geonmin-kim
commited on
Commit
•
a78d3ba
1
Parent(s):
4789993
Upload folder using huggingface_hub
Browse files- merges.txt +0 -0
- mlc-chat-config.json +84 -0
- ndarray-cache.json +481 -0
- params_shard_0.bin +3 -0
- params_shard_1.bin +3 -0
- params_shard_10.bin +3 -0
- params_shard_11.bin +3 -0
- params_shard_12.bin +3 -0
- params_shard_13.bin +3 -0
- params_shard_14.bin +3 -0
- params_shard_15.bin +3 -0
- params_shard_16.bin +3 -0
- params_shard_17.bin +3 -0
- params_shard_2.bin +3 -0
- params_shard_3.bin +3 -0
- params_shard_4.bin +3 -0
- params_shard_5.bin +3 -0
- params_shard_6.bin +3 -0
- params_shard_7.bin +3 -0
- params_shard_8.bin +3 -0
- params_shard_9.bin +3 -0
- tokenizer.json +0 -0
- tokenizer_config.json +40 -0
- vocab.json +0 -0
merges.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
mlc-chat-config.json
ADDED
@@ -0,0 +1,84 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"version": "0.1.0",
|
3 |
+
"model_type": "qwen2",
|
4 |
+
"quantization": "q0f16",
|
5 |
+
"model_config": {
|
6 |
+
"hidden_act": "silu",
|
7 |
+
"hidden_size": 3584,
|
8 |
+
"intermediate_size": 18944,
|
9 |
+
"num_attention_heads": 28,
|
10 |
+
"num_hidden_layers": 4,
|
11 |
+
"num_key_value_heads": 4,
|
12 |
+
"rms_norm_eps": 1e-06,
|
13 |
+
"rope_theta": 1000000.0,
|
14 |
+
"vocab_size": 152064,
|
15 |
+
"tie_word_embeddings": false,
|
16 |
+
"context_window_size": 32768,
|
17 |
+
"prefill_chunk_size": 8192,
|
18 |
+
"tensor_parallel_shards": 1,
|
19 |
+
"head_dim": 128,
|
20 |
+
"dtype": "float32",
|
21 |
+
"max_batch_size": 128
|
22 |
+
},
|
23 |
+
"vocab_size": 152064,
|
24 |
+
"context_window_size": 32768,
|
25 |
+
"sliding_window_size": -1,
|
26 |
+
"prefill_chunk_size": 8192,
|
27 |
+
"attention_sink_size": -1,
|
28 |
+
"tensor_parallel_shards": 1,
|
29 |
+
"pipeline_parallel_stages": 1,
|
30 |
+
"temperature": 0.7,
|
31 |
+
"presence_penalty": 0.0,
|
32 |
+
"frequency_penalty": 0.0,
|
33 |
+
"repetition_penalty": 1.05,
|
34 |
+
"top_p": 0.8,
|
35 |
+
"tokenizer_files": [
|
36 |
+
"tokenizer.json",
|
37 |
+
"vocab.json",
|
38 |
+
"merges.txt",
|
39 |
+
"tokenizer_config.json"
|
40 |
+
],
|
41 |
+
"tokenizer_info": {
|
42 |
+
"token_postproc_method": "byte_level",
|
43 |
+
"prepend_space_in_encode": false,
|
44 |
+
"strip_space_in_decode": false
|
45 |
+
},
|
46 |
+
"conv_template": {
|
47 |
+
"name": "qwen2",
|
48 |
+
"system_template": "<|im_start|>system\n{system_message}<|im_end|>\n",
|
49 |
+
"system_message": "You are a helpful assistant.",
|
50 |
+
"system_prefix_token_ids": null,
|
51 |
+
"add_role_after_system_message": true,
|
52 |
+
"roles": {
|
53 |
+
"user": "<|im_start|>user",
|
54 |
+
"assistant": "<|im_start|>assistant"
|
55 |
+
},
|
56 |
+
"role_templates": {
|
57 |
+
"user": "{user_message}",
|
58 |
+
"assistant": "{assistant_message}",
|
59 |
+
"tool": "{tool_message}"
|
60 |
+
},
|
61 |
+
"messages": [],
|
62 |
+
