Shubhra Pandit commited on
Commit
c94e540
·
1 Parent(s): 6ee3e98

Upload model files

Browse files
.gitattributes CHANGED
@@ -33,3 +33,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ tokenizer.json filter=lfs diff=lfs merge=lfs -text
37
+
chat_template.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ {
2
+ "chat_template": "{%- set today = strftime_now(\"%Y-%m-%d\") %}\n{%- set default_system_message = \"You are Pixtral, a Large Language Model (LLM) created by Mistral AI, a French startup headquartered in Paris.\\nYou power an AI assistant called Le Chat.\\nYour knowledge base was last updated on 2023-10-01.\\nThe current date is \" + today + \"\\n\\nWhen you're not sure about some information, you say that you don't have the information and don't make up anything.\\nIf the user's question is not clear, ambiguous, or does not provide enough context for you to accurately answer the question, you do not try to answer it right away and you rather ask the user to clarify their request (e.g. \\\"What are some good restaurants around me?\\\" => \\\"Where are you?\\\" or \\\"When is the next flight to Tokyo\\\" => \\\"Where do you travel from?\\\").\\nYou are always very attentive to dates, in particular you try to resolve dates (e.g. \\\"today\\\" is \" + today + \") and when asked about information at specific dates, you discard information that is at another date.\\nYou follow these instructions in all languages, and always respond to the user in the language they use or request.\\nNext sections describe the capabilities that you have.\\n\\n# WEB BROWSING INSTRUCTIONS\\n\\nYou cannot perform any web search or access internet to open URLs, links etc. If it seems like the user is expecting you to do so, you clarify the situation and ask the user to copy paste the text directly in the chat.\\n\\n# MULTI-MODAL INSTRUCTIONS\\n\\nYou have the ability to read images, but you cannot generate images. You also cannot transcribe audio files or videos.\\nYou cannot read nor transcribe audio files or videos and you cannot read images.\" %}\n\n\n\n{{- bos_token }}\n{%- if messages[0]['role'] == 'system' %}\n {%- set system_message = messages[0]['content'] %}\n {%- set loop_messages = messages[1:] %}\n{%- else %}\n {%- set system_message = default_system_message %}\n {%- set loop_messages = messages %}\n{%- endif %}\n\n{{- '[SYSTEM_PROMPT] ' + system_message + '[/SYSTEM_PROMPT]' }}\n\n{%- for message in loop_messages %}\n {%- if message['role'] == 'user' %}\n {%- if message[\"content\"] is not string %}\n {{- '[INST]' }}\n {%- set ns = namespace(text_count=0, text_index=-1) %}\n {%- for chunk in message[\"content\"] %}\n {%- if chunk[\"type\"] == \"text\" %}\n {%- set ns.text_count = ns.text_count + 1 %}\n {%- set ns.text_index = loop.index0 %}\n {%- endif %}\n {%- endfor %}\n {#- When there is only one text chunk, put it at the end to match the behaviour of mistral-common #}\n {%- if ns.text_count == 1 and message[\"content\"]|length > 1 %}\n {%- set content = message[\"content\"][:ns.text_index] + message[\"content\"][ns.text_index + 1:] + message[\"content\"][ns.text_index: ns.text_index + 1] %}\n {%- else %}\n {%- set content = message[\"content\"] %}\n {%- endif %}\n {%- for chunk in content %}\n {%- if chunk[\"type\"] == \"text\" %}\n {{- \" \" + chunk[\"text\"]|trim }}\n {%- elif chunk[\"type\"] == \"image\" %}\n {{- \"[IMG]\" }}\n {%- else %}\n {{- raise_exception(\"Unrecognized content type!\") }}\n {%- endif %}\n {%- endfor %}\n {%- else %}\n {{- '[INST] ' }}\n {{- message[\"content\"] }}\n {%- endif %}\n {{- '[/INST]' }}\n {%- elif message['role'] == 'assistant' %}\n {{- ' ' + message['content'] + eos_token }}\n {%- else %}\n {{- raise_exception('Only user and assistant roles are supported, plus an optional initial system message!') }}\n {%- endif %}\n{%- endfor %}\n"
3
+ }
config.json ADDED
@@ -0,0 +1,369 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "/network/shubhra/Pixtral-Large-Instruct-2411-pr23/",
3
+ "architectures": [
4
+ "LlavaForConditionalGeneration"
5
+ ],
6
+ "ignore_index": -100,
7
+ "image_seq_length": 1,
8
+ "image_token_index": 10,
9
+ "model_type": "llava",
10
+ "multimodal_projector_bias": false,
11
+ "projector_hidden_act": "gelu",
12
+ "quantization_config": {
13
+ "config_groups": {
14
+ "group_0": {
15
+ "input_activations": {
16
+ "actorder": null,
17
+ "block_structure": null,
18
+ "dynamic": true,
19
+ "group_size": null,
20
+ "num_bits": 8,
21
+ "observer": null,
22
+ "observer_kwargs": {},
23
+ "strategy": "token",
24
+ "symmetric": true,
25
+ "type": "int"
26
+ },
27
+ "output_activations": null,
28
+ "targets": [
29
+ "Linear"
30
+ ],
31
+ "weights": {
32
+ "actorder": null,
33
+ "block_structure": null,
34
+ "dynamic": false,
35
+ "group_size": null,
36
+ "num_bits": 8,
37
+ "observer": "minmax",
38
+ "observer_kwargs": {},
39
+ "strategy": "channel",
40
+ "symmetric": true,
41
+ "type": "int"
42
+ }
43
+ }
44
+ },
45
+ "format": "int-quantized",
46
+ "global_compression_ratio": 1.3136315653954034,
47
+ "ignore": [
48
+ "vision_tower.transformer.layers.0.feed_forward.gate_proj",
49
+ "vision_tower.transformer.layers.0.feed_forward.up_proj",
50
+ "vision_tower.transformer.layers.0.feed_forward.down_proj",
51
+ "vision_tower.transformer.layers.0.attention.k_proj",
52
+ "vision_tower.transformer.layers.0.attention.v_proj",
53
+ "vision_tower.transformer.layers.0.attention.q_proj",
54
+ "vision_tower.transformer.layers.0.attention.o_proj",
55
+ "vision_tower.transformer.layers.1.feed_forward.gate_proj",
56
+ "vision_tower.transformer.layers.1.feed_forward.up_proj",
57
+ "vision_tower.transformer.layers.1.feed_forward.down_proj",
58
+ "vision_tower.transformer.layers.1.attention.k_proj",
59
+ "vision_tower.transformer.layers.1.attention.v_proj",
60
+ "vision_tower.transformer.layers.1.attention.q_proj",
61
+ "vision_tower.transformer.layers.1.attention.o_proj",
62
