Hugging Face's logo

Yosemat
/
designvlm

Visual Question Answering
Transformers
PyTorch
Safetensors
internlm2
text-generation
custom_code
Model card Files Community
Yosemat commited on
Commit
d251895
·
verified ·
1 Parent(s): fd8a4dd

Upload folder using huggingface_hub

Browse files
Files changed (11) hide show
  1. added_tokens.json +385 -6
  2. config.json +9 -9
  3. generation_config.json +1 -1
  4. ixc_layout.py +671 -0
  5. model-00001-of-00005.safetensors +3 -0
  6. model-00002-of-00005.safetensors +3 -0
  7. model-00003-of-00005.safetensors +3 -0
  8. model-00004-of-00005.safetensors +3 -0
  9. model-00005-of-00005.safetensors +3 -0
  10. model.safetensors.index.json +0 -0
  11. tokenizer_config.json +3080 -0
added_tokens.json CHANGED
@@ -1,8 +1,387 @@
1
  {
2
- "<|action_end|>": 92547,
3
- "<|action_start|>": 92546,
4
- "<|im_end|>": 92545,
5
- "<|im_start|>": 92544,
6
- "<|interpreter|>": 92548,
7
- "<|plugin|>": 92549
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
8
  }
 
1
  {
2
+ "<|-100|>": 92572,
3
+ "<|-101|>": 92571,
4
+ "<|-102|>": 92570,
5
+ "<|-103|>": 92569,
6
+ "<|-104|>": 92568,
7
+ "<|-105|>": 92567,
8
+ "<|-106|>": 92566,
9
+ "<|-107|>": 92565,
10
+ "<|-108|>": 92564,
11
+ "<|-109|>": 92563,
12
+ "<|-10|>": 92662,
13
+ "<|-110|>": 92562,
14
+ "<|-111|>": 92561,
15
+ "<|-112|>": 92560,
16
+ "<|-113|>": 92559,
17
+ "<|-114|>": 92558,
18
+ "<|-115|>": 92557,
19
+ "<|-116|>": 92556,
20
+ "<|-117|>": 92555,
21
+ "<|-118|>": 92554,
22
+ "<|-119|>": 92553,
23
+ "<|-11|>": 92661,
24
+ "<|-120|>": 92552,
25
+ "<|-121|>": 92551,
26
+ "<|-122|>": 92550,
27
+ "<|-123|>": 92549,
28
+ "<|-124|>": 92548,
29
+ "<|-125|>": 92547,
30
+ "<|-126|>": 92546,
31
+ "<|-127|>": 92545,
32
+ "<|-128|>": 92544,
33
+ "<|-12|>": 92660,
34
+ "<|-13|>": 92659,
35
+ "<|-14|>": 92658,
36
+ "<|-15|>": 92657,
37
+ "<|-16|>": 92656,
38
+ "<|-17|>": 92655,
39
+ "<|-18|>": 92654,
40
+ "<|-19|>": 92653,
41
+ "<|-1|>": 92671,
42
+ "<|-20|>": 92652,
43
+ "<|-21|>": 92651,
44
+ "<|-22|>": 92650,
45
+ "<|-23|>": 92649,
46
+ "<|-24|>": 92648,
47
+ "<|-25|>": 92647,
48
+ "<|-26|>": 92646,
49
+ "<|-27|>": 92645,
50
+ "<|-28|>": 92644,
51
+ "<|-29|>": 92643,
52
+ "<|-2|>": 92670,
53
+ "<|-30|>": 92642,
54
+ "<|-31|>": 92641,
55
+ "<|-32|>": 92640,
56
+ "<|-33|>": 92639,
57
+ "<|-34|>": 92638,
58
+ "<|-35|>": 92637,
59
+ "<|-36|>": 92636,
60
+ "<|-37|>": 92635,
61
+ "<|-38|>": 92634,
62
+ "<|-39|>": 92633,
63
+ "<|-3|>": 92669,
64
+ "<|-40|>": 92632,
65
+ "<|-41|>": 92631,
66
+ "<|-42|>": 92630,
67
+ "<|-43|>": 92629,
68
+ "<|-44|>": 92628,
69
+ "<|-45|>": 92627,
70
+ "<|-46|>": 92626,
71
+ "<|-47|>": 92625,
72
+ "<|-48|>": 92624,
73
+ "<|-49|>": 92623,
74
+ "<|-4|>": 92668,
75
+ "<|-50|>": 92622,
76
+ "<|-51|>": 92621,
77
+ "<|-52|>": 92620,
78
+ "<|-53|>": 92619,
79
+ "<|-54|>": 92618,
80
+ "<|-55|>": 92617,
81
+ "<|-56|>": 92616,
82
+ "<|-57|>": 92615,
83
+ "<|-58|>": 92614,
84
+ "<|-59|>": 92613,
85
+ "<|-5|>": 92667,
86
+ "<|-60|>": 92612,
87
+ "<|-61|>": 92611,
88
+ "<|-62|>": 92610,
89
+ "<|-63|>": 92609,
90
+ "<|-64|>": 92608,
91
+ "<|-65|>": 92607,
92
+ "<|-66|>": 92606,
93
+ "<|-67|>": 92605,
94
+ "<|-68|>": 92604,
95
+ "<|-69|>": 92603,
96
+ "<|-6|>": 92666,
97
+ "<|-70|>": 92602,
98
+ "<|-71|>": 92601,
99
+ "<|-72|>": 92600,
100
+ "<|-73|>": 92599,
101
+ "<|-74|>": 92598,
102
+ "<|-75|>": 92597,
103
+ "<|-76|>": 92596,
104
+ "<|-77|>": 92595,
105
+ "<|-78|>": 92594,
106
+ "<|-79|>": 92593,
107
+ "<|-7|>": 92665,
108
+ "<|-80|>": 92592,
109
+ "<|-81|>": 92591,
110
+ "<|-82|>": 92590,
111
+ "<|-83|>": 92589,
112
+ "<|-84|>": 92588,
113
+ "<|-85|>": 92587,
114
+ "<|-86|>": 92586,
115
+ "<|-87|>": 92585,
116
+ "<|-88|>": 92584,
117
+ "<|-89|>": 92583,
118
+ "<|-8|>": 92664,
119
+ "<|-90|>": 92582,
120
+ "<|-91|>": 92581,
121
+ "<|-92|>": 92580,
122
+ "<|-93|>": 92579,
123
+ "<|-94|>": 92578,
124
+ "<|-95|>": 92577,
125
+ "<|-96|>": 92576,
126
+ "<|-97|>": 92575,
127
+ "<|-98|>": 92574,
128
+ "<|-99|>": 92573,
129
+ "<|-9|>": 92663,
130
+ "<|0|>": 92672,
131
+ "<|100|>": 92772,
132
+ "<|101|>": 92773,
133
+ "<|102|>": 92774,
134
+ "<|103|>": 92775,
135
+ "<|104|>": 92776,
136
+ "<|105|>": 92777,
137
+ "<|106|>": 92778,
138
+ "<|107|>": 92779,
139
+ "<|108|>": 92780,
140
+ "<|109|>": 92781,
141
+ "<|10|>": 92682,
142
+ "<|110|>": 92782,
143
+ "<|111|>": 92783,
144
+ "<|112|>": 92784,
145
+ "<|113|>": 92785,
146
+ "<|114|>": 92786,
147
+ "<|115|>": 92787,
148
+ "<|116|>": 92788,
149
+ "<|117|>": 92789,
150
+ "<|118|>": 92790,
151
+ "<|119|>": 92791,
152
+ "<|11|>": 92683,
153
+ "<|120|>": 92792,
154
+ "<|121|>": 92793,
155
+ "<|122|>": 92794,
156
+ "<|123|>": 92795,
157
+ "<|124|>": 92796,
158
+ "<|125|>": 92797,
159
+ "<|126|>": 92798,
160
+ "<|127|>": 92799,
161
+ "<|128|>": 92800,
162
+ "<|129|>": 92801,
163
+ "<|12|>": 92684,
164
+ "<|130|>": 92802,
165
+ "<|131|>": 92803,
166
+ "<|132|>": 92804,
167
+ "<|133|>": 92805,
168
+ "<|134|>": 92806,
169
+ "<|135|>": 92807,
170
+ "<|136|>": 92808,
171
+ "<|137|>": 92809,
172
+ "<|138|>": 92810,
173
+ "<|139|>": 92811,
174
+ "<|13|>": 92685,
175
+ "<|140|>": 92812,
176
+ "<|141|>": 92813,
177
+ "<|142|>": 92814,
178
+ "<|143|>": 92815,
179
+ "<|144|>": 92816,
180
+ "<|145|>": 92817,
181
+ "<|146|>": 92818,
182
+ "<|147|>": 92819,
183
+ "<|148|>": 92820,
184
+ "<|149|>": 92821,
185
+ "<|14|>": 92686,
186
+ "<|150|>": 92822,
187
+ "<|151|>": 92823,
188
+ "<|152|>": 92824,
189
+ "<|153|>": 92825,
190
+ "<|154|>": 92826,
191
+ "<|155|>": 92827,
192
+ "<|156|>": 92828,
193
+ "<|157|>": 92829,
194
+ "<|158|>": 92830,
195
+ "<|159|>": 92831,
196
+ "<|15|>": 92687,
197
+ "<|160|>": 92832,
198
+ "<|161|>": 92833,
199
+ "<|162|>": 92834,
200
+ "<|163|>": 92835,
201
+ "<|164|>": 92836,
202
+ "<|165|>": 92837,
203
+ "<|166|>": 92838,
204
+ "<|167|>": 92839,
205
+ "<|168|>": 92840,
206
+ "<|169|>": 92841,
207
+ "<|16|>": 92688,
208
+ "<|170|>": 92842,
209
+ "<|171|>": 92843,
210
+ "<|172|>": 92844,
211
+ "<|173|>": 92845,
212
+ "<|174|>": 92846,
213
+ "<|175|>": 92847,
214
+ "<|176|>": 92848,
215
+ "<|177|>": 92849,
216
+ "<|178|>": 92850,
217
+ "<|179|>": 92851,
218
+ "<|17|>": 92689,
219
+ "<|180|>": 92852,
220
+ "<|181|>": 92853,
221
+ "<|182|>": 92854,
222
+ "<|183|>": 92855,
223
+ "<|184|>": 92856,
224
+ "<|185|>": 92857,
225
+ "<|186|>": 92858,
226
+ "<|187|>": 92859,
227
+ "<|188|>": 92860,
228
+ "<|189|>": 92861,
229
+ "<|18|>": 92690,
230
+ "<|190|>": 92862,
231
+ "<|191|>": 92863,
232
+ "<|192|>": 92864,
233
+ "<|193|>": 92865,
234
+ "<|194|>": 92866,
235
+ "<|195|>": 92867,
236
+ "<|196|>": 92868,
237
+ "<|197|>": 92869,
238
+ "<|198|>": 92870,
239
+ "<|199|>": 92871,
240
+ "<|19|>": 92691,
241
+ "<|1|>": 92673,
242
+ "<|200|>": 92872,
243
+ "<|201|>": 92873,
244
+ "<|202|>": 92874,
245
+ "<|203|>": 92875,
246
+ "<|204|>": 92876,
247
+ "<|205|>": 92877,
248
+ "<|206|>": 92878,
249
+ "<|207|>": 92879,
250
+ "<|208|>": 92880,
251
+ "<|209|>": 92881,
252
+ "<|20|>": 92692,
253
+ "<|210|>": 92882,
254
+ "<|211|>": 92883,
255
+ "<|212|>": 92884,
256
+ "<|213|>": 92885,
257
+ "<|214|>": 92886,
258
+ "<|215|>": 92887,
259
+ "<|216|>": 92888,
260
+ "<|217|>": 92889,
261
+ "<|218|>": 92890,
262
+ "<|219|>": 92891,
263
+ "<|21|>": 92693,
264
+ "<|220|>": 92892,
265
+ "<|221|>": 92893,
266
+ "<|222|>": 92894,
267
+ "<|223|>": 92895,
268
+ "<|224|>": 92896,
269
+ "<|225|>": 92897,
270
+ "<|226|>": 92898,
271
+ "<|227|>": 92899,
272
+ "<|228|>": 92900,
273
+ "<|229|>": 92901,
274
+ "<|22|>": 92694,
275
+ "<|230|>": 92902,
276
+ "<|231|>": 92903,
277
+ "<|232|>": 92904,
278
+ "<|233|>": 92905,
279
+ "<|234|>": 92906,
280
+ "<|235|>": 92907,
281
+ "<|236|>": 92908,
282
+ "<|237|>": 92909,
283
+ "<|238|>": 92910,
284
+ "<|239|>": 92911,
285
+ "<|23|>": 92695,
286
+ "<|240|>": 92912,
287
+ "<|241|>": 92913,
288
+ "<|242|>": 92914,
289
+ "<|243|>": 92915,
290
+ "<|244|>": 92916,
291
+ "<|245|>": 92917,
292
+ "<|246|>": 92918,
293
+ "<|247|>": 92919,
294
+ "<|248|>": 92920,
295
+ "<|249|>": 92921,
296
+ "<|24|>": 92696,
297
+ "<|250|>": 92922,
298
+ "<|251|>": 92923,
299
+ "<|252|>": 92924,
300
+ "<|253|>": 92925,
301
+ "<|254|>": 92926,
302
+ "<|255|>": 92927,
303
+ "<|256|>": 92928,
304
+ "<|25|>": 92697,
305
+ "<|26|>": 92698,
306
+ "<|27|>": 92699,
307
+ "<|28|>": 92700,
308
+ "<|29|>": 92701,
309
+ "<|2|>": 92674,
310
+ "<|30|>": 92702,
311
+ "<|31|>": 92703,
312
+ "<|32|>": 92704,
313
+ "<|33|>": 92705,
314
+ "<|34|>": 92706,
315
+ "<|35|>": 92707,
316
+ "<|36|>": 92708,
317
+ "<|37|>": 92709,
318
+ "<|38|>": 92710,
319
+ "<|39|>": 92711,
320
+ "<|3|>": 92675,
321
+ "<|40|>": 92712,
322
+ "<|41|>": 92713,
323
+ "<|42|>": 92714,
324
+ "<|43|>": 92715,
325
+ "<|44|>": 92716,
326
+ "<|45|>": 92717,
327
+ "<|46|>": 92718,
328
+ "<|47|>": 92719,
329
+ "<|48|>": 92720,
330
+ "<|49|>": 92721,
331
+ "<|4|>": 92676,
332
+ "<|50|>": 92722,
333
+ "<|51|>": 92723,
334
+ "<|52|>": 92724,
335
+ "<|53|>": 92725,
336
+ "<|54|>": 92726,
337
+ "<|55|>": 92727,
338
+ "<|56|>": 92728,
339
+ "<|57|>": 92729,
340
+ "<|58|>": 92730,
341
+ "<|59|>": 92731,
342
+ "<|5|>": 92677,
343
+ "<|60|>": 92732,
344
+ "<|61|>": 92733,
345
+ "<|62|>": 92734,
346
+ "<|63|>": 92735,
347
+ "<|64|>": 92736,
348
+ "<|65|>": 92737,
349
+ "<|66|>": 92738,
350
+ "<|67|>": 92739,
351
+ "<|68|>": 92740,
352
+ "<|69|>": 92741,
353
+ "<|6|>": 92678,
354
+ "<|70|>": 92742,
355
+ "<|71|>": 92743,
356
+ "<|72|>": 92744,
357
+ "<|73|>": 92745,
358
+ "<|74|>": 92746,
359
+ "<|75|>": 92747,
360
+ "<|76|>": 92748,
361
+ "<|77|>": 92749,
362
+ "<|78|>": 92750,
363
+ "<|79|>": 92751,
364
+ "<|7|>": 92679,
365
+ "<|80|>": 92752,
366
+ "<|81|>": 92753,
367
+ "<|82|>": 92754,
368
+ "<|83|>": 92755,
369
+ "<|84|>": 92756,
370
+ "<|85|>": 92757,
371
+ "<|86|>": 92758,
372
+ "<|87|>": 92759,
373
+ "<|88|>": 92760,
374
+ "<|89|>": 92761,
375
+ "<|8|>": 92680,
376
+ "<|90|>": 92762,
377
+ "<|91|>": 92763,
378
+ "<|92|>": 92764,
379
+ "<|93|>": 92765,
380
+ "<|94|>": 92766,
381
+ "<|95|>": 92767,
382
+ "<|96|>": 92768,
383
+ "<|97|>": 92769,
384
+ "<|98|>": 92770,
385
+ "<|99|>": 92771,
386
+ "<|9|>": 92681
387
  }
config.json CHANGED
@@ -1,16 +1,16 @@
1
  {
2
- "_name_or_path": "/mnt/petrelfs/dongxiaoyi/gittest/temp_model/internlm-xcomposer2d5-7b",
3
  "architectures": [
4
- "InternLMXComposer2ForCausalLM"
5
  ],
6
  "attn_implementation": "flash_attention_2",
7
  "auto_map": {
8
- "AutoConfig": "configuration_internlm_xcomposer2.InternLMXcomposer2Config",
9
- "AutoModel": "modeling_internlm_xcomposer2.InternLMXComposer2ForCausalLM",
10
- "AutoModelForCausalLM": "modeling_internlm_xcomposer2.InternLMXComposer2ForCausalLM"
11
  },
12
  "bias": false,
13
  "bos_token_id": 1,
 
14
  "eos_token_id": 2,
15
  "hidden_act": "silu",
16
  "hidden_size": 4096,
@@ -25,13 +25,13 @@
25
  "pad_token_id": 2,
26
  "rms_norm_eps": 1e-05,
27
  "rope_scaling": {
28
- "type": "dynamic",
29
- "factor": 2.0
30
  },
31
  "rope_theta": 1000000,
32
  "tie_word_embeddings": false,
33
  "torch_dtype": "bfloat16",
34
- "transformers_version": "4.33.1",
35
  "use_cache": false,
36
- "vocab_size": 92544
37
  }
 
1
  {
2
+ "_name_or_path": "yosematvlm",
3
  "architectures": [
4
+ "IXCLayout"
5
  ],
6
  "attn_implementation": "flash_attention_2",
7
  "auto_map": {
8
+ "AutoConfig": "ixc_layout.IXCLayoutConfig",
9
+ "AutoModelForCausalLM": "ixc_layout.IXCLayout"
 
10
  },
11
  "bias": false,
12
  "bos_token_id": 1,
13
+ "discrete_coordinate_tokens": 128,
14
  "eos_token_id": 2,
15
  "hidden_act": "silu",
16
  "hidden_size": 4096,
 
25
  "pad_token_id": 2,
26
  "rms_norm_eps": 1e-05,
27
  "rope_scaling": {
28
+ "factor": 2.0,
29
+ "type": "dynamic"
30
  },
31
  "rope_theta": 1000000,
32
  "tie_word_embeddings": false,
33
  "torch_dtype": "bfloat16",
34
+ "transformers_version": "4.43.3",
35
  "use_cache": false,
36
+ "vocab_size": 92929
37
  }
generation_config.json CHANGED
@@ -4,6 +4,6 @@
4
  "eos_token_id": 2,
5
  "max_length": 16384,
6
  "pad_token_id": 2,
7
- "transformers_version": "4.33.1",
8
  "use_cache": false
9
  }
 
4
  "eos_token_id": 2,
5
  "max_length": 16384,
6
  "pad_token_id": 2,
7
+ "transformers_version": "4.43.3",
8
  "use_cache": false
9
  }
ixc_layout.py ADDED
@@ -0,0 +1,671 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ import copy
2
+
3
+ import numpy as np
4
+ import torch
5
+ import torch.utils.checkpoint
6
+ from peft.mapping import get_peft_model
7
+ from peft.peft_model import PeftModel
8
+ from peft.tuners.lora import LoraConfig
9
+ from torch import nn
10
+ from torch.nn import CrossEntropyLoss
11
+ from transformers import AutoConfig, AutoTokenizer
12
+ from transformers.generation.streamers import BaseStreamer
13
+ from transformers.modeling_outputs import (
14
+ BaseModelOutputWithPast,
15
+ CausalLMOutputWithPast,
16
+ )
17
+ from transformers.utils import (
18
+ replace_return_docstrings,
19
+ )
20
+
21
+ from designvlm.crello_dataset import int_wrap
22
+ from designvlm.loading import LoraArguments
23
+ from yosematvlm.configuration_internlm_xcomposer2 import InternLMXcomposer2Config
24
+ from yosematvlm.modeling_internlm_xcomposer2 import (
25
+ FROM_TOKEN_2,
26
+ InternLMXComposer2ForCausalLM,
27
+ )
28
+
29
+ IM_END_TOKEN = 92542
30
+ EOS = 2 # </s>
31
+ SWITCH_TOKEN_LENGTH = 6 # after <|im_end|>, there are 5 additional tokens ['\n', '<|im_start|>', 'ass', 'istant', '\n']. You don't want the model to learn these tokens.