"seps": [
|
63 |
+
"<|im_end|>\n"
|
64 |
+
],
|
65 |
+
"role_content_sep": "\n",
|
66 |
+
"role_empty_sep": "\n",
|
67 |
+
"stop_str": [
|
68 |
+
"<|endoftext|>",
|
69 |
+
"<|im_end|>"
|
70 |
+
],
|
71 |
+
"stop_token_ids": [
|
72 |
+
151643,
|
73 |
+
151645
|
74 |
+
],
|
75 |
+
"function_string": "",
|
76 |
+
"use_function_calling": false
|
77 |
+
},
|
78 |
+
"pad_token_id": 151643,
|
79 |
+
"bos_token_id": 151643,
|
80 |
+
"eos_token_id": [
|
81 |
+
151645,
|
82 |
+
151643
|
83 |
+
]
|
84 |
+
}
|
ndarray-cache.json
ADDED
@@ -0,0 +1,481 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"metadata": {
|
3 |
+
"ParamSize": 31,
|
4 |
+
"ParamBytes": 4044459008.0,
|
5 |
+
"BitsPerParam": 16.0
|
6 |
+
},
|
7 |
+
"records": [
|
8 |
+
{
|
9 |
+
"dataPath": "params_shard_0.bin",
|
10 |
+
"format": "raw-shard",
|
11 |
+
"nbytes": 1089994752,
|
12 |
+
"records": [
|
13 |
+
{
|
14 |
+
"name": "model.embed_tokens.weight",
|
15 |
+
"shape": [
|
16 |
+
152064,
|
17 |
+
3584
|
18 |
+
],
|
19 |
+
"dtype": "float16",
|
20 |
+
"format": "f32-to-bf16",
|
21 |
+
"nbytes": 1089994752,
|
22 |
+
"byteOffset": 0
|
23 |
+
}
|
24 |
+
],
|
25 |
+
"md5sum": "1096c3043fce52d07e686e7d70c5687e"
|
26 |
+
},
|
27 |
+
{
|
28 |
+
"dataPath": "params_shard_1.bin",
|
29 |
+
"format": "raw-shard",
|
30 |
+
"nbytes": 25690112,
|
31 |
+
"records": [
|
32 |
+
{
|
33 |
+
"name": "model.layers.0.self_attn.o_proj.weight",
|
34 |
+
"shape": [
|
35 |
+
3584,
|
36 |
+
3584
|
37 |
+
],
|
38 |
+
"dtype": "float16",
|
39 |
+
"format": "f32-to-bf16",
|
40 |
+
"nbytes": 25690112,
|
41 |
+
"byteOffset": 0
|
42 |
+
}
|
43 |
+
],
|
44 |
+
"md5sum": "fa12518f256dab0e627e73cd62f21382"
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"dataPath": "params_shard_2.bin",
|
48 |
+
"format": "raw-shard",
|
49 |
+
"nbytes": 271581184,
|
50 |
+
"records": [
|
51 |
+
{
|
52 |
+
"name": "model.layers.0.mlp.gate_up_proj.weight",
|
53 |
+
"shape": [
|
54 |
+
37888,
|
55 |
+
3584
|
56 |
+
],
|
57 |
+
"dtype": "float16",
|
58 |
+
"format": "f32-to-bf16",
|
59 |
+
"nbytes": 271581184,
|
60 |
+
"byteOffset": 0
|
61 |
+
}
|
62 |
+
],
|
63 |
+
"md5sum": "7c93523d9c9ec575cc22618781e70be9"
|
64 |
+
},
|
65 |
+
{
|
66 |
+
"dataPath": "params_shard_3.bin",
|
67 |
+
"format": "raw-shard",
|
68 |
+
"nbytes": 135790592,
|
69 |
+
"records": [
|
70 |
+
{
|
71 |
+
"name": "model.layers.0.mlp.down_proj.weight",
|
72 |
+
"shape": [
|
73 |
+
3584,
|
74 |
+
18944
|
75 |
+
],
|
76 |
+
"dtype": "float16",
|
77 |
+
"format": "f32-to-bf16",
|
78 |
+
"nbytes": 135790592,
|
79 |
+
"byteOffset": 0
|
80 |
+
}
|
81 |
+
],
|
82 |
+
"md5sum": "24b1add61255660dbf0e160b5f01eae0"
|
83 |
+
},
|
84 |
+
{
|
85 |
+
"dataPath": "params_shard_4.bin",
|
86 |
+
"format": "raw-shard",
|
87 |
+
"nbytes": 33030144,
|
88 |
+
"records": [
|
89 |
+
{
|
90 |
+
"name": "model.layers.1.self_attn.c_attn.weight",
|
91 |
+
"shape": [
|
92 |
+
4608,
|
93 |