+ "vision_tower.transformer.layers.2.feed_forward.gate_proj",
63
+ "vision_tower.transformer.layers.2.feed_forward.up_proj",
64
+ "vision_tower.transformer.layers.2.feed_forward.down_proj",
65
+ "vision_tower.transformer.layers.2.attention.k_proj",
66
+ "vision_tower.transformer.layers.2.attention.v_proj",
67
+ "vision_tower.transformer.layers.2.attention.q_proj",
68
+ "vision_tower.transformer.layers.2.attention.o_proj",
69
+ "vision_tower.transformer.layers.3.feed_forward.gate_proj",
70
+ "vision_tower.transformer.layers.3.feed_forward.up_proj",
71
+ "vision_tower.transformer.layers.3.feed_forward.down_proj",
72
+ "vision_tower.transformer.layers.3.attention.k_proj",
73
+ "vision_tower.transformer.layers.3.attention.v_proj",
74
+ "vision_tower.transformer.layers.3.attention.q_proj",
75
+ "vision_tower.transformer.layers.3.attention.o_proj",
76
+ "vision_tower.transformer.layers.4.feed_forward.gate_proj",
77
+ "vision_tower.transformer.layers.4.feed_forward.up_proj",
78
+ "vision_tower.transformer.layers.4.feed_forward.down_proj",
79
+ "vision_tower.transformer.layers.4.attention.k_proj",
80
+ "vision_tower.transformer.layers.4.attention.v_proj",
81
+ "vision_tower.transformer.layers.4.attention.q_proj",
82
+ "vision_tower.transformer.layers.4.attention.o_proj",
83
+ "vision_tower.transformer.layers.5.feed_forward.gate_proj",
84
+ "vision_tower.transformer.layers.5.feed_forward.up_proj",
85
+ "vision_tower.transformer.layers.5.feed_forward.down_proj",
86
+ "vision_tower.transformer.layers.5.attention.k_proj",
87
+ "vision_tower.transformer.layers.5.attention.v_proj",
88
+ "vision_tower.transformer.layers.5.attention.q_proj",
89
+ "vision_tower.transformer.layers.5.attention.o_proj",
90
+ "vision_tower.transformer.layers.6.feed_forward.gate_proj",
91
+ "vision_tower.transformer.layers.6.feed_forward.up_proj",
92
+ "vision_tower.transformer.layers.6.feed_forward.down_proj",
93
+ "vision_tower.transformer.layers.6.attention.k_proj",
94
+ "vision_tower.transformer.layers.6.attention.v_proj",
95
+ "vision_tower.transformer.layers.6.attention.q_proj",
96
+ "vision_tower.transformer.layers.6.attention.o_proj",
97
+ "vision_tower.transformer.layers.7.feed_forward.gate_proj",
98
+ "vision_tower.transformer.layers.7.feed_forward.up_proj",
99
+ "vision_tower.transformer.layers.7.feed_forward.down_proj",
100
+ "vision_tower.transformer.layers.7.attention.k_proj",
101
+ "vision_tower.transformer.layers.7.attention.v_proj",
102
+ "vision_tower.transformer.layers.7.attention.q_proj",
103
+ "vision_tower.transformer.layers.7.attention.o_proj",
104
+ "vision_tower.transformer.layers.8.feed_forward.gate_proj",
105
+ "vision_tower.transformer.layers.8.feed_forward.up_proj",
106
+ "vision_tower.transformer.layers.8.feed_forward.down_proj",
107
+ "vision_tower.transformer.layers.8.attention.k_proj",
108
+ "vision_tower.transformer.layers.8.attention.v_proj",
109
+ "vision_tower.transformer.layers.8.attention.q_proj",
110
+ "vision_tower.transformer.layers.8.attention.o_proj",
111
+ "vision_tower.transformer.layers.9.feed_forward.gate_proj",
112
+ "vision_tower.transformer.layers.9.feed_forward.up_proj",
113
+ "vision_tower.transformer.layers.9.feed_forward.down_proj",
114
+ "vision_tower.transformer.layers.9.attention.k_proj",
115
+ "vision_tower.transformer.layers.9.attention.v_proj",
116
+ "vision_tower.transformer.layers.9.attention.q_proj",
117
+ "vision_tower.transformer.layers.9.attention.o_proj",
118
+ "vision_tower.transformer.layers.10.feed_forward.gate_proj",
119
+ "vision_tower.transformer.layers.10.feed_forward.up_proj",
120
+ "vision_tower.transformer.layers.10.feed_forward.down_proj",
121
+ "vision_tower.transformer.layers.10.attention.k_proj",
122
+ "vision_tower.transformer.layers.10.attention.v_proj",
123
+ "vision_tower.transformer.layers.10.attention.q_proj",
124
+ "vision_tower.transformer.layers.10.attention.o_proj",
125
+ "vision_tower.transformer.layers.11.feed_forward.gate_proj",
126
+ "vision_tower.transformer.layers.11.feed_forward.up_proj",
127
+ "vision_tower.transformer.layers.11.feed_forward.down_proj",
128
+ "vision_tower.transformer.layers.11.attention.k_proj",
129
+ "vision_tower.transformer.layers.11.attention.v_proj",
130
+ "vision_tower.transformer.layers.11.attention.q_proj",
131
+ "vision_tower.transformer.layers.11.attention.o_proj",
132
+ "vision_tower.transformer.layers.12.feed_forward.gate_proj",
133
+ "vision_tower.transformer.layers.12.feed_forward.up_proj",
134
+ "vision_tower.transformer.layers.12.feed_forward.down_proj",
135
+ "vision_tower.transformer.layers.12.attention.k_proj",
136
+ "vision_tower.transformer.layers.12.attention.v_proj",
137
+ "vision_tower.transformer.layers.12.attention.q_proj",
138
+ "vision_tower.transformer.layers.12.attention.o_proj",
139
+ "vision_tower.transformer.layers.13.feed_forward.gate_proj",
140
+ "vision_tower.transformer.layers.13.feed_forward.up_proj",
141
+ "vision_tower.transformer.layers.13.feed_forward.down_proj",
142
+ "vision_tower.transformer.layers.13.attention.k_proj",
143
+ "vision_tower.transformer.layers.13.attention.v_proj",
144
+ "vision_tower.transformer.layers.13.attention.q_proj",
145
+ "vision_tower.transformer.layers.13.attention.o_proj",
146
+ "vision_tower.transformer.layers.14.feed_forward.gate_proj",
147
+ "vision_tower.transformer.layers.14.feed_forward.up_proj",
148
+ "vision_tower.transformer.layers.14.feed_forward.down_proj",
149
+ "vision_tower.transformer.layers.14.attention.k_proj",
150
+ "vision_tower.transformer.layers.14.attention.v_proj",
151
+ "vision_tower.transformer.layers.14.attention.q_proj",
152
+ "vision_tower.transformer.layers.14.attention.o_proj",
153
+ "vision_tower.transformer.layers.15.feed_forward.gate_proj",
154
+ "vision_tower.transformer.layers.15.feed_forward.up_proj",
155
+ "vision_tower.transformer.layers.15.feed_forward.down_proj",
156
+ "vision_tower.transformer.layers.15.attention.k_proj",
157
+ "vision_tower.transformer.layers.15.attention.v_proj",