32
+ MASK_ID = -100
33
+
34
+
35
+ class IXCLayoutConfig(InternLMXcomposer2Config):
36
+ def __init__(
37
+ self,
38
+ vocab_size=103168,
39
+ hidden_size=4096,
40
+ intermediate_size=11008,
41
+ num_hidden_layers=32,
42
+ num_attention_heads=32,
43
+ num_key_value_heads=None,
44
+ hidden_act="silu",
45
+ max_position_embeddings=2048,
46
+ initializer_range=0.02,
47
+ rms_norm_eps=1e-6,
48
+ use_cache=True,
49
+ pad_token_id=0,
50
+ bos_token_id=1,
51
+ eos_token_id=2,
52
+ tie_word_embeddings=False,
53
+ bias=True,
54
+ rope_theta=10000,
55
+ rope_scaling=None,
56
+ attn_implementation="flash_attention_2",
57
+ max_length: int = 16384,
58
+ discrete_coordinate_tokens: int | None = None,
59
+ **kwargs,
60
+ ):
61
+ self.discrete_coordinate_tokens = discrete_coordinate_tokens
62
+ super().__init__(
63
+ vocab_size=vocab_size,
64
+ hidden_size=hidden_size,
65
+ intermediate_size=intermediate_size,
66
+ num_hidden_layers=num_hidden_layers,
67
+ num_attention_heads=num_attention_heads,
68
+ num_key_value_heads=num_key_value_heads,
69
+ hidden_act=hidden_act,
70
+ max_position_embeddings=max_position_embeddings,
71
+ initializer_range=initializer_range,
72
+ rms_norm_eps=rms_norm_eps,
73
+ use_cache=use_cache,
74
+ pad_token_id=pad_token_id,
75
+ bos_token_id=bos_token_id,
76
+ eos_token_id=eos_token_id,
77
+ tie_word_embeddings=tie_word_embeddings,
78
+ bias=bias,
79
+ rope_theta=rope_theta,
80
+ rope_scaling=rope_scaling,
81
+ attn_implementation=attn_implementation,
82
+ **kwargs,
83
+ )
84
+ self.max_length = max_length
85
+ self.use_cache = False
86
+
87
+ @classmethod
88
+ def with_internlm_config(
89
+ cls,
90
+ config: InternLMXcomposer2Config,
91
+ discrete_custom_tokens: int | None = None,
92
+ max_length: int = 16384,
93
+ ):
94
+ return cls(
95
+ vocab_size=config.vocab_size,
96
+ hidden_size=config.hidden_size,
97
+ intermediate_size=config.intermediate_size,
98
+ num_hidden_layers=config.num_hidden_layers,
99
+ num_attention_heads=config.num_attention_heads,
100
+ num_key_value_heads=config.num_key_value_heads,
101
+ hidden_act=config.hidden_act,
102
+ max_position_embeddings=config.max_position_embeddings,
103
+ initializer_range=config.initializer_range,
104
+ rms_norm_eps=config.rms_norm_eps,
105
+ use_cache=config.use_cache,
106
+ pad_token_id=config.pad_token_id,
107
+ bos_token_id=config.bos_token_id,
108
+ eos_token_id=config.eos_token_id,
109
+ tie_word_embeddings=config.tie_word_embeddings,
110
+ bias=config.bias,
111
+ rope_theta=config.rope_theta,
112
+ rope_scaling=config.rope_scaling,
113
+ attn_implementation=config.attn_implementation,
114
+ discrete_coordinate_tokens=discrete_custom_tokens,
115
+ max_length=max_length,
116
+ )
117
+
118
+
119
+ class IXCLayout(InternLMXComposer2ForCausalLM):
120
+ config_class = IXCLayoutConfig
121
+
122
+ def __init__(self, config: IXCLayoutConfig):
123
+ super().__init__(config)
124
+
125
+ self.vit.vision_tower.vision_model.post_layernorm = torch.nn.Identity()
126
+ self.tokenizer = AutoTokenizer.from_pretrained(
127
+ "yosematvlm",
128
+ padding_side="right",
129
+ use_fast=False,
130
+ trust_remote_code=True,
131
+ ) # type: ignore
132
+
133
+ # Add coordinate tokens
134
+ self.coordinate_token_ids: set[int] = set()
135
+ if config.discrete_coordinate_tokens is not None:
136
+ self.add_coordinate_tokens(config.discrete_coordinate_tokens)
137
+ self.config = config
138
+
139
+ @replace_return_docstrings(
140
+ output_type=CausalLMOutputWithPast, config_class=InternLMXcomposer2Config
141
+ )
142
+ def forward(
143
+ self,
144
+ input_ids: torch.LongTensor | None = None,
145
+ attention_mask: torch.Tensor | None = None,
146
+ position_ids: torch.LongTensor | None = None,
147
+ past_key_values: list[torch.FloatTensor] | None = None,
148
+ inputs_embeds: torch.FloatTensor | None = None,
149
+ labels: torch.LongTensor | None = None,
150
+ use_cache: bool | None = None,
151
+ output_attentions: bool | None = None,
152
+ output_hidden_states: bool | None = None,
153
+ return_dict: bool | None = None,
154
+ **kwargs,
155
+ ) -> CausalLMOutputWithPast:
156
+ r"""
157
+ Args:
158
+ labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*):
159
+ Labels for computing the masked language modeling loss. Indices should either be in `[0, ...,
160
+ config.vocab_size]` or -100 (see `input_ids` docstring). Tokens with indices set to `-100` are ignored
161
+ (masked), the loss is only computed for the tokens with labels in `[0, ..., config.vocab_size]`.
162
+ Returns:
163
+ """
164
+ infer_mode = "base"
165
+ return_dict, output_attentions, output_hidden_states = self.or_config(
166
+ return_dict=return_dict,
167
+ output_attentions=output_attentions,
168
+ output_hidden_states=output_hidden_states,
169
+ )
170
+ if "samples" in kwargs:
171
+ # Training
172
+ samples = kwargs["samples"]
173
+ # encode text
174
+ text = samples["text_input"]
175
+ # encode image
176
+ image = samples["image"][0]
177
+ bs = len(samples["text_input"][0])
178
+ image_nums = []
179
+ temp_image = []
180
+ for im in image:
181
+ if type(im) is list:
182
+ image_nums.append(len(im))
183
+ temp_image.extend(im)
184
+ else:
185
+ image_nums.append(1)
186
+ temp_image.append(im)
187
+ image = temp_image
188
+ assert type(image) is list and len(image_nums) == bs
189
+
190
+ to_regress_embeds, attention_mask, targets, im_mask = self.interleav_wrap(
191
+ image, text, image_nums
192
+ )
193
+
194
+ inputs_embeds = to_regress_embeds[:, : self.max_length] # type: ignore
195
+ attention_mask = attention_mask[:, : self.max_length] # type: ignore
196
+ targets = targets[:, : self.max_length]
197
+ im_mask = im_mask[:, : self.max_length].bool()
198
+ labels = targets # type: ignore
199
+
200
+ elif inputs_embeds is not None or input_ids is not None:
201
+ im_mask = kwargs["im_mask"]
202
+ if im_mask is None and inputs_embeds is not None:
203
+ im_mask = torch.zeros(inputs_embeds.shape[:2]).to(inputs_embeds.device)
204
+ im_mask = im_mask.bool()
205
+ else:
206
+ raise ValueError(
207
+ "Either samples, inputs_embeds or input_ids should be provided."
208
+ )
209
+
210
+ # decoder outputs consists of (dec_features, layer_state, dec_hidden, dec_attn)
211
+ outputs: BaseModelOutputWithPast = self.model(
212
+ input_ids=input_ids,
213
+ attention_mask=attention_mask,
214
+ position_ids=position_ids,
215
+ past_key_values=past_key_values,
216
+ inputs_embeds=inputs_embeds,
217
+ use_cache=use_cache,
218
+ output_attentions=output_attentions,
219
+ output_hidden_states=output_hidden_states,
220
+ return_dict=True,
221
+ im_mask=im_mask,
222
+ infer_mode=infer_mode,
223
+ )
224
+ logits: torch.Tensor = self.output(
225
+ outputs.last_hidden_state
226
+ ).float() # B x L x V
227
+
228
+ if labels is not None:
229
+ # Shift so that tokens < n predict n
230
+ shift_logits = logits[..., :-1, :].contiguous()
231
+ shift_labels = labels[..., 1:].contiguous()
232
+ # Flatten the tokens
233
+ loss_fct = CrossEntropyLoss()
234
+ shift_logits = shift_logits.view(-1, self.vocab_size)
235
+ shift_labels = shift_labels.view(-1)
236
+ # Enable model parallelism
237
+ shift_labels = shift_labels.to(shift_logits.device)
238
+ ce_loss: torch.Tensor = loss_fct(shift_logits, shift_labels)
239
+ assert not ce_loss.isnan().any()
240
+ kl_loss = self.coordinate_kl_loss(logits, labels)
241
+ assert not kl_loss.isnan().any()
242
+ loss = ce_loss + kl_loss
243
+
244
+ else:
245
+ loss = None
246
+ return_dict = (
247
+ return_dict if return_dict is not None else self.config.use_return_dict
248
+ )
249
+
250
+ if not return_dict:
251
+ output = (logits,) + outputs[1:]
252
+ return (ce_loss,) + output if ce_loss is not None else output
253
+
254
+ return CausalLMOutputWithPast(
255
+ loss=loss, # type: ignore
256
+ logits=logits, # type: ignore
257
+ past_key_values=outputs.past_key_values,
258
+ hidden_states=outputs.hidden_states,
259
+ attentions=outputs.attentions,
260
+ )
261
+
262
+ def coordinate_kl_loss(
263
+ self, logits: torch.Tensor, labels: torch.Tensor, eps: float = 1e-9
264
+ ) -> torch.Tensor:
265
+ """
266
+ For coordinate token, calculate the KL loss between the predicted logits and the target labels.
267
+ Instead of one-hot vector, we assume that the target labels are the probability distribution of the target token.
268
+ The distribution is a discrete gaussian distribution with mean at the target token and variance of 1.
269
+
270
+ Args:
271
+ logits: B x T x V; The predicted logits of the model.
272
+ labels: B x T; The target labels of the model.