+
3584
|
94 |
+
],
|
95 |
+
"dtype": "float16",
|
96 |
+
"format": "f32-to-bf16",
|
97 |
+
"nbytes": 33030144,
|
98 |
+
"byteOffset": 0
|
99 |
+
}
|
100 |
+
],
|
101 |
+
"md5sum": "c0eb62c7ee3c6ab51d17be1cedd4f64f"
|
102 |
+
},
|
103 |
+
{
|
104 |
+
"dataPath": "params_shard_5.bin",
|
105 |
+
"format": "raw-shard",
|
106 |
+
"nbytes": 25690112,
|
107 |
+
"records": [
|
108 |
+
{
|
109 |
+
"name": "model.layers.1.self_attn.o_proj.weight",
|
110 |
+
"shape": [
|
111 |
+
3584,
|
112 |
+
3584
|
113 |
+
],
|
114 |
+
"dtype": "float16",
|
115 |
+
"format": "f32-to-bf16",
|
116 |
+
"nbytes": 25690112,
|
117 |
+
"byteOffset": 0
|
118 |
+
}
|
119 |
+
],
|
120 |
+
"md5sum": "9722bac502585ec42f5af5a0a7b00046"
|
121 |
+
},
|
122 |
+
{
|
123 |
+
"dataPath": "params_shard_6.bin",
|
124 |
+
"format": "raw-shard",
|
125 |
+
"nbytes": 271581184,
|
126 |
+
"records": [
|
127 |
+
{
|
128 |
+
"name": "model.layers.1.mlp.gate_up_proj.weight",
|
129 |
+
"shape": [
|
130 |
+
37888,
|
131 |
+
3584
|
132 |
+
],
|
133 |
+
"dtype": "float16",
|
134 |
+
"format": "f32-to-bf16",
|
135 |
+
"nbytes": 271581184,
|
136 |
+
"byteOffset": 0
|
137 |
+
}
|
138 |
+
],
|
139 |
+
"md5sum": "a19d16581a1cbb884505e130b635797e"
|
140 |
+
},
|
141 |
+
{
|
142 |
+
"dataPath": "params_shard_7.bin",
|
143 |
+
"format": "raw-shard",
|
144 |
+
"nbytes": 135790592,
|
145 |
+
"records": [
|
146 |
+
{
|
147 |
+
"name": "model.layers.1.mlp.down_proj.weight",
|
148 |
+
"shape": [
|
149 |
+
3584,
|
150 |
+
18944
|
151 |
+
],
|
152 |
+
"dtype": "float16",
|
153 |
+
"format": "f32-to-bf16",
|
154 |
+
"nbytes": 135790592,
|
155 |
+
"byteOffset": 0
|
156 |
+
}
|
157 |
+
],
|
158 |
+
"md5sum": "14ccce743b3fc0660e2b0132db110ba5"
|
159 |
+
},
|
160 |
+
{
|
161 |
+
"dataPath": "params_shard_8.bin",
|
162 |
+
"format": "raw-shard",
|
163 |
+
"nbytes": 33030144,
|
164 |
+
"records": [
|
165 |
+
{
|
166 |
+
"name": "model.layers.2.self_attn.c_attn.weight",
|
167 |
+
"shape": [
|
168 |
+
4608,
|
169 |
+
3584
|
170 |
+
],
|
171 |
+
"dtype": "float16",
|
172 |
+
"format": "f32-to-bf16",
|
173 |
+
"nbytes": 33030144,
|
174 |
+
"byteOffset": 0
|
175 |
+
}
|
176 |
+
],
|
177 |
+
"md5sum": "f2a26b2d7170a6ed2cf070f9c0815f33"
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"dataPath": "params_shard_9.bin",
|
181 |
+
"format": "raw-shard",
|
182 |
+
"nbytes": 25690112,
|
183 |
+
"records": [
|
184 |
+
{
|
185 |
+
"name": "model.layers.2.self_attn.o_proj.weight",
|
186 |
+
"shape": [
|
187 |
+
3584,
|
188 |
+
3584
|
189 |
+
],
|
190 |
+
"dtype": "float16",
|
191 |
+
"format": "f32-to-bf16",
|
192 |
+
"nbytes": 25690112,
|
193 |
+
"byteOffset": 0
|
194 |
+
}
|
195 |
+
],
|
196 |
+
"md5sum": "ff56b50721f25ec8693491f6776565bc"
|
197 |
+
},
|
198 |
+
{
|
199 |
+
"dataPath": "params_shard_10.bin",
|
200 |
+
"format": "raw-shard",
|
201 |
+
"nbytes": 271581184,
|
202 |
+
"records": [
|
203 |
+
{
|
204 |
+
