.mlp.up_proj.weight": "model-00025-of-00059.safetensors",
342
+ "transformer.h.39.post_attention_layernorm.weight": "model-00024-of-00059.safetensors",
343
+ "transformer.h.39.self_attention.dense.bias": "model-00024-of-00059.safetensors",
344
+ "transformer.h.39.self_attention.dense.weight": "model-00024-of-00059.safetensors",
345
+ "transformer.h.39.self_attention.key_value.weight": "model-00024-of-00059.safetensors",
346
+ "transformer.h.39.self_attention.query.weight": "model-00024-of-00059.safetensors",
347
+ "transformer.h.4.input_layernorm.weight": "model-00003-of-00059.safetensors",
348
+ "transformer.h.4.mlp.down_proj.bias": "model-00004-of-00059.safetensors",
349
+ "transformer.h.4.mlp.down_proj.weight": "model-00004-of-00059.safetensors",
350
+ "transformer.h.4.mlp.gate_proj.weight": "model-00004-of-00059.safetensors",
351
+ "transformer.h.4.mlp.up_proj.weight": "model-00004-of-00059.safetensors",
352
+ "transformer.h.4.post_attention_layernorm.weight": "model-00003-of-00059.safetensors",
353
+ "transformer.h.4.self_attention.dense.bias": "model-00003-of-00059.safetensors",
354
+ "transformer.h.4.self_attention.dense.weight": "model-00003-of-00059.safetensors",
355
+ "transformer.h.4.self_attention.key_value.weight": "model-00003-of-00059.safetensors",
356
+ "transformer.h.4.self_attention.query.weight": "model-00003-of-00059.safetensors",
357
+ "transformer.h.40.input_layernorm.weight": "model-00025-of-00059.safetensors",
358
+ "transformer.h.40.mlp.down_proj.bias": "model-00026-of-00059.safetensors",
359
+ "transformer.h.40.mlp.down_proj.weight": "model-00026-of-00059.safetensors",
360
+ "transformer.h.40.mlp.gate_proj.weight": "model-00025-of-00059.safetensors",
361
+ "transformer.h.40.mlp.up_proj.weight": "model-00025-of-00059.safetensors",
362
+ "transformer.h.40.post_attention_layernorm.weight": "model-00025-of-00059.safetensors",
363
+ "transformer.h.40.self_attention.dense.bias": "model-00025-of-00059.safetensors",
364
+ "transformer.h.40.self_attention.dense.weight": "model-00025-of-00059.safetensors",
365
+ "transformer.h.40.self_attention.key_value.weight": "model-00025-of-00059.safetensors",
366
+ "transformer.h.40.self_attention.query.weight": "model-00025-of-00059.safetensors",
367
+ "transformer.h.41.input_layernorm.weight": "model-00026-of-00059.safetensors",
368
+ "transformer.h.41.mlp.down_proj.bias": "model-00026-of-00059.safetensors",
369
+ "transformer.h.41.mlp.down_proj.weight": "model-00026-of-00059.safetensors",
370
+ "transformer.h.41.mlp.gate_proj.weight": "model-00026-of-00059.safetensors",
371
+ "transformer.h.41.mlp.up_proj.weight": "model-00026-of-00059.safetensors",
372
+ "transformer.h.41.post_attention_layernorm.weight": "model-00026-of-00059.safetensors",
373
+ "transformer.h.41.self_attention.dense.bias": "model-00026-of-00059.safetensors",
374
+ "transformer.h.41.self_attention.dense.weight": "model-00026-of-00059.safetensors",
375
+ "transformer.h.41.self_attention.key_value.weight": "model-00026-of-00059.safetensors",
376
+ "transformer.h.41.self_attention.query.weight": "model-00026-of-00059.safetensors",
377
+ "transformer.h.42.input_layernorm.weight": "model-00026-of-00059.safetensors",
378
+ "transformer.h.42.mlp.down_proj.bias": "model-00027-of-00059.safetensors",
379
+ "transformer.h.42.mlp.down_proj.weight": "model-00027-of-00059.safetensors",
380
+ "transformer.h.42.mlp.gate_proj.weight": "model-00026-of-00059.safetensors",
381
+ "transformer.h.42.mlp.up_proj.weight": "model-00027-of-00059.safetensors",
382
+ "transformer.h.42.post_attention_layernorm.weight": "model-00026-of-00059.safetensors",
383
+ "transformer.h.42.self_attention.dense.bias": "model-00026-of-00059.safetensors",
384
+ "transformer.h.42.self_attention.dense.weight": "model-00026-of-00059.safetensors",
385
+ "transformer.h.42.self_attention.key_value.weight": "model-00026-of-00059.safetensors",
386
+ "transformer.h.42.self_attention.query.weight": "model-00026-of-00059.safetensors",
387
+ "transformer.h.43.input_layernorm.weight": "model-00027-of-00059.safetensors",
388
+ "transformer.h.43.mlp.down_proj.bias": "model-00027-of-00059.safetensors",
389
+ "transformer.h.43.mlp.down_proj.weight": "model-00027-of-00059.safetensors",
390
+ "transformer.h.43.mlp.gate_proj.weight": "model-00027-of-00059.safetensors",
391
+ "transformer.h.43.mlp.up_proj.weight": "model-00027-of-00059.safetensors",
392
+ "transformer.h.43.post_attention_layernorm.weight": "model-00027-of-00059.safetensors",
393
+ "transformer.h.43.self_attention.dense.bias": "model-00027-of-00059.safetensors",
394
+ "transformer.h.43.self_attention.dense.weight": "model-00027-of-00059.safetensors",
395
+ "transformer.h.43.self_attention.key_value.weight": "model-00027-of-00059.safetensors",
396
+ "transformer.h.43.self_attention.query.weight": "model-00027-of-00059.safetensors",
397
+ "transformer.h.44.input_layernorm.weight": "model-00027-of-00059.safetensors",
398
+ "transformer.h.44.mlp.down_proj.bias": "model-00028-of-00059.safetensors",
399
+ "transformer.h.44.mlp.down_proj.weight": "model-00028-of-00059.safetensors",
400
+ "transformer.h.44.mlp.gate_proj.weight": "model-00028-of-00059.safetensors",
401
+ "transformer.h.44.mlp.up_proj.weight": "model-00028-of-00059.safetensors",
402
+ "transformer.h.44.post_attention_layernorm.weight": "model-00027-of-00059.safetensors",
403
+ "transformer.h.44.self_attention.dense.bias": "model-00027-of-00059.safetensors",
404
+ "transformer.h.44.self_attention.dense.weight": "model-00027-of-00059.safetensors",
405
+ "transformer.h.44.self_attention.key_value.weight": "model-00027-of-00059.safetensors",
406
+ "transformer.h.44.self_attention.query.weight": "model-00027-of-00059.safetensors",
407
+ "transformer.h.45.input_layernorm.weight": "model-00028-of-00059.safetensors",
408
+ "transformer.h.45.mlp.down_proj.bias": "model-00029-of-00059.safetensors",
409
+ "transformer.h.45.mlp.down_proj.weight": "model-00029-of-00059.safetensors",
410
+ "transformer.h.45.mlp.gate_proj.weight": "model-00028-of-00059.safetensors",
411
+ "transformer.h.45.mlp.up_proj.weight": "model-00028-of-00059.safetensors",
412
+ "transformer.h.45.post_attention_layernorm.weight": "model-00028-of-00059.safetensors",
413
+ "transformer.h.45.self_attention.dense.bias": "model-00028-of-00059.safetensors",
414
+ "transformer.h.45.self_attention.dense.weight": "model-00028-of-00059.safetensors",
415
+ "transformer.h.45.self_attention.key_value.weight": "model-00028-of-00059.safetensors",
416
+ "transformer.h.45.self_attention.query.weight": "model-00028-of-00059.safetensors",
417
+ "transformer.h.46.input_layernorm.weight": "model-00029-of-00059.safetensors",
418
+ "transformer.h.46.mlp.down_proj.bias": "model-00029-of-00059.safetensors",
419
+ "transformer.h.46.mlp.down_proj.weight": "model-00029-of-00059.safetensors",
420
+ "transformer.h.46.mlp.gate_proj.weight": "model-00029-of-00059.safetensors",
421
+ "transformer.h.46.mlp.up_proj.weight": "model-00029-of-00059.safetensors",
422
+ "transformer.h.46.post_attention_layernorm.weight": "model-00029-of-00059.safetensors",
423
+ "transformer.h.46.self_attention.dense.bias": "model-00029-of-00059.safetensors",
424
+ "transformer.h.46.self_attention.dense.weight": "model-00029-of-00059.safetensors",
425
+ "transformer.h.46.self_attention.key_value.weight": "model-00029-of-00059.safetensors",
426
+ "transformer.h.46.self_attention.query.weight": "model-00029-of-00059.safetensors",