273
+ """
274
+ label_std_dev = 2.0
275
+ coordinate_token_ids = torch.tensor(
276
+ list(self.coordinate_token_ids), device=labels.device, dtype=labels.dtype
277
+ )
278
+ assert len(self.coordinate_token_ids) > 0
279
+ # Get the mask of the coordinate tokens
280
+ is_label_coordinate = torch.isin(
281
+ labels,
282
+ coordinate_token_ids,
283
+ ).type_as(labels) # B x T
284
+ # Get the target labels of the coordinate tokens
285
+ # Range of indices
286
+ indices = torch.arange(
287
+ 0, self.vocab_size, dtype=labels.dtype, device=logits.device
288
+ ).repeat(labels.shape[0], labels.shape[1], 1) # B x T x V
289
+
290
+ # Indices that are not coordinate tokens are set to 0
291
+ # To reduce memory consumption, we iterate over the sequence length
292
+ is_indice_coordinate_token = torch.stack(
293
+ [
294
+ torch.isin(indices[:, idx], coordinate_token_ids)
295
+ for idx in range(indices.size(1))
296
+ ],
297
+ dim=1,
298
+ ) # B x T x V
299
+
300
+ total_mask = is_label_coordinate.unsqueeze(-1) * is_indice_coordinate_token
301
+
302
+ # Create a Gaussian distribution centered at the label index
303
+ gauss_label = torch.exp(
304
+ -0.5 * ((indices - labels.unsqueeze(-1)) / label_std_dev) ** 2
305
+ ) # B x T x V
306
+
307
+ # Apply the mask so that only the coordinate tokens are considered
308
+ gauss_label = gauss_label * total_mask + eps # Add eps for numerical stability
309
+
310
+ # Normalize the distribution
311
+ gauss_label /= gauss_label.sum(dim=-1, keepdim=True)
312
+
313
+ pointwise_kl = (
314
+ nn.functional.kl_div(
315
+ logits.log_softmax(dim=-1), gauss_label, reduction="none"
316
+ ).nan_to_num()
317
+ * total_mask
318
+ )
319
+ kl_loss = pointwise_kl.sum(-1).mean()
320
+ assert kl_loss > 0
321
+ return kl_loss
322
+
323
+ def interleav_wrap(
324
+ self,
325
+ img_list: list[torch.Tensor], # V x 1 x 3 x H x W
326
+ text_list_list: list[list[str]], # B x 1 x str
327
+ image_nums: list[int], # B
328
+ ):
329
+ temp_embeds = []
330
+ temp_im_mask = []
331
+ temp_tars = []
332
+
333
+ # encode_image
334
+ if len(img_list) > 0:
335
+ img_embeds, img_split = self.vit(
336
+ img_list, self.plora_glb_GN, self.plora_sub_GN
337
+ )
338
+ img_embeds = self.vision_proj(img_embeds)
339
+ else:
340
+ img_embeds = None
341
+ img_split = []
342
+
343
+ text_list = text_list_list[0]
344
+ for idx, text in enumerate(text_list):
345
+ image_num = image_nums[idx]
346
+ im_id = int(np.sum(image_nums[:idx]))
347
+ images = []
348
+ for i in range(image_nums[idx]):
349
+ st = int(np.sum(img_split[: im_id + i]))
350
+ sp = img_split[im_id + i]
351
+ temp_img = img_embeds[:, st : st + sp] # type: ignore
352
+ images.append(temp_img)
353
+
354
+ if image_num == 1 and text.find("<ImageHere>") == -1:
355
+ text = "<ImageHere>" + text
356
+ parts = text.split("<ImageHere>")
357
+
358
+ wrap_tokens, wrap_embeds, wrap_im_mask = [], [], []
359
+ temp_len = 0
360
+ need_bos = True
361
+ for idx, part in enumerate(parts):
362
+ if len(part) > 0:
363
+ part_tokens = self.tokenizer(
364
+ part,
365
+ return_tensors="pt",
366
+ padding="longest",
367
+ add_special_tokens=need_bos,
368
+ ).to(self.device) # type: ignore
369
+ if need_bos:
370
+ need_bos = False
371
+ wrap_tokens.append(part_tokens.input_ids)
372
+ part_embeds = self.model.tok_embeddings(part_tokens.input_ids)
373
+ wrap_embeds.append(part_embeds)
374
+ wrap_im_mask.append(
375
+ torch.zeros(part_embeds.shape[:2]).to(self.device)
376
+ )
377
+ temp_len += part_embeds.shape[1]
378
+ if idx < image_num:
379
+ wrap_embeds.append(images[idx])
380
+ wrap_token = (
381
+ torch.ones(images[idx].shape[:2], dtype=torch.long).to(
382
+ self.device
383
+ )
384
+ * -100
385
+ )
386
+ wrap_tokens.append(wrap_token)
387
+ wrap_im_mask.append(
388
+ torch.ones(images[idx].shape[:2]).to(self.device)
389
+ )
390
+ temp_len += images[idx].shape[1]
391
+ if temp_len > self.max_length:
392
+ break
393
+ wrap_tokens = torch.cat(wrap_tokens, dim=1)
394
+ wrap_embeds = torch.cat(wrap_embeds, dim=1)
395
+ wrap_im_mask = torch.cat(wrap_im_mask, dim=1)
396
+
397
+ wrap_target = self.mask_human_targets(wrap_tokens).to(self.device)
398
+
399
+ temp_embeds.append(wrap_embeds)
400
+ temp_im_mask.append(wrap_im_mask)
401
+ temp_tars.append(wrap_target)
402
+
403
+ temp_max_len = np.max([i.shape[1] for i in temp_embeds])
404
+ temp_max_len = min(temp_max_len, self.max_length)
405
+
406
+ final_input, final_atts, final_tars, final_mask = [], [], [], []
407
+ pad = torch.ones([1, 1]) * self.tokenizer.pad_token_id # type: ignore
408
+ pad = pad.long().to(self.device)
409
+ pad_emb = self.model.tok_embeddings(pad)
410
+
411
+ for idx in range(len(temp_embeds)):
412
+ temp_len = temp_embeds[idx].shape[1]
413
+ if temp_len >= temp_max_len:
414
+ final_input.append(temp_embeds[idx][:, :temp_max_len])
415
+ final_atts.append(
416
+ torch.ones(1, temp_max_len).to(wrap_target.dtype).to(self.device)
417
+ )
418
+ final_tars.append(temp_tars[idx][:, :temp_max_len])
419
+ final_mask.append(temp_im_mask[idx][:, :temp_max_len])
420
+ else:
421
+ final_input.append(
422
+ torch.cat(
423
+ [
424
+ temp_embeds[idx],
425
+ pad_emb.repeat(1, temp_max_len - temp_len, 1),
426
+ ],
427
+ dim=1,
428
+ )
429
+ )
430
+ final_atts.append(
431
+ torch.cat(
432
+ [
433
+ torch.ones(1, temp_len),
434
+ torch.zeros(1, temp_max_len - temp_len),
435
+ ],
436
+ dim=1,
437
+ )
438
+ .to(wrap_target.dtype)
439
+ .to(self.device)
440
+ )
441
+ final_tars.append(
442
+ torch.cat(
443
+ [
444
+ temp_tars[idx],
445
+ (torch.ones(1, temp_max_len - temp_len) * MASK_ID)
446
+ .to(wrap_target.dtype)
447
+ .to(self.device),
448
+ ],
449
+ dim=1,
450
+ )
451
+ )
452
+ final_mask.append(
453
+ torch.cat(
454
+ [
455
+ temp_im_mask[idx],
456
+ (torch.zeros(1, temp_max_len - temp_len))
457
+ .to(wrap_target.dtype)
458
+ .to(self.device),
459
+ ],
460
+ dim=1,
461
+ )
462
+ )
463
+
464
+ inputs_embeds = torch.cat(final_input, dim=0)
465
+ attention_mask = torch.cat(final_atts, dim=0)
466
+ targets = torch.cat(final_tars, dim=0)
467
+ im_mask = torch.cat(final_mask, dim=0)
468
+
469
+ return inputs_embeds, attention_mask, targets, im_mask
470
+
471
+ def mask_human_targets(self, input_ids: torch.Tensor) -> torch.Tensor:
472
+ target_batch = []
473
+ for bs in range(input_ids.shape[0]):
474
+ ids = input_ids[bs]
475
+ targets = copy.deepcopy(ids)
476
+ end_count = 0
477
+ last_eoa = 0
478
+ for i, temp_id in enumerate(ids):
479
+ # Counterintuitively, IM_END_TOKEN is the token for the end of utterance
480
+ # In the whole source code, "[UNUSED_TOKEN_145]" corresponds to IM_END_TOKEN
481
+ if temp_id == IM_END_TOKEN:
482
+ if end_count % 2 == 0:
483
+ targets[last_eoa : i + SWITCH_TOKEN_LENGTH] = MASK_ID
484
+ else:
485
+ last_eoa = i + 1
486
+ end_count += 1
487
+ # # eos and following pad
488
+ elif temp_id == EOS:
489
+ # loss on eos, but not on pad
490
+ targets[i + 1 :] = MASK_ID
491
+ break
492
+ target_batch.append(targets.unsqueeze(0))
493
+ target_batch = torch.cat(target_batch, dim=0)
494
+ return target_batch
495
+
496
+ @classmethod
497
+ def from_ixc_pretrained(
498
+ cls,
499
+ ixc_pretrained_model_name_or_path: str,
500
+ max_length: int,
501
+ img_size: int,
502
+ discrete_custom_tokens: int | None = 128,
503
+ ) -> torch.nn.Module:
504
+ r"""
505
+ Instantiate a pretrained InternLMXComposer2 model from a pre-trained model configuration.
506
+ """
507
+ config: IXCLayoutConfig = AutoConfig.from_pretrained(
508
+ ixc_pretrained_model_name_or_path,
509
+ trust_remote_code=True,
510
+ )
511
+ config = IXCLayoutConfig.with_internlm_config(
512
+ config,
513
+ max_length=max_length,
514
+ discrete_custom_tokens=discrete_custom_tokens,
515
+ )
516
+
517
+ model: IXCLayout = super().from_pretrained(
518
+ ixc_pretrained_model_name_or_path,
519
+ config=config,
520
+ torch_dtype=torch.bfloat16,
521
+ ) # type: ignore
522
+ if img_size != 336:
523
+ model.vit.resize_pos()
524
+ model.vit.requires_grad_(False)
525
+ model.vision_proj.requires_grad_(True)
526
+
527
+ return model
528
+
529
+ def or_config(
530
+ self,
531
+ return_dict: bool | None,
532
+ output_attentions: bool | None,
533
+ output_hidden_states: bool | None,
534
+ ):
535
+ output_attentions = (
536
+ output_attentions
537
+ if output_attentions is not None
538
+ else self.config.output_attentions
539
+ )
540
+ output_hidden_states = (
541
+ output_hidden_states
542
+ if output_hidden_states is not None
543
+ else self.config.output_hidden_states
544
+ )
545
+ return_dict = (
546
+ return_dict if return_dict is not None else self.config.use_return_dict
547
+ )
548
+ return return_dict, output_attentions, output_hidden_states
549
+
550
+ @torch.no_grad()
551
+ def chat(
552
+ self,
553
+ query: str,
554
+ image: list[tuple[str, str]] | torch.Tensor = [],
555
+ hd_num: int = 24,
556
+ history: list[tuple[str, str]] = [],
557
+ streamer: BaseStreamer | None = None,
558
+ max_new_tokens: int = 1024,
559
+ do_sample: bool = True,
560
+ num_beams: int = 1,
561
+ temperature: float = 1.0,
562
+ top_p: float = 0.8,
563
+ repetition_penalty: float = 1.005,
564
+ infer_mode: str = "base",
565
+ use_meta: bool = False,
566
+ meta_instruction: str = "You are an AI assistant whose name is InternLM-XComposer (浦语·灵笔).\n"
567
+ "- InternLM-XComposer (浦语·灵笔) is a multi-modality conversational language model that is developed by Shanghai AI Laboratory (上海人工智能实验室). It is designed to be helpful, honest, and harmless.\n"
568
+ "- InternLM-XComposer (浦语·灵笔) can understand and communicate fluently in the language chosen by the user such as English and 中文.\n"
569
+ "- InternLM-XComposer (浦语·灵笔) is capable of comprehending and articulating responses effectively based on the provided image.",
570
+ **kwargs,
571
+ ):
572
+ if not use_meta:
573
+ meta_instruction = ""
574
+
575
+ inputs, im_mask, _ = self.interleav_wrap_chat(
576
+ query,
577
+ image,
578
+ history=history,
579
+ meta_instruction=meta_instruction,
580
+ hd_num=hd_num,
581
+ )
582
+ inputs = {k: v.to(self.device) for k, v in inputs.items() if torch.is_tensor(v)}
583
+ # also add end-of-assistant token in eos token id to avoid unnecessary generation
584
+ eos_token_id = [
585
+ self.tokenizer.eos_token_id,
586
+ self.tokenizer.convert_tokens_to_ids([FROM_TOKEN_2])[0], # type: ignore
587
+ ]
588
+ outputs = self.generate(
589
+ **inputs,
590
+ streamer=streamer,
591
+ max_new_tokens=max_new_tokens,
592
+ num_beams=num_beams,
593
+ do_sample=do_sample,
594
+ temperature=temperature,
595
+ top_p=top_p,
596
+ eos_token_id=eos_token_id,
597
+ repetition_penalty=repetition_penalty,
598
+ im_mask=im_mask,
599
+ infer_mode=infer_mode,
600
+ **kwargs,
601
+ )
602
+ outputs = outputs[0].cpu().tolist()
603
+ response = self.tokenizer.decode(outputs, skip_special_tokens=True)
604
+ response = response.split(FROM_TOKEN_2)[0]
605
+ history = history + [(query, response)]
606
+ return response, history
607
+
608
+ def add_coordinate_tokens(self, bins: int):
609
+ start = -bins
610
+ end = bins * 2
611
+ new_tokens = [int_wrap(idx) for idx in range(start, end + 1)]
612
+ new_token_ids = self._add_tokens(new_tokens)
613
+ self.coordinate_token_ids.update(new_token_ids)
614
+ self.config.discrete_coordinate_tokens = end // 2
615
+
616
+ def _add_tokens(self, new_tokens: list[str]) -> list[int]:
617
+ prev_vocab_size = len(self.tokenizer)
618
+ vocab = self.tokenizer.get_vocab()
619
+ new_tokens = [token for token in new_tokens if token not in vocab]
620
+ self.tokenizer.add_tokens(new_tokens) # type: ignore
621
+ self.model.resize_token_embeddings(len(self.tokenizer))
622
+ self.vocab_size = len(self.tokenizer)
623
+
624
+ # self.output needs to be resized accordingly but without loosing the weight
625
+ new_output = nn.Linear(
626
+ self.model.config.hidden_size,
627
+ self.vocab_size,
628
+ bias=False,
629
+ dtype=self.output.weight.dtype,
630
+ device=self.output.weight.device,
631
+ ).to(self.device)
632
+ new_output.weight.data[: self.output.weight.shape[0]] = self.output.weight.data
633
+ self.output = new_output
634
+ return list(range(prev_vocab_size, self.vocab_size))
635
+
636
+
637
+ def setup_ixclayout(
638
+ model_name_or_path: str,
639
+ max_length: int,
640
+ img_size: int,
641
+ use_lora: bool,
642
+ gradient_checkpointing: bool,
643
+ discrete_custom_tokens: int | None = 128,
644
+ ) -> IXCLayout | PeftModel:
645
+ model = IXCLayout.from_ixc_pretrained(
646
+ ixc_pretrained_model_name_or_path=model_name_or_path,
647
+ max_length=max_length,
648
+ img_size=img_size,
649
+ discrete_custom_tokens=discrete_custom_tokens,
650
+ )
651
+ if use_lora:
652
+ lora_args = LoraArguments()
653
+ lora_config = LoraConfig(
654
+ r=lora_args.lora_r,
655
+ lora_alpha=lora_args.lora_alpha,
656
+ target_modules=lora_args.lora_target_modules,
657
+ lora_dropout=lora_args.lora_dropout,
658
+ bias=lora_args.lora_bias,
659
+ task_type="CAUSAL_LM", # type: ignore
660
+ )
661
+
662
+ model = get_peft_model(model, lora_config) # type: ignore
663
+ model.print_trainable_parameters()
664
+
665
+ if gradient_checkpointing:
666
+ model.enable_input_require_grads()
667
+ model.vit.vision_tower.gradient_checkpointing_enable(
668
+ {"use_reentrant": True}
669
+ )
670
+
671
+ return model # type: ignore
model-00001-of-00005.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:36b7014c0e8503b8a713268de8174320082f530651aa8fab0d1071132bd3cb45
3
+ size 4971460952
model-00002-of-00005.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:adcaf40ef3dd4003f490997dc868821699a8b0a6138c068389c8e42299eedd02
3
+ size 4999787752
model-00003-of-00005.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9820557f3f087aaa9da4351751be0363a219046d53bbb8aeaa2fee251571139d
3
+ size 4999787912
model-00004-of-00005.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3cdf7273cef864a992e08bd02e5bd5346ca8870196640cd5600d1525cddab8cc
3
+ size 4999787912
model-00005-of-00005.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:816b10acaf1d2345c699f8aa70a335961249e06de832f4869a7e5f0976802634
3
+ size 2227189176
model.safetensors.index.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json CHANGED
@@ -71,6 +71,3086 @@