"name": "model.layers.2.mlp.gate_up_proj.weight",
|
205 |
+
"shape": [
|
206 |
+
37888,
|
207 |
+
3584
|
208 |
+
],
|
209 |
+
"dtype": "float16",
|
210 |
+
"format": "f32-to-bf16",
|
211 |
+
"nbytes": 271581184,
|
212 |
+
"byteOffset": 0
|
213 |
+
}
|
214 |
+
],
|
215 |
+
"md5sum": "1de3679fdbffb14849545694266a49a4"
|
216 |
+
},
|
217 |
+
{
|
218 |
+
"dataPath": "params_shard_11.bin",
|
219 |
+
"format": "raw-shard",
|
220 |
+
"nbytes": 135790592,
|
221 |
+
"records": [
|
222 |
+
{
|
223 |
+
"name": "model.layers.2.mlp.down_proj.weight",
|
224 |
+
"shape": [
|
225 |
+
3584,
|
226 |
+
18944
|
227 |
+
],
|
228 |
+
"dtype": "float16",
|
229 |
+
"format": "f32-to-bf16",
|
230 |
+
"nbytes": 135790592,
|
231 |
+
"byteOffset": 0
|
232 |
+
}
|
233 |
+
],
|
234 |
+
"md5sum": "a1ac20b3b6a4dbac225c68dd1e484e0f"
|
235 |
+
},
|
236 |
+
{
|
237 |
+
"dataPath": "params_shard_12.bin",
|
238 |
+
"format": "raw-shard",
|
239 |
+
"nbytes": 33030144,
|
240 |
+
"records": [
|
241 |
+
{
|
242 |
+
"name": "model.layers.3.self_attn.c_attn.weight",
|
243 |
+
"shape": [
|
244 |
+
4608,
|
245 |
+
3584
|
246 |
+
],
|
247 |
+
"dtype": "float16",
|
248 |
+
"format": "f32-to-bf16",
|
249 |
+
"nbytes": 33030144,
|
250 |
+
"byteOffset": 0
|
251 |
+
}
|
252 |
+
],
|
253 |
+
"md5sum": "b87a624913f69fc8783afa80e04d5655"
|
254 |
+
},
|
255 |
+
{
|
256 |
+
"dataPath": "params_shard_13.bin",
|
257 |
+
"format": "raw-shard",
|
258 |
+
"nbytes": 25690112,
|
259 |
+
"records": [
|
260 |
+
{
|
261 |
+
"name": "model.layers.3.self_attn.o_proj.weight",
|
262 |
+
"shape": [
|
263 |
+
3584,
|
264 |
+
3584
|
265 |
+
],
|
266 |
+
"dtype": "float16",
|
267 |
+
"format": "f32-to-bf16",
|
268 |
+
"nbytes": 25690112,
|
269 |
+
"byteOffset": 0
|
270 |
+
}
|
271 |
+
],
|
272 |
+
"md5sum": "13d09067604b7f0c1d63c98937b4e174"
|
273 |
+
},
|
274 |
+
{
|
275 |
+
"dataPath": "params_shard_14.bin",
|
276 |
+
"format": "raw-shard",
|
277 |
+
"nbytes": 271581184,
|
278 |
+
"records": [
|
279 |
+
{
|
280 |
+
"name": "model.layers.3.mlp.gate_up_proj.weight",
|
281 |
+
"shape": [
|
282 |
+
37888,
|
283 |
+
3584
|
284 |
+
],
|
285 |
+
"dtype": "float16",
|
286 |
+
"format": "f32-to-bf16",
|
287 |
+
"nbytes": 271581184,
|
288 |
+
"byteOffset": 0
|
289 |
+
}
|
290 |
+
],
|
291 |
+
"md5sum": "b6e2d8049e71f27bdd1a704f6e8f537a"
|
292 |
+
},
|
293 |
+
{
|
294 |
+
"dataPath": "params_shard_15.bin",
|
295 |
+
"format": "raw-shard",
|
296 |
+
"nbytes": 135790592,
|
297 |
+
"records": [
|
298 |
+
{
|
299 |
+
"name": "model.layers.3.mlp.down_proj.weight",
|
300 |
+
"shape": [
|
301 |
+
3584,
|
302 |
+
18944
|
303 |
+
],
|
304 |
+
"dtype": "float16",
|
305 |
+
"format": "f32-to-bf16",
|
306 |
+
"nbytes": 135790592,
|
307 |
+
"byteOffset": 0
|
308 |
+
}
|
309 |
+
],
|
310 |
+
"md5sum": "165dc6852e9590980d6b93e38b55e09f"
|
311 |
+
},
|
312 |
+
{
|
313 |
+
"dataPath": "params_shard_16.bin",
|
314 |
+
"format": "raw-shard",
|