427
+ "transformer.h.47.input_layernorm.weight": "model-00029-of-00059.safetensors",
428
+ "transformer.h.47.mlp.down_proj.bias": "model-00030-of-00059.safetensors",
429
+ "transformer.h.47.mlp.down_proj.weight": "model-00030-of-00059.safetensors",
430
+ "transformer.h.47.mlp.gate_proj.weight": "model-00029-of-00059.safetensors",
431
+ "transformer.h.47.mlp.up_proj.weight": "model-00030-of-00059.safetensors",
432
+ "transformer.h.47.post_attention_layernorm.weight": "model-00029-of-00059.safetensors",
433
+ "transformer.h.47.self_attention.dense.bias": "model-00029-of-00059.safetensors",
434
+ "transformer.h.47.self_attention.dense.weight": "model-00029-of-00059.safetensors",
435
+ "transformer.h.47.self_attention.key_value.weight": "model-00029-of-00059.safetensors",
436
+ "transformer.h.47.self_attention.query.weight": "model-00029-of-00059.safetensors",
437
+ "transformer.h.48.input_layernorm.weight": "model-00030-of-00059.safetensors",
438
+ "transformer.h.48.mlp.down_proj.bias": "model-00030-of-00059.safetensors",
439
+ "transformer.h.48.mlp.down_proj.weight": "model-00030-of-00059.safetensors",
440
+ "transformer.h.48.mlp.gate_proj.weight": "model-00030-of-00059.safetensors",
441
+ "transformer.h.48.mlp.up_proj.weight": "model-00030-of-00059.safetensors",
442
+ "transformer.h.48.post_attention_layernorm.weight": "model-00030-of-00059.safetensors",
443
+ "transformer.h.48.self_attention.dense.bias": "model-00030-of-00059.safetensors",
444
+ "transformer.h.48.self_attention.dense.weight": "model-00030-of-00059.safetensors",
445
+ "transformer.h.48.self_attention.key_value.weight": "model-00030-of-00059.safetensors",
446
+ "transformer.h.48.self_attention.query.weight": "model-00030-of-00059.safetensors",
447
+ "transformer.h.49.input_layernorm.weight": "model-00030-of-00059.safetensors",
448
+ "transformer.h.49.mlp.down_proj.bias": "model-00031-of-00059.safetensors",
449
+ "transformer.h.49.mlp.down_proj.weight": "model-00031-of-00059.safetensors",
450
+ "transformer.h.49.mlp.gate_proj.weight": "model-00031-of-00059.safetensors",
451
+ "transformer.h.49.mlp.up_proj.weight": "model-00031-of-00059.safetensors",
452
+ "transformer.h.49.post_attention_layernorm.weight": "model-00030-of-00059.safetensors",
453
+ "transformer.h.49.self_attention.dense.bias": "model-00030-of-00059.safetensors",
454
+ "transformer.h.49.self_attention.dense.weight": "model-00030-of-00059.safetensors",
455
+ "transformer.h.49.self_attention.key_value.weight": "model-00030-of-00059.safetensors",
456
+ "transformer.h.49.self_attention.query.weight": "model-00030-of-00059.safetensors",
457
+ "transformer.h.5.input_layernorm.weight": "model-00004-of-00059.safetensors",
458
+ "transformer.h.5.mlp.down_proj.bias": "model-00005-of-00059.safetensors",
459
+ "transformer.h.5.mlp.down_proj.weight": "model-00005-of-00059.safetensors",
460
+ "transformer.h.5.mlp.gate_proj.weight": "model-00004-of-00059.safetensors",
461
+ "transformer.h.5.mlp.up_proj.weight": "model-00004-of-00059.safetensors",
462
+ "transformer.h.5.post_attention_layernorm.weight": "model-00004-of-00059.safetensors",
463
+ "transformer.h.5.self_attention.dense.bias": "model-00004-of-00059.safetensors",
464
+ "transformer.h.5.self_attention.dense.weight": "model-00004-of-00059.safetensors",
465
+ "transformer.h.5.self_attention.key_value.weight": "model-00004-of-00059.safetensors",
466
+ "transformer.h.5.self_attention.query.weight": "model-00004-of-00059.safetensors",
467
+ "transformer.h.50.input_layernorm.weight": "model-00031-of-00059.safetensors",
468
+ "transformer.h.50.mlp.down_proj.bias": "model-00032-of-00059.safetensors",
469
+ "transformer.h.50.mlp.down_proj.weight": "model-00032-of-00059.safetensors",
470
+ "transformer.h.50.mlp.gate_proj.weight": "model-00031-of-00059.safetensors",
471
+ "transformer.h.50.mlp.up_proj.weight": "model-00031-of-00059.safetensors",
472
+ "transformer.h.50.post_attention_layernorm.weight": "model-00031-of-00059.safetensors",
473
+ "transformer.h.50.self_attention.dense.bias": "model-00031-of-00059.safetensors",
474
+ "transformer.h.50.self_attention.dense.weight": "model-00031-of-00059.safetensors",
475
+ "transformer.h.50.self_attention.key_value.weight": "model-00031-of-00059.safetensors",
476
+ "transformer.h.50.self_attention.query.weight": "model-00031-of-00059.safetensors",
477
+ "transformer.h.51.input_layernorm.weight": "model-00032-of-00059.safetensors",
478
+ "transformer.h.51.mlp.down_proj.bias": "model-00032-of-00059.safetensors",
479
+ "transformer.h.51.mlp.down_proj.weight": "model-00032-of-00059.safetensors",
480
+ "transformer.h.51.mlp.gate_proj.weight": "model-00032-of-00059.safetensors",
481
+ "transformer.h.51.mlp.up_proj.weight": "model-00032-of-00059.safetensors",
482
+ "transformer.h.51.post_attention_layernorm.weight": "model-00032-of-00059.safetensors",
483
+ "transformer.h.51.self_attention.dense.bias": "model-00032-of-00059.safetensors",
484
+ "transformer.h.51.self_attention.dense.weight": "model-00032-of-00059.safetensors",
485
+ "transformer.h.51.self_attention.key_value.weight": "model-00032-of-00059.safetensors",
486
+ "transformer.h.51.self_attention.query.weight": "model-00032-of-00059.safetensors",
487
+ "transformer.h.52.input_layernorm.weight": "model-00032-of-00059.safetensors",
488
+ "transformer.h.52.mlp.down_proj.bias": "model-00033-of-00059.safetensors",
489
+ "transformer.h.52.mlp.down_proj.weight": "model-00033-of-00059.safetensors",
490
+ "transformer.h.52.mlp.gate_proj.weight": "model-00032-of-00059.safetensors",
491
+ "transformer.h.52.mlp.up_proj.weight": "model-00033-of-00059.safetensors",
492
+ "transformer.h.52.post_attention_layernorm.weight": "model-00032-of-00059.safetensors",
493
+ "transformer.h.52.self_attention.dense.bias": "model-00032-of-00059.safetensors",
494
+ "transformer.h.52.self_attention.dense.weight": "model-00032-of-00059.safetensors",
495
+ "transformer.h.52.self_attention.key_value.weight": "model-00032-of-00059.safetensors",
496
+ "transformer.h.52.self_attention.query.weight": "model-00032-of-00059.safetensors",
497
+ "transformer.h.53.input_layernorm.weight": "model-00033-of-00059.safetensors",
498
+ "transformer.h.53.mlp.down_proj.bias": "model-00033-of-00059.safetensors",
499
+ "transformer.h.53.mlp.down_proj.weight": "model-00033-of-00059.safetensors",
500
+ "transformer.h.53.mlp.gate_proj.weight": "model-00033-of-00059.safetensors",
501
+ "transformer.h.53.mlp.up_proj.weight": "model-00033-of-00059.safetensors",
502
+ "transformer.h.53.post_attention_layernorm.weight": "model-00033-of-00059.safetensors",
503
+ "transformer.h.53.self_attention.dense.bias": "model-00033-of-00059.safetensors",
504
+ "transformer.h.53.self_attention.dense.weight": "model-00033-of-00059.safetensors",
505
+ "transformer.h.53.self_attention.key_value.weight": "model-00033-of-00059.safetensors",
506
+ "transformer.h.53.self_attention.query.weight": "model-00033-of-00059.safetensors",
507
+ "transformer.h.54.input_layernorm.weight": "model-00033-of-00059.safetensors",
508
+ "transformer.h.54.mlp.down_proj.bias": "model-00034-of-00059.safetensors",
509
+ "transformer.h.54.mlp.down_proj.weight": "model-00034-of-00059.safetensors",
510
+ "transformer.h.54.mlp.gate_proj.weight": "model-00034-of-00059.safetensors",
511
+ "transformer.h.54.mlp.up_proj.weight": "model-00034-of-00059.safetensors",
512
+ "transformer.h.54.post_attention_layernorm.weight": "model-00033-of-00059.safetensors",
513
