Severian commited on
Commit
7a3f73b
1 Parent(s): a7b6b5f

Upload model

Browse files
config.json CHANGED
@@ -1,3 +1,28 @@
1
  {
2
- "model_type": "mistral"
3
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  {
2
+ "_name_or_path": "/content/Mistral-Base-v0.2-Nexus-IKMv5-7B",
3
+ "architectures": [
4
+ "MistralModel"
5
+ ],
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 1,
8
+ "eos_token_id": 2,
9
+ "hidden_act": "silu",
10
+ "hidden_size": 4096,
11
+ "initializer_range": 0.02,
12
+ "intermediate_size": 14336,
13
+ "max_position_embeddings": 32768,
14
+ "model_type": "mistral",
15
+ "num_attention_heads": 32,
16
+ "num_hidden_layers": 32,
17
+ "num_key_value_heads": 8,
18
+ "pad_token_id": 0,
19
+ "rms_norm_eps": 1e-05,
20
+ "rope_theta": 1000000.0,
21
+ "sliding_window": 32768,
22
+ "tie_word_embeddings": false,
23
+ "torch_dtype": "float16",
24
+ "transformers_version": "4.38.2",
25
+ "unsloth_version": "2024.3",
26
+ "use_cache": true,
27
+ "vocab_size": 32000
28
+ }
model-00001-of-00003.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a4219409f70ae0ae06c4e0c7a74c7a14842aaa8fa2a8a09e84b7a50d4525ddd1
3
+ size 4943161664
model-00002-of-00003.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:549e79a59b4b0cc998bdb622a6902c17fa806e882e6d7e6fc83605d365610088
3
+ size 4999818600
model-00003-of-00003.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e78295c2befce64ad37dd98f0c2cc24d616d6e9ca3b22a00a04205882f3c6f11
3
+ size 4278371624
model.safetensors.index.json ADDED
@@ -0,0 +1,297 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 14221320192
4
+ },
5
+ "weight_map": {
6
+ "embed_tokens.weight": "model-00001-of-00003.safetensors",
7
+ "layers.0.input_layernorm.weight": "model-00001-of-00003.safetensors",
8
+ "layers.0.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
9
+ "layers.0.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
10
+ "layers.0.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
11
+ "layers.0.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
12
+ "layers.0.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
13
+ "layers.0.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
14
+ "layers.0.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
15
+ "layers.0.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
16
+ "layers.1.input_layernorm.weight": "model-00001-of-00003.safetensors",
17
+ "layers.1.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
18
+ "layers.1.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
19
+ "layers.1.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
20
+ "layers.1.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
21
+ "layers.1.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
22
+ "layers.1.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
23
+ "layers.1.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
24
+ "layers.1.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
25
+ "layers.10.input_layernorm.weight": "model-00002-of-00003.safetensors",
26
+ "layers.10.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
27
+ "layers.10.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
28
+ "layers.10.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
29
+ "layers.10.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
30
+ "layers.10.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
31
+ "layers.10.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
32
+ "layers.10.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
33
+ "layers.10.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
34
+ "layers.11.input_layernorm.weight": "model-00002-of-00003.safetensors",
35
+ "layers.11.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
36
+ "layers.11.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
37
+ "layers.11.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
38
+ "layers.11.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
39
+ "layers.11.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
40
+ "layers.11.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
41
+ "layers.11.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
42
+ "layers.11.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
43
+ "layers.12.input_layernorm.weight": "model-00002-of-00003.safetensors",
44
+ "layers.12.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
45
+ "layers.12.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
46
+ "layers.12.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
47
+ "layers.12.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
48
+ "layers.12.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
49
+ "layers.12.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
50
+ "layers.12.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
51
+ "layers.12.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
52
+ "layers.13.input_layernorm.weight": "model-00002-of-00003.safetensors",
53
+ "layers.13.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
54
+ "layers.13.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
55
+ "layers.13.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
56
+ "layers.13.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
57
+ "layers.13.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
58
+ "layers.13.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
59
+ "layers.13.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
60
+ "layers.13.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
61
+ "layers.14.input_layernorm.weight": "model-00002-of-00003.safetensors",
62
+ "layers.14.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
63
+ "layers.14.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
64
+ "layers.14.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
65
+ "layers.14.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
66
+ "layers.14.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
67
+ "layers.14.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
68
+ "layers.14.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
69
+ "layers.14.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
70
+ "layers.15.input_layernorm.weight": "model-00002-of-00003.safetensors",
71
+ "layers.15.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
72
+ "layers.15.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
73
+ "layers.15.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
74
+ "layers.15.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
75
+ "layers.15.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
76
+ "layers.15.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
77
+ "layers.15.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
78
+ "layers.15.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
79
+ "layers.16.input_layernorm.weight": "model-00002-of-00003.safetensors",
80
+ "layers.16.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
81
+ "layers.16.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
82
+ "layers.16.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
83
+ "layers.16.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
84
+ "layers.16.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
85
+ "layers.16.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
86
+ "layers.16.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
87
+ "layers.16.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
88