315 |
+
"nbytes": 1089994752,
|
316 |
+
"records": [
|
317 |
+
{
|
318 |
+
"name": "lm_head.weight",
|
319 |
+
"shape": [
|
320 |
+
152064,
|
321 |
+
3584
|
322 |
+
],
|
323 |
+
"dtype": "float16",
|
324 |
+
"format": "f32-to-bf16",
|
325 |
+
"nbytes": 1089994752,
|
326 |
+
"byteOffset": 0
|
327 |
+
}
|
328 |
+
],
|
329 |
+
"md5sum": "b0a68a612362cfcfb3d746bbd9bbb73d"
|
330 |
+
},
|
331 |
+
{
|
332 |
+
"dataPath": "params_shard_17.bin",
|
333 |
+
"format": "raw-shard",
|
334 |
+
"nbytes": 33131520,
|
335 |
+
"records": [
|
336 |
+
{
|
337 |
+
"name": "model.layers.0.self_attn.c_attn.weight",
|
338 |
+
"shape": [
|
339 |
+
4608,
|
340 |
+
3584
|
341 |
+
],
|
342 |
+
"dtype": "float16",
|
343 |
+
"format": "f32-to-bf16",
|
344 |
+
"nbytes": 33030144,
|
345 |
+
"byteOffset": 0
|
346 |
+
},
|
347 |
+
{
|
348 |
+
"name": "model.layers.0.self_attn.c_attn.bias",
|
349 |
+
"shape": [
|
350 |
+
4608
|
351 |
+
],
|
352 |
+
"dtype": "float16",
|
353 |
+
"format": "f32-to-bf16",
|
354 |
+
"nbytes": 9216,
|
355 |
+
"byteOffset": 33030144
|
356 |
+
},
|
357 |
+
{
|
358 |
+
"name": "model.layers.0.input_layernorm.weight",
|
359 |
+
"shape": [
|
360 |
+
3584
|
361 |
+
],
|
362 |
+
"dtype": "float16",
|
363 |
+
"format": "f32-to-bf16",
|
364 |
+
"nbytes": 7168,
|
365 |
+
"byteOffset": 33039360
|
366 |
+
},
|
367 |
+
{
|
368 |
+
"name": "model.layers.0.post_attention_layernorm.weight",
|
369 |
+
"shape": [
|
370 |
+
3584
|
371 |
+
],
|
372 |
+
"dtype": "float16",
|
373 |
+
"format": "f32-to-bf16",
|
374 |
+
"nbytes": 7168,
|
375 |
+
"byteOffset": 33046528
|
376 |
+
},
|
377 |
+
{
|
378 |
+
"name": "model.layers.1.self_attn.c_attn.bias",
|
379 |
+
"shape": [
|
380 |
+
4608
|
381 |
+
],
|
382 |
+
"dtype": "float16",
|
383 |
+
"format": "f32-to-bf16",
|
384 |
+
"nbytes": 9216,
|
385 |
+
"byteOffset": 33053696
|
386 |
+
},
|
387 |
+
{
|
388 |
+
"name": "model.layers.1.input_layernorm.weight",
|
389 |
+
"shape": [
|
390 |
+
3584
|
391 |
+
],
|
392 |
+
"dtype": "float16",
|
393 |
+
"format": "f32-to-bf16",
|
394 |
+
"nbytes": 7168,
|
395 |
+
"byteOffset": 33062912
|
396 |
+
},
|
397 |
+
{
|
398 |
+
"name": "model.layers.1.post_attention_layernorm.weight",
|
399 |
+
"shape": [
|
400 |
+
3584
|
401 |
+
],
|
402 |
+
"dtype": "float16",
|
403 |
+
"format": "f32-to-bf16",
|
404 |
+
"nbytes": 7168,
|
405 |
+
"byteOffset": 33070080
|
406 |
+
},
|
407 |
+
{
|
408 |
+
"name": "model.layers.2.self_attn.c_attn.bias",
|
409 |
+
"shape": [
|
410 |
+
4608
|
411 |
+
],
|
412 |
+
"dtype": "float16",
|
413 |
+
"format": "f32-to-bf16",
|
414 |
+
"nbytes": 9216,
|
415 |
+
"byteOffset": 33077248
|
416 |
+
},
|
417 |
+
{
|
418 |
+
"name": "model.layers.2.input_layernorm.weight",
|
419 |
+
"shape": [
|
420 |
+
3584
|
421 |
+
],
|
422 |
+
"dtype": "float16",
|
423 |
+
"format": "f32-to-bf16",
|
424 |
+
"nbytes": 7168,
|
425 |
+
"byteOffset": 33086464
|
426 |
+
},
|
427 |
+
{
|
428 |
+