+ "transformer.h.54.self_attention.dense.bias": "model-00033-of-00059.safetensors",
514
+ "transformer.h.54.self_attention.dense.weight": "model-00033-of-00059.safetensors",
515
+ "transformer.h.54.self_attention.key_value.weight": "model-00033-of-00059.safetensors",
516
+ "transformer.h.54.self_attention.query.weight": "model-00033-of-00059.safetensors",
517
+ "transformer.h.55.input_layernorm.weight": "model-00034-of-00059.safetensors",
518
+ "transformer.h.55.mlp.down_proj.bias": "model-00035-of-00059.safetensors",
519
+ "transformer.h.55.mlp.down_proj.weight": "model-00035-of-00059.safetensors",
520
+ "transformer.h.55.mlp.gate_proj.weight": "model-00034-of-00059.safetensors",
521
+ "transformer.h.55.mlp.up_proj.weight": "model-00034-of-00059.safetensors",
522
+ "transformer.h.55.post_attention_layernorm.weight": "model-00034-of-00059.safetensors",
523
+ "transformer.h.55.self_attention.dense.bias": "model-00034-of-00059.safetensors",
524
+ "transformer.h.55.self_attention.dense.weight": "model-00034-of-00059.safetensors",
525
+ "transformer.h.55.self_attention.key_value.weight": "model-00034-of-00059.safetensors",
526
+ "transformer.h.55.self_attention.query.weight": "model-00034-of-00059.safetensors",
527
+ "transformer.h.56.input_layernorm.weight": "model-00035-of-00059.safetensors",
528
+ "transformer.h.56.mlp.down_proj.bias": "model-00035-of-00059.safetensors",
529
+ "transformer.h.56.mlp.down_proj.weight": "model-00035-of-00059.safetensors",
530
+ "transformer.h.56.mlp.gate_proj.weight": "model-00035-of-00059.safetensors",
531
+ "transformer.h.56.mlp.up_proj.weight": "model-00035-of-00059.safetensors",
532
+ "transformer.h.56.post_attention_layernorm.weight": "model-00035-of-00059.safetensors",
533
+ "transformer.h.56.self_attention.dense.bias": "model-00035-of-00059.safetensors",
534
+ "transformer.h.56.self_attention.dense.weight": "model-00035-of-00059.safetensors",
535
+ "transformer.h.56.self_attention.key_value.weight": "model-00035-of-00059.safetensors",
536
+ "transformer.h.56.self_attention.query.weight": "model-00035-of-00059.safetensors",
537
+ "transformer.h.57.input_layernorm.weight": "model-00035-of-00059.safetensors",
538
+ "transformer.h.57.mlp.down_proj.bias": "model-00036-of-00059.safetensors",
539
+ "transformer.h.57.mlp.down_proj.weight": "model-00036-of-00059.safetensors",
540
+ "transformer.h.57.mlp.gate_proj.weight": "model-00035-of-00059.safetensors",
541
+ "transformer.h.57.mlp.up_proj.weight": "model-00036-of-00059.safetensors",
542
+ "transformer.h.57.post_attention_layernorm.weight": "model-00035-of-00059.safetensors",
543
+ "transformer.h.57.self_attention.dense.bias": "model-00035-of-00059.safetensors",
544
+ "transformer.h.57.self_attention.dense.weight": "model-00035-of-00059.safetensors",
545
+ "transformer.h.57.self_attention.key_value.weight": "model-00035-of-00059.safetensors",
546
+ "transformer.h.57.self_attention.query.weight": "model-00035-of-00059.safetensors",
547
+ "transformer.h.58.input_layernorm.weight": "model-00036-of-00059.safetensors",
548
+ "transformer.h.58.mlp.down_proj.bias": "model-00036-of-00059.safetensors",
549
+ "transformer.h.58.mlp.down_proj.weight": "model-00036-of-00059.safetensors",
550
+ "transformer.h.58.mlp.gate_proj.weight": "model-00036-of-00059.safetensors",
551
+ "transformer.h.58.mlp.up_proj.weight": "model-00036-of-00059.safetensors",
552
+ "transformer.h.58.post_attention_layernorm.weight": "model-00036-of-00059.safetensors",
553
+ "transformer.h.58.self_attention.dense.bias": "model-00036-of-00059.safetensors",
554
+ "transformer.h.58.self_attention.dense.weight": "model-00036-of-00059.safetensors",
555
+ "transformer.h.58.self_attention.key_value.weight": "model-00036-of-00059.safetensors",
556
+ "transformer.h.58.self_attention.query.weight": "model-00036-of-00059.safetensors",
557
+ "transformer.h.59.input_layernorm.weight": "model-00036-of-00059.safetensors",
558
+ "transformer.h.59.mlp.down_proj.bias": "model-00037-of-00059.safetensors",
559
+ "transformer.h.59.mlp.down_proj.weight": "model-00037-of-00059.safetensors",
560
+ "transformer.h.59.mlp.gate_proj.weight": "model-00037-of-00059.safetensors",
561
+ "transformer.h.59.mlp.up_proj.weight": "model-00037-of-00059.safetensors",
562
+ "transformer.h.59.post_attention_layernorm.weight": "model-00036-of-00059.safetensors",
563
+ "transformer.h.59.self_attention.dense.bias": "model-00036-of-00059.safetensors",
564
+ "transformer.h.59.self_attention.dense.weight": "model-00036-of-00059.safetensors",
565
+ "transformer.h.59.self_attention.key_value.weight": "model-00036-of-00059.safetensors",
566
+ "transformer.h.59.self_attention.query.weight": "model-00036-of-00059.safetensors",
567
+ "transformer.h.6.input_layernorm.weight": "model-00005-of-00059.safetensors",
568
+ "transformer.h.6.mlp.down_proj.bias": "model-00005-of-00059.safetensors",
569
+ "transformer.h.6.mlp.down_proj.weight": "model-00005-of-00059.safetensors",
570
+ "transformer.h.6.mlp.gate_proj.weight": "model-00005-of-00059.safetensors",
571
+ "transformer.h.6.mlp.up_proj.weight": "model-00005-of-00059.safetensors",
572
+ "transformer.h.6.post_attention_layernorm.weight": "model-00005-of-00059.safetensors",
573
+ "transformer.h.6.self_attention.dense.bias": "model-00005-of-00059.safetensors",
574
+ "transformer.h.6.self_attention.dense.weight": "model-00005-of-00059.safetensors",
575
+ "transformer.h.6.self_attention.key_value.weight": "model-00005-of-00059.safetensors",
576
+ "transformer.h.6.self_attention.query.weight": "model-00005-of-00059.safetensors",
577
+ "transformer.h.60.input_layernorm.weight": "model-00037-of-00059.safetensors",
578
+ "transformer.h.60.mlp.down_proj.bias": "model-00038-of-00059.safetensors",
579
+ "transformer.h.60.mlp.down_proj.weight": "model-00038-of-00059.safetensors",
580
+ "transformer.h.60.mlp.gate_proj.weight": "model-00037-of-00059.safetensors",
581
+ "transformer.h.60.mlp.up_proj.weight": "model-00037-of-00059.safetensors",
582
+ "transformer.h.60.post_attention_layernorm.weight": "model-00037-of-00059.safetensors",
583
+ "transformer.h.60.self_attention.dense.bias": "model-00037-of-00059.safetensors",
584
+ "transformer.h.60.self_attention.dense.weight": "model-00037-of-00059.safetensors",
585
+ "transformer.h.60.self_attention.key_value.weight": "model-00037-of-00059.safetensors",
586
+ "transformer.h.60.self_attention.query.weight": "model-00037-of-00059.safetensors",
587
+ "transformer.h.61.input_layernorm.weight": "model-00038-of-00059.safetensors",
588
+ "transformer.h.61.mlp.down_proj.bias": "model-00038-of-00059.safetensors",
589
+ "transformer.h.61.mlp.down_proj.weight": "model-00038-of-00059.safetensors",
590
+ "transformer.h.61.mlp.gate_proj.weight": "model-00038-of-00059.safetensors",
591
+ "transformer.h.61.mlp.up_proj.weight": "model-00038-of-00059.safetensors",
592
+ "transformer.h.61.post_attention_layernorm.weight": "model-00038-of-00059.safetensors",
593
+ "transformer.h.61.self_attention.dense.bias": "model-00038-of-00059.safetensors",
594
+ "transformer.h.61.self_attention.dense.weight": "model-00038-of-00059.safetensors",
595
+ "transformer.h.61.self_attention.key_value.weight": "model-00038-of-00059.safetensors",
596
+ "transformer.h.61.self_attention.query.weight": "model-00038-of-00059.safetensors",
597
+ "transformer.h.62.input_layernorm.weight": "model-00038-of-00059.safetensors",
598
+ "transformer.h.62.mlp.down_proj.bias": "model-00039-of-00059.safetensors",