+ "layers.17.input_layernorm.weight": "model-00002-of-00003.safetensors",
89
+ "layers.17.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
90
+ "layers.17.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
91
+ "layers.17.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
92
+ "layers.17.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
93
+ "layers.17.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
94
+ "layers.17.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
95
+ "layers.17.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
96
+ "layers.17.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
97
+ "layers.18.input_layernorm.weight": "model-00002-of-00003.safetensors",
98
+ "layers.18.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
99
+ "layers.18.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
100
+ "layers.18.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
101
+ "layers.18.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
102
+ "layers.18.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
103
+ "layers.18.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
104
+ "layers.18.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
105
+ "layers.18.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
106
+ "layers.19.input_layernorm.weight": "model-00002-of-00003.safetensors",
107
+ "layers.19.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
108
+ "layers.19.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
109
+ "layers.19.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
110
+ "layers.19.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
111
+ "layers.19.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
112
+ "layers.19.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
113
+ "layers.19.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
114
+ "layers.19.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
115
+ "layers.2.input_layernorm.weight": "model-00001-of-00003.safetensors",
116
+ "layers.2.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
117
+ "layers.2.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
118
+ "layers.2.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
119
+ "layers.2.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
120
+ "layers.2.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
121
+ "layers.2.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
122
+ "layers.2.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
123
+ "layers.2.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
124
+ "layers.20.input_layernorm.weight": "model-00002-of-00003.safetensors",
125
+ "layers.20.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
126
+ "layers.20.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
127
+ "layers.20.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
128
+ "layers.20.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
129
+ "layers.20.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
130
+ "layers.20.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
131
+ "layers.20.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
132
+ "layers.20.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
133
+ "layers.21.input_layernorm.weight": "model-00002-of-00003.safetensors",
134
+ "layers.21.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
135
+ "layers.21.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
136
+ "layers.21.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
137
+ "layers.21.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
138
+ "layers.21.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
139
+ "layers.21.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
140
+ "layers.21.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
141
+ "layers.21.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
142
+ "layers.22.input_layernorm.weight": "model-00003-of-00003.safetensors",
143
+ "layers.22.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
144
+ "layers.22.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
145
+ "layers.22.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
146
+ "layers.22.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
147
+ "layers.22.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
148
+ "layers.22.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
149
+ "layers.22.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
150
+ "layers.22.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
151
+ "layers.23.input_layernorm.weight": "model-00003-of-00003.safetensors",
152
+ "layers.23.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
153
+ "layers.23.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
154
+ "layers.23.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
155
+ "layers.23.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
156
+ "layers.23.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
157
+ "layers.23.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
158
+ "layers.23.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
159
+ "layers.23.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
160
+ "layers.24.input_layernorm.weight": "model-00003-of-00003.safetensors",
161
+ "layers.24.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
162
+ "layers.24.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
163
+ "layers.24.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
164
+ "layers.24.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
165
+ "layers.24.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
166
+ "layers.24.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
167
+ "layers.24.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
168
+ "layers.24.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
169
+ "layers.25.input_layernorm.weight": "model-00003-of-00003.safetensors",
170
+ "layers.25.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
171
+ "layers.25.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
172
+ "layers.25.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
173
+ "layers.25.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
174
+ "layers.25.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
175
+ "layers.25.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
176
+ "layers.25.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
177
+ "layers.25.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
178
+ "layers.26.input_layernorm.weight": "model-00003-of-00003.safetensors",
179
+ "layers.26.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
180
+ "layers.26.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
181
+ "layers.26.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
182
+ "layers.26.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
183
+ "layers.26.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
184
+ "layers.26.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
185
+ "layers.26.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
186
+ "layers.26.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
187
+ "layers.27.input_layernorm.weight": "model-00003-of-00003.safetensors",
188
+ "layers.27.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
189
+ "layers.27.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
190
+ "layers.27.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
191
+ "layers.27.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