"name": "model.layers.2.post_attention_layernorm.weight",
|
429 |
+
"shape": [
|
430 |
+
3584
|
431 |
+
],
|
432 |
+
"dtype": "float16",
|
433 |
+
"format": "f32-to-bf16",
|
434 |
+
"nbytes": 7168,
|
435 |
+
"byteOffset": 33093632
|
436 |
+
},
|
437 |
+
{
|
438 |
+
"name": "model.layers.3.self_attn.c_attn.bias",
|
439 |
+
"shape": [
|
440 |
+
4608
|
441 |
+
],
|
442 |
+
"dtype": "float16",
|
443 |
+
"format": "f32-to-bf16",
|
444 |
+
"nbytes": 9216,
|
445 |
+
"byteOffset": 33100800
|
446 |
+
},
|
447 |
+
{
|
448 |
+
"name": "model.layers.3.input_layernorm.weight",
|
449 |
+
"shape": [
|
450 |
+
3584
|
451 |
+
],
|
452 |
+
"dtype": "float16",
|
453 |
+
"format": "f32-to-bf16",
|
454 |
+
"nbytes": 7168,
|
455 |
+
"byteOffset": 33110016
|
456 |
+
},
|
457 |
+
{
|
458 |
+
"name": "model.layers.3.post_attention_layernorm.weight",
|
459 |
+
"shape": [
|
460 |
+
3584
|
461 |
+
],
|
462 |
+
"dtype": "float16",
|
463 |
+
"format": "f32-to-bf16",
|
464 |
+
"nbytes": 7168,
|
465 |
+
"byteOffset": 33117184
|
466 |
+
},
|
467 |
+
{
|
468 |
+
"name": "model.norm.weight",
|
469 |
+
"shape": [
|
470 |
+
3584
|
471 |
+
],
|
472 |
+
"dtype": "float16",
|
473 |
+
"format": "f32-to-bf16",
|
474 |
+
"nbytes": 7168,
|
475 |
+
"byteOffset": 33124352
|
476 |
+
}
|
477 |
+
],
|
478 |
+
"md5sum": "418e357fec9f143413cbad405043e1a3"
|
479 |
+
}
|
480 |
+
]
|
481 |
+
}
|
params_shard_0.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:36da690306d53161bf9bafc286ba6e3f7cdf86087e4108f222808eef912b95e9
|
3 |
+
size 1089994752
|
params_shard_1.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a95701b4ac52cf96783ae59da8cc4bda3108a3735f7acbeb8dc94dc090162602
|
3 |
+
size 25690112
|
params_shard_10.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:4c97007a80f5d663c8f29200aecf25dc4fe86d1577f5419007097a1704ec31ad
|
3 |
+
size 271581184
|
params_shard_11.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:6bed4ca735d520d7b65d3c3749e3c7ab57ba8084f293cb360e7ad9b7b28a690e
|
3 |
+
size 135790592
|
params_shard_12.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:243b5570437c89ec3390cfd78d7516be819dd63e47d63dca30168f92b8508175
|
3 |
+
size 33030144
|
params_shard_13.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f59e59ab6e7aac7019a2a49cc5edc0ee345305c4a3df8d44f223c2e250ee86ac
|
3 |
+
size 25690112
|
params_shard_14.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:68124542428386161db487e9847876100a03a9439b4b498c97b50ddccac01fd4
|
3 |
+
size 271581184
|
params_shard_15.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:146c9e250131dfbc9dbd9f9471725dd52a0e8ccaa871fc50f27630f69122b71b
|
3 |
+
size 135790592
|
params_shard_16.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:4e9086fecf0ed82fb9bacf4c0d7b0df7bccbf6913c96b66d0aea0d584f9b6a07
|
3 |
+
size 1089994752
|
params_shard_17.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ebfc478a389b18f545317ef4ac22c0e3b41c7a7256a990298074a17571d86595