599
+ "transformer.h.62.mlp.down_proj.weight": "model-00039-of-00059.safetensors",
600
+ "transformer.h.62.mlp.gate_proj.weight": "model-00038-of-00059.safetensors",
601
+ "transformer.h.62.mlp.up_proj.weight": "model-00039-of-00059.safetensors",
602
+ "transformer.h.62.post_attention_layernorm.weight": "model-00038-of-00059.safetensors",
603
+ "transformer.h.62.self_attention.dense.bias": "model-00038-of-00059.safetensors",
604
+ "transformer.h.62.self_attention.dense.weight": "model-00038-of-00059.safetensors",
605
+ "transformer.h.62.self_attention.key_value.weight": "model-00038-of-00059.safetensors",
606
+ "transformer.h.62.self_attention.query.weight": "model-00038-of-00059.safetensors",
607
+ "transformer.h.63.input_layernorm.weight": "model-00039-of-00059.safetensors",
608
+ "transformer.h.63.mlp.down_proj.bias": "model-00039-of-00059.safetensors",
609
+ "transformer.h.63.mlp.down_proj.weight": "model-00039-of-00059.safetensors",
610
+ "transformer.h.63.mlp.gate_proj.weight": "model-00039-of-00059.safetensors",
611
+ "transformer.h.63.mlp.up_proj.weight": "model-00039-of-00059.safetensors",
612
+ "transformer.h.63.post_attention_layernorm.weight": "model-00039-of-00059.safetensors",
613
+ "transformer.h.63.self_attention.dense.bias": "model-00039-of-00059.safetensors",
614
+ "transformer.h.63.self_attention.dense.weight": "model-00039-of-00059.safetensors",
615
+ "transformer.h.63.self_attention.key_value.weight": "model-00039-of-00059.safetensors",
616
+ "transformer.h.63.self_attention.query.weight": "model-00039-of-00059.safetensors",
617
+ "transformer.h.64.input_layernorm.weight": "model-00039-of-00059.safetensors",
618
+ "transformer.h.64.mlp.down_proj.bias": "model-00040-of-00059.safetensors",
619
+ "transformer.h.64.mlp.down_proj.weight": "model-00040-of-00059.safetensors",
620
+ "transformer.h.64.mlp.gate_proj.weight": "model-00040-of-00059.safetensors",
621
+ "transformer.h.64.mlp.up_proj.weight": "model-00040-of-00059.safetensors",
622
+ "transformer.h.64.post_attention_layernorm.weight": "model-00039-of-00059.safetensors",
623
+ "transformer.h.64.self_attention.dense.bias": "model-00039-of-00059.safetensors",
624
+ "transformer.h.64.self_attention.dense.weight": "model-00039-of-00059.safetensors",
625
+ "transformer.h.64.self_attention.key_value.weight": "model-00039-of-00059.safetensors",
626
+ "transformer.h.64.self_attention.query.weight": "model-00039-of-00059.safetensors",
627
+ "transformer.h.65.input_layernorm.weight": "model-00040-of-00059.safetensors",
628
+ "transformer.h.65.mlp.down_proj.bias": "model-00041-of-00059.safetensors",
629
+ "transformer.h.65.mlp.down_proj.weight": "model-00041-of-00059.safetensors",
630
+ "transformer.h.65.mlp.gate_proj.weight": "model-00040-of-00059.safetensors",
631
+ "transformer.h.65.mlp.up_proj.weight": "model-00040-of-00059.safetensors",
632
+ "transformer.h.65.post_attention_layernorm.weight": "model-00040-of-00059.safetensors",
633
+ "transformer.h.65.self_attention.dense.bias": "model-00040-of-00059.safetensors",
634
+ "transformer.h.65.self_attention.dense.weight": "model-00040-of-00059.safetensors",
635
+ "transformer.h.65.self_attention.key_value.weight": "model-00040-of-00059.safetensors",
636
+ "transformer.h.65.self_attention.query.weight": "model-00040-of-00059.safetensors",
637
+ "transformer.h.66.input_layernorm.weight": "model-00041-of-00059.safetensors",
638
+ "transformer.h.66.mlp.down_proj.bias": "model-00041-of-00059.safetensors",
639
+ "transformer.h.66.mlp.down_proj.weight": "model-00041-of-00059.safetensors",
640
+ "transformer.h.66.mlp.gate_proj.weight": "model-00041-of-00059.safetensors",
641
+ "transformer.h.66.mlp.up_proj.weight": "model-00041-of-00059.safetensors",
642
+ "transformer.h.66.post_attention_layernorm.weight": "model-00041-of-00059.safetensors",
643
+ "transformer.h.66.self_attention.dense.bias": "model-00041-of-00059.safetensors",
644
+ "transformer.h.66.self_attention.dense.weight": "model-00041-of-00059.safetensors",
645
+ "transformer.h.66.self_attention.key_value.weight": "model-00041-of-00059.safetensors",
646
+ "transformer.h.66.self_attention.query.weight": "model-00041-of-00059.safetensors",
647
+ "transformer.h.67.input_layernorm.weight": "model-00041-of-00059.safetensors",
648
+ "transformer.h.67.mlp.down_proj.bias": "model-00042-of-00059.safetensors",
649
+ "transformer.h.67.mlp.down_proj.weight": "model-00042-of-00059.safetensors",
650
+ "transformer.h.67.mlp.gate_proj.weight": "model-00041-of-00059.safetensors",
651
+ "transformer.h.67.mlp.up_proj.weight": "model-00042-of-00059.safetensors",
652
+ "transformer.h.67.post_attention_layernorm.weight": "model-00041-of-00059.safetensors",
653
+ "transformer.h.67.self_attention.dense.bias": "model-00041-of-00059.safetensors",
654
+ "transformer.h.67.self_attention.dense.weight": "model-00041-of-00059.safetensors",
655
+ "transformer.h.67.self_attention.key_value.weight": "model-00041-of-00059.safetensors",
656
+ "transformer.h.67.self_attention.query.weight": "model-00041-of-00059.safetensors",
657
+ "transformer.h.68.input_layernorm.weight": "model-00042-of-00059.safetensors",
658
+ "transformer.h.68.mlp.down_proj.bias": "model-00042-of-00059.safetensors",
659
+ "transformer.h.68.mlp.down_proj.weight": "model-00042-of-00059.safetensors",
660
+ "transformer.h.68.mlp.gate_proj.weight": "model-00042-of-00059.safetensors",
661
+ "transformer.h.68.mlp.up_proj.weight": "model-00042-of-00059.safetensors",
662
+ "transformer.h.68.post_attention_layernorm.weight": "model-00042-of-00059.safetensors",
663
+ "transformer.h.68.self_attention.dense.bias": "model-00042-of-00059.safetensors",
664
+ "transformer.h.68.self_attention.dense.weight": "model-00042-of-00059.safetensors",
665
+ "transformer.h.68.self_attention.key_value.weight": "model-00042-of-00059.safetensors",
666
+ "transformer.h.68.self_attention.query.weight": "model-00042-of-00059.safetensors",
667
+ "transformer.h.69.input_layernorm.weight": "model-00042-of-00059.safetensors",
668
+ "transformer.h.69.mlp.down_proj.bias": "model-00043-of-00059.safetensors",
669
+ "transformer.h.69.mlp.down_proj.weight": "model-00043-of-00059.safetensors",
670
+ "transformer.h.69.mlp.gate_proj.weight": "model-00043-of-00059.safetensors",
671
+ "transformer.h.69.mlp.up_proj.weight": "model-00043-of-00059.safetensors",
672
+ "transformer.h.69.post_attention_layernorm.weight": "model-00042-of-00059.safetensors",
673
+ "transformer.h.69.self_attention.dense.bias": "model-00042-of-00059.safetensors",
674
+ "transformer.h.69.self_attention.dense.weight": "model-00042-of-00059.safetensors",
675
+ "transformer.h.69.self_attention.key_value.weight": "model-00042-of-00059.safetensors",
676
+ "transformer.h.69.self_attention.query.weight": "model-00042-of-00059.safetensors",
677
+ "transformer.h.7.input_layernorm.weight": "model-00005-of-00059.safetensors",
678
+ "transformer.h.7.mlp.down_proj.bias": "model-00006-of-00059.safetensors",
679
+ "transformer.h.7.mlp.down_proj.weight": "model-00006-of-00059.safetensors",
680
+ "transformer.h.7.mlp.gate_proj.weight": "model-00005-of-00059.safetensors",
681
+ "transformer.h.7.mlp.up_proj.weight": "model-00006-of-00059.safetensors",
682
+ "transformer.h.7.post_attention_layernorm.weight": "model-00005-of-00059.safetensors",
683
+ "transformer.h.7.self_attention.dense.bias": "model-00005-of-00059.safetensors",
684
+ "transformer.h.7.self_attention.dense.weight": "model-00005-of-00059.safetensors",