71
  "rstrip": false,
72
  "single_word": false,
73
  "special": true
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
74
  }
75
  },
76
  "additional_special_tokens": [
 
71
  "rstrip": false,
72
  "single_word": false,
73
  "special": true
74
+ },
75
+ "92544": {
76
+ "content": "<|-128|>",
77
+ "lstrip": false,
78
+ "normalized": true,
79
+ "rstrip": false,
80
+ "single_word": false,
81
+ "special": false
82
+ },
83
+ "92545": {
84
+ "content": "<|-127|>",
85
+ "lstrip": false,
86
+ "normalized": true,
87
+ "rstrip": false,
88
+ "single_word": false,
89
+ "special": false
90
+ },
91
+ "92546": {
92
+ "content": "<|-126|>",
93
+ "lstrip": false,
94
+ "normalized": true,
95
+ "rstrip": false,
96
+ "single_word": false,
97
+ "special": false
98
+ },
99
+ "92547": {
100
+ "content": "<|-125|>",
101
+ "lstrip": false,
102
+ "normalized": true,
103
+ "rstrip": false,
104
+ "single_word": false,
105
+ "special": false
106
+ },
107
+ "92548": {
108
+ "content": "<|-124|>",
109
+ "lstrip": false,
110
+ "normalized": true,
111
+ "rstrip": false,
112
+ "single_word": false,
113
+ "special": false
114
+ },
115
+ "92549": {
116
+ "content": "<|-123|>",
117
+ "lstrip": false,
118
+ "normalized": true,
119
+ "rstrip": false,
120
+ "single_word": false,
121
+ "special": false
122
+ },
123
+ "92550": {
124
+ "content": "<|-122|>",
125
+ "lstrip": false,
126
+ "normalized": true,
127
+ "rstrip": false,
128
+ "single_word": false,
129
+ "special": false
130
+ },
131
+ "92551": {
132
+ "content": "<|-121|>",
133
+ "lstrip": false,
134
+ "normalized": true,
135
+ "rstrip": false,
136
+ "single_word": false,
137
+ "special": false
138
+ },
139
+ "92552": {
140
+ "content": "<|-120|>",
141
+ "lstrip": false,
142
+ "normalized": true,
143
+ "rstrip": false,
144
+ "single_word": false,
145
+ "special": false
146
+ },
147
+ "92553": {
148
+ "content": "<|-119|>",
149
+ "lstrip": false,
150
+ "normalized": true,
151
+ "rstrip": false,
152
+ "single_word": false,
153
+ "special": false
154
+ },
155
+ "92554": {
156
+ "content": "<|-118|>",
157
+ "lstrip": false,
158
+ "normalized": true,
159
+ "rstrip": false,
160
+ "single_word": false,
161
+ "special": false
162
+ },
163
+ "92555": {
164
+ "content": "<|-117|>",
165
+ "lstrip": false,
166
+ "normalized": true,
167
+ "rstrip": false,
168
+ "single_word": false,
169
+ "special": false
170
+ },
171
+ "92556": {
172
+ "content": "<|-116|>",
173
+ "lstrip": false,
174
+ "normalized": true,
175
+ "rstrip": false,
176
+ "single_word": false,
177
+ "special": false
178
+ },
179
+ "92557": {
180
+ "content": "<|-115|>",
181
+ "lstrip": false,
182
+ "normalized": true,
183
+ "rstrip": false,
184
+ "single_word": false,
185
+ "special": false
186
+ },
187
+ "92558": {
188
+ "content": "<|-114|>",
189
+ "lstrip": false,
190
+ "normalized": true,
191
+ "rstrip": false,
192
+ "single_word": false,
193
+ "special": false
194
+ },
195
+ "92559": {
196
+ "content": "<|-113|>",
197
+ "lstrip": false,
198
+ "normalized": true,
199
+ "rstrip": false,
200
+ "single_word": false,
201
+ "special": false
202
+ },
203
+ "92560": {
204
+ "content": "<|-112|>",
205
+ "lstrip": false,
206
+ "normalized": true,
207
+ "rstrip": false,
208
+ "single_word": false,
209
+ "special": false
210
+ },
211
+ "92561": {
212
+ "content": "<|-111|>",
213
+ "lstrip": false,
214
+ "normalized": true,
215
+ "rstrip": false,
216
+ "single_word": false,
217
+ "special": false
218
+ },
219
+ "92562": {
220
+ "content": "<|-110|>",
221
+ "lstrip": false,
222
+ "normalized": true,
223
+ "rstrip": false,
224
+ "single_word": false,
225
+ "special": false
226
+ },
227
+ "92563": {
228
+ "content": "<|-109|>",
229
+ "lstrip": false,
230
+ "normalized": true,
231
+ "rstrip": false,
232
+ "single_word": false,
233
+ "special": false
234
+ },
235
+ "92564": {
236
+ "content": "<|-108|>",
237
+ "lstrip": false,
238
+ "normalized": true,
239
+ "rstrip": false,
240
+ "single_word": false,
241
+ "special": false
242
+ },
243
+ "92565": {
244
+ "content": "<|-107|>",
245
+ "lstrip": false,
246
+ "normalized": true,
247
+ "rstrip": false,
248
+ "single_word": false,
249
+ "special": false
250
+ },
251
+ "92566": {
252
+ "content": "<|-106|>",
253
+ "lstrip": false,
254
+ "normalized": true,
255
+ "rstrip": false,
256
+ "single_word": false,
257
+ "special": false
258
+ },
259
+ "92567": {
260
+ "content": "<|-105|>",
261
+ "lstrip": false,
262
+ "normalized": true,
263
+ "rstrip": false,
264
+ "single_word": false,
265
+ "special": false
266
+ },
267
+ "92568": {
268
+ "content": "<|-104|>",
269
+ "lstrip": false,
270
+ "normalized": true,
271
+ "rstrip": false,
272
+ "single_word": false,
273
+ "special": false
274
+ },
275
+ "92569": {
276
+ "content": "<|-103|>",
277
+ "lstrip": false,
278
+ "normalized": true,
279
+ "rstrip": false,
280
+ "single_word": false,
281
+ "special": false
282
+ },
283
+ "92570": {
284
+ "content": "<|-102|>",
285
+ "lstrip": false,
286
+ "normalized": true,
287
+ "rstrip": false,
288
+ "single_word": false,
289
+ "special": false
290
+ },
291
+ "92571": {
292
+ "content": "<|-101|>",
293
+ "lstrip": false,
294
+ "normalized": true,
295
+ "rstrip": false,
296
+ "single_word": false,
297
+ "special": false
298
+ },
299
+ "92572": {
300
+ "content": "<|-100|>",
301
+ "lstrip": false,
302
+ "normalized": true,
303
+ "rstrip": false,
304
+ "single_word": false,
305
+ "special": false
306
+ },
307
+ "92573": {
308
+ "content": "<|-99|>",
309
+ "lstrip": false,
310
+ "normalized": true,
311
+ "rstrip": false,
312
+ "single_word": false,
313
+ "special": false
314
+ },
315
+ "92574": {
316
+ "content": "<|-98|>",
317
+ "lstrip": false,
318
+ "normalized": true,
319
+ "rstrip": false,
320
+ "single_word": false,
321
+ "special": false
322
+ },
323
+ "92575": {
324
+ "content": "<|-97|>",
325
+ "lstrip": false,
326
+ "normalized": true,
327
+ "rstrip": false,
328
+ "single_word": false,
329
+ "special": false
330
+ },
331
+ "92576": {
332
+ "content": "<|-96|>",
333
+ "lstrip": false,
334
+ "normalized": true,
335
+ "rstrip": false,
336
+ "single_word": false,
337
+ "special": false
338
+ },
339
+ "92577": {
340
+ "content": "<|-95|>",
341
+ "lstrip": false,
342
+ "normalized": true,
343
+ "rstrip": false,
344
+ "single_word": false,
345
+ "special": false
346
+ },
347
+ "92578": {
348
+ "content": "<|-94|>",
349
+ "lstrip": false,
350
+ "normalized": true,
351
+ "rstrip": false,
352
+ "single_word": false,
353
+ "special": false
354
+ },
355
+ "92579": {
356
+ "content": "<|-93|>",
357
+ "lstrip": false,
358
+ "normalized": true,
359
+ "rstrip": false,
360
+ "single_word": false,
361
+ "special": false
362
+ },
363
+ "92580": {
364
+ "content": "<|-92|>",
365
+ "lstrip": false,
366
+ "normalized": true,
367
+ "rstrip": false,
368
+ "single_word": false,
369
+ "special": false
370
+ },
371
+ "92581": {
372
+ "content": "<|-91|>",
373
+ "lstrip": false,
374
+ "normalized": true,
375
+ "rstrip": false,
376
+ "single_word": false,
377
+ "special": false
378
+ },
379
+ "92582": {
380
+ "content": "<|-90|>",
381
+ "lstrip": false,
382
+ "normalized": true,
383
+ "rstrip": false,
384
+ "single_word": false,
385
+ "special": false
386
+ },
387
+ "92583": {
388
+ "content": "<|-89|>",
389
+ "lstrip": false,
390
+ "normalized": true,
391
+ "rstrip": false,
392
+ "single_word": false,
393
+ "special": false
394
+ },
395
+ "92584": {
396
+ "content": "<|-88|>",
397
+ "lstrip": false,
398
+ "normalized": true,
399
+ "rstrip": false,
400
+ "single_word": false,
401
+ "special": false
402
+ },
403
+ "92585": {
404
+ "content": "<|-87|>",
405
+ "lstrip": false,
406
+ "normalized": true,
407
+ "rstrip": false,
408
+ "single_word": false,
409
+ "special": false
410
+ },
411
+ "92586": {
412
+ "content": "<|-86|>",
413
+ "lstrip": false,
414
+ "normalized": true,
415
+ "rstrip": false,
416
+ "single_word": false,
417
+ "special": false
418
+ },
419
+ "92587": {
420
+ "content": "<|-85|>",
421
+ "lstrip": false,
422
+ "normalized": true,
423
+ "rstrip": false,
424
+ "single_word": false,
425
+ "special": false
426
+ },
427
+ "92588": {
428
+ "content": "<|-84|>",
429
+ "lstrip": false,
430
+ "normalized": true,
431
+ "rstrip": false,
432
+ "single_word": false,
433
+ "special": false
434
+ },
435
+ "92589": {
436
+ "content": "<|-83|>",
437
+ "lstrip": false,
438
+ "normalized": true,
439
+ "rstrip": false,
440
+ "single_word": false,
441
+ "special": false
442
+ },
443
+ "92590": {
444
+ "content": "<|-82|>",
445
+ "lstrip": false,
446
+ "normalized": true,
447
+ "rstrip": false,
448
+ "single_word": false,
449
+ "special": false
450
+ },
451
+ "92591": {
452
+ "content": "<|-81|>",
453
+ "lstrip": false,
454
+ "normalized": true,
455
+ "rstrip": false,
456
+ "single_word": false,
457
+ "special": false
458
+ },
459
+ "92592": {
460
+ "content": "<|-80|>",
461
+ "lstrip": false,
462
+ "normalized": true,
463
+ "rstrip": false,
464
+ "single_word": false,
465
+ "special": false
466
+ },
467
+ "92593": {
468
+ "content": "<|-79|>",
469
+ "lstrip": false,
470
+ "normalized": true,
471
+ "rstrip": false,
472
+ "single_word": false,
473
+ "special": false
474
+ },
475
+ "92594": {
476
+ "content": "<|-78|>",
477
+ "lstrip": false,
478
+ "normalized": true,
479
+ "rstrip": false,
480
+ "single_word": false,
481
+ "special": false
482
+ },
483
+ "92595": {
484
+ "content": "<|-77|>",
485
+ "lstrip": false,
486
+ "normalized": true,
487
+ "rstrip": false,
488
+ "single_word": false,
489
+ "special": false
490
+ },
491
+ "92596": {
492
+ "content": "<|-76|>",
493
+ "lstrip": false,
494
+ "normalized": true,
495
+ "rstrip": false,
496
+ "single_word": false,
497
+ "special": false
498
+ },
499
+ "92597": {
500
+ "content": "<|-75|>",
501
+ "lstrip": false,
502
+ "normalized": true,
503
+ "rstrip": false,
504
+ "single_word": false,
505
+ "special": false
506
+ },
507
+ "92598": {
508
+ "content": "<|-74|>",
509
+ "lstrip": false,
510
+ "normalized": true,
511
+ "rstrip": false,
512
+ "single_word": false,
513
+ "special": false
514
+ },
515
+ "92599": {
516
+ "content": "<|-73|>",
517
+ "lstrip": false,
518
+ "normalized": true,
519
+ "rstrip": false,
520
+ "single_word": false,
521
+ "special": false
522
+ },
523
+ "92600": {
524
+ "content": "<|-72|>",
525
+ "lstrip": false,
526
+ "normalized": true,
527
+ "rstrip": false,
528
+ "single_word": false,
529
+ "special": false
530
+ },
531
+ "92601": {
532
+ "content": "<|-71|>",
533
+ "lstrip": false,
534
+ "normalized": true,
535
+ "rstrip": false,
536
+ "single_word": false,
537
+ "special": false
538
+ },
539
+ "92602": {
540
+ "content": "<|-70|>",
541
+ "lstrip": false,
542
+ "normalized": true,
543
+ "rstrip": false,
544
+ "single_word": false,
545
+ "special": false
546
+ },
547
+ "92603": {
548
+ "content": "<|-69|>",
549
+ "lstrip": false,
550
+ "normalized": true,
551
+ "rstrip": false,
552
+ "single_word": false,
553
+ "special": false
554
+ },
555
+ "92604": {
556
+ "content": "<|-68|>",
557
+ "lstrip": false,
558
+ "normalized": true,
559
+ "rstrip": false,
560
+ "single_word": false,
561
+ "special": false
562
+ },
563
+ "92605": {
564
+ "content": "<|-67|>",
565
+ "lstrip": false,
566
+ "normalized": true,
567
+ "rstrip": false,
568
+ "single_word": false,
569
+ "special": false
570
+ },
571
+ "92606": {
572
+ "content": "<|-66|>",
573
+ "lstrip": false,
574
+ "normalized": true,
575
+ "rstrip": false,
576
+ "single_word": false,
577
+ "special": false
578
+ },
579
+ "92607": {
580
+ "content": "<|-65|>",
581
+ "lstrip": false,
582
+ "normalized": true,
583
+ "rstrip": false,
584
+ "single_word": false,
585
+ "special": false
586
+ },
587
+ "92608": {
588
+ "content": "<|-64|>",
589
+ "lstrip": false,
590
+ "normalized": true,
591
+ "rstrip": false,
592
+ "single_word": false,
593
+ "special": false
594
+ },
595
+ "92609": {
596
+ "content": "<|-63|>",
597
+ "lstrip": false,
598
+ "normalized": true,
599
+ "rstrip": false,
600
+ "single_word": false,
601
+ "special": false
602
+ },
603
+ "92610": {
604
+ "content": "<|-62|>",
605
+ "lstrip": false,
606
+ "normalized": true,
607
+ "rstrip": false,
608
+ "single_word": false,
609
+ "special": false
610
+ },
611
+ "92611": {
612
+ "content": "<|-61|>",
613
+ "lstrip": false,
614
+ "normalized": true,
615
+ "rstrip": false,
616
+ "single_word": false,
617
+ "special": false
618
+ },
619
+ "92612": {
620
+ "content": "<|-60|>",
621
+ "lstrip": false,
622
+ "normalized": true,
623
+ "rstrip": false,
624
+ "single_word": false,
625
+ "special": false
626
+ },
627
+ "92613": {
628
+ "content": "<|-59|>",
629
+ "lstrip": false,
630
+ "normalized": true,
631
+ "rstrip": false,
632
+ "single_word": false,
633
+ "special": false
634
+ },
635
+ "92614": {
636
+ "content": "<|-58|>",
637
+ "lstrip": false,
638
+ "normalized": true,
639
+ "rstrip": false,
640
+ "single_word": false,
641
+ "special": false
642
+ },
643
+ "92615": {
644
+ "content": "<|-57|>",
645
+ "lstrip": false,
646
+ "normalized": true,
647
+ "rstrip": false,
648
+ "single_word": false,
649
+ "special": false
650
+ },
651
+ "92616": {
652
+ "content": "<|-56|>",
653
+ "lstrip": false,
654
+ "normalized": true,
655
+ "rstrip": false,
656
+ "single_word": false,
657
+ "special": false
658
+ },
659
+ "92617": {
660
+ "content": "<|-55|>",
661
+ "lstrip": false,
662
+ "normalized": true,
663
+ "rstrip": false,
664
+ "single_word": false,
665
+ "special": false
666
+ },
667
+ "92618": {
668
+ "content": "<|-54|>",
669
+ "lstrip": false,
670
+ "normalized": true,
671
+ "rstrip": false,
672
+ "single_word": false,
673
+ "special": false
674
+ },
675
+ "92619": {
676
+ "content": "<|-53|>",
677
+ "lstrip": false,
678
+ "normalized": true,
679
+ "rstrip": false,
680
+ "single_word": false,
681
+ "special": false
682
+ },
683
+ "92620": {
684
+ "content": "<|-52|>",
685
+ "lstrip": false,
686
+ "normalized": true,
687
+ "rstrip": false,
688
+ "single_word": false,
689
+ "special": false
690
+ },
691
+ "92621": {
692
+ "content": "<|-51|>",
693
+ "lstrip": false,
694
+ "normalized": true,
695
+ "rstrip": false,
696
+ "single_word": false,
697
+ "special": false
698
+ },
699
+ "92622": {
700