158
+ "vision_tower.transformer.layers.15.attention.q_proj",
159
+ "vision_tower.transformer.layers.15.attention.o_proj",
160
+ "vision_tower.transformer.layers.16.feed_forward.gate_proj",
161
+ "vision_tower.transformer.layers.16.feed_forward.up_proj",
162
+ "vision_tower.transformer.layers.16.feed_forward.down_proj",
163
+ "vision_tower.transformer.layers.16.attention.k_proj",
164
+ "vision_tower.transformer.layers.16.attention.v_proj",
165
+ "vision_tower.transformer.layers.16.attention.q_proj",
166
+ "vision_tower.transformer.layers.16.attention.o_proj",
167
+ "vision_tower.transformer.layers.17.feed_forward.gate_proj",
168
+ "vision_tower.transformer.layers.17.feed_forward.up_proj",
169
+ "vision_tower.transformer.layers.17.feed_forward.down_proj",
170
+ "vision_tower.transformer.layers.17.attention.k_proj",
171
+ "vision_tower.transformer.layers.17.attention.v_proj",
172
+ "vision_tower.transformer.layers.17.attention.q_proj",
173
+ "vision_tower.transformer.layers.17.attention.o_proj",
174
+ "vision_tower.transformer.layers.18.feed_forward.gate_proj",
175
+ "vision_tower.transformer.layers.18.feed_forward.up_proj",
176
+ "vision_tower.transformer.layers.18.feed_forward.down_proj",
177
+ "vision_tower.transformer.layers.18.attention.k_proj",
178
+ "vision_tower.transformer.layers.18.attention.v_proj",
179
+ "vision_tower.transformer.layers.18.attention.q_proj",
180
+ "vision_tower.transformer.layers.18.attention.o_proj",
181
+ "vision_tower.transformer.layers.19.feed_forward.gate_proj",
182
+ "vision_tower.transformer.layers.19.feed_forward.up_proj",
183
+ "vision_tower.transformer.layers.19.feed_forward.down_proj",
184
+ "vision_tower.transformer.layers.19.attention.k_proj",
185
+ "vision_tower.transformer.layers.19.attention.v_proj",
186
+ "vision_tower.transformer.layers.19.attention.q_proj",
187
+ "vision_tower.transformer.layers.19.attention.o_proj",
188
+ "vision_tower.transformer.layers.20.feed_forward.gate_proj",
189
+ "vision_tower.transformer.layers.20.feed_forward.up_proj",
190
+ "vision_tower.transformer.layers.20.feed_forward.down_proj",
191
+ "vision_tower.transformer.layers.20.attention.k_proj",
192
+ "vision_tower.transformer.layers.20.attention.v_proj",
193
+ "vision_tower.transformer.layers.20.attention.q_proj",
194
+ "vision_tower.transformer.layers.20.attention.o_proj",
195
+ "vision_tower.transformer.layers.21.feed_forward.gate_proj",
196
+ "vision_tower.transformer.layers.21.feed_forward.up_proj",
197
+ "vision_tower.transformer.layers.21.feed_forward.down_proj",
198
+ "vision_tower.transformer.layers.21.attention.k_proj",
199
+ "vision_tower.transformer.layers.21.attention.v_proj",
200
+ "vision_tower.transformer.layers.21.attention.q_proj",
201
+ "vision_tower.transformer.layers.21.attention.o_proj",
202
+ "vision_tower.transformer.layers.22.feed_forward.gate_proj",
203
+ "vision_tower.transformer.layers.22.feed_forward.up_proj",
204
+ "vision_tower.transformer.layers.22.feed_forward.down_proj",
205
+ "vision_tower.transformer.layers.22.attention.k_proj",
206
+ "vision_tower.transformer.layers.22.attention.v_proj",
207
+ "vision_tower.transformer.layers.22.attention.q_proj",
208
+ "vision_tower.transformer.layers.22.attention.o_proj",
209
+ "vision_tower.transformer.layers.23.feed_forward.gate_proj",
210
+ "vision_tower.transformer.layers.23.feed_forward.up_proj",
211
+ "vision_tower.transformer.layers.23.feed_forward.down_proj",
212
+ "vision_tower.transformer.layers.23.attention.k_proj",
213
+ "vision_tower.transformer.layers.23.attention.v_proj",
214
+ "vision_tower.transformer.layers.23.attention.q_proj",
215
+ "vision_tower.transformer.layers.23.attention.o_proj",
216
+ "vision_tower.transformer.layers.24.feed_forward.gate_proj",
217
+ "vision_tower.transformer.layers.24.feed_forward.up_proj",
218
+ "vision_tower.transformer.layers.24.feed_forward.down_proj",
219
+ "vision_tower.transformer.layers.24.attention.k_proj",
220
+ "vision_tower.transformer.layers.24.attention.v_proj",
221
+ "vision_tower.transformer.layers.24.attention.q_proj",
222
+ "vision_tower.transformer.layers.24.attention.o_proj",
223
+ "vision_tower.transformer.layers.25.feed_forward.gate_proj",
224
+ "vision_tower.transformer.layers.25.feed_forward.up_proj",
225
+ "vision_tower.transformer.layers.25.feed_forward.down_proj",
226
+ "vision_tower.transformer.layers.25.attention.k_proj",
227
+ "vision_tower.transformer.layers.25.attention.v_proj",
228
+ "vision_tower.transformer.layers.25.attention.q_proj",
229
+ "vision_tower.transformer.layers.25.attention.o_proj",
230
+ "vision_tower.transformer.layers.26.feed_forward.gate_proj",
231
+ "vision_tower.transformer.layers.26.feed_forward.up_proj",
232
+ "vision_tower.transformer.layers.26.feed_forward.down_proj",
233
+ "vision_tower.transformer.layers.26.attention.k_proj",
234
+ "vision_tower.transformer.layers.26.attention.v_proj",
235
+ "vision_tower.transformer.layers.26.attention.q_proj",
236
+ "vision_tower.transformer.layers.26.attention.o_proj",
237
+ "vision_tower.transformer.layers.27.feed_forward.gate_proj",
238
+ "vision_tower.transformer.layers.27.feed_forward.up_proj",
239
+ "vision_tower.transformer.layers.27.feed_forward.down_proj",
240
+ "vision_tower.transformer.layers.27.attention.k_proj",
241
+ "vision_tower.transformer.layers.27.attention.v_proj",
242
+ "vision_tower.transformer.layers.27.attention.q_proj",
243
+ "vision_tower.transformer.layers.27.attention.o_proj",
244
+ "vision_tower.transformer.layers.28.feed_forward.gate_proj",
245
+ "vision_tower.transformer.layers.28.feed_forward.up_proj",
246
+ "vision_tower.transformer.layers.28.feed_forward.down_proj",
247
+ "vision_tower.transformer.layers.28.attention.k_proj",
248
+ "vision_tower.transformer.layers.28.attention.v_proj",
249
+ "vision_tower.transformer.layers.28.attention.q_proj",
250
+ "vision_tower.transformer.layers.28.attention.o_proj",
251
+ "vision_tower.transformer.layers.29.feed_forward.gate_proj",