685
+ "transformer.h.7.self_attention.key_value.weight": "model-00005-of-00059.safetensors",
686
+ "transformer.h.7.self_attention.query.weight": "model-00005-of-00059.safetensors",
687
+ "transformer.h.70.input_layernorm.weight": "model-00043-of-00059.safetensors",
688
+ "transformer.h.70.mlp.down_proj.bias": "model-00044-of-00059.safetensors",
689
+ "transformer.h.70.mlp.down_proj.weight": "model-00044-of-00059.safetensors",
690
+ "transformer.h.70.mlp.gate_proj.weight": "model-00043-of-00059.safetensors",
691
+ "transformer.h.70.mlp.up_proj.weight": "model-00043-of-00059.safetensors",
692
+ "transformer.h.70.post_attention_layernorm.weight": "model-00043-of-00059.safetensors",
693
+ "transformer.h.70.self_attention.dense.bias": "model-00043-of-00059.safetensors",
694
+ "transformer.h.70.self_attention.dense.weight": "model-00043-of-00059.safetensors",
695
+ "transformer.h.70.self_attention.key_value.weight": "model-00043-of-00059.safetensors",
696
+ "transformer.h.70.self_attention.query.weight": "model-00043-of-00059.safetensors",
697
+ "transformer.h.71.input_layernorm.weight": "model-00044-of-00059.safetensors",
698
+ "transformer.h.71.mlp.down_proj.bias": "model-00044-of-00059.safetensors",
699
+ "transformer.h.71.mlp.down_proj.weight": "model-00044-of-00059.safetensors",
700
+ "transformer.h.71.mlp.gate_proj.weight": "model-00044-of-00059.safetensors",
701
+ "transformer.h.71.mlp.up_proj.weight": "model-00044-of-00059.safetensors",
702
+ "transformer.h.71.post_attention_layernorm.weight": "model-00044-of-00059.safetensors",
703
+ "transformer.h.71.self_attention.dense.bias": "model-00044-of-00059.safetensors",
704
+ "transformer.h.71.self_attention.dense.weight": "model-00044-of-00059.safetensors",
705
+ "transformer.h.71.self_attention.key_value.weight": "model-00044-of-00059.safetensors",
706
+ "transformer.h.71.self_attention.query.weight": "model-00044-of-00059.safetensors",
707
+ "transformer.h.72.input_layernorm.weight": "model-00044-of-00059.safetensors",
708
+ "transformer.h.72.mlp.down_proj.bias": "model-00045-of-00059.safetensors",
709
+ "transformer.h.72.mlp.down_proj.weight": "model-00045-of-00059.safetensors",
710
+ "transformer.h.72.mlp.gate_proj.weight": "model-00044-of-00059.safetensors",
711
+ "transformer.h.72.mlp.up_proj.weight": "model-00045-of-00059.safetensors",
712
+ "transformer.h.72.post_attention_layernorm.weight": "model-00044-of-00059.safetensors",
713
+ "transformer.h.72.self_attention.dense.bias": "model-00044-of-00059.safetensors",
714
+ "transformer.h.72.self_attention.dense.weight": "model-00044-of-00059.safetensors",
715
+ "transformer.h.72.self_attention.key_value.weight": "model-00044-of-00059.safetensors",
716
+ "transformer.h.72.self_attention.query.weight": "model-00044-of-00059.safetensors",
717
+ "transformer.h.73.input_layernorm.weight": "model-00045-of-00059.safetensors",
718
+ "transformer.h.73.mlp.down_proj.bias": "model-00045-of-00059.safetensors",
719
+ "transformer.h.73.mlp.down_proj.weight": "model-00045-of-00059.safetensors",
720
+ "transformer.h.73.mlp.gate_proj.weight": "model-00045-of-00059.safetensors",
721
+ "transformer.h.73.mlp.up_proj.weight": "model-00045-of-00059.safetensors",
722
+ "transformer.h.73.post_attention_layernorm.weight": "model-00045-of-00059.safetensors",
723
+ "transformer.h.73.self_attention.dense.bias": "model-00045-of-00059.safetensors",
724
+ "transformer.h.73.self_attention.dense.weight": "model-00045-of-00059.safetensors",
725
+ "transformer.h.73.self_attention.key_value.weight": "model-00045-of-00059.safetensors",
726
+ "transformer.h.73.self_attention.query.weight": "model-00045-of-00059.safetensors",
727
+ "transformer.h.74.input_layernorm.weight": "model-00045-of-00059.safetensors",
728
+ "transformer.h.74.mlp.down_proj.bias": "model-00046-of-00059.safetensors",
729
+ "transformer.h.74.mlp.down_proj.weight": "model-00046-of-00059.safetensors",
730
+ "transformer.h.74.mlp.gate_proj.weight": "model-00046-of-00059.safetensors",
731
+ "transformer.h.74.mlp.up_proj.weight": "model-00046-of-00059.safetensors",
732
+ "transformer.h.74.post_attention_layernorm.weight": "model-00045-of-00059.safetensors",
733
+ "transformer.h.74.self_attention.dense.bias": "model-00045-of-00059.safetensors",
734
+ "transformer.h.74.self_attention.dense.weight": "model-00045-of-00059.safetensors",
735
+ "transformer.h.74.self_attention.key_value.weight": "model-00045-of-00059.safetensors",
736
+ "transformer.h.74.self_attention.query.weight": "model-00045-of-00059.safetensors",
737
+ "transformer.h.75.input_layernorm.weight": "model-00046-of-00059.safetensors",
738
+ "transformer.h.75.mlp.down_proj.bias": "model-00047-of-00059.safetensors",
739
+ "transformer.h.75.mlp.down_proj.weight": "model-00047-of-00059.safetensors",
740
+ "transformer.h.75.mlp.gate_proj.weight": "model-00046-of-00059.safetensors",
741
+ "transformer.h.75.mlp.up_proj.weight": "model-00046-of-00059.safetensors",
742
+ "transformer.h.75.post_attention_layernorm.weight": "model-00046-of-00059.safetensors",
743
+ "transformer.h.75.self_attention.dense.bias": "model-00046-of-00059.safetensors",
744
+ "transformer.h.75.self_attention.dense.weight": "model-00046-of-00059.safetensors",
745
+ "transformer.h.75.self_attention.key_value.weight": "model-00046-of-00059.safetensors",
746
+ "transformer.h.75.self_attention.query.weight": "model-00046-of-00059.safetensors",
747
+ "transformer.h.76.input_layernorm.weight": "model-00047-of-00059.safetensors",
748
+ "transformer.h.76.mlp.down_proj.bias": "model-00047-of-00059.safetensors",
749
+ "transformer.h.76.mlp.down_proj.weight": "model-00047-of-00059.safetensors",
750
+ "transformer.h.76.mlp.gate_proj.weight": "model-00047-of-00059.safetensors",
751
+ "transformer.h.76.mlp.up_proj.weight": "model-00047-of-00059.safetensors",
752
+ "transformer.h.76.post_attention_layernorm.weight": "model-00047-of-00059.safetensors",
753
+ "transformer.h.76.self_attention.dense.bias": "model-00047-of-00059.safetensors",
754
+ "transformer.h.76.self_attention.dense.weight": "model-00047-of-00059.safetensors",
755
+ "transformer.h.76.self_attention.key_value.weight": "model-00047-of-00059.safetensors",
756
+ "transformer.h.76.self_attention.query.weight": "model-00047-of-00059.safetensors",
757
+ "transformer.h.77.input_layernorm.weight": "model-00047-of-00059.safetensors",
758
+ "transformer.h.77.mlp.down_proj.bias": "model-00048-of-00059.safetensors",
759
+ "transformer.h.77.mlp.down_proj.weight": "model-00048-of-00059.safetensors",
760
+ "transformer.h.77.mlp.gate_proj.weight": "model-00047-of-00059.safetensors",
761
+ "transformer.h.77.mlp.up_proj.weight": "model-00048-of-00059.safetensors",
762
+ "transformer.h.77.post_attention_layernorm.weight": "model-00047-of-00059.safetensors",
763
+ "transformer.h.77.self_attention.dense.bias": "model-00047-of-00059.safetensors",
764
+ "transformer.h.77.self_attention.dense.weight": "model-00047-of-00059.safetensors",
765
+ "transformer.h.77.self_attention.key_value.weight": "model-00047-of-00059.safetensors",
766
+ "transformer.h.77.self_attention.query.weight": "model-00047-of-00059.safetensors",
767
+ "transformer.h.78.input_layernorm.weight": "model-00048-of-00059.safetensors",
768
+ "transformer.h.78.mlp.down_proj.bias": "model-00048-of-00059.safetensors",
769
+ "transformer.h.78.mlp.down_proj.weight": "model-00048-of-00059.safetensors",
770
+ "transformer.h.78.mlp.gate_proj.weight": "model-00048-of-00059.safetensors",
