Hugging Face's logo

helenai
/
allam-7b-instruct-ov-int4-sym

OpenVINO
llama
Model card Files Community
helenai commited on
Commit
d8e0b5b
·
1 Parent(s): 3a7dc79

Update model

Browse files
Files changed (7) hide show
  1. openvino_config.json +1 -1
  2. openvino_detokenizer.bin +2 -2
  3. openvino_detokenizer.xml +130 -65
  4. openvino_model.bin +2 -2
  5. openvino_model.xml +0 -0
  6. openvino_tokenizer.bin +2 -2
  7. openvino_tokenizer.xml +180 -164
openvino_config.json CHANGED
@@ -9,7 +9,7 @@
9
  "dataset": "auto",
10
  "dtype": "int4",
11
  "gptq": null,
12
- "group_size": 128,
13
  "ignored_scope": null,
14
  "lora_correction": null,
15
  "num_samples": null,
 
9
  "dataset": "auto",
10
  "dtype": "int4",
11
  "gptq": null,
12
+ "group_size": -1,
13
  "ignored_scope": null,
14
  "lora_correction": null,
15
  "num_samples": null,
openvino_detokenizer.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:427e98d5f122128c2462e75ce2bc68ce16a1623b08c59efc12c7d51da8e9dd3e
3
- size 1151430
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9daf33716e93cb85e42070664c9f6661922124ca3c81892acf9a119a41265ac9
3
+ size 895458
openvino_detokenizer.xml CHANGED
@@ -1,16 +1,16 @@
1
  <?xml version="1.0"?>
2
  <net name="detokenizer" version="11">
3
  <layers>
4
- <layer id="0" name="Parameter_1339275" type="Parameter" version="opset1">
5
  <data shape="?,?" element_type="i64" />
6
  <output>
7
- <port id="0" precision="I64" names="Parameter_1339275">
8
  <dim>-1</dim>
9
  <dim>-1</dim>
10
  </port>
11
  </output>
12
  </layer>
13
- <layer id="1" name="Convert_1339367" type="Convert" version="opset1">
14
  <data destination_type="i32" />
15
  <input>
16
  <port id="0" precision="I64">
@@ -25,39 +25,98 @@
25
  </port>
26
  </output>
27
  </layer>
28
- <layer id="2" name="Constant_1339233" type="Const" version="opset1">
29
- <data element_type="i32" shape="64000" offset="0" size="256000" />
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
30
  <output>
31
  <port id="0" precision="I32">
32
- <dim>64000</dim>
33
  </port>
34
  </output>
35
  </layer>
36
- <layer id="3" name="Constant_1339235" type="Const" version="opset1">
37
- <data element_type="i32" shape="64000" offset="256000" size="256000" />
38
  <output>
39
  <port id="0" precision="I32">
40
- <dim>64000</dim>
41
  </port>
42
  </output>
43
  </layer>
44
- <layer id="4" name="Constant_1339237" type="Const" version="opset1">
45
- <data element_type="u8" shape="639411" offset="512000" size="639411" />
46
  <output>
47
- <port id="0" precision="U8">
48
- <dim>639411</dim>
49
  </port>
50
  </output>
51
  </layer>
52
- <layer id="5" name="Slice_1339280" type="Const" version="opset1">
53
- <data element_type="i32" shape="3" offset="1151411" size="12" />
54
  <output>
55
  <port id="0" precision="I32">
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
56
  <dim>3</dim>
57
  </port>
58
  </output>
59
  </layer>
60
- <layer id="6" name="VocabDecoder_1339282" type="VocabDecoder" version="extension">
61
  <data skip_tokens="" />
62
  <input>
63
  <port id="0" precision="I32">
@@ -65,13 +124,13 @@
65
  <dim>-1</dim>
66
  </port>
67
  <port id="1" precision="I32">
68
- <dim>64000</dim>
69
  </port>
70
  <port id="2" precision="I32">
71
- <dim>64000</dim>
72
  </port>
73
  <port id="3" precision="U8">
74
- <dim>639411</dim>
75
  </port>
76
  <port id="4" precision="I32">
77
  <dim>3</dim>
@@ -95,23 +154,23 @@
95
  </port>
96
  </output>
97
  </layer>
98
- <layer id="7" name="Constant_1339284" type="Const" version="opset1">
99
- <data element_type="u8" shape="3" offset="1151423" size="3" />
100
  <output>
101
  <port id="0" precision="U8">
102
  <dim>3</dim>
103
  </port>
104
  </output>
105
  </layer>
106
- <layer id="8" name="Constant_1339286" type="Const" version="opset1">
107
- <data element_type="u8" shape="1" offset="1151426" size="1" />
108
  <output>
109
  <port id="0" precision="U8">
110
  <dim>1</dim>
111
  </port>
112
  </output>
113
  </layer>
114
- <layer id="9" name="RegexNormalization_1339287" type="RegexNormalization" version="extension">
115
  <data global_replace="true" />
116
  <input>
117
  <port id="0" precision="I32">
@@ -142,7 +201,7 @@
142
  </port>
143
  </output>
144
  </layer>
145
- <layer id="10" name="ByteFallback_1339288" type="ByteFallback" version="extension">
146
  <input>
147
  <port id="0" precision="I32">
148
  <dim>-1</dim>
@@ -166,7 +225,7 @@
166
  </port>
167
  </output>
168
  </layer>
169
- <layer id="11" name="FuzeRagged_1339289" type="FuzeRagged" version="extension">
170
  <input>
171
  <port id="0" precision="I32">
172
  <dim>-1</dim>
@@ -190,23 +249,23 @@
190
  </port>
191
  </output>
192
  </layer>
193
- <layer id="12" name="Constant_1339291" type="Const" version="opset1">
194
- <data element_type="u8" shape="2" offset="1151427" size="2" />
195
  <output>
196
  <port id="0" precision="U8">
197
  <dim>2</dim>
198
  </port>
199
  </output>
200
  </layer>
201
- <layer id="13" name="Constant_1339293" type="Const" version="opset1">
202
- <data element_type="u8" shape="0" offset="1151429" size="1" />
203
  <output>
204
  <port id="0" precision="U8">
205
  <dim>0</dim>
206
  </port>
207
  </output>
208
  </layer>
209
- <layer id="14" name="RegexNormalization_1339294" type="RegexNormalization" version="extension">
210
  <data global_replace="true" />
211
  <input>
212
  <port id="0" precision="I32">
@@ -237,7 +296,7 @@
237
  </port>
238
  </output>
239
  </layer>
240
- <layer id="15" name="UTF8Validate_1339295" type="UTF8Validate" version="extension">
241
  <data replace_mode="true" />
242
  <input>
243
  <port id="0" precision="I32">
@@ -262,7 +321,8 @@
262
  </port>
263
  </output>
264
  </layer>
265
- <layer id="16" name="StringTensorPack_1339296" type="StringTensorPack" version="opset15">
 