252
+ "vision_tower.transformer.layers.29.feed_forward.up_proj",
253
+ "vision_tower.transformer.layers.29.feed_forward.down_proj",
254
+ "vision_tower.transformer.layers.29.attention.k_proj",
255
+ "vision_tower.transformer.layers.29.attention.v_proj",
256
+ "vision_tower.transformer.layers.29.attention.q_proj",
257
+ "vision_tower.transformer.layers.29.attention.o_proj",
258
+ "vision_tower.transformer.layers.30.feed_forward.gate_proj",
259
+ "vision_tower.transformer.layers.30.feed_forward.up_proj",
260
+ "vision_tower.transformer.layers.30.feed_forward.down_proj",
261
+ "vision_tower.transformer.layers.30.attention.k_proj",
262
+ "vision_tower.transformer.layers.30.attention.v_proj",
263
+ "vision_tower.transformer.layers.30.attention.q_proj",
264
+ "vision_tower.transformer.layers.30.attention.o_proj",
265
+ "vision_tower.transformer.layers.31.feed_forward.gate_proj",
266
+ "vision_tower.transformer.layers.31.feed_forward.up_proj",
267
+ "vision_tower.transformer.layers.31.feed_forward.down_proj",
268
+ "vision_tower.transformer.layers.31.attention.k_proj",
269
+ "vision_tower.transformer.layers.31.attention.v_proj",
270
+ "vision_tower.transformer.layers.31.attention.q_proj",
271
+ "vision_tower.transformer.layers.31.attention.o_proj",
272
+ "vision_tower.transformer.layers.32.feed_forward.gate_proj",
273
+ "vision_tower.transformer.layers.32.feed_forward.up_proj",
274
+ "vision_tower.transformer.layers.32.feed_forward.down_proj",
275
+ "vision_tower.transformer.layers.32.attention.k_proj",
276
+ "vision_tower.transformer.layers.32.attention.v_proj",
277
+ "vision_tower.transformer.layers.32.attention.q_proj",
278
+ "vision_tower.transformer.layers.32.attention.o_proj",
279
+ "vision_tower.transformer.layers.33.feed_forward.gate_proj",
280
+ "vision_tower.transformer.layers.33.feed_forward.up_proj",
281
+ "vision_tower.transformer.layers.33.feed_forward.down_proj",
282
+ "vision_tower.transformer.layers.33.attention.k_proj",
283
+ "vision_tower.transformer.layers.33.attention.v_proj",
284
+ "vision_tower.transformer.layers.33.attention.q_proj",
285
+ "vision_tower.transformer.layers.33.attention.o_proj",
286
+ "vision_tower.transformer.layers.34.feed_forward.gate_proj",
287
+ "vision_tower.transformer.layers.34.feed_forward.up_proj",
288
+ "vision_tower.transformer.layers.34.feed_forward.down_proj",
289
+ "vision_tower.transformer.layers.34.attention.k_proj",
290
+ "vision_tower.transformer.layers.34.attention.v_proj",
291
+ "vision_tower.transformer.layers.34.attention.q_proj",
292
+ "vision_tower.transformer.layers.34.attention.o_proj",
293
+ "vision_tower.transformer.layers.35.feed_forward.gate_proj",
294
+ "vision_tower.transformer.layers.35.feed_forward.up_proj",
295
+ "vision_tower.transformer.layers.35.feed_forward.down_proj",
296
+ "vision_tower.transformer.layers.35.attention.k_proj",
297
+ "vision_tower.transformer.layers.35.attention.v_proj",
298
+ "vision_tower.transformer.layers.35.attention.q_proj",
299
+ "vision_tower.transformer.layers.35.attention.o_proj",
300
+ "vision_tower.transformer.layers.36.feed_forward.gate_proj",
301
+ "vision_tower.transformer.layers.36.feed_forward.up_proj",
302
+ "vision_tower.transformer.layers.36.feed_forward.down_proj",
303
+ "vision_tower.transformer.layers.36.attention.k_proj",
304
+ "vision_tower.transformer.layers.36.attention.v_proj",
305
+ "vision_tower.transformer.layers.36.attention.q_proj",
306
+ "vision_tower.transformer.layers.36.attention.o_proj",
307
+ "vision_tower.transformer.layers.37.feed_forward.gate_proj",
308
+ "vision_tower.transformer.layers.37.feed_forward.up_proj",
309
+ "vision_tower.transformer.layers.37.feed_forward.down_proj",
310
+ "vision_tower.transformer.layers.37.attention.k_proj",
311
+ "vision_tower.transformer.layers.37.attention.v_proj",
312
+ "vision_tower.transformer.layers.37.attention.q_proj",
313
+ "vision_tower.transformer.layers.37.attention.o_proj",
314
+ "vision_tower.transformer.layers.38.feed_forward.gate_proj",
315
+ "vision_tower.transformer.layers.38.feed_forward.up_proj",
316
+ "vision_tower.transformer.layers.38.feed_forward.down_proj",
317
+ "vision_tower.transformer.layers.38.attention.k_proj",
318
+ "vision_tower.transformer.layers.38.attention.v_proj",
319
+ "vision_tower.transformer.layers.38.attention.q_proj",
320
+ "vision_tower.transformer.layers.38.attention.o_proj",
321
+ "vision_tower.transformer.layers.39.feed_forward.gate_proj",
322
+ "vision_tower.transformer.layers.39.feed_forward.up_proj",
323
+ "vision_tower.transformer.layers.39.feed_forward.down_proj",
324
+ "vision_tower.transformer.layers.39.attention.k_proj",
325
+ "vision_tower.transformer.layers.39.attention.v_proj",
326
+ "vision_tower.transformer.layers.39.attention.q_proj",
327
+ "vision_tower.transformer.layers.39.attention.o_proj",
328
+ "multi_modal_projector.linear_1",
329
+ "multi_modal_projector.linear_2",
330
+ "language_model.lm_head"
331
+ ],
332
+ "kv_cache_scheme": null,
333
+ "quant_method": "compressed-tensors",
334
+ "quantization_status": "compressed"
335
+ },
336
+ "text_config": {
337
+ "hidden_size": 12288,
338
+ "intermediate_size": 28672,
339
+ "is_composition": true,
340
+ "max_position_embeddings": 131072,
341
+ "model_type": "mistral",
342
+ "norm_eps": 1e-05,
343
+ "num_attention_heads": 96,
344
+ "num_hidden_layers": 88,
345
+ "num_key_value_heads": 8,
346
+ "rms_norm_eps": 1e-05,
347
+ "rope_theta": 1000000000.0,
348
+ "sliding_window": null,
349
+ "vocab_size": 32768
350
+ },
351
+ "torch_dtype": "bfloat16",
352
+ "transformers_version": "4.48.2",
353
+ "vision_config": {
354
+ "head_dim": 88,
355
+ "hidden_act": "silu",
356
+ "hidden_size": 1408,
357
+ "image_break_token_id": 14,
358
+ "image_end_token_id": 15,
359
+ "image_size": 1024,
360
+ "image_token_id": 10,
361
+ "intermediate_size": 6144,
362
+ "model_type": "pixtral",
363
+ "num_hidden_layers": 40,
364
+ "patch_size": 16,
365