771
+ "transformer.h.78.mlp.up_proj.weight": "model-00048-of-00059.safetensors",
772
+ "transformer.h.78.post_attention_layernorm.weight": "model-00048-of-00059.safetensors",
773
+ "transformer.h.78.self_attention.dense.bias": "model-00048-of-00059.safetensors",
774
+ "transformer.h.78.self_attention.dense.weight": "model-00048-of-00059.safetensors",
775
+ "transformer.h.78.self_attention.key_value.weight": "model-00048-of-00059.safetensors",
776
+ "transformer.h.78.self_attention.query.weight": "model-00048-of-00059.safetensors",
777
+ "transformer.h.79.input_layernorm.weight": "model-00048-of-00059.safetensors",
778
+ "transformer.h.79.mlp.down_proj.bias": "model-00049-of-00059.safetensors",
779
+ "transformer.h.79.mlp.down_proj.weight": "model-00049-of-00059.safetensors",
780
+ "transformer.h.79.mlp.gate_proj.weight": "model-00049-of-00059.safetensors",
781
+ "transformer.h.79.mlp.up_proj.weight": "model-00049-of-00059.safetensors",
782
+ "transformer.h.79.post_attention_layernorm.weight": "model-00048-of-00059.safetensors",
783
+ "transformer.h.79.self_attention.dense.bias": "model-00048-of-00059.safetensors",
784
+ "transformer.h.79.self_attention.dense.weight": "model-00048-of-00059.safetensors",
785
+ "transformer.h.79.self_attention.key_value.weight": "model-00048-of-00059.safetensors",
786
+ "transformer.h.79.self_attention.query.weight": "model-00048-of-00059.safetensors",
787
+ "transformer.h.8.input_layernorm.weight": "model-00006-of-00059.safetensors",
788
+ "transformer.h.8.mlp.down_proj.bias": "model-00006-of-00059.safetensors",
789
+ "transformer.h.8.mlp.down_proj.weight": "model-00006-of-00059.safetensors",
790
+ "transformer.h.8.mlp.gate_proj.weight": "model-00006-of-00059.safetensors",
791
+ "transformer.h.8.mlp.up_proj.weight": "model-00006-of-00059.safetensors",
792
+ "transformer.h.8.post_attention_layernorm.weight": "model-00006-of-00059.safetensors",
793
+ "transformer.h.8.self_attention.dense.bias": "model-00006-of-00059.safetensors",
794
+ "transformer.h.8.self_attention.dense.weight": "model-00006-of-00059.safetensors",
795
+ "transformer.h.8.self_attention.key_value.weight": "model-00006-of-00059.safetensors",
796
+ "transformer.h.8.self_attention.query.weight": "model-00006-of-00059.safetensors",
797
+ "transformer.h.80.input_layernorm.weight": "model-00049-of-00059.safetensors",
798
+ "transformer.h.80.mlp.down_proj.bias": "model-00050-of-00059.safetensors",
799
+ "transformer.h.80.mlp.down_proj.weight": "model-00050-of-00059.safetensors",
800
+ "transformer.h.80.mlp.gate_proj.weight": "model-00049-of-00059.safetensors",
801
+ "transformer.h.80.mlp.up_proj.weight": "model-00049-of-00059.safetensors",
802
+ "transformer.h.80.post_attention_layernorm.weight": "model-00049-of-00059.safetensors",
803
+ "transformer.h.80.self_attention.dense.bias": "model-00049-of-00059.safetensors",
804
+ "transformer.h.80.self_attention.dense.weight": "model-00049-of-00059.safetensors",
805
+ "transformer.h.80.self_attention.key_value.weight": "model-00049-of-00059.safetensors",
806
+ "transformer.h.80.self_attention.query.weight": "model-00049-of-00059.safetensors",
807
+ "transformer.h.81.input_layernorm.weight": "model-00050-of-00059.safetensors",
808
+ "transformer.h.81.mlp.down_proj.bias": "model-00050-of-00059.safetensors",
809
+ "transformer.h.81.mlp.down_proj.weight": "model-00050-of-00059.safetensors",
810
+ "transformer.h.81.mlp.gate_proj.weight": "model-00050-of-00059.safetensors",
811
+ "transformer.h.81.mlp.up_proj.weight": "model-00050-of-00059.safetensors",
812
+ "transformer.h.81.post_attention_layernorm.weight": "model-00050-of-00059.safetensors",
813
+ "transformer.h.81.self_attention.dense.bias": "model-00050-of-00059.safetensors",
814
+ "transformer.h.81.self_attention.dense.weight": "model-00050-of-00059.safetensors",
815
+ "transformer.h.81.self_attention.key_value.weight": "model-00050-of-00059.safetensors",
816
+ "transformer.h.81.self_attention.query.weight": "model-00050-of-00059.safetensors",
817
+ "transformer.h.82.input_layernorm.weight": "model-00050-of-00059.safetensors",
818
+ "transformer.h.82.mlp.down_proj.bias": "model-00051-of-00059.safetensors",
819
+ "transformer.h.82.mlp.down_proj.weight": "model-00051-of-00059.safetensors",
820
+ "transformer.h.82.mlp.gate_proj.weight": "model-00050-of-00059.safetensors",
821
+ "transformer.h.82.mlp.up_proj.weight": "model-00051-of-00059.safetensors",
822
+ "transformer.h.82.post_attention_layernorm.weight": "model-00050-of-00059.safetensors",
823
+ "transformer.h.82.self_attention.dense.bias": "model-00050-of-00059.safetensors",
824
+ "transformer.h.82.self_attention.dense.weight": "model-00050-of-00059.safetensors",
825
+ "transformer.h.82.self_attention.key_value.weight": "model-00050-of-00059.safetensors",
826
+ "transformer.h.82.self_attention.query.weight": "model-00050-of-00059.safetensors",
827
+ "transformer.h.83.input_layernorm.weight": "model-00051-of-00059.safetensors",
828
+ "transformer.h.83.mlp.down_proj.bias": "model-00051-of-00059.safetensors",
829
+ "transformer.h.83.mlp.down_proj.weight": "model-00051-of-00059.safetensors",
830
+ "transformer.h.83.mlp.gate_proj.weight": "model-00051-of-00059.safetensors",
831
+ "transformer.h.83.mlp.up_proj.weight": "model-00051-of-00059.safetensors",
832
+ "transformer.h.83.post_attention_layernorm.weight": "model-00051-of-00059.safetensors",
833
+ "transformer.h.83.self_attention.dense.bias": "model-00051-of-00059.safetensors",
834
+ "transformer.h.83.self_attention.dense.weight": "model-00051-of-00059.safetensors",
835
+ "transformer.h.83.self_attention.key_value.weight": "model-00051-of-00059.safetensors",
836
+ "transformer.h.83.self_attention.query.weight": "model-00051-of-00059.safetensors",
837
+ "transformer.h.84.input_layernorm.weight": "model-00051-of-00059.safetensors",
838
+ "transformer.h.84.mlp.down_proj.bias": "model-00052-of-00059.safetensors",
839
+ "transformer.h.84.mlp.down_proj.weight": "model-00052-of-00059.safetensors",
840
+ "transformer.h.84.mlp.gate_proj.weight": "model-00052-of-00059.safetensors",
841
+ "transformer.h.84.mlp.up_proj.weight": "model-00052-of-00059.safetensors",
842
+ "transformer.h.84.post_attention_layernorm.weight": "model-00051-of-00059.safetensors",
843
+ "transformer.h.84.self_attention.dense.bias": "model-00051-of-00059.safetensors",
844
+ "transformer.h.84.self_attention.dense.weight": "model-00051-of-00059.safetensors",
845
+ "transformer.h.84.self_attention.key_value.weight": "model-00051-of-00059.safetensors",
846
+ "transformer.h.84.self_attention.query.weight": "model-00051-of-00059.safetensors",
847
+ "transformer.h.85.input_layernorm.weight": "model-00052-of-00059.safetensors",
848
+ "transformer.h.85.mlp.down_proj.bias": "model-00053-of-00059.safetensors",
849
+ "transformer.h.85.mlp.down_proj.weight": "model-00053-of-00059.safetensors",
850
+ "transformer.h.85.mlp.gate_proj.weight": "model-00052-of-00059.safetensors",
851
+ "transformer.h.85.mlp.up_proj.weight": "model-00052-of-00059.safetensors",
852
+ "transformer.h.85.post_attention_layernorm.weight": "model-00052-of-00059.safetensors",
853
+ "transformer.h.85.self_attention.dense.bias": "model-00052-of-00059.safetensors",
854
+ "transformer.h.85.self_attention.dense.weight": "model-00052-of-00059.safetensors",
855
+ "transformer.h.85.self_attention.key_value.weight": "model-00052-of-00059.safetensors",
856
+ "transformer.h.85.self_attention.query.weight": "model-00052-of-00059.safetensors",