266
  <input>
267
  <port id="0" precision="I32">
268
  <dim>-1</dim>
@@ -275,12 +335,12 @@
275
  </port>
276
  </input>
277
  <output>
278
- <port id="3" precision="STRING" names="Result_1339297,string_output">
279
  <dim>-1</dim>
280
  </port>
281
  </output>
282
  </layer>
283
- <layer id="17" name="Result_1339297" type="Result" version="opset1" output_names="Result_1339297,string_output">
284
  <input>
285
  <port id="0" precision="STRING">
286
  <dim>-1</dim>
@@ -290,35 +350,41 @@
290
  </layers>
291
  <edges>
292
  <edge from-layer="0" from-port="0" to-layer="1" to-port="0" />
293
- <edge from-layer="1" from-port="1" to-layer="6" to-port="0" />
294
- <edge from-layer="2" from-port="0" to-layer="6" to-port="1" />
295
- <edge from-layer="3" from-port="0" to-layer="6" to-port="2" />
296
- <edge from-layer="4" from-port="0" to-layer="6" to-port="3" />
297
- <edge from-layer="5" from-port="0" to-layer="6" to-port="4" />
298
- <edge from-layer="6" from-port="6" to-layer="11" to-port="1" />
299
- <edge from-layer="6" from-port="5" to-layer="11" to-port="0" />
300
- <edge from-layer="6" from-port="8" to-layer="9" to-port="1" />
301
- <edge from-layer="6" from-port="7" to-layer="9" to-port="0" />
302
- <edge from-layer="6" from-port="9" to-layer="9" to-port="2" />
303
  <edge from-layer="7" from-port="0" to-layer="9" to-port="3" />
304
  <edge from-layer="8" from-port="0" to-layer="9" to-port="4" />
305
- <edge from-layer="9" from-port="5" to-layer="10" to-port="0" />
306
- <edge from-layer="9" from-port="6" to-layer="10" to-port="1" />
307
- <edge from-layer="9" from-port="7" to-layer="10" to-port="2" />
308
- <edge from-layer="10" from-port="3" to-layer="11" to-port="2" />
309
- <edge from-layer="10" from-port="4" to-layer="11" to-port="3" />
310
- <edge from-layer="10" from-port="5" to-layer="14" to-port="2" />
311
- <edge from-layer="11" from-port="5" to-layer="14" to-port="1" />
312
- <edge from-layer="11" from-port="4" to-layer="14" to-port="0" />
313
- <edge from-layer="12" from-port="0" to-layer="14" to-port="3" />
314
- <edge from-layer="13" from-port="0" to-layer="14" to-port="4" />
315
- <edge from-layer="14" from-port="5" to-layer="15" to-port="0" />
316
- <edge from-layer="14" from-port="6" to-layer="15" to-port="1" />
317
- <edge from-layer="14" from-port="7" to-layer="15" to-port="2" />
318
- <edge from-layer="15" from-port="3" to-layer="16" to-port="0" />
319
- <edge from-layer="15" from-port="4" to-layer="16" to-port="1" />
320
- <edge from-layer="15" from-port="5" to-layer="16" to-port="2" />
321
- <edge from-layer="16" from-port="3" to-layer="17" to-port="0" />
 