+ "content": "<|-50|>",
701
+ "lstrip": false,
702
+ "normalized": true,
703
+ "rstrip": false,
704
+ "single_word": false,
705
+ "special": false
706
+ },
707
+ "92623": {
708
+ "content": "<|-49|>",
709
+ "lstrip": false,
710
+ "normalized": true,
711
+ "rstrip": false,
712
+ "single_word": false,
713
+ "special": false
714
+ },
715
+ "92624": {
716
+ "content": "<|-48|>",
717
+ "lstrip": false,
718
+ "normalized": true,
719
+ "rstrip": false,
720
+ "single_word": false,
721
+ "special": false
722
+ },
723
+ "92625": {
724
+ "content": "<|-47|>",
725
+ "lstrip": false,
726
+ "normalized": true,
727
+ "rstrip": false,
728
+ "single_word": false,
729
+ "special": false
730
+ },
731
+ "92626": {
732
+ "content": "<|-46|>",
733
+ "lstrip": false,
734
+ "normalized": true,
735
+ "rstrip": false,
736
+ "single_word": false,
737
+ "special": false
738
+ },
739
+ "92627": {
740
+ "content": "<|-45|>",
741
+ "lstrip": false,
742
+ "normalized": true,
743
+ "rstrip": false,
744
+ "single_word": false,
745
+ "special": false
746
+ },
747
+ "92628": {
748
+ "content": "<|-44|>",
749
+ "lstrip": false,
750
+ "normalized": true,
751
+ "rstrip": false,
752
+ "single_word": false,
753
+ "special": false
754
+ },
755
+ "92629": {
756
+ "content": "<|-43|>",
757
+ "lstrip": false,
758
+ "normalized": true,
759
+ "rstrip": false,
760
+ "single_word": false,
761
+ "special": false
762
+ },
763
+ "92630": {
764
+ "content": "<|-42|>",
765
+ "lstrip": false,
766
+ "normalized": true,
767
+ "rstrip": false,
768
+ "single_word": false,
769
+ "special": false
770
+ },
771
+ "92631": {
772
+ "content": "<|-41|>",
773
+ "lstrip": false,
774
+ "normalized": true,
775
+ "rstrip": false,
776
+ "single_word": false,
777
+ "special": false
778
+ },
779
+ "92632": {
780
+ "content": "<|-40|>",
781
+ "lstrip": false,
782
+ "normalized": true,
783
+ "rstrip": false,
784
+ "single_word": false,
785
+ "special": false
786
+ },
787
+ "92633": {
788
+ "content": "<|-39|>",
789
+ "lstrip": false,
790
+ "normalized": true,
791
+ "rstrip": false,
792
+ "single_word": false,
793
+ "special": false
794
+ },
795
+ "92634": {
796
+ "content": "<|-38|>",
797
+ "lstrip": false,
798
+ "normalized": true,
799
+ "rstrip": false,
800
+ "single_word": false,
801
+ "special": false
802
+ },
803
+ "92635": {
804
+ "content": "<|-37|>",
805
+ "lstrip": false,
806
+ "normalized": true,
807
+ "rstrip": false,
808
+ "single_word": false,
809
+ "special": false
810
+ },
811
+ "92636": {
812
+ "content": "<|-36|>",
813
+ "lstrip": false,
814
+ "normalized": true,
815
+ "rstrip": false,
816
+ "single_word": false,
817
+ "special": false
818
+ },
819
+ "92637": {
820
+ "content": "<|-35|>",
821
+ "lstrip": false,
822
+ "normalized": true,
823
+ "rstrip": false,
824
+ "single_word": false,
825
+ "special": false
826
+ },
827
+ "92638": {
828
+ "content": "<|-34|>",
829
+ "lstrip": false,
830
+ "normalized": true,
831
+ "rstrip": false,
832
+ "single_word": false,
833
+ "special": false
834
+ },
835
+ "92639": {
836
+ "content": "<|-33|>",
837
+ "lstrip": false,
838
+ "normalized": true,
839
+ "rstrip": false,
840
+ "single_word": false,
841
+ "special": false
842
+ },
843
+ "92640": {
844
+ "content": "<|-32|>",
845
+ "lstrip": false,
846
+ "normalized": true,
847
+ "rstrip": false,
848
+ "single_word": false,
849
+ "special": false
850
+ },
851
+ "92641": {
852
+ "content": "<|-31|>",
853
+ "lstrip": false,
854
+ "normalized": true,
855
+ "rstrip": false,
856
+ "single_word": false,
857
+ "special": false
858
+ },
859
+ "92642": {
860
+ "content": "<|-30|>",
861
+ "lstrip": false,
862
+ "normalized": true,
863
+ "rstrip": false,
864
+ "single_word": false,
865
+ "special": false
866
+ },
867
+ "92643": {
868
+ "content": "<|-29|>",
869
+ "lstrip": false,
870
+ "normalized": true,
871
+ "rstrip": false,
872
+ "single_word": false,
873
+ "special": false
874
+ },
875
+ "92644": {
876
+ "content": "<|-28|>",
877
+ "lstrip": false,
878
+ "normalized": true,
879
+ "rstrip": false,
880
+ "single_word": false,
881
+ "special": false
882
+ },
883
+ "92645": {
884
+ "content": "<|-27|>",
885
+ "lstrip": false,
886
+ "normalized": true,
887
+ "rstrip": false,
888
+ "single_word": false,
889
+ "special": false
890
+ },
891
+ "92646": {
892
+ "content": "<|-26|>",
893
+ "lstrip": false,
894
+ "normalized": true,
895
+ "rstrip": false,
896
+ "single_word": false,
897
+ "special": false
898
+ },
899
+ "92647": {
900
+ "content": "<|-25|>",
901
+ "lstrip": false,
902
+ "normalized": true,
903
+ "rstrip": false,
904
+ "single_word": false,
905
+ "special": false
906
+ },
907
+ "92648": {
908
+ "content": "<|-24|>",
909
+ "lstrip": false,
910
+ "normalized": true,
911
+ "rstrip": false,
912
+ "single_word": false,
913
+ "special": false
914
+ },
915
+ "92649": {
916
+ "content": "<|-23|>",
917
+ "lstrip": false,
918
+ "normalized": true,
919
+ "rstrip": false,
920
+ "single_word": false,
921
+ "special": false
922
+ },
923
+ "92650": {
924
+ "content": "<|-22|>",
925
+ "lstrip": false,
926
+ "normalized": true,
927
+ "rstrip": false,
928
+ "single_word": false,
929
+ "special": false
930
+ },
931
+ "92651": {
932
+ "content": "<|-21|>",
933
+ "lstrip": false,
934
+ "normalized": true,
935
+ "rstrip": false,
936
+ "single_word": false,
937
+ "special": false
938
+ },
939
+ "92652": {
940
+ "content": "<|-20|>",
941
+ "lstrip": false,
942
+ "normalized": true,
943
+ "rstrip": false,
944
+ "single_word": false,
945
+ "special": false
946
+ },
947
+ "92653": {
948
+ "content": "<|-19|>",
949
+ "lstrip": false,
950
+ "normalized": true,
951
+ "rstrip": false,
952
+ "single_word": false,
953
+ "special": false
954
+ },
955
+ "92654": {
956
+ "content": "<|-18|>",
957
+ "lstrip": false,
958
+ "normalized": true,
959
+ "rstrip": false,
960
+ "single_word": false,
961
+ "special": false
962
+ },
963
+ "92655": {
964
+ "content": "<|-17|>",
965
+ "lstrip": false,
966
+ "normalized": true,
967
+ "rstrip": false,
968
+ "single_word": false,
969
+ "special": false
970
+ },
971
+ "92656": {
972
+ "content": "<|-16|>",
973
+ "lstrip": false,
974
+ "normalized": true,
975
+ "rstrip": false,
976
+ "single_word": false,
977
+ "special": false
978
+ },
979
+ "92657": {
980
+ "content": "<|-15|>",
981
+ "lstrip": false,
982
+ "normalized": true,
983
+ "rstrip": false,
984
+ "single_word": false,
985
+ "special": false
986
+ },
987
+ "92658": {
988
+ "content": "<|-14|>",
989
+ "lstrip": false,
990
+ "normalized": true,
991
+ "rstrip": false,
992
+ "single_word": false,
993
+ "special": false
994
+ },
995
+ "92659": {
996
+ "content": "<|-13|>",
997
+ "lstrip": false,
998
+ "normalized": true,
999
+ "rstrip": false,
1000
+ "single_word": false,
1001
+ "special": false
1002
+ },
1003
+ "92660": {
1004
+ "content": "<|-12|>",
1005
+ "lstrip": false,
1006
+ "normalized": true,
1007
+ "rstrip": false,
1008
+ "single_word": false,
1009
+ "special": false
1010
+ },
1011
+ "92661": {
1012
+ "content": "<|-11|>",
1013
+ "lstrip": false,
1014
+ "normalized": true,
1015
+ "rstrip": false,
1016
+ "single_word": false,
1017
+ "special": false
1018
+ },
1019
+ "92662": {
1020
+ "content": "<|-10|>",
1021
+ "lstrip": false,
1022
+ "normalized": true,
1023
+ "rstrip": false,
1024
+ "single_word": false,
1025
+ "special": false
1026
+ },
1027
+ "92663": {
1028
+ "content": "<|-9|>",
1029
+ "lstrip": false,
1030
+ "normalized": true,
1031
+ "rstrip": false,
1032
+ "single_word": false,
1033
+ "special": false
1034
+ },
1035
+ "92664": {
1036
+ "content": "<|-8|>",
1037
+ "lstrip": false,
1038
+ "normalized": true,
1039
+ "rstrip": false,
1040
+ "single_word": false,
1041
+ "special": false
1042
+ },
1043
+ "92665": {
1044
+ "content": "<|-7|>",
1045
+ "lstrip": false,
1046
+ "normalized": true,
1047
+ "rstrip": false,
1048
+ "single_word": false,
1049
+ "special": false
1050
+ },
1051
+ "92666": {
1052
+ "content": "<|-6|>",
1053
+ "lstrip": false,
1054
+ "normalized": true,
1055
+ "rstrip": false,
1056
+ "single_word": false,
1057
+ "special": false
1058
+ },
1059
+ "92667": {
1060
+ "content": "<|-5|>",
1061
+ "lstrip": false,
1062
+ "normalized": true,
1063
+ "rstrip": false,
1064
+ "single_word": false,
1065
+ "special": false
1066
+ },
1067
+ "92668": {
1068
+ "content": "<|-4|>",
1069
+ "lstrip": false,
1070
+ "normalized": true,
1071
+ "rstrip": false,
1072
+ "single_word": false,
1073
+ "special": false
1074
+ },
1075
+ "92669": {
1076
+ "content": "<|-3|>",
1077
+ "lstrip": false,
1078
+ "normalized": true,
1079
+ "rstrip": false,
1080
+ "single_word": false,
1081
+ "special": false
1082
+ },
1083
+ "92670": {
1084
+ "content": "<|-2|>",
1085
+ "lstrip": false,
1086
+ "normalized": true,
1087
+ "rstrip": false,
1088
+ "single_word": false,
1089
+ "special": false
1090
+ },
1091
+ "92671": {
1092
+ "content": "<|-1|>",
1093
+ "lstrip": false,
1094
+ "normalized": true,
1095
+ "rstrip": false,
1096
+ "single_word": false,
1097
+ "special": false
1098
+ },
1099
+ "92672": {
1100
+ "content": "<|0|>",
1101
+ "lstrip": false,
1102
+ "normalized": true,
1103
+ "rstrip": false,
1104
+ "single_word": false,
1105
+ "special": false
1106
+ },
1107
+ "92673": {
1108
+ "content": "<|1|>",
1109
+ "lstrip": false,
1110
+ "normalized": true,
1111
+ "rstrip": false,
1112
+ "single_word": false,
1113
+ "special": false
1114
+ },
1115
+ "92674": {
1116
+ "content": "<|2|>",
1117
+ "lstrip": false,
1118
+ "normalized": true,
1119
+ "rstrip": false,
1120
+ "single_word": false,
1121
+ "special": false
1122
+ },
1123
+ "92675": {
1124
+ "content": "<|3|>",
1125
+ "lstrip": false,
1126
+ "normalized": true,
1127
+ "rstrip": false,
1128
+ "single_word": false,
1129
+ "special": false
1130
+ },
1131
+ "92676": {
1132
+ "content": "<|4|>",
1133
+ "lstrip": false,
1134
+ "normalized": true,
1135
+ "rstrip": false,
1136
+ "single_word": false,
1137
+ "special": false
1138
+ },
1139
+ "92677": {
1140
+ "content": "<|5|>",
1141
+ "lstrip": false,
1142
+ "normalized": true,
1143
+ "rstrip": false,
1144
+ "single_word": false,
1145
+ "special": false
1146
+ },
1147
+ "92678": {
1148
+ "content": "<|6|>",
1149
+ "lstrip": false,
1150
+ "normalized": true,
1151
+ "rstrip": false,
1152
+ "single_word": false,
1153
+ "special": false
1154
+ },
1155
+ "92679": {
1156
+ "content": "<|7|>",
1157
+ "lstrip": false,
1158
+ "normalized": true,
1159
+ "rstrip": false,
1160
+ "single_word": false,
1161
+ "special": false
1162
+ },
1163
+ "92680": {
1164
+ "content": "<|8|>",
1165
+ "lstrip": false,
1166
+ "normalized": true,
1167
+ "rstrip": false,
1168
+ "single_word": false,
1169
+ "special": false
1170
+ },
1171
+ "92681": {
1172
+ "content": "<|9|>",
1173
+ "lstrip": false,
1174
+ "normalized": true,
1175
+ "rstrip": false,
1176
+ "single_word": false,
1177
+ "special": false
1178
+ },
1179
+ "92682": {
1180
+ "content": "<|10|>",
1181
+ "lstrip": false,
1182
+ "normalized": true,
1183
+ "rstrip": false,
1184
+ "single_word": false,
1185
+ "special": false
1186
+ },
1187
+ "92683": {
1188
+ "content": "<|11|>",
1189
+ "lstrip": false,
1190
+ "normalized": true,
1191
+ "rstrip": false,
1192
+ "single_word": false,
1193
+ "special": false
1194
+ },
1195
+ "92684": {
1196
+ "content": "<|12|>",
1197
+ "lstrip": false,
1198
+ "normalized": true,
1199
+ "rstrip": false,
1200
+ "single_word": false,
1201
+ "special": false
1202
+ },
1203
+ "92685": {
1204
+ "content": "<|13|>",
1205
+ "lstrip": false,
1206
+ "normalized": true,
1207
+ "rstrip": false,
1208
+ "single_word": false,
1209
+ "special": false
1210
+ },
1211
+ "92686": {
1212
+ "content": "<|14|>",
1213
+ "lstrip": false,
1214
+ "normalized": true,
1215
+ "rstrip": false,
1216
+ "single_word": false,
1217
+ "special": false
1218
+ },
1219
+ "92687": {
1220
+ "content": "<|15|>",
1221
+ "lstrip": false,
1222
+ "normalized": true,
1223
+ "rstrip": false,
1224
+ "single_word": false,
1225
+ "special": false
1226
+ },
1227
+ "92688": {
1228
+ "content": "<|16|>",
1229
+ "lstrip": false,
1230
+ "normalized": true,
1231
+ "rstrip": false,
1232
+ "single_word": false,
1233
+ "special": false
1234
+ },
1235
+ "92689": {
1236
+ "content": "<|17|>",
1237
+ "lstrip": false,
1238
+ "normalized": true,
1239
+ "rstrip": false,
1240
+ "single_word": false,
1241
+ "special": false
1242
+ },
1243
+ "92690": {
1244
+ "content": "<|18|>",
1245
+ "lstrip": false,
1246
+ "normalized": true,
1247
+ "rstrip": false,
1248
+ "single_word": false,
1249
+ "special": false
1250
+ },
1251
+ "92691": {
1252
+ "content": "<|19|>",
1253
+ "lstrip": false,
1254
+ "normalized": true,
1255
+ "rstrip": false,
1256
+ "single_word": false,
1257
+ "special": false
1258
+ },
1259
+ "92692": {
1260
+ "content": "<|20|>",
1261
+ "lstrip": false,
1262
+ "normalized": true,
1263
+ "rstrip": false,
1264
+ "single_word": false,
1265
+ "special": false
1266
+ },
1267
+ "92693": {
1268
+ "content": "<|21|>",
1269
+ "lstrip": false,
1270
+ "normalized": true,
1271
+ "rstrip": false,
1272
+ "single_word": false,
1273
+ "special": false
1274
+ },
1275
+ "92694": {
1276
+ "content": "<|22|>",
1277
+ "lstrip": false,
1278
+ "normalized": true,
1279
+ "rstrip": false,
1280
+ "single_word": false,
1281
+ "special": false
1282
+ },
1283
+ "92695": {
1284
+ "content": "<|23|>",
1285
+ "lstrip": false,
1286
+ "normalized": true,
1287
+ "rstrip": false,
1288
+ "single_word": false,
1289
+ "special": false
1290
+ },
1291
+ "92696": {
1292
+ "content": "<|24|>",
1293
+ "lstrip": false,
1294
+ "normalized": true,
1295
+ "rstrip": false,
1296
+ "single_word": false,
1297
+ "special": false
1298
+ },
1299
+ "92697": {
1300
+ "content": "<|25|>",
1301
+ "lstrip": false,
1302
+ "normalized": true,
1303
+ "rstrip": false,
1304
+ "single_word": false,
1305
+ "special": false
1306
+ },
1307
+ "92698": {
1308
+ "content": "<|26|>",
1309
+ "lstrip": false,
1310
+ "normalized": true,
1311
+ "rstrip": false,
1312
+ "single_word": false,
1313
+ "special": false
1314
+ },
1315
+ "92699": {
1316
+ "content": "<|27|>",
1317
+ "lstrip": false,
1318
+ "normalized": true,
1319
+ "rstrip": false,
1320
+ "single_word": false,
1321
+ "special": false
1322
+ },
1323
+ "92700": {
1324
+ "content": "<|28|>",