857
+ "transformer.h.86.input_layernorm.weight": "model-00053-of-00059.safetensors",
858
+ "transformer.h.86.mlp.down_proj.bias": "model-00053-of-00059.safetensors",
859
+ "transformer.h.86.mlp.down_proj.weight": "model-00053-of-00059.safetensors",
860
+ "transformer.h.86.mlp.gate_proj.weight": "model-00053-of-00059.safetensors",
861
+ "transformer.h.86.mlp.up_proj.weight": "model-00053-of-00059.safetensors",
862
+ "transformer.h.86.post_attention_layernorm.weight": "model-00053-of-00059.safetensors",
863
+ "transformer.h.86.self_attention.dense.bias": "model-00053-of-00059.safetensors",
864
+ "transformer.h.86.self_attention.dense.weight": "model-00053-of-00059.safetensors",
865
+ "transformer.h.86.self_attention.key_value.weight": "model-00053-of-00059.safetensors",
866
+ "transformer.h.86.self_attention.query.weight": "model-00053-of-00059.safetensors",
867
+ "transformer.h.87.input_layernorm.weight": "model-00053-of-00059.safetensors",
868
+ "transformer.h.87.mlp.down_proj.bias": "model-00054-of-00059.safetensors",
869
+ "transformer.h.87.mlp.down_proj.weight": "model-00054-of-00059.safetensors",
870
+ "transformer.h.87.mlp.gate_proj.weight": "model-00053-of-00059.safetensors",
871
+ "transformer.h.87.mlp.up_proj.weight": "model-00054-of-00059.safetensors",
872
+ "transformer.h.87.post_attention_layernorm.weight": "model-00053-of-00059.safetensors",
873
+ "transformer.h.87.self_attention.dense.bias": "model-00053-of-00059.safetensors",
874
+ "transformer.h.87.self_attention.dense.weight": "model-00053-of-00059.safetensors",
875
+ "transformer.h.87.self_attention.key_value.weight": "model-00053-of-00059.safetensors",
876
+ "transformer.h.87.self_attention.query.weight": "model-00053-of-00059.safetensors",
877
+ "transformer.h.88.input_layernorm.weight": "model-00054-of-00059.safetensors",
878
+ "transformer.h.88.mlp.down_proj.bias": "model-00054-of-00059.safetensors",
879
+ "transformer.h.88.mlp.down_proj.weight": "model-00054-of-00059.safetensors",
880
+ "transformer.h.88.mlp.gate_proj.weight": "model-00054-of-00059.safetensors",
881
+ "transformer.h.88.mlp.up_proj.weight": "model-00054-of-00059.safetensors",
882
+ "transformer.h.88.post_attention_layernorm.weight": "model-00054-of-00059.safetensors",
883
+ "transformer.h.88.self_attention.dense.bias": "model-00054-of-00059.safetensors",
884
+ "transformer.h.88.self_attention.dense.weight": "model-00054-of-00059.safetensors",
885
+ "transformer.h.88.self_attention.key_value.weight": "model-00054-of-00059.safetensors",
886
+ "transformer.h.88.self_attention.query.weight": "model-00054-of-00059.safetensors",
887
+ "transformer.h.89.input_layernorm.weight": "model-00054-of-00059.safetensors",
888
+ "transformer.h.89.mlp.down_proj.bias": "model-00055-of-00059.safetensors",
889
+ "transformer.h.89.mlp.down_proj.weight": "model-00055-of-00059.safetensors",
890
+ "transformer.h.89.mlp.gate_proj.weight": "model-00055-of-00059.safetensors",
891
+ "transformer.h.89.mlp.up_proj.weight": "model-00055-of-00059.safetensors",
892
+ "transformer.h.89.post_attention_layernorm.weight": "model-00054-of-00059.safetensors",
893
+ "transformer.h.89.self_attention.dense.bias": "model-00054-of-00059.safetensors",
894
+ "transformer.h.89.self_attention.dense.weight": "model-00054-of-00059.safetensors",
895
+ "transformer.h.89.self_attention.key_value.weight": "model-00054-of-00059.safetensors",
896
+ "transformer.h.89.self_attention.query.weight": "model-00054-of-00059.safetensors",
897
+ "transformer.h.9.input_layernorm.weight": "model-00006-of-00059.safetensors",
898
+ "transformer.h.9.mlp.down_proj.bias": "model-00007-of-00059.safetensors",
899
+ "transformer.h.9.mlp.down_proj.weight": "model-00007-of-00059.safetensors",
900
+ "transformer.h.9.mlp.gate_proj.weight": "model-00007-of-00059.safetensors",
901
+ "transformer.h.9.mlp.up_proj.weight": "model-00007-of-00059.safetensors",
902
+ "transformer.h.9.post_attention_layernorm.weight": "model-00006-of-00059.safetensors",
903
+ "transformer.h.9.self_attention.dense.bias": "model-00006-of-00059.safetensors",
904
+ "transformer.h.9.self_attention.dense.weight": "model-00006-of-00059.safetensors",
905
+ "transformer.h.9.self_attention.key_value.weight": "model-00006-of-00059.safetensors",
906
+ "transformer.h.9.self_attention.query.weight": "model-00006-of-00059.safetensors",
907
+ "transformer.h.90.input_layernorm.weight": "model-00055-of-00059.safetensors",
908
+ "transformer.h.90.mlp.down_proj.bias": "model-00056-of-00059.safetensors",
909
+ "transformer.h.90.mlp.down_proj.weight": "model-00056-of-00059.safetensors",
910
+ "transformer.h.90.mlp.gate_proj.weight": "model-00055-of-00059.safetensors",
911
+ "transformer.h.90.mlp.up_proj.weight": "model-00055-of-00059.safetensors",
912
+ "transformer.h.90.post_attention_layernorm.weight": "model-00055-of-00059.safetensors",
913
+ "transformer.h.90.self_attention.dense.bias": "model-00055-of-00059.safetensors",
914
+ "transformer.h.90.self_attention.dense.weight": "model-00055-of-00059.safetensors",
915
+ "transformer.h.90.self_attention.key_value.weight": "model-00055-of-00059.safetensors",
916
+ "transformer.h.90.self_attention.query.weight": "model-00055-of-00059.safetensors",
917
+ "transformer.h.91.input_layernorm.weight": "model-00056-of-00059.safetensors",
918
+ "transformer.h.91.mlp.down_proj.bias": "model-00056-of-00059.safetensors",
919
+ "transformer.h.91.mlp.down_proj.weight": "model-00056-of-00059.safetensors",
920
+ "transformer.h.91.mlp.gate_proj.weight": "model-00056-of-00059.safetensors",
921
+ "transformer.h.91.mlp.up_proj.weight": "model-00056-of-00059.safetensors",
922
+ "transformer.h.91.post_attention_layernorm.weight": "model-00056-of-00059.safetensors",
923
+ "transformer.h.91.self_attention.dense.bias": "model-00056-of-00059.safetensors",
924
+ "transformer.h.91.self_attention.dense.weight": "model-00056-of-00059.safetensors",
925
+ "transformer.h.91.self_attention.key_value.weight": "model-00056-of-00059.safetensors",
926
+ "transformer.h.91.self_attention.query.weight": "model-00056-of-00059.safetensors",
927
+ "transformer.h.92.input_layernorm.weight": "model-00056-of-00059.safetensors",
928
+ "transformer.h.92.mlp.down_proj.bias": "model-00057-of-00059.safetensors",
929
+ "transformer.h.92.mlp.down_proj.weight": "model-00057-of-00059.safetensors",
930
+ "transformer.h.92.mlp.gate_proj.weight": "model-00056-of-00059.safetensors",
931
+ "transformer.h.92.mlp.up_proj.weight": "model-00057-of-00059.safetensors",
932
+ "transformer.h.92.post_attention_layernorm.weight": "model-00056-of-00059.safetensors",
933
+ "transformer.h.92.self_attention.dense.bias": "model-00056-of-00059.safetensors",
934
+ "transformer.h.92.self_attention.dense.weight": "model-00056-of-00059.safetensors",
935
+ "transformer.h.92.self_attention.key_value.weight": "model-00056-of-00059.safetensors",
936
+ "transformer.h.92.self_attention.query.weight": "model-00056-of-00059.safetensors",
937
+ "transformer.h.93.input_layernorm.weight": "model-00057-of-00059.safetensors",
938
+ "transformer.h.93.mlp.down_proj.bias": "model-00057-of-00059.safetensors",
939
+ "transformer.h.93.mlp.down_proj.weight": "model-00057-of-00059.safetensors",
940
+ "transformer.h.93.mlp.gate_proj.weight": "model-00057-of-00059.safetensors",
941
+ "transformer.h.93.mlp.up_proj.weight": "model-00057-of-00059.safetensors",
942
+ "transformer.h.93.post_attention_layernorm.weight": "model-00057-of-00059.safetensors",