|
3 |
+
size 33131520
|
params_shard_2.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:9395feaec3fe572f6a6897e71ed4d03bd1c14af64ab276b004be27717e229299
|
3 |
+
size 271581184
|
params_shard_3.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a6069620d271b1b3ab4c7b9de62818c3696473df8e3bc4c25bddeb5eae920475
|
3 |
+
size 135790592
|
params_shard_4.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:7b45d4d3d3fcaa4fa0f37f548f3b621267c9dfae22139e6db1e0b238482aa9a7
|
3 |
+
size 33030144
|
params_shard_5.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:eeb491ebaf8a3548196dcb0f9d1a6b6daf4026f58dd5122ad00682660dc43e38
|
3 |
+
size 25690112
|
params_shard_6.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:4c68843cacfd63f02cdc10845444bbb5e231402b8f6050e5bd44aebbc4a5141a
|
3 |
+
size 271581184
|
params_shard_7.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:882c9177364e87fc41a818b4c1483d37693c8877e8d9772f37522851ad7faf49
|
3 |
+
size 135790592
|
params_shard_8.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:935739307f975adf6512eb1192068918a29675314f7e8fdffa267d14591c60a6
|
3 |
+
size 33030144
|
params_shard_9.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ea386c8e0b5cc577e38921f05ea61c7332d26abec37350b8352740c7f5ffa697
|
3 |
+
size 25690112
|
tokenizer.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
tokenizer_config.json
ADDED
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"add_prefix_space": false,
|
3 |
+
"added_tokens_decoder": {
|
4 |
+
"151643": {
|
5 |
+
"content": "<|endoftext|>",
|
6 |
+
"lstrip": false,
|
7 |
+
"normalized": false,
|
8 |
+
"rstrip": false,
|
9 |
+
"single_word": false,
|
10 |
+
"special": true
|
11 |
+
},
|
12 |
+
"151644": {
|
13 |
+
"content": "<|im_start|>",
|
14 |
+
"lstrip": false,
|
15 |
+
"normalized": false,
|
16 |
+
"rstrip": false,
|
17 |
+
"single_word": false,
|
18 |
+
"special": true
|
19 |
+
},
|
20 |
+
"151645": {
|
21 |
+
"content": "<|im_end|>",
|
22 |
+
"lstrip": false,
|
23 |
+
"normalized": false,
|
24 |
+
"rstrip": false,
|
25 |
+
"single_word": false,
|
26 |
+
"special": true
|
27 |
+
}
|
28 |
+
},
|
29 |
+
"additional_special_tokens": ["<|im_start|>", "<|im_end|>"],
|
30 |
+
"bos_token": null,
|
31 |
+
"chat_template": "{% for message in messages %}{% if loop.first and messages[0]['role'] != 'system' %}{{ '<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n' }}{% endif %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
|
32 |
+
"clean_up_tokenization_spaces": false,
|
33 |
+
"eos_token": "<|im_end|>",
|
34 |
+
"errors": "replace",
|
35 |
+
"model_max_length": 131072,
|
36 |
+
"pad_token": "<|endoftext|>",
|
37 |
+
"split_special_tokens": false,
|
38 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
39 |
+
"unk_token": null
|
40 |
+
}
|
vocab.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|