+ "rope_theta": 10000.0
366
+ },
367
+ "vision_feature_layer": -1,
368
+ "vision_feature_select_strategy": "full"
369
+ }
generation_config.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 1,
4
+ "eos_token_id": 2,
5
+ "transformers_version": "4.48.2"
6
+ }
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
model-00001-of-00027.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7fd731e5adec594d150a4e67b84daaa22536873bfae852c686bef4782398115f
3
+ size 4886876976
model-00002-of-00027.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4776bd5931ed3cffcc5b79cdfe0a0297468690b701994739cfa0d9687b06e6e0
3
+ size 4832697856
model-00003-of-00027.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5b87222774ec59d0c6d467b586ee0e678f1efb20b48edcf34033a5e630b94b76
3
+ size 4857850368
model-00004-of-00027.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3345698eb73f1755c16fccc7997d8e06e94bc53127c3d024ee03509bc20c6bf8
3
+ size 4832697888
model-00005-of-00027.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dc4b5121809f97f5426173ab844e384b737ee391c2909b56eef1cdc55040e62f
3
+ size 4857850416
model-00006-of-00027.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6b0098de5f2b6285b9a08972041b8e64979e9fe34e230399a939d1d0859105d1
3
+ size 4832697920
model-00007-of-00027.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:936173f51767443021ded8a24f1e32feb9f3c9b079d38ecae653bc56bc030e0b
3
+ size 4857850416
model-00008-of-00027.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7edc53ad3bcd38e759ec718fbfd4023d8c002d70dc7c50e96dec49a34b5ad6d8
3
+ size 4832697920
model-00009-of-00027.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:951db249c56cd5ed44c36eb2be3e76128570f784c6d7d2939bd8d445de1ef4c4
3
+ size 4857850416
model-00010-of-00027.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bfc1c7882f18670f591dd90a7920b2e586ad71ab4481fd7675c1333abaab171c
3
+ size 4832697920
model-00011-of-00027.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e8c6faae1e05ab568049847503468fcc738c02fe0adf278bd2e97b7e3e8c167c
3
+ size 4857850416
model-00012-of-00027.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a69c572a4a8e7c29674a9789264d4c252bdb2be144a8660217e005f9f3cb30d8
3
+ size 4832697920
model-00013-of-00027.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ac05bd435a39c47d5d597c77a447dbcf78896e2fc3038c523a34a454da73cb68
3
+ size 4857850416
model-00014-of-00027.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5625522241eb5c741e7524077e0304a25ede3568594bed4006c21cc98796e48e
3
+ size 4832697920
model-00015-of-00027.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a3de4f18568f7f5200d0d53dbdbf6c862f38c87ac0bff1319f941d06fc60d6f4
3
+ size 4857850416
model-00016-of-00027.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d3eceb4e08ed1dad05861da3d7ed1cbcf4e534fcfe3b328585222896aa11b6b8
3
+ size 4832697920
model-00017-of-00027.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:39ab24461e62bbe440cd82841fff541fef716af4d7bf1031dcec55c27f448de8
3
+ size 4857850416
model-00018-of-00027.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:068a2e4b5a82f843f9f23afbb47de893e0d377a4000f3fed817dd07dca333b33
3
+ size 4832697920
model-00019-of-00027.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8434537ff2955941dbdd601fdfa55ed92d0087301fec8e00dff6dff0e97a34af
3
+ size 4857850416
model-00020-of-00027.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9387b9ba622c6e595dd4fbf85d7f74db818011b7978dba59d0ac4a81f45c6f4f
3
+ size 4832697920
model-00021-of-00027.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f081734f4dfbaf09abf2e27b7fd8ffe4c31cb632d852f391c6ec7774c2fb8f92
3
+ size 4857850416
model-00022-of-00027.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5c9426fea39248f7477c6a668cd3603a5c0e8e8fd439b74a9845f2aa0f508825
3
+ size 4832697920
model-00023-of-00027.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9f988909a9780e98237467d95e80a74cfb80a77c1c7e545b3adca668fd0dfa92
3
+ size 4857850416
model-00024-of-00027.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3f3522e7b2098a22cd9e44a04daf384b2ce7d7a75a4143d1f62816a7c4694157
3
+ size 4832697920
model-00025-of-00027.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:04adc6736bb637bf8676f68ea1a3ccb95c973c028a36fceddb295c2fdfe7f9f4
3
+ size 4857850416
model-00026-of-00027.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3381bee55dfc21dfa965630a51a98c7e4f893cd2ba97bf8e2396bccf0b31a71a
3
+ size 4505512592
model-00027-of-00027.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:317efe3355bc4f792d02386b1a062e53807219d5815a2cd6c5a59e01f44ba16c
3
+ size 805306512
model.safetensors.index.json ADDED
The diff for this file is too large to render. See raw diff
 
preprocessor_config.json ADDED
@@ -0,0 +1,27 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_convert_rgb": true,
3
+ "do_normalize": true,
4
+ "do_rescale": true,
5
+ "do_resize": true,
6
+ "image_mean": [
7
+ 0.48145466,
8
+ 0.4578275,
9
+ 0.40821073
10
+ ],
11
+ "image_processor_type": "PixtralImageProcessor",
12
+ "image_std": [
13
+ 0.26862954,
14
+ 0.26130258,
15
+ 0.27577711
16
+ ],
17
+ "patch_size": {
18
+ "height": 16,
19
+ "width": 16
20
+ },
21
+ "processor_class": "PixtralProcessor",
22
+ "resample": 3,
23
+ "rescale_factor": 0.00392156862745098,
24
+ "size": {
25
+ "longest_edge": 1024
26
+ }
27
+ }
processor_config.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "image_break_token": "[IMG_BREAK]",
3
+ "image_end_token": "[IMG_END]",
4
+ "image_token": "[IMG]",
5
+ "patch_size": 16,
6
+ "processor_class": "PixtralProcessor"
7
+ }
recipe.yaml ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ DEFAULT_stage:
2
+ DEFAULT_modifiers:
3
+ GPTQModifier:
4
+ sequential_targets: [MistralDecoderLayer]
5
+ scheme: W8A8
6
+ targets: Linear
7
+ ignore: ['re:.*lm_head', 're:vision_tower.*', 're:multi_modal_projector.*']
special_tokens_map.json ADDED
@@ -0,0 +1,773 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<s>",