943
+ "transformer.h.93.self_attention.dense.bias": "model-00057-of-00059.safetensors",
944
+ "transformer.h.93.self_attention.dense.weight": "model-00057-of-00059.safetensors",
945
+ "transformer.h.93.self_attention.key_value.weight": "model-00057-of-00059.safetensors",
946
+ "transformer.h.93.self_attention.query.weight": "model-00057-of-00059.safetensors",
947
+ "transformer.h.94.input_layernorm.weight": "model-00057-of-00059.safetensors",
948
+ "transformer.h.94.mlp.down_proj.bias": "model-00058-of-00059.safetensors",
949
+ "transformer.h.94.mlp.down_proj.weight": "model-00058-of-00059.safetensors",
950
+ "transformer.h.94.mlp.gate_proj.weight": "model-00058-of-00059.safetensors",
951
+ "transformer.h.94.mlp.up_proj.weight": "model-00058-of-00059.safetensors",
952
+ "transformer.h.94.post_attention_layernorm.weight": "model-00057-of-00059.safetensors",
953
+ "transformer.h.94.self_attention.dense.bias": "model-00057-of-00059.safetensors",
954
+ "transformer.h.94.self_attention.dense.weight": "model-00057-of-00059.safetensors",
955
+ "transformer.h.94.self_attention.key_value.weight": "model-00057-of-00059.safetensors",
956
+ "transformer.h.94.self_attention.query.weight": "model-00057-of-00059.safetensors",
957
+ "transformer.h.95.input_layernorm.weight": "model-00058-of-00059.safetensors",
958
+ "transformer.h.95.mlp.down_proj.bias": "model-00059-of-00059.safetensors",
959
+ "transformer.h.95.mlp.down_proj.weight": "model-00059-of-00059.safetensors",
960
+ "transformer.h.95.mlp.gate_proj.weight": "model-00058-of-00059.safetensors",
961
+ "transformer.h.95.mlp.up_proj.weight": "model-00058-of-00059.safetensors",
962
+ "transformer.h.95.post_attention_layernorm.weight": "model-00058-of-00059.safetensors",
963
+ "transformer.h.95.self_attention.dense.bias": "model-00058-of-00059.safetensors",
964
+ "transformer.h.95.self_attention.dense.weight": "model-00058-of-00059.safetensors",
965
+ "transformer.h.95.self_attention.key_value.weight": "model-00058-of-00059.safetensors",
966
+ "transformer.h.95.self_attention.query.weight": "model-00058-of-00059.safetensors",
967
+ "transformer.ln_f.weight": "model-00059-of-00059.safetensors",
968
+ "transformer.word_embeddings.weight": "model-00001-of-00059.safetensors"
969
+ }
970
+ }
special_tokens_map.json ADDED
@@ -0,0 +1,35 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<_start>",
4
+ "<_end>",
5
+ "<_pad>",
6
+ "<_user>",
7
+ "<_bot>",
8
+ "<_system>",
9
+ "<tool_call>",
10
+ "</tool_call>",
11
+ "<tool_response>",
12
+ "</tool_response>"
13
+ ],
14
+ "bos_token": {
15
+ "content": "<_start>",
16
+ "lstrip": false,
17
+ "normalized": false,
18
+ "rstrip": false,
19
+ "single_word": false
20
+ },
21
+ "eos_token": {
22
+ "content": "<_end>",
23
+ "lstrip": false,
24
+ "normalized": false,
25
+ "rstrip": false,
26
+ "single_word": false
27
+ },
28
+ "pad_token": {
29
+ "content": "<_pad>",
30
+ "lstrip": false,
31
+ "normalized": false,
32
+ "rstrip": false,
33
+ "single_word": false
34
+ }
35
+ }
tokenizer_config.json ADDED
@@ -0,0 +1,114 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": false,
3
+ "add_eos_token": false,
4
+ "added_tokens_decoder": {
5
+ "1": {
6
+ "content": "<_start>",
7
+ "lstrip": false,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false,
11
+ "special": true
12
+ },
13
+ "2": {
14
+ "content": "<_end>",
15
+ "lstrip": false,
16
+ "normalized": false,
17
+ "rstrip": false,
18
+ "single_word": false,
19
+ "special": true
20
+ },
21
+ "3": {
22
+ "content": "<_pad>",
23
+ "lstrip": false,
24
+ "normalized": false,
25
+ "rstrip": false,
26
+ "single_word": false,
27
+ "special": true
28
+ },
29
+ "4": {
30
+ "content": "<_user>",
31
+ "lstrip": false,
32
+ "normalized": false,
33
+ "rstrip": false,
34
+ "single_word": false,
35
+ "special": true
36
+ },
37
+ "5": {
38
+ "content": "<_bot>",
39
+ "lstrip": false,
40
+ "normalized": false,
41
+ "rstrip": false,
42
+ "single_word": false,
43
+ "special": true
44
+ },
45
+ "6": {
46
+ "content": "<_system>",
47
+ "lstrip": false,
48
+ "normalized": false,
49
+ "rstrip": false,
50
+ "single_word": false,
51
+ "special": true
52
+ },
53
+ "9": {
54
+ "content": "<tool_call>",
55
+ "lstrip": false,
56
+ "normalized": false,
57
+ "rstrip": false,
58
+ "single_word": false,
59
+ "special": true
60
+ },
61
+ "10": {
62
+ "content": "</tool_call>",
63
+ "lstrip": false,
64
+ "normalized": false,
65
+ "rstrip": false,
66
+ "single_word": false,
67
+ "special": true
68
+ },
69
+ "11": {
70
+ "content": "<tool_response>",
71
+ "lstrip": false,
72
+ "normalized": false,
73
+ "rstrip": false,
74
+ "single_word": false,
75
+ "special": true
76
+ },
77
+ "12": {
78
+ "content": "</tool_response>",
79
+ "lstrip": false,
80
+ "normalized": false,
81
+ "rstrip": false,
82
+ "single_word": false,
83
+ "special": true
84
+ }
85
+ },
86
+ "additional_special_tokens": [
87
+ "<_start>",
88
+ "<_end>",
89
+ "<_pad>",
90
+ "<_user>",
91
+ "<_bot>",
92
+ "<_system>",
93
+ "<tool_call>",
94
+ "</tool_call>",
95
+ "<tool_response>",
96
+ "</tool_response>"
97
+ ],
98
+ "auto_map": {
99
+ "AutoTokenizer": [
100
+ "tokenization_telechat2.Telechat2Tokenizer",
101
+ null
102
+ ]
103
+ },
104
+ "bos_token": "<_start>",
105
+ "chat_template": "{%- if tools %}\n {%- if messages[0]['role'] == 'system' %}\n {{-'<_system>'+messages[0]['content'] }}\n {%- else %}\n {{- '<_system>'+'你是中国电信星辰语义大模型，英文名是TeleChat，你是由中电信人工智能科技有限公司和中国电信人工智能研究院（TeleAI）研发的人工智能助手。' }}\n {%- endif %}\n {{- '\\n\\n# 可用工具\\n你可以调用<tools></tools>标签中包含的一个或多个工具来辅助你回答问题,以下是可用工具详情：\\n<tools>\\n' }}\n {%- for tool in tools %}\n {{- tool | tojson }}\n {{-'\\n'}}\n {%- endfor %}\n {{- '</tools>\\n\\n# 调用方法\\n你需要遵循工具的要求，使用json格式返回工具名称及参数，并用<tool_call></tool_call>包含。下方是一个调用模板：\\n<tool_call>\\n{\\\"name\\\": <function-name>, \\\"arguments\\\": <args-json-object>}\\n</tool_call>\\n\\n' }}\n{%- else %}\n {%- if messages[0]['role'] == 'system' %}\n {{- '<_system>' + messages[0]['content'] + '\\n' }}\n {%- else %}\n {{- '<_system>'+'你是中国电信星辰语义大模型，英文名是TeleChat，你是由中电信人工智能科技有限公司和中国电信人工智能研究院（TeleAI）研发的人工智能助手。\\n' }}\n {%- endif %}\n{%- endif %}\n{%- for message in messages %}\n {%- if (message.role == 'user') %}\n {{- '<_user>' + message.content }}\n {%- elif message.role == 'bot' or message.role == 'assistant' %}\n {{- '<_bot>' }}\n {%- if message.content %}\n {{- message.content }}\n {%- endif %}\n {%- for tool_call in message.tool_calls %}\n {%- if tool_call.function is defined %}\n {%- set tool_call = tool_call.function %}\n {%- endif %}\n {%- if loop.index0 == 0 %}\n {{-'<tool_call>'}}\n {%- else %}\n {{-'\\n<tool_call>'}}\n {%- endif %}\n {{- '\\n{\"name\": \"' }}{{ tool_call.name }}\n {{- '\", \"arguments\": ' }}\n {{- tool_call.arguments | tojson }}\n {{- '}\\n</tool_call>' }}\n {%- endfor %}\n {{- '<_end>\\n' }}\n {%- elif message.role == 'tool' %}\n {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != 'tool') %}\n {{- '<_user>'+'<tool_response>\\n' }}\n {%- else %}\n {{- '\\n<tool_response>\\n' }}\n {%- endif %}\n {{- message.content }}\n {{- '\\n</tool_response>' }}\n {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n {{- '<_bot>' }}\n{%- endif %}",
106
+ "clean_up_tokenization_spaces": false,
107
+ "eos_token": "<_end>",
108
+ "model_max_length": 100000000,
109
+ "pad_token": "<_pad>",
110
+ "sp_model_kwargs": {},
111
+ "split_special_tokens": false,
112
+ "tokenizer_class": "Telechat2Tokenizer",
113
+ "use_fast": false
114
+ }