1325
+ "lstrip": false,
1326
+ "normalized": true,
1327
+ "rstrip": false,
1328
+ "single_word": false,
1329
+ "special": false
1330
+ },
1331
+ "92701": {
1332
+ "content": "<|29|>",
1333
+ "lstrip": false,
1334
+ "normalized": true,
1335
+ "rstrip": false,
1336
+ "single_word": false,
1337
+ "special": false
1338
+ },
1339
+ "92702": {
1340
+ "content": "<|30|>",
1341
+ "lstrip": false,
1342
+ "normalized": true,
1343
+ "rstrip": false,
1344
+ "single_word": false,
1345
+ "special": false
1346
+ },
1347
+ "92703": {
1348
+ "content": "<|31|>",
1349
+ "lstrip": false,
1350
+ "normalized": true,
1351
+ "rstrip": false,
1352
+ "single_word": false,
1353
+ "special": false
1354
+ },
1355
+ "92704": {
1356
+ "content": "<|32|>",
1357
+ "lstrip": false,
1358
+ "normalized": true,
1359
+ "rstrip": false,
1360
+ "single_word": false,
1361
+ "special": false
1362
+ },
1363
+ "92705": {
1364
+ "content": "<|33|>",
1365
+ "lstrip": false,
1366
+ "normalized": true,
1367
+ "rstrip": false,
1368
+ "single_word": false,
1369
+ "special": false
1370
+ },
1371
+ "92706": {
1372
+ "content": "<|34|>",
1373
+ "lstrip": false,
1374
+ "normalized": true,
1375
+ "rstrip": false,
1376
+ "single_word": false,
1377
+ "special": false
1378
+ },
1379
+ "92707": {
1380
+ "content": "<|35|>",
1381
+ "lstrip": false,
1382
+ "normalized": true,
1383
+ "rstrip": false,
1384
+ "single_word": false,
1385
+ "special": false
1386
+ },
1387
+ "92708": {
1388
+ "content": "<|36|>",
1389
+ "lstrip": false,
1390
+ "normalized": true,
1391
+ "rstrip": false,
1392
+ "single_word": false,
1393
+ "special": false
1394
+ },
1395
+ "92709": {
1396
+ "content": "<|37|>",
1397
+ "lstrip": false,
1398
+ "normalized": true,
1399
+ "rstrip": false,
1400
+ "single_word": false,
1401
+ "special": false
1402
+ },
1403
+ "92710": {
1404
+ "content": "<|38|>",
1405
+ "lstrip": false,
1406
+ "normalized": true,
1407
+ "rstrip": false,
1408
+ "single_word": false,
1409
+ "special": false
1410
+ },
1411
+ "92711": {
1412
+ "content": "<|39|>",
1413
+ "lstrip": false,
1414
+ "normalized": true,
1415
+ "rstrip": false,
1416
+ "single_word": false,
1417
+ "special": false
1418
+ },
1419
+ "92712": {
1420
+ "content": "<|40|>",
1421
+ "lstrip": false,
1422
+ "normalized": true,
1423
+ "rstrip": false,
1424
+ "single_word": false,
1425
+ "special": false
1426
+ },
1427
+ "92713": {
1428
+ "content": "<|41|>",
1429
+ "lstrip": false,
1430
+ "normalized": true,
1431
+ "rstrip": false,
1432
+ "single_word": false,
1433
+ "special": false
1434
+ },
1435
+ "92714": {
1436
+ "content": "<|42|>",
1437
+ "lstrip": false,
1438
+ "normalized": true,
1439
+ "rstrip": false,
1440
+ "single_word": false,
1441
+ "special": false
1442
+ },
1443
+ "92715": {
1444
+ "content": "<|43|>",
1445
+ "lstrip": false,
1446
+ "normalized": true,
1447
+ "rstrip": false,
1448
+ "single_word": false,
1449
+ "special": false
1450
+ },
1451
+ "92716": {
1452
+ "content": "<|44|>",
1453
+ "lstrip": false,
1454
+ "normalized": true,
1455
+ "rstrip": false,
1456
+ "single_word": false,
1457
+ "special": false
1458
+ },
1459
+ "92717": {
1460
+ "content": "<|45|>",
1461
+ "lstrip": false,
1462
+ "normalized": true,
1463
+ "rstrip": false,
1464
+ "single_word": false,
1465
+ "special": false
1466
+ },
1467
+ "92718": {
1468
+ "content": "<|46|>",
1469
+ "lstrip": false,
1470
+ "normalized": true,
1471
+ "rstrip": false,
1472
+ "single_word": false,
1473
+ "special": false
1474
+ },
1475
+ "92719": {
1476
+ "content": "<|47|>",
1477
+ "lstrip": false,
1478
+ "normalized": true,
1479
+ "rstrip": false,
1480
+ "single_word": false,
1481
+ "special": false
1482
+ },
1483
+ "92720": {
1484
+ "content": "<|48|>",
1485
+ "lstrip": false,
1486
+ "normalized": true,
1487
+ "rstrip": false,
1488
+ "single_word": false,
1489
+ "special": false
1490
+ },
1491
+ "92721": {
1492
+ "content": "<|49|>",
1493
+ "lstrip": false,
1494
+ "normalized": true,
1495
+ "rstrip": false,
1496
+ "single_word": false,
1497
+ "special": false
1498
+ },
1499
+ "92722": {
1500
+ "content": "<|50|>",
1501
+ "lstrip": false,
1502
+ "normalized": true,
1503
+ "rstrip": false,
1504
+ "single_word": false,
1505
+ "special": false
1506
+ },
1507
+ "92723": {
1508
+ "content": "<|51|>",
1509
+ "lstrip": false,
1510
+ "normalized": true,
1511
+ "rstrip": false,
1512
+ "single_word": false,
1513
+ "special": false
1514
+ },
1515
+ "92724": {
1516
+ "content": "<|52|>",
1517
+ "lstrip": false,
1518
+ "normalized": true,
1519
+ "rstrip": false,
1520
+ "single_word": false,
1521
+ "special": false
1522
+ },
1523
+ "92725": {
1524
+ "content": "<|53|>",
1525
+ "lstrip": false,
1526
+ "normalized": true,
1527
+ "rstrip": false,
1528
+ "single_word": false,
1529
+ "special": false
1530
+ },
1531
+ "92726": {
1532
+ "content": "<|54|>",
1533
+ "lstrip": false,
1534
+ "normalized": true,
1535
+ "rstrip": false,
1536
+ "single_word": false,
1537
+ "special": false
1538
+ },
1539
+ "92727": {
1540
+ "content": "<|55|>",
1541
+ "lstrip": false,
1542
+ "normalized": true,
1543
+ "rstrip": false,
1544
+ "single_word": false,
1545
+ "special": false
1546
+ },
1547
+ "92728": {
1548
+ "content": "<|56|>",
1549
+ "lstrip": false,
1550
+ "normalized": true,
1551
+ "rstrip": false,
1552
+ "single_word": false,
1553
+ "special": false
1554
+ },
1555
+ "92729": {
1556
+ "content": "<|57|>",
1557
+ "lstrip": false,
1558
+ "normalized": true,
1559
+ "rstrip": false,
1560
+ "single_word": false,
1561
+ "special": false
1562
+ },
1563
+ "92730": {
1564
+ "content": "<|58|>",
1565
+ "lstrip": false,
1566
+ "normalized": true,
1567
+ "rstrip": false,
1568
+ "single_word": false,
1569
+ "special": false
1570
+ },
1571
+ "92731": {
1572
+ "content": "<|59|>",
1573
+ "lstrip": false,
1574
+ "normalized": true,
1575
+ "rstrip": false,
1576
+ "single_word": false,
1577
+ "special": false
1578
+ },
1579
+ "92732": {
1580
+ "content": "<|60|>",
1581
+ "lstrip": false,
1582
+ "normalized": true,
1583
+ "rstrip": false,
1584
+ "single_word": false,
1585
+ "special": false
1586
+ },
1587
+ "92733": {
1588
+ "content": "<|61|>",
1589
+ "lstrip": false,
1590
+ "normalized": true,
1591
+ "rstrip": false,
1592
+ "single_word": false,
1593
+ "special": false
1594
+ },
1595
+ "92734": {
1596
+ "content": "<|62|>",
1597
+ "lstrip": false,
1598
+ "normalized": true,
1599
+ "rstrip": false,
1600
+ "single_word": false,
1601
+ "special": false
1602
+ },
1603
+ "92735": {
1604
+ "content": "<|63|>",
1605
+ "lstrip": false,
1606
+ "normalized": true,
1607
+ "rstrip": false,
1608
+ "single_word": false,
1609
+ "special": false
1610
+ },
1611
+ "92736": {
1612
+ "content": "<|64|>",
1613
+ "lstrip": false,
1614
+ "normalized": true,
1615
+ "rstrip": false,
1616
+ "single_word": false,
1617
+ "special": false
1618
+ },
1619
+ "92737": {
1620
+ "content": "<|65|>",
1621
+ "lstrip": false,
1622
+ "normalized": true,
1623
+ "rstrip": false,
1624
+ "single_word": false,
1625
+ "special": false
1626
+ },
1627
+ "92738": {
1628
+ "content": "<|66|>",
1629
+ "lstrip": false,
1630
+ "normalized": true,
1631
+ "rstrip": false,
1632
+ "single_word": false,
1633
+ "special": false
1634
+ },
1635
+ "92739": {
1636
+ "content": "<|67|>",
1637
+ "lstrip": false,
1638
+ "normalized": true,
1639
+ "rstrip": false,
1640
+ "single_word": false,
1641
+ "special": false
1642
+ },
1643
+ "92740": {
1644
+ "content": "<|68|>",
1645
+ "lstrip": false,
1646
+ "normalized": true,
1647
+ "rstrip": false,
1648
+ "single_word": false,
1649
+ "special": false
1650
+ },
1651
+ "92741": {
1652
+ "content": "<|69|>",
1653
+ "lstrip": false,
1654
+ "normalized": true,
1655
+ "rstrip": false,
1656
+ "single_word": false,
1657
+ "special": false
1658
+ },
1659
+ "92742": {
1660
+ "content": "<|70|>",
1661
+ "lstrip": false,
1662
+ "normalized": true,
1663
+ "rstrip": false,
1664
+ "single_word": false,
1665
+ "special": false
1666
+ },
1667
+ "92743": {
1668
+ "content": "<|71|>",
1669
+ "lstrip": false,
1670
+ "normalized": true,
1671
+ "rstrip": false,
1672
+ "single_word": false,
1673
+ "special": false
1674
+ },
1675
+ "92744": {
1676
+ "content": "<|72|>",
1677
+ "lstrip": false,
1678
+ "normalized": true,
1679
+ "rstrip": false,
1680
+ "single_word": false,
1681
+ "special": false
1682
+ },
1683
+ "92745": {
1684
+ "content": "<|73|>",
1685
+ "lstrip": false,
1686
+ "normalized": true,
1687
+ "rstrip": false,
1688
+ "single_word": false,
1689
+ "special": false
1690
+ },
1691
+ "92746": {
1692
+ "content": "<|74|>",
1693
+ "lstrip": false,
1694
+ "normalized": true,
1695
+ "rstrip": false,
1696
+ "single_word": false,
1697
+ "special": false
1698
+ },
1699
+ "92747": {
1700
+ "content": "<|75|>",
1701
+ "lstrip": false,
1702
+ "normalized": true,
1703
+ "rstrip": false,
1704
+ "single_word": false,
1705
+ "special": false
1706
+ },
1707
+ "92748": {
1708
+ "content": "<|76|>",
1709
+ "lstrip": false,
1710
+ "normalized": true,
1711
+ "rstrip": false,
1712
+ "single_word": false,
1713
+ "special": false
1714
+ },
1715
+ "92749": {
1716
+ "content": "<|77|>",
1717
+ "lstrip": false,
1718
+ "normalized": true,
1719
+ "rstrip": false,
1720
+ "single_word": false,
1721
+ "special": false
1722
+ },
1723
+ "92750": {
1724
+ "content": "<|78|>",
1725
+ "lstrip": false,
1726
+ "normalized": true,
1727
+ "rstrip": false,
1728
+ "single_word": false,
1729
+ "special": false
1730
+ },
1731
+ "92751": {
1732
+ "content": "<|79|>",
1733
+ "lstrip": false,
1734
+ "normalized": true,
1735
+ "rstrip": false,
1736
+ "single_word": false,
1737
+ "special": false
1738
+ },
1739
+ "92752": {
1740
+ "content": "<|80|>",
1741
+ "lstrip": false,
1742
+ "normalized": true,
1743
+ "rstrip": false,
1744
+ "single_word": false,
1745
+ "special": false
1746
+ },
1747
+ "92753": {
1748
+ "content": "<|81|>",
1749
+ "lstrip": false,
1750
+ "normalized": true,
1751
+ "rstrip": false,
1752
+ "single_word": false,
1753
+ "special": false
1754
+ },
1755
+ "92754": {
1756
+ "content": "<|82|>",
1757
+ "lstrip": false,
1758
+ "normalized": true,
1759
+ "rstrip": false,
1760
+ "single_word": false,
1761
+ "special": false
1762
+ },
1763
+ "92755": {
1764
+ "content": "<|83|>",
1765
+ "lstrip": false,
1766
+ "normalized": true,
1767
+ "rstrip": false,
1768
+ "single_word": false,
1769
+ "special": false
1770
+ },
1771
+ "92756": {
1772
+ "content": "<|84|>",
1773
+ "lstrip": false,
1774
+ "normalized": true,
1775
+ "rstrip": false,
1776
+ "single_word": false,
1777
+ "special": false
1778
+ },
1779
+ "92757": {
1780
+ "content": "<|85|>",
1781
+ "lstrip": false,
1782
+ "normalized": true,
1783
+ "rstrip": false,
1784
+ "single_word": false,
1785
+ "special": false
1786
+ },
1787
+ "92758": {
1788
+ "content": "<|86|>",
1789
+ "lstrip": false,
1790
+ "normalized": true,
1791
+ "rstrip": false,
1792
+ "single_word": false,
1793
+ "special": false
1794
+ },
1795
+ "92759": {
1796
+ "content": "<|87|>",
1797
+ "lstrip": false,
1798
+ "normalized": true,
1799
+ "rstrip": false,
1800
+ "single_word": false,
1801
+ "special": false
1802
+ },
1803
+ "92760": {
1804
+ "content": "<|88|>",
1805
+ "lstrip": false,
1806
+ "normalized": true,
1807
+ "rstrip": false,
1808
+ "single_word": false,
1809
+ "special": false
1810
+ },
1811
+ "92761": {
1812
+ "content": "<|89|>",
1813
+ "lstrip": false,
1814
+ "normalized": true,
1815
+ "rstrip": false,
1816
+ "single_word": false,
1817
+ "special": false
1818
+ },
1819
+ "92762": {
1820
+ "content": "<|90|>",
1821
+ "lstrip": false,
1822
+ "normalized": true,
1823
+ "rstrip": false,
1824
+ "single_word": false,
1825
+ "special": false
1826
+ },
1827
+ "92763": {
1828
+ "content": "<|91|>",
1829
+ "lstrip": false,
1830
+ "normalized": true,
1831
+ "rstrip": false,
1832
+ "single_word": false,
1833
+ "special": false
1834
+ },
1835
+ "92764": {
1836
+ "content": "<|92|>",
1837
+ "lstrip": false,
1838
+ "normalized": true,
1839
+ "rstrip": false,
1840
+ "single_word": false,
1841
+ "special": false
1842
+ },
1843
+ "92765": {
1844
+ "content": "<|93|>",
1845
+ "lstrip": false,
1846
+ "normalized": true,
1847
+ "rstrip": false,
1848
+ "single_word": false,
1849
+ "special": false
1850
+ },
1851
+ "92766": {
1852
+ "content": "<|94|>",
1853
+ "lstrip": false,
1854
+ "normalized": true,
1855
+ "rstrip": false,
1856
+ "single_word": false,
1857
+ "special": false
1858
+ },
1859
+ "92767": {
1860
+ "content": "<|95|>",
1861
+ "lstrip": false,
1862
+ "normalized": true,
1863
+ "rstrip": false,
1864
+ "single_word": false,
1865
+ "special": false
1866
+ },
1867
+ "92768": {
1868
+ "content": "<|96|>",
1869
+ "lstrip": false,
1870
+ "normalized": true,
1871
+ "rstrip": false,
1872
+ "single_word": false,
1873
+ "special": false
1874
+ },
1875
+ "92769": {
1876
+ "content": "<|97|>",
1877
+ "lstrip": false,
1878
+ "normalized": true,
1879
+ "rstrip": false,
1880
+ "single_word": false,
1881
+ "special": false
1882
+ },
1883
+ "92770": {
1884
+ "content": "<|98|>",
1885
+ "lstrip": false,
1886
+ "normalized": true,
1887
+ "rstrip": false,
1888
+ "single_word": false,
1889
+ "special": false
1890
+ },
1891
+ "92771": {
1892
+ "content": "<|99|>",
1893
+ "lstrip": false,
1894
+ "normalized": true,
1895
+ "rstrip": false,
1896
+ "single_word": false,
1897
+ "special": false
1898
+ },
1899
+ "92772": {
1900
+ "content": "<|100|>",
1901
+ "lstrip": false,
1902
+ "normalized": true,
1903
+ "rstrip": false,
1904
+ "single_word": false,
1905
+ "special": false
1906
+ },
1907
+ "92773": {
1908
+ "content": "<|101|>",
1909
+ "lstrip": false,
1910
+ "normalized": true,
1911
+ "rstrip": false,
1912
+ "single_word": false,
1913
+ "special": false
1914
+ },
1915
+ "92774": {
1916
+ "content": "<|102|>",
1917
+ "lstrip": false,
1918
+ "normalized": true,
1919
+ "rstrip": false,
1920
+ "single_word": false,
1921
+ "special": false
1922
+ },
1923
+ "92775": {
1924
+ "content": "<|103|>",
1925
+ "lstrip": false,
1926
+ "normalized": true,
1927
+ "rstrip": false,
1928
+ "single_word": false,
1929
+ "special": false
1930
+ },
1931
+ "92776": {
1932
+ "content": "<|104|>",
1933
+ "lstrip": false,
1934
+ "normalized": true,
1935
+ "rstrip": false,
1936
+ "single_word": false,
1937