4
+ "</s>",
5
+ "[INST]",
6
+ "[/INST]",
7
+ "[TOOL_CALLS]",
8
+ "[AVAILABLE_TOOLS]",
9
+ "[/AVAILABLE_TOOLS]",
10
+ "[TOOL_RESULTS]",
11
+ "[/TOOL_RESULTS]",
12
+ "[IMG]",
13
+ "[PREFIX]",
14
+ "[MIDDLE]",
15
+ "[SUFFIX]",
16
+ "[IMG_BREAK]",
17
+ "[IMG_END]",
18
+ "[SYSTEM_PROMPT]",
19
+ "[/SYSTEM_PROMPT]",
20
+ "[TOOL_CONTENT]",
21
+ "[control_17]",
22
+ "[control_18]",
23
+ "[control_19]",
24
+ "[control_20]",
25
+ "[control_21]",
26
+ "[control_22]",
27
+ "[control_23]",
28
+ "[control_24]",
29
+ "[control_25]",
30
+ "[control_26]",
31
+ "[control_27]",
32
+ "[control_28]",
33
+ "[control_29]",
34
+ "[control_30]",
35
+ "[control_31]",
36
+ "[control_32]",
37
+ "[control_33]",
38
+ "[control_34]",
39
+ "[control_35]",
40
+ "[control_36]",
41
+ "[control_37]",
42
+ "[control_38]",
43
+ "[control_39]",
44
+ "[control_40]",
45
+ "[control_41]",
46
+ "[control_42]",
47
+ "[control_43]",
48
+ "[control_44]",
49
+ "[control_45]",
50
+ "[control_46]",
51
+ "[control_47]",
52
+ "[control_48]",
53
+ "[control_49]",
54
+ "[control_50]",
55
+ "[control_51]",
56
+ "[control_52]",
57
+ "[control_53]",
58
+ "[control_54]",
59
+ "[control_55]",
60
+ "[control_56]",
61
+ "[control_57]",
62
+ "[control_58]",
63
+ "[control_59]",
64
+ "[control_60]",
65
+ "[control_61]",
66
+ "[control_62]",
67
+ "[control_63]",
68
+ "[control_64]",
69
+ "[control_65]",
70
+ "[control_66]",
71
+ "[control_67]",
72
+ "[control_68]",
73
+ "[control_69]",
74
+ "[control_70]",
75
+ "[control_71]",
76
+ "[control_72]",
77
+ "[control_73]",
78
+ "[control_74]",
79
+ "[control_75]",
80
+ "[control_76]",
81
+ "[control_77]",
82
+ "[control_78]",
83
+ "[control_79]",
84
+ "[control_80]",
85
+ "[control_81]",
86
+ "[control_82]",
87
+ "[control_83]",
88
+ "[control_84]",
89
+ "[control_85]",
90
+ "[control_86]",
91
+ "[control_87]",
92
+ "[control_88]",
93
+ "[control_89]",
94
+ "[control_90]",
95
+ "[control_91]",
96
+ "[control_92]",
97
+ "[control_93]",
98
+ "[control_94]",
99
+ "[control_95]",
100
+ "[control_96]",
101
+ "[control_97]",
102
+ "[control_98]",
103
+ "[control_99]",
104
+ "[control_100]",
105
+ "[control_101]",
106
+ "[control_102]",
107
+ "[control_103]",
108
+ "[control_104]",
109
+ "[control_105]",
110
+ "[control_106]",
111
+ "[control_107]",
112
+ "[control_108]",
113
+ "[control_109]",
114
+ "[control_110]",
115
+ "[control_111]",
116
+ "[control_112]",
117
+ "[control_113]",
118
+ "[control_114]",
119
+ "[control_115]",
120
+ "[control_116]",
121
+ "[control_117]",
122
+ "[control_118]",
123
+ "[control_119]",
124
+ "[control_120]",
125
+ "[control_121]",
126
+ "[control_122]",
127
+ "[control_123]",
128
+ "[control_124]",
129
+ "[control_125]",
130
+ "[control_126]",
131
+ "[control_127]",
132
+ "[control_128]",
133
+ "[control_129]",
134
+ "[control_130]",
135
+ "[control_131]",
136
+ "[control_132]",
137
+ "[control_133]",
138
+ "[control_134]",
139
+ "[control_135]",
140
+ "[control_136]",
141
+ "[control_137]",
142
+ "[control_138]",
143
+ "[control_139]",
144
+ "[control_140]",
145
+ "[control_141]",
146
+ "[control_142]",
147
+ "[control_143]",
148
+ "[control_144]",
149
+ "[control_145]",
150
+ "[control_146]",
151
+ "[control_147]",
152
+ "[control_148]",
153
+ "[control_149]",
154
+ "[control_150]",
155
+ "[control_151]",
156
+ "[control_152]",
157
+ "[control_153]",
158
+ "[control_154]",
159
+ "[control_155]",
160
+ "[control_156]",
161
+ "[control_157]",
162
+ "[control_158]",
163
+ "[control_159]",
164
+ "[control_160]",
165
+ "[control_161]",
166
+ "[control_162]",
167
+ "[control_163]",
168
+ "[control_164]",
169
+ "[control_165]",
170
+ "[control_166]",
171
+ "[control_167]",
172
+ "[control_168]",
173
+ "[control_169]",
174
+ "[control_170]",
175
+ "[control_171]",
176
+ "[control_172]",
177
+ "[control_173]",
178
+ "[control_174]",
179
+ "[control_175]",
180
+ "[control_176]",
181
+ "[control_177]",
182
+ "[control_178]",
183
+ "[control_179]",
184
+ "[control_180]",
185
+ "[control_181]",
186
+ "[control_182]",
187
+ "[control_183]",
188
+ "[control_184]",
189
+ "[control_185]",
190
+ "[control_186]",
191
+ "[control_187]",
192
+ "[control_188]",
193
+ "[control_189]",
194
+ "[control_190]",
195
+ "[control_191]",
196
+ "[control_192]",
197
+ "[control_193]",
198
+ "[control_194]",
199
+ "[control_195]",
200
+ "[control_196]",
201
+ "[control_197]",
202
+ "[control_198]",
203
+ "[control_199]",
204
+ "[control_200]",
205
+ "[control_201]",
206
+ "[control_202]",
207
+ "[control_203]",
208
+ "[control_204]",
209
+ "[control_205]",
210
+ "[control_206]",
211
+ "[control_207]",
212
+ "[control_208]",
213
+ "[control_209]",
214
+ "[control_210]",
215
+ "[control_211]",
216
+ "[control_212]",
217
+ "[control_213]",
218
+ "[control_214]",
219
+ "[control_215]",
220
+ "[control_216]",
221
+ "[control_217]",
222
+ "[control_218]",
223
+ "[control_219]",
224
+ "[control_220]",
225
+ "[control_221]",
226
+ "[control_222]",
227
+ "[control_223]",
228
+ "[control_224]",
229
+ "[control_225]",
230
+ "[control_226]",
231
+ "[control_227]",
232
+ "[control_228]",
233
+ "[control_229]",
234
+ "[control_230]",
235
+ "[control_231]",
236
+ "[control_232]",
237
+ "[control_233]",
238
+ "[control_234]",
239
+ "[control_235]",
240
+ "[control_236]",
241
+ "[control_237]",
242
+ "[control_238]",
243
+ "[control_239]",
244
+ "[control_240]",
245
+ "[control_241]",
246
+ "[control_242]",
247
+ "[control_243]",
248
+ "[control_244]",
249
+ "[control_245]",
250
+ "[control_246]",
251
+ "[control_247]",
252
+ "[control_248]",
253
+ "[control_249]",
254
+ "[control_250]",
255
+ "[control_251]",
256
+ "[control_252]",
257
+ "[control_253]",
258
+ "[control_254]",
259
+ "[control_255]",
260
+ "[control_256]",
261
+ "[control_257]",
262
+ "[control_258]",
263
+ "[control_259]",
264
+ "[control_260]",
265
+ "[control_261]",
266
+ "[control_262]",
267
+ "[control_263]",
268
+ "[control_264]",
269
+ "[control_265]",
270
+ "[control_266]",