192
+ "layers.27.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
193
+ "layers.27.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
194
+ "layers.27.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
195
+ "layers.27.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
196
+ "layers.28.input_layernorm.weight": "model-00003-of-00003.safetensors",
197
+ "layers.28.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
198
+ "layers.28.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
199
+ "layers.28.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
200
+ "layers.28.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
201
+ "layers.28.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
202
+ "layers.28.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
203
+ "layers.28.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
204
+ "layers.28.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
205
+ "layers.29.input_layernorm.weight": "model-00003-of-00003.safetensors",
206
+ "layers.29.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
207
+ "layers.29.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
208
+ "layers.29.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
209
+ "layers.29.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
210
+ "layers.29.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
211
+ "layers.29.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
212
+ "layers.29.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
213
+ "layers.29.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
214
+ "layers.3.input_layernorm.weight": "model-00001-of-00003.safetensors",
215
+ "layers.3.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
216
+ "layers.3.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
217
+ "layers.3.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
218
+ "layers.3.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
219
+ "layers.3.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
220
+ "layers.3.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
221
+ "layers.3.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
222
+ "layers.3.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
223
+ "layers.30.input_layernorm.weight": "model-00003-of-00003.safetensors",
224
+ "layers.30.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
225
+ "layers.30.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
226
+ "layers.30.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
227
+ "layers.30.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
228
+ "layers.30.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
229
+ "layers.30.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
230
+ "layers.30.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
231
+ "layers.30.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
232
+ "layers.31.input_layernorm.weight": "model-00003-of-00003.safetensors",
233
+ "layers.31.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
234
+ "layers.31.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
235
+ "layers.31.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
236
+ "layers.31.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
237
+ "layers.31.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
238
+ "layers.31.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
239
+ "layers.31.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
240
+ "layers.31.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
241
+ "layers.4.input_layernorm.weight": "model-00001-of-00003.safetensors",
242
+ "layers.4.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
243
+ "layers.4.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
244
+ "layers.4.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
245
+ "layers.4.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
246
+ "layers.4.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
247
+ "layers.4.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
248
+ "layers.4.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
249
+ "layers.4.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
250
+ "layers.5.input_layernorm.weight": "model-00001-of-00003.safetensors",
251
+ "layers.5.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
252
+ "layers.5.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
253
+ "layers.5.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
254
+ "layers.5.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
255
+ "layers.5.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
256
+ "layers.5.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
257
+ "layers.5.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
258
+ "layers.5.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
259
+ "layers.6.input_layernorm.weight": "model-00001-of-00003.safetensors",
260
+ "layers.6.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
261
+ "layers.6.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
262
+ "layers.6.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
263
+ "layers.6.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
264
+ "layers.6.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
265
+ "layers.6.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
266
+ "layers.6.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
267
+ "layers.6.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
268
+ "layers.7.input_layernorm.weight": "model-00001-of-00003.safetensors",
269
+ "layers.7.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
270
+ "layers.7.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
271
+ "layers.7.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
272
+ "layers.7.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
273
+ "layers.7.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
274
+ "layers.7.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
275
+ "layers.7.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
276
+ "layers.7.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
277
+ "layers.8.input_layernorm.weight": "model-00001-of-00003.safetensors",
278
+ "layers.8.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
279
+ "layers.8.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
280
+ "layers.8.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
281
+ "layers.8.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
282
+ "layers.8.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
283
+ "layers.8.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
284
+ "layers.8.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
285
+ "layers.8.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
286
+ "layers.9.input_layernorm.weight": "model-00001-of-00003.safetensors",
287
+ "layers.9.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
288
+ "layers.9.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
289
+ "layers.9.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
290
+ "layers.9.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
291
+ "layers.9.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
292
+ "layers.9.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
293
+ "layers.9.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
294
+ "layers.9.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
295
+ "norm.weight": "model-00003-of-00003.safetensors"
296
+ }
297
+ }