+ "special": false
1938
+ },
1939
+ "92777": {
1940
+ "content": "<|105|>",
1941
+ "lstrip": false,
1942
+ "normalized": true,
1943
+ "rstrip": false,
1944
+ "single_word": false,
1945
+ "special": false
1946
+ },
1947
+ "92778": {
1948
+ "content": "<|106|>",
1949
+ "lstrip": false,
1950
+ "normalized": true,
1951
+ "rstrip": false,
1952
+ "single_word": false,
1953
+ "special": false
1954
+ },
1955
+ "92779": {
1956
+ "content": "<|107|>",
1957
+ "lstrip": false,
1958
+ "normalized": true,
1959
+ "rstrip": false,
1960
+ "single_word": false,
1961
+ "special": false
1962
+ },
1963
+ "92780": {
1964
+ "content": "<|108|>",
1965
+ "lstrip": false,
1966
+ "normalized": true,
1967
+ "rstrip": false,
1968
+ "single_word": false,
1969
+ "special": false
1970
+ },
1971
+ "92781": {
1972
+ "content": "<|109|>",
1973
+ "lstrip": false,
1974
+ "normalized": true,
1975
+ "rstrip": false,
1976
+ "single_word": false,
1977
+ "special": false
1978
+ },
1979
+ "92782": {
1980
+ "content": "<|110|>",
1981
+ "lstrip": false,
1982
+ "normalized": true,
1983
+ "rstrip": false,
1984
+ "single_word": false,
1985
+ "special": false
1986
+ },
1987
+ "92783": {
1988
+ "content": "<|111|>",
1989
+ "lstrip": false,
1990
+ "normalized": true,
1991
+ "rstrip": false,
1992
+ "single_word": false,
1993
+ "special": false
1994
+ },
1995
+ "92784": {
1996
+ "content": "<|112|>",
1997
+ "lstrip": false,
1998
+ "normalized": true,
1999
+ "rstrip": false,
2000
+ "single_word": false,
2001
+ "special": false
2002
+ },
2003
+ "92785": {
2004
+ "content": "<|113|>",
2005
+ "lstrip": false,
2006
+ "normalized": true,
2007
+ "rstrip": false,
2008
+ "single_word": false,
2009
+ "special": false
2010
+ },
2011
+ "92786": {
2012
+ "content": "<|114|>",
2013
+ "lstrip": false,
2014
+ "normalized": true,
2015
+ "rstrip": false,
2016
+ "single_word": false,
2017
+ "special": false
2018
+ },
2019
+ "92787": {
2020
+ "content": "<|115|>",
2021
+ "lstrip": false,
2022
+ "normalized": true,
2023
+ "rstrip": false,
2024
+ "single_word": false,
2025
+ "special": false
2026
+ },
2027
+ "92788": {
2028
+ "content": "<|116|>",
2029
+ "lstrip": false,
2030
+ "normalized": true,
2031
+ "rstrip": false,
2032
+ "single_word": false,
2033
+ "special": false
2034
+ },
2035
+ "92789": {
2036
+ "content": "<|117|>",
2037
+ "lstrip": false,
2038
+ "normalized": true,
2039
+ "rstrip": false,
2040
+ "single_word": false,
2041
+ "special": false
2042
+ },
2043
+ "92790": {
2044
+ "content": "<|118|>",
2045
+ "lstrip": false,
2046
+ "normalized": true,
2047
+ "rstrip": false,
2048
+ "single_word": false,
2049
+ "special": false
2050
+ },
2051
+ "92791": {
2052
+ "content": "<|119|>",
2053
+ "lstrip": false,
2054
+ "normalized": true,
2055
+ "rstrip": false,
2056
+ "single_word": false,
2057
+ "special": false
2058
+ },
2059
+ "92792": {
2060
+ "content": "<|120|>",
2061
+ "lstrip": false,
2062
+ "normalized": true,
2063
+ "rstrip": false,
2064
+ "single_word": false,
2065
+ "special": false
2066
+ },
2067
+ "92793": {
2068
+ "content": "<|121|>",
2069
+ "lstrip": false,
2070
+ "normalized": true,
2071
+ "rstrip": false,
2072
+ "single_word": false,
2073
+ "special": false
2074
+ },
2075
+ "92794": {
2076
+ "content": "<|122|>",
2077
+ "lstrip": false,
2078
+ "normalized": true,
2079
+ "rstrip": false,
2080
+ "single_word": false,
2081
+ "special": false
2082
+ },
2083
+ "92795": {
2084
+ "content": "<|123|>",
2085
+ "lstrip": false,
2086
+ "normalized": true,
2087
+ "rstrip": false,
2088
+ "single_word": false,
2089
+ "special": false
2090
+ },
2091
+ "92796": {
2092
+ "content": "<|124|>",
2093
+ "lstrip": false,
2094
+ "normalized": true,
2095
+ "rstrip": false,
2096
+ "single_word": false,
2097
+ "special": false
2098
+ },
2099
+ "92797": {
2100
+ "content": "<|125|>",
2101
+ "lstrip": false,
2102
+ "normalized": true,
2103
+ "rstrip": false,
2104
+ "single_word": false,
2105
+ "special": false
2106
+ },
2107
+ "92798": {
2108
+ "content": "<|126|>",
2109
+ "lstrip": false,
2110
+ "normalized": true,
2111
+ "rstrip": false,
2112
+ "single_word": false,
2113
+ "special": false
2114
+ },
2115
+ "92799": {
2116
+ "content": "<|127|>",
2117
+ "lstrip": false,
2118
+ "normalized": true,
2119
+ "rstrip": false,
2120
+ "single_word": false,
2121
+ "special": false
2122
+ },
2123
+ "92800": {
2124
+ "content": "<|128|>",
2125
+ "lstrip": false,
2126
+ "normalized": true,
2127
+ "rstrip": false,
2128
+ "single_word": false,
2129
+ "special": false
2130
+ },
2131
+ "92801": {
2132
+ "content": "<|129|>",
2133
+ "lstrip": false,
2134
+ "normalized": true,
2135
+ "rstrip": false,
2136
+ "single_word": false,
2137
+ "special": false
2138
+ },
2139
+ "92802": {
2140
+ "content": "<|130|>",
2141
+ "lstrip": false,
2142
+ "normalized": true,
2143
+ "rstrip": false,
2144
+ "single_word": false,
2145
+ "special": false
2146
+ },
2147
+ "92803": {
2148
+ "content": "<|131|>",
2149
+ "lstrip": false,
2150
+ "normalized": true,
2151
+ "rstrip": false,
2152
+ "single_word": false,
2153
+ "special": false
2154
+ },
2155
+ "92804": {
2156
+ "content": "<|132|>",
2157
+ "lstrip": false,
2158
+ "normalized": true,
2159
+ "rstrip": false,
2160
+ "single_word": false,
2161
+ "special": false
2162
+ },
2163
+ "92805": {
2164
+ "content": "<|133|>",
2165
+ "lstrip": false,
2166
+ "normalized": true,
2167
+ "rstrip": false,
2168
+ "single_word": false,
2169
+ "special": false
2170
+ },
2171
+ "92806": {
2172
+ "content": "<|134|>",
2173
+ "lstrip": false,
2174
+ "normalized": true,
2175
+ "rstrip": false,
2176
+ "single_word": false,
2177
+ "special": false
2178
+ },
2179
+ "92807": {
2180
+ "content": "<|135|>",
2181
+ "lstrip": false,
2182
+ "normalized": true,
2183
+ "rstrip": false,
2184
+ "single_word": false,
2185
+ "special": false
2186
+ },
2187
+ "92808": {
2188
+ "content": "<|136|>",
2189
+ "lstrip": false,
2190
+ "normalized": true,
2191
+ "rstrip": false,
2192
+ "single_word": false,
2193
+ "special": false
2194
+ },
2195
+ "92809": {
2196
+ "content": "<|137|>",
2197
+ "lstrip": false,
2198
+ "normalized": true,
2199
+ "rstrip": false,
2200
+ "single_word": false,
2201
+ "special": false
2202
+ },
2203
+ "92810": {
2204
+ "content": "<|138|>",
2205
+ "lstrip": false,
2206
+ "normalized": true,
2207
+ "rstrip": false,
2208
+ "single_word": false,
2209
+ "special": false
2210
+ },
2211
+ "92811": {
2212
+ "content": "<|139|>",
2213
+ "lstrip": false,
2214
+ "normalized": true,
2215
+ "rstrip": false,
2216
+ "single_word": false,
2217
+ "special": false
2218
+ },
2219
+ "92812": {
2220
+ "content": "<|140|>",
2221
+ "lstrip": false,
2222
+ "normalized": true,
2223
+ "rstrip": false,
2224
+ "single_word": false,
2225
+ "special": false
2226
+ },
2227
+ "92813": {
2228
+ "content": "<|141|>",
2229
+ "lstrip": false,
2230
+ "normalized": true,
2231
+ "rstrip": false,
2232
+ "single_word": false,
2233
+ "special": false
2234
+ },
2235
+ "92814": {
2236
+ "content": "<|142|>",
2237
+ "lstrip": false,
2238
+ "normalized": true,
2239
+ "rstrip": false,
2240
+ "single_word": false,
2241
+ "special": false
2242
+ },
2243
+ "92815": {
2244
+ "content": "<|143|>",
2245
+ "lstrip": false,
2246
+ "normalized": true,
2247
+ "rstrip": false,
2248
+ "single_word": false,
2249
+ "special": false
2250
+ },
2251
+ "92816": {
2252
+ "content": "<|144|>",
2253
+ "lstrip": false,
2254
+ "normalized": true,
2255
+ "rstrip": false,
2256
+ "single_word": false,
2257
+ "special": false
2258
+ },
2259
+ "92817": {
2260
+ "content": "<|145|>",
2261
+ "lstrip": false,
2262
+ "normalized": true,
2263
+ "rstrip": false,
2264
+ "single_word": false,
2265
+ "special": false
2266
+ },
2267
+ "92818": {
2268
+ "content": "<|146|>",
2269
+ "lstrip": false,
2270
+ "normalized": true,
2271
+ "rstrip": false,
2272
+ "single_word": false,
2273
+ "special": false
2274
+ },
2275
+ "92819": {
2276
+ "content": "<|147|>",
2277
+ "lstrip": false,
2278
+ "normalized": true,
2279
+ "rstrip": false,
2280
+ "single_word": false,
2281
+ "special": false
2282
+ },
2283
+ "92820": {
2284
+ "content": "<|148|>",
2285
+ "lstrip": false,
2286
+ "normalized": true,
2287
+ "rstrip": false,
2288
+ "single_word": false,
2289
+ "special": false
2290
+ },
2291
+ "92821": {
2292
+ "content": "<|149|>",
2293
+ "lstrip": false,
2294
+ "normalized": true,
2295
+ "rstrip": false,
2296
+ "single_word": false,
2297
+ "special": false
2298
+ },
2299
+ "92822": {
2300
+ "content": "<|150|>",
2301
+ "lstrip": false,
2302
+ "normalized": true,
2303
+ "rstrip": false,
2304
+ "single_word": false,
2305
+ "special": false
2306
+ },
2307
+ "92823": {
2308
+ "content": "<|151|>",
2309
+ "lstrip": false,
2310
+ "normalized": true,
2311
+ "rstrip": false,
2312
+ "single_word": false,
2313
+ "special": false
2314
+ },
2315
+ "92824": {
2316
+ "content": "<|152|>",
2317
+ "lstrip": false,
2318
+ "normalized": true,
2319
+ "rstrip": false,
2320
+ "single_word": false,
2321
+ "special": false
2322
+ },
2323
+ "92825": {
2324
+ "content": "<|153|>",
2325
+ "lstrip": false,
2326
+ "normalized": true,
2327
+ "rstrip": false,
2328
+ "single_word": false,
2329
+ "special": false
2330
+ },
2331
+ "92826": {
2332
+ "content": "<|154|>",
2333
+ "lstrip": false,
2334
+ "normalized": true,
2335
+ "rstrip": false,
2336
+ "single_word": false,
2337
+ "special": false
2338
+ },
2339
+ "92827": {
2340
+ "content": "<|155|>",
2341
+ "lstrip": false,
2342
+ "normalized": true,
2343
+ "rstrip": false,
2344
+ "single_word": false,
2345
+ "special": false
2346
+ },
2347
+ "92828": {
2348
+ "content": "<|156|>",
2349
+ "lstrip": false,
2350
+ "normalized": true,
2351
+ "rstrip": false,
2352
+ "single_word": false,
2353
+ "special": false
2354
+ },
2355
+ "92829": {
2356
+ "content": "<|157|>",
2357
+ "lstrip": false,
2358
+ "normalized": true,
2359
+ "rstrip": false,
2360
+ "single_word": false,
2361
+ "special": false
2362
+ },
2363
+ "92830": {
2364
+ "content": "<|158|>",
2365
+ "lstrip": false,
2366
+ "normalized": true,
2367
+ "rstrip": false,
2368
+ "single_word": false,
2369
+ "special": false
2370
+ },
2371
+ "92831": {
2372
+ "content": "<|159|>",
2373
+ "lstrip": false,
2374
+ "normalized": true,
2375
+ "rstrip": false,
2376
+ "single_word": false,
2377
+ "special": false
2378
+ },
2379
+ "92832": {
2380
+ "content": "<|160|>",
2381
+ "lstrip": false,
2382
+ "normalized": true,
2383
+ "rstrip": false,
2384
+ "single_word": false,
2385
+ "special": false
2386
+ },
2387
+ "92833": {
2388
+ "content": "<|161|>",
2389
+ "lstrip": false,
2390
+ "normalized": true,
2391
+ "rstrip": false,
2392
+ "single_word": false,
2393
+ "special": false
2394
+ },
2395
+ "92834": {
2396
+ "content": "<|162|>",
2397
+ "lstrip": false,
2398
+ "normalized": true,
2399
+ "rstrip": false,
2400
+ "single_word": false,
2401
+ "special": false
2402
+ },
2403
+ "92835": {
2404
+ "content": "<|163|>",
2405
+ "lstrip": false,
2406
+ "normalized": true,
2407
+ "rstrip": false,
2408
+ "single_word": false,
2409
+ "special": false
2410
+ },
2411
+ "92836": {
2412
+ "content": "<|164|>",
2413
+ "lstrip": false,
2414
+ "normalized": true,
2415
+ "rstrip": false,
2416
+ "single_word": false,
2417
+ "special": false
2418
+ },
2419
+ "92837": {
2420
+ "content": "<|165|>",
2421
+ "lstrip": false,
2422
+ "normalized": true,
2423
+ "rstrip": false,
2424
+ "single_word": false,
2425
+ "special": false
2426
+ },
2427
+ "92838": {
2428
+ "content": "<|166|>",
2429
+ "lstrip": false,
2430
+ "normalized": true,
2431
+ "rstrip": false,
2432
+ "single_word": false,
2433
+ "special": false
2434
+ },
2435
+ "92839": {
2436
+ "content": "<|167|>",
2437
+ "lstrip": false,
2438
+ "normalized": true,
2439
+ "rstrip": false,
2440
+ "single_word": false,
2441
+ "special": false
2442
+ },
2443
+ "92840": {
2444
+ "content": "<|168|>",
2445
+ "lstrip": false,
2446
+ "normalized": true,
2447
+ "rstrip": false,
2448
+ "single_word": false,
2449
+ "special": false
2450
+ },
2451
+ "92841": {
2452
+ "content": "<|169|>",
2453
+ "lstrip": false,
2454
+ "normalized": true,
2455
+ "rstrip": false,
2456
+ "single_word": false,
2457
+ "special": false
2458
+ },
2459
+ "92842": {
2460
+ "content": "<|170|>",
2461
+ "lstrip": false,
2462
+ "normalized": true,
2463
+ "rstrip": false,
2464
+ "single_word": false,
2465
+ "special": false
2466
+ },
2467
+ "92843": {
2468
+ "content": "<|171|>",
2469
+ "lstrip": false,
2470
+ "normalized": true,
2471
+ "rstrip": false,
2472
+ "single_word": false,
2473
+ "special": false
2474
+ },
2475
+ "92844": {
2476
+ "content": "<|172|>",
2477
+ "lstrip": false,
2478
+ "normalized": true,
2479
+ "rstrip": false,
2480
+ "single_word": false,
2481
+ "special": false
2482
+ },
2483
+ "92845": {
2484
+ "content": "<|173|>",
2485
+ "lstrip": false,
2486
+ "normalized": true,
2487
+ "rstrip": false,
2488
+ "single_word": false,
2489
+ "special": false
2490
+ },
2491
+ "92846": {
2492
+ "content": "<|174|>",
2493
+ "lstrip": false,
2494
+ "normalized": true,
2495
+ "rstrip": false,
2496
+ "single_word": false,
2497
+ "special": false
2498
+ },
2499
+ "92847": {
2500
+ "content": "<|175|>",
2501
+ "lstrip": false,
2502
+ "normalized": true,
2503
+ "rstrip": false,
2504
+ "single_word": false,
2505
+ "special": false
2506
+ },
2507
+ "92848": {
2508
+ "content": "<|176|>",
2509
+ "lstrip": false,
2510
+ "normalized": true,
2511
+ "rstrip": false,
2512
+ "single_word": false,
2513
+ "special": false
2514
+ },
2515
+ "92849": {
2516
+ "content": "<|177|>",
2517
+ "lstrip": false,
2518
+ "normalized": true,
2519
+ "rstrip": false,
2520
+ "single_word": false,
2521
+ "special": false
2522
+ },
2523
+ "92850": {
2524
+ "content": "<|178|>",
2525
+ "lstrip": false,
2526
+ "normalized": true,
2527
+ "rstrip": false,
2528
+ "single_word": false,
2529
+ "special": false
2530
+ },
2531
+ "92851": {
2532
+ "content": "<|179|>",
2533
+ "lstrip": false,
2534
+ "normalized": true,
2535
+ "rstrip": false,
2536
+ "single_word": false,
2537
+ "special": false
2538
+ },
2539
+ "92852": {
2540
+ "content": "<|180|>",
2541
+ "lstrip": false,
2542
+ "normalized": true,
2543
+ "rstrip": false,
2544
+ "single_word": false,
2545
+ "special": false
2546
+ },
2547