271
+ "[control_267]",
272
+ "[control_268]",
273
+ "[control_269]",
274
+ "[control_270]",
275
+ "[control_271]",
276
+ "[control_272]",
277
+ "[control_273]",
278
+ "[control_274]",
279
+ "[control_275]",
280
+ "[control_276]",
281
+ "[control_277]",
282
+ "[control_278]",
283
+ "[control_279]",
284
+ "[control_280]",
285
+ "[control_281]",
286
+ "[control_282]",
287
+ "[control_283]",
288
+ "[control_284]",
289
+ "[control_285]",
290
+ "[control_286]",
291
+ "[control_287]",
292
+ "[control_288]",
293
+ "[control_289]",
294
+ "[control_290]",
295
+ "[control_291]",
296
+ "[control_292]",
297
+ "[control_293]",
298
+ "[control_294]",
299
+ "[control_295]",
300
+ "[control_296]",
301
+ "[control_297]",
302
+ "[control_298]",
303
+ "[control_299]",
304
+ "[control_300]",
305
+ "[control_301]",
306
+ "[control_302]",
307
+ "[control_303]",
308
+ "[control_304]",
309
+ "[control_305]",
310
+ "[control_306]",
311
+ "[control_307]",
312
+ "[control_308]",
313
+ "[control_309]",
314
+ "[control_310]",
315
+ "[control_311]",
316
+ "[control_312]",
317
+ "[control_313]",
318
+ "[control_314]",
319
+ "[control_315]",
320
+ "[control_316]",
321
+ "[control_317]",
322
+ "[control_318]",
323
+ "[control_319]",
324
+ "[control_320]",
325
+ "[control_321]",
326
+ "[control_322]",
327
+ "[control_323]",
328
+ "[control_324]",
329
+ "[control_325]",
330
+ "[control_326]",
331
+ "[control_327]",
332
+ "[control_328]",
333
+ "[control_329]",
334
+ "[control_330]",
335
+ "[control_331]",
336
+ "[control_332]",
337
+ "[control_333]",
338
+ "[control_334]",
339
+ "[control_335]",
340
+ "[control_336]",
341
+ "[control_337]",
342
+ "[control_338]",
343
+ "[control_339]",
344
+ "[control_340]",
345
+ "[control_341]",
346
+ "[control_342]",
347
+ "[control_343]",
348
+ "[control_344]",
349
+ "[control_345]",
350
+ "[control_346]",
351
+ "[control_347]",
352
+ "[control_348]",
353
+ "[control_349]",
354
+ "[control_350]",
355
+ "[control_351]",
356
+ "[control_352]",
357
+ "[control_353]",
358
+ "[control_354]",
359
+ "[control_355]",
360
+ "[control_356]",
361
+ "[control_357]",
362
+ "[control_358]",
363
+ "[control_359]",
364
+ "[control_360]",
365
+ "[control_361]",
366
+ "[control_362]",
367
+ "[control_363]",
368
+ "[control_364]",
369
+ "[control_365]",
370
+ "[control_366]",
371
+ "[control_367]",
372
+ "[control_368]",
373
+ "[control_369]",
374
+ "[control_370]",
375
+ "[control_371]",
376
+ "[control_372]",
377
+ "[control_373]",
378
+ "[control_374]",
379
+ "[control_375]",
380
+ "[control_376]",
381
+ "[control_377]",
382
+ "[control_378]",
383
+ "[control_379]",
384
+ "[control_380]",
385
+ "[control_381]",
386
+ "[control_382]",
387
+ "[control_383]",
388
+ "[control_384]",
389
+ "[control_385]",
390
+ "[control_386]",
391
+ "[control_387]",
392
+ "[control_388]",
393
+ "[control_389]",
394
+ "[control_390]",
395
+ "[control_391]",
396
+ "[control_392]",
397
+ "[control_393]",
398
+ "[control_394]",
399
+ "[control_395]",
400
+ "[control_396]",
401
+ "[control_397]",
402
+ "[control_398]",
403
+ "[control_399]",
404
+ "[control_400]",
405
+ "[control_401]",
406
+ "[control_402]",
407
+ "[control_403]",
408
+ "[control_404]",
409
+ "[control_405]",
410
+ "[control_406]",
411
+ "[control_407]",
412
+ "[control_408]",
413
+ "[control_409]",
414
+ "[control_410]",
415
+ "[control_411]",
416
+ "[control_412]",
417
+ "[control_413]",
418
+ "[control_414]",
419
+ "[control_415]",
420
+ "[control_416]",
421
+ "[control_417]",
422
+ "[control_418]",
423
+ "[control_419]",
424
+ "[control_420]",
425
+ "[control_421]",
426
+ "[control_422]",
427
+ "[control_423]",
428
+ "[control_424]",
429
+ "[control_425]",
430
+ "[control_426]",
431
+ "[control_427]",
432
+ "[control_428]",
433
+ "[control_429]",
434
+ "[control_430]",
435
+ "[control_431]",
436
+ "[control_432]",
437
+ "[control_433]",
438
+ "[control_434]",
439
+ "[control_435]",
440
+ "[control_436]",
441
+ "[control_437]",
442
+ "[control_438]",
443
+ "[control_439]",
444
+ "[control_440]",
445
+ "[control_441]",
446
+ "[control_442]",
447
+ "[control_443]",
448
+ "[control_444]",
449
+ "[control_445]",
450
+ "[control_446]",
451
+ "[control_447]",
452
+ "[control_448]",
453
+ "[control_449]",
454
+ "[control_450]",
455
+ "[control_451]",
456
+ "[control_452]",
457
+ "[control_453]",
458
+ "[control_454]",
459
+ "[control_455]",
460
+ "[control_456]",
461
+ "[control_457]",
462
+ "[control_458]",
463
+ "[control_459]",
464
+ "[control_460]",
465
+ "[control_461]",
466
+ "[control_462]",
467
+ "[control_463]",
468
+ "[control_464]",
469
+ "[control_465]",
470
+ "[control_466]",
471
+ "[control_467]",
472
+ "[control_468]",
473
+ "[control_469]",
474
+ "[control_470]",
475
+ "[control_471]",
476
+ "[control_472]",
477
+ "[control_473]",
478
+ "[control_474]",
479
+ "[control_475]",
480
+ "[control_476]",
481
+ "[control_477]",
482
+ "[control_478]",
483
+ "[control_479]",
484
+ "[control_480]",
485
+ "[control_481]",
486
+ "[control_482]",
487
+ "[control_483]",
488
+ "[control_484]",
489
+ "[control_485]",
490
+ "[control_486]",
491
+ "[control_487]",
492
+ "[control_488]",
493
+ "[control_489]",
494
+ "[control_490]",
495
+ "[control_491]",
496
+ "[control_492]",
497
+ "[control_493]",
498
+ "[control_494]",
499
+ "[control_495]",
500
+ "[control_496]",
501
+ "[control_497]",
502
+ "[control_498]",
503
+ "[control_499]",
504
+ "[control_500]",
505
+ "[control_501]",
506
+ "[control_502]",
507
+ "[control_503]",
508
+ "[control_504]",
509
+ "[control_505]",
510
+ "[control_506]",
511
+ "[control_507]",
512
+ "[control_508]",
513
+ "[control_509]",
514
+ "[control_510]",
515
+ "[control_511]",
516
+ "[control_512]",
517
+ "[control_513]",
518
+ "[control_514]",
519
+ "[control_515]",
520
+ "[control_516]",
521
+ "[control_517]",
522
+ "[control_518]",
523
+ "[control_519]",
524
+ "[control_520]",
525
+ "[control_521]",
526
+ "[control_522]",
527
+ "[control_523]",
528
+ "[control_524]",
529