+ "92853": {
2548
+ "content": "<|181|>",
2549
+ "lstrip": false,
2550
+ "normalized": true,
2551
+ "rstrip": false,
2552
+ "single_word": false,
2553
+ "special": false
2554
+ },
2555
+ "92854": {
2556
+ "content": "<|182|>",
2557
+ "lstrip": false,
2558
+ "normalized": true,
2559
+ "rstrip": false,
2560
+ "single_word": false,
2561
+ "special": false
2562
+ },
2563
+ "92855": {
2564
+ "content": "<|183|>",
2565
+ "lstrip": false,
2566
+ "normalized": true,
2567
+ "rstrip": false,
2568
+ "single_word": false,
2569
+ "special": false
2570
+ },
2571
+ "92856": {
2572
+ "content": "<|184|>",
2573
+ "lstrip": false,
2574
+ "normalized": true,
2575
+ "rstrip": false,
2576
+ "single_word": false,
2577
+ "special": false
2578
+ },
2579
+ "92857": {
2580
+ "content": "<|185|>",
2581
+ "lstrip": false,
2582
+ "normalized": true,
2583
+ "rstrip": false,
2584
+ "single_word": false,
2585
+ "special": false
2586
+ },
2587
+ "92858": {
2588
+ "content": "<|186|>",
2589
+ "lstrip": false,
2590
+ "normalized": true,
2591
+ "rstrip": false,
2592
+ "single_word": false,
2593
+ "special": false
2594
+ },
2595
+ "92859": {
2596
+ "content": "<|187|>",
2597
+ "lstrip": false,
2598
+ "normalized": true,
2599
+ "rstrip": false,
2600
+ "single_word": false,
2601
+ "special": false
2602
+ },
2603
+ "92860": {
2604
+ "content": "<|188|>",
2605
+ "lstrip": false,
2606
+ "normalized": true,
2607
+ "rstrip": false,
2608
+ "single_word": false,
2609
+ "special": false
2610
+ },
2611
+ "92861": {
2612
+ "content": "<|189|>",
2613
+ "lstrip": false,
2614
+ "normalized": true,
2615
+ "rstrip": false,
2616
+ "single_word": false,
2617
+ "special": false
2618
+ },
2619
+ "92862": {
2620
+ "content": "<|190|>",
2621
+ "lstrip": false,
2622
+ "normalized": true,
2623
+ "rstrip": false,
2624
+ "single_word": false,
2625
+ "special": false
2626
+ },
2627
+ "92863": {
2628
+ "content": "<|191|>",
2629
+ "lstrip": false,
2630
+ "normalized": true,
2631
+ "rstrip": false,
2632
+ "single_word": false,
2633
+ "special": false
2634
+ },
2635
+ "92864": {
2636
+ "content": "<|192|>",
2637
+ "lstrip": false,
2638
+ "normalized": true,
2639
+ "rstrip": false,
2640
+ "single_word": false,
2641
+ "special": false
2642
+ },
2643
+ "92865": {
2644
+ "content": "<|193|>",
2645
+ "lstrip": false,
2646
+ "normalized": true,
2647
+ "rstrip": false,
2648
+ "single_word": false,
2649
+ "special": false
2650
+ },
2651
+ "92866": {
2652
+ "content": "<|194|>",
2653
+ "lstrip": false,
2654
+ "normalized": true,
2655
+ "rstrip": false,
2656
+ "single_word": false,
2657
+ "special": false
2658
+ },
2659
+ "92867": {
2660
+ "content": "<|195|>",
2661
+ "lstrip": false,
2662
+ "normalized": true,
2663
+ "rstrip": false,
2664
+ "single_word": false,
2665
+ "special": false
2666
+ },
2667
+ "92868": {
2668
+ "content": "<|196|>",
2669
+ "lstrip": false,
2670
+ "normalized": true,
2671
+ "rstrip": false,
2672
+ "single_word": false,
2673
+ "special": false
2674
+ },
2675
+ "92869": {
2676
+ "content": "<|197|>",
2677
+ "lstrip": false,
2678
+ "normalized": true,
2679
+ "rstrip": false,
2680
+ "single_word": false,
2681
+ "special": false
2682
+ },
2683
+ "92870": {
2684
+ "content": "<|198|>",
2685
+ "lstrip": false,
2686
+ "normalized": true,
2687
+ "rstrip": false,
2688
+ "single_word": false,
2689
+ "special": false
2690
+ },
2691
+ "92871": {
2692
+ "content": "<|199|>",
2693
+ "lstrip": false,
2694
+ "normalized": true,
2695
+ "rstrip": false,
2696
+ "single_word": false,
2697
+ "special": false
2698
+ },
2699
+ "92872": {
2700
+ "content": "<|200|>",
2701
+ "lstrip": false,
2702
+ "normalized": true,
2703
+ "rstrip": false,
2704
+ "single_word": false,
2705
+ "special": false
2706
+ },
2707
+ "92873": {
2708
+ "content": "<|201|>",
2709
+ "lstrip": false,
2710
+ "normalized": true,
2711
+ "rstrip": false,
2712
+ "single_word": false,
2713
+ "special": false
2714
+ },
2715
+ "92874": {
2716
+ "content": "<|202|>",
2717
+ "lstrip": false,
2718
+ "normalized": true,
2719
+ "rstrip": false,
2720
+ "single_word": false,
2721
+ "special": false
2722
+ },
2723
+ "92875": {
2724
+ "content": "<|203|>",
2725
+ "lstrip": false,
2726
+ "normalized": true,
2727
+ "rstrip": false,
2728
+ "single_word": false,
2729
+ "special": false
2730
+ },
2731
+ "92876": {
2732
+ "content": "<|204|>",
2733
+ "lstrip": false,
2734
+ "normalized": true,
2735
+ "rstrip": false,
2736
+ "single_word": false,
2737
+ "special": false
2738
+ },
2739
+ "92877": {
2740
+ "content": "<|205|>",
2741
+ "lstrip": false,
2742
+ "normalized": true,
2743
+ "rstrip": false,
2744
+ "single_word": false,
2745
+ "special": false
2746
+ },
2747
+ "92878": {
2748
+ "content": "<|206|>",
2749
+ "lstrip": false,
2750
+ "normalized": true,
2751
+ "rstrip": false,
2752
+ "single_word": false,
2753
+ "special": false
2754
+ },
2755
+ "92879": {
2756
+ "content": "<|207|>",
2757
+ "lstrip": false,
2758
+ "normalized": true,
2759
+ "rstrip": false,
2760
+ "single_word": false,
2761
+ "special": false
2762
+ },
2763
+ "92880": {
2764
+ "content": "<|208|>",
2765
+ "lstrip": false,
2766
+ "normalized": true,
2767
+ "rstrip": false,
2768
+ "single_word": false,
2769
+ "special": false
2770
+ },
2771
+ "92881": {
2772
+ "content": "<|209|>",
2773
+ "lstrip": false,
2774
+ "normalized": true,
2775
+ "rstrip": false,
2776
+ "single_word": false,
2777
+ "special": false
2778
+ },
2779
+ "92882": {
2780
+ "content": "<|210|>",
2781
+ "lstrip": false,
2782
+ "normalized": true,
2783
+ "rstrip": false,
2784
+ "single_word": false,
2785
+ "special": false
2786
+ },
2787
+ "92883": {
2788
+ "content": "<|211|>",
2789
+ "lstrip": false,
2790
+ "normalized": true,
2791
+ "rstrip": false,
2792
+ "single_word": false,
2793
+ "special": false
2794
+ },
2795
+ "92884": {
2796
+ "content": "<|212|>",
2797
+ "lstrip": false,
2798
+ "normalized": true,
2799
+ "rstrip": false,
2800
+ "single_word": false,
2801
+ "special": false
2802
+ },
2803
+ "92885": {
2804
+ "content": "<|213|>",
2805
+ "lstrip": false,
2806
+ "normalized": true,
2807
+ "rstrip": false,
2808
+ "single_word": false,
2809
+ "special": false
2810
+ },
2811
+ "92886": {
2812
+ "content": "<|214|>",
2813
+ "lstrip": false,
2814
+ "normalized": true,
2815
+ "rstrip": false,
2816
+ "single_word": false,
2817
+ "special": false
2818
+ },
2819
+ "92887": {
2820
+ "content": "<|215|>",
2821
+ "lstrip": false,
2822
+ "normalized": true,
2823
+ "rstrip": false,
2824
+ "single_word": false,
2825
+ "special": false
2826
+ },
2827
+ "92888": {
2828
+ "content": "<|216|>",
2829
+ "lstrip": false,
2830
+ "normalized": true,
2831
+ "rstrip": false,
2832
+ "single_word": false,
2833
+ "special": false
2834
+ },
2835
+ "92889": {
2836
+ "content": "<|217|>",
2837
+ "lstrip": false,
2838
+ "normalized": true,
2839
+ "rstrip": false,
2840
+ "single_word": false,
2841
+ "special": false
2842
+ },
2843
+ "92890": {
2844
+ "content": "<|218|>",
2845
+ "lstrip": false,
2846
+ "normalized": true,
2847
+ "rstrip": false,
2848
+ "single_word": false,
2849
+ "special": false
2850
+ },
2851
+ "92891": {
2852
+ "content": "<|219|>",
2853
+ "lstrip": false,
2854
+ "normalized": true,
2855
+ "rstrip": false,
2856
+ "single_word": false,
2857
+ "special": false
2858
+ },
2859
+ "92892": {
2860
+ "content": "<|220|>",
2861
+ "lstrip": false,
2862
+ "normalized": true,
2863
+ "rstrip": false,
2864
+ "single_word": false,
2865
+ "special": false
2866
+ },
2867
+ "92893": {
2868
+ "content": "<|221|>",
2869
+ "lstrip": false,
2870
+ "normalized": true,
2871
+ "rstrip": false,
2872
+ "single_word": false,
2873
+ "special": false
2874
+ },
2875
+ "92894": {
2876
+ "content": "<|222|>",
2877
+ "lstrip": false,
2878
+ "normalized": true,
2879
+ "rstrip": false,
2880
+ "single_word": false,
2881
+ "special": false
2882
+ },
2883
+ "92895": {
2884
+ "content": "<|223|>",
2885
+ "lstrip": false,
2886
+ "normalized": true,
2887
+ "rstrip": false,
2888
+ "single_word": false,
2889
+ "special": false
2890
+ },
2891
+ "92896": {
2892
+ "content": "<|224|>",
2893
+ "lstrip": false,
2894
+ "normalized": true,
2895
+ "rstrip": false,
2896
+ "single_word": false,
2897
+ "special": false
2898
+ },
2899
+ "92897": {
2900
+ "content": "<|225|>",
2901
+ "lstrip": false,
2902
+ "normalized": true,
2903
+ "rstrip": false,
2904
+ "single_word": false,
2905
+ "special": false
2906
+ },
2907
+ "92898": {
2908
+ "content": "<|226|>",
2909
+ "lstrip": false,
2910
+ "normalized": true,
2911
+ "rstrip": false,
2912
+ "single_word": false,
2913
+ "special": false
2914
+ },
2915
+ "92899": {
2916
+ "content": "<|227|>",
2917
+ "lstrip": false,
2918
+ "normalized": true,
2919
+ "rstrip": false,
2920
+ "single_word": false,
2921
+ "special": false
2922
+ },
2923
+ "92900": {
2924
+ "content": "<|228|>",
2925
+ "lstrip": false,
2926
+ "normalized": true,
2927
+ "rstrip": false,
2928
+ "single_word": false,
2929
+ "special": false
2930
+ },
2931
+ "92901": {
2932
+ "content": "<|229|>",
2933
+ "lstrip": false,
2934
+ "normalized": true,
2935
+ "rstrip": false,
2936
+ "single_word": false,
2937
+ "special": false
2938
+ },
2939
+ "92902": {
2940
+ "content": "<|230|>",
2941
+ "lstrip": false,
2942
+ "normalized": true,
2943
+ "rstrip": false,
2944
+ "single_word": false,
2945
+ "special": false
2946
+ },
2947
+ "92903": {
2948
+ "content": "<|231|>",
2949
+ "lstrip": false,
2950
+ "normalized": true,
2951
+ "rstrip": false,
2952
+ "single_word": false,
2953
+ "special": false
2954
+ },
2955
+ "92904": {
2956
+ "content": "<|232|>",
2957
+ "lstrip": false,
2958
+ "normalized": true,
2959
+ "rstrip": false,
2960
+ "single_word": false,
2961
+ "special": false
2962
+ },
2963
+ "92905": {
2964
+ "content": "<|233|>",
2965
+ "lstrip": false,
2966
+ "normalized": true,
2967
+ "rstrip": false,
2968
+ "single_word": false,
2969
+ "special": false
2970
+ },
2971
+ "92906": {
2972
+ "content": "<|234|>",
2973
+ "lstrip": false,
2974
+ "normalized": true,
2975
+ "rstrip": false,
2976
+ "single_word": false,
2977
+ "special": false
2978
+ },
2979
+ "92907": {
2980
+ "content": "<|235|>",
2981
+ "lstrip": false,
2982
+ "normalized": true,
2983
+ "rstrip": false,
2984
+ "single_word": false,
2985
+ "special": false
2986
+ },
2987
+ "92908": {
2988
+ "content": "<|236|>",
2989
+ "lstrip": false,
2990
+ "normalized": true,
2991
+ "rstrip": false,
2992
+ "single_word": false,
2993
+ "special": false
2994
+ },
2995
+ "92909": {
2996
+ "content": "<|237|>",
2997
+ "lstrip": false,
2998
+ "normalized": true,
2999
+ "rstrip": false,
3000
+ "single_word": false,
3001
+ "special": false
3002
+ },
3003
+ "92910": {
3004
+ "content": "<|238|>",
3005
+ "lstrip": false,
3006
+ "normalized": true,
3007
+ "rstrip": false,
3008
+ "single_word": false,
3009
+ "special": false
3010
+ },
3011
+ "92911": {
3012
+ "content": "<|239|>",
3013
+ "lstrip": false,
3014
+ "normalized": true,
3015
+ "rstrip": false,
3016
+ "single_word": false,
3017
+ "special": false
3018
+ },
3019
+ "92912": {
3020
+ "content": "<|240|>",
3021
+ "lstrip": false,
3022
+ "normalized": true,
3023
+ "rstrip": false,
3024
+ "single_word": false,
3025
+ "special": false
3026
+ },
3027
+ "92913": {
3028
+ "content": "<|241|>",
3029
+ "lstrip": false,
3030
+ "normalized": true,
3031
+ "rstrip": false,
3032
+ "single_word": false,
3033
+ "special": false
3034
+ },
3035
+ "92914": {
3036
+ "content": "<|242|>",
3037
+ "lstrip": false,
3038
+ "normalized": true,
3039
+ "rstrip": false,
3040
+ "single_word": false,
3041
+ "special": false
3042
+ },
3043
+ "92915": {
3044
+ "content": "<|243|>",
3045
+ "lstrip": false,
3046
+ "normalized": true,
3047
+ "rstrip": false,
3048
+ "single_word": false,
3049
+ "special": false
3050
+ },
3051
+ "92916": {
3052
+ "content": "<|244|>",
3053
+ "lstrip": false,
3054
+ "normalized": true,
3055
+ "rstrip": false,
3056
+ "single_word": false,
3057
+ "special": false
3058
+ },
3059
+ "92917": {
3060
+ "content": "<|245|>",
3061
+ "lstrip": false,
3062
+ "normalized": true,
3063
+ "rstrip": false,
3064
+ "single_word": false,
3065
+ "special": false
3066
+ },
3067
+ "92918": {
3068
+ "content": "<|246|>",
3069
+ "lstrip": false,
3070
+ "normalized": true,
3071
+ "rstrip": false,
3072
+ "single_word": false,
3073
+ "special": false
3074
+ },
3075
+ "92919": {
3076
+ "content": "<|247|>",
3077
+ "lstrip": false,
3078
+ "normalized": true,
3079
+ "rstrip": false,
3080
+ "single_word": false,
3081
+ "special": false
3082
+ },
3083
+ "92920": {
3084
+ "content": "<|248|>",
3085
+ "lstrip": false,
3086
+ "normalized": true,
3087
+ "rstrip": false,
3088
+ "single_word": false,
3089
+ "special": false
3090
+ },
3091
+ "92921": {
3092
+ "content": "<|249|>",
3093
+ "lstrip": false,
3094
+ "normalized": true,
3095
+ "rstrip": false,
3096
+ "single_word": false,
3097
+ "special": false
3098
+ },
3099
+ "92922": {
3100
+ "content": "<|250|>",
3101
+ "lstrip": false,
3102
+ "normalized": true,
3103
+ "rstrip": false,
3104
+ "single_word": false,
3105
+ "special": false
3106
+ },
3107
+ "92923": {
3108
+ "content": "<|251|>",
3109
+ "lstrip": false,
3110
+ "normalized": true,
3111
+ "rstrip": false,
3112
+ "single_word": false,
3113
+ "special": false
3114
+ },
3115
+ "92924": {
3116
+ "content": "<|252|>",
3117
+ "lstrip": false,
3118
+ "normalized": true,
3119
+ "rstrip": false,
3120
+ "single_word": false,
3121
+ "special": false
3122
+ },
3123
+ "92925": {
3124
+ "content": "<|253|>",
3125
+ "lstrip": false,
3126
+ "normalized": true,
3127
+ "rstrip": false,
3128
+ "single_word": false,
3129
+ "special": false
3130
+ },
3131
+ "92926": {
3132
+ "content": "<|254|>",
3133
+ "lstrip": false,
3134
+ "normalized": true,
3135
+ "rstrip": false,
3136
+ "single_word": false,
3137
+ "special": false
3138
+ },
3139
+ "92927": {
3140
+ "content": "<|255|>",
3141
+ "lstrip": false,
3142
+ "normalized": true,
3143
+ "rstrip": false,
3144
+ "single_word": false,
3145
+ "special": false
3146
+ },
3147
+ "92928": {
3148
+ "content": "<|256|>",
3149
+ "lstrip": false,
3150
+ "normalized": true,
3151
+ "rstrip": false,
3152
+ "single_word": false,
3153
+ "special": false
3154
  }
3155
  },
3156
  "additional_special_tokens": [