+ "[control_525]",
530
+ "[control_526]",
531
+ "[control_527]",
532
+ "[control_528]",
533
+ "[control_529]",
534
+ "[control_530]",
535
+ "[control_531]",
536
+ "[control_532]",
537
+ "[control_533]",
538
+ "[control_534]",
539
+ "[control_535]",
540
+ "[control_536]",
541
+ "[control_537]",
542
+ "[control_538]",
543
+ "[control_539]",
544
+ "[control_540]",
545
+ "[control_541]",
546
+ "[control_542]",
547
+ "[control_543]",
548
+ "[control_544]",
549
+ "[control_545]",
550
+ "[control_546]",
551
+ "[control_547]",
552
+ "[control_548]",
553
+ "[control_549]",
554
+ "[control_550]",
555
+ "[control_551]",
556
+ "[control_552]",
557
+ "[control_553]",
558
+ "[control_554]",
559
+ "[control_555]",
560
+ "[control_556]",
561
+ "[control_557]",
562
+ "[control_558]",
563
+ "[control_559]",
564
+ "[control_560]",
565
+ "[control_561]",
566
+ "[control_562]",
567
+ "[control_563]",
568
+ "[control_564]",
569
+ "[control_565]",
570
+ "[control_566]",
571
+ "[control_567]",
572
+ "[control_568]",
573
+ "[control_569]",
574
+ "[control_570]",
575
+ "[control_571]",
576
+ "[control_572]",
577
+ "[control_573]",
578
+ "[control_574]",
579
+ "[control_575]",
580
+ "[control_576]",
581
+ "[control_577]",
582
+ "[control_578]",
583
+ "[control_579]",
584
+ "[control_580]",
585
+ "[control_581]",
586
+ "[control_582]",
587
+ "[control_583]",
588
+ "[control_584]",
589
+ "[control_585]",
590
+ "[control_586]",
591
+ "[control_587]",
592
+ "[control_588]",
593
+ "[control_589]",
594
+ "[control_590]",
595
+ "[control_591]",
596
+ "[control_592]",
597
+ "[control_593]",
598
+ "[control_594]",
599
+ "[control_595]",
600
+ "[control_596]",
601
+ "[control_597]",
602
+ "[control_598]",
603
+ "[control_599]",
604
+ "[control_600]",
605
+ "[control_601]",
606
+ "[control_602]",
607
+ "[control_603]",
608
+ "[control_604]",
609
+ "[control_605]",
610
+ "[control_606]",
611
+ "[control_607]",
612
+ "[control_608]",
613
+ "[control_609]",
614
+ "[control_610]",
615
+ "[control_611]",
616
+ "[control_612]",
617
+ "[control_613]",
618
+ "[control_614]",
619
+ "[control_615]",
620
+ "[control_616]",
621
+ "[control_617]",
622
+ "[control_618]",
623
+ "[control_619]",
624
+ "[control_620]",
625
+ "[control_621]",
626
+ "[control_622]",
627
+ "[control_623]",
628
+ "[control_624]",
629
+ "[control_625]",
630
+ "[control_626]",
631
+ "[control_627]",
632
+ "[control_628]",
633
+ "[control_629]",
634
+ "[control_630]",
635
+ "[control_631]",
636
+ "[control_632]",
637
+ "[control_633]",
638
+ "[control_634]",
639
+ "[control_635]",
640
+ "[control_636]",
641
+ "[control_637]",
642
+ "[control_638]",
643
+ "[control_639]",
644
+ "[control_640]",
645
+ "[control_641]",
646
+ "[control_642]",
647
+ "[control_643]",
648
+ "[control_644]",
649
+ "[control_645]",
650
+ "[control_646]",
651
+ "[control_647]",
652
+ "[control_648]",
653
+ "[control_649]",
654
+ "[control_650]",
655
+ "[control_651]",
656
+ "[control_652]",
657
+ "[control_653]",
658
+ "[control_654]",
659
+ "[control_655]",
660
+ "[control_656]",
661
+ "[control_657]",
662
+ "[control_658]",
663
+ "[control_659]",
664
+ "[control_660]",
665
+ "[control_661]",
666
+ "[control_662]",
667
+ "[control_663]",
668
+ "[control_664]",
669
+ "[control_665]",
670
+ "[control_666]",
671
+ "[control_667]",
672
+ "[control_668]",
673
+ "[control_669]",
674
+ "[control_670]",
675
+ "[control_671]",
676
+ "[control_672]",
677
+ "[control_673]",
678
+ "[control_674]",
679
+ "[control_675]",
680
+ "[control_676]",
681
+ "[control_677]",
682
+ "[control_678]",
683
+ "[control_679]",
684
+ "[control_680]",
685
+ "[control_681]",
686
+ "[control_682]",
687
+ "[control_683]",
688
+ "[control_684]",
689
+ "[control_685]",
690
+ "[control_686]",
691
+ "[control_687]",
692
+ "[control_688]",
693
+ "[control_689]",
694
+ "[control_690]",
695
+ "[control_691]",
696
+ "[control_692]",
697
+ "[control_693]",
698
+ "[control_694]",
699
+ "[control_695]",
700
+ "[control_696]",
701
+ "[control_697]",
702
+ "[control_698]",
703
+ "[control_699]",
704
+ "[control_700]",
705
+ "[control_701]",
706
+ "[control_702]",
707
+ "[control_703]",
708
+ "[control_704]",
709
+ "[control_705]",
710
+ "[control_706]",
711
+ "[control_707]",
712
+ "[control_708]",
713
+ "[control_709]",
714
+ "[control_710]",
715
+ "[control_711]",
716
+ "[control_712]",
717
+ "[control_713]",
718
+ "[control_714]",
719
+ "[control_715]",
720
+ "[control_716]",
721
+ "[control_717]",
722
+ "[control_718]",
723
+ "[control_719]",
724
+ "[control_720]",
725
+ "[control_721]",
726
+ "[control_722]",
727
+ "[control_723]",
728
+ "[control_724]",
729
+ "[control_725]",
730
+ "[control_726]",
731
+ "[control_727]",
732
+ "[control_728]",
733
+ "[control_729]",
734
+ "[control_730]",
735
+ "[control_731]",
736
+ "[control_732]",
737
+ "[control_733]",
738
+ "[control_734]",
739
+ "[control_735]",
740
+ "[control_736]",
741
+ "[control_737]",
742
+ "[control_738]",
743
+ "[control_739]",
744
+ "[control_740]",
745
+ "[control_741]",
746
+ "[control_742]",
747
+ "[control_743]",
748
+ "[control_744]",
749
+ "[control_745]",
750
+ "[control_746]"
751
+ ],
752
+ "bos_token": {
753
+ "content": "<s>",
754
+ "lstrip": false,
755
+ "normalized": false,
756
+ "rstrip": false,
757
+ "single_word": false
758
+ },
759
+ "eos_token": {
760
+ "content": "</s>",
761
+ "lstrip": false,
762
+ "normalized": false,
763
+ "rstrip": false,
764
+ "single_word": false
765
+ },
766
+ "unk_token": {
767
+ "content": "<unk>",
768
+ "lstrip": false,
769
+ "normalized": false,
770
+ "rstrip": false,
771
+ "single_word": false
772
+ }
773
+ }
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:93346fbb7e676ad3c177ae3ed72700dd93ffeb380d9b083b58227cb0ea4a75da
3
+ size 3672259
tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1b968b8dc352f42192367337c78ccc61e1eaddc6d641a579372d4f20694beb7a
3
+ size 587562
tokenizer_config.json ADDED
The diff for this file is too large to render. See raw diff