 
 
 
 
 
 
 
322
  </edges>
323
  <rt_info>
324
  <add_attention_mask value="True" />
@@ -330,10 +396,9 @@
330
  <detokenizer_input_type value="i64" />
331
  <eos_token_id value="2" />
332
  <handle_special_tokens_with_re />
333
- <max_length />
334
  <number_of_inputs value="1" />
335
- <openvino_tokenizers_version value="2025.1.0.0-513-7881453f285" />
336
- <openvino_version value="2025.1.0-18378-fcb9c9d5d6d" />
337
  <original_tokenizer_class value="&lt;class 'transformers.models.llama.tokenization_llama_fast.LlamaTokenizerFast'>" />
338
  <sentencepiece_version value="0.2.0" />
339
  <skip_special_tokens value="True" />
 
1
  <?xml version="1.0"?>
2
  <net name="detokenizer" version="11">
3
  <layers>
4
+ <layer id="0" name="Parameter_1310401" type="Parameter" version="opset1">
5
  <data shape="?,?" element_type="i64" />
6
  <output>
7
+ <port id="0" precision="I64" names="Parameter_1310401">
8
  <dim>-1</dim>
9
  <dim>-1</dim>
10
  </port>
11
  </output>
12
  </layer>
13
+ <layer id="1" name="Convert_1310428" type="Convert" version="opset1">
14
  <data destination_type="i32" />
15
  <input>
16
  <port id="0" precision="I64">
 
25
  </port>
26
  </output>
27
  </layer>
28
+ <layer id="2" name="Constant_1310371" type="Const" version="opset1">
29
+ <data element_type="u8" shape="895419" offset="0" size="895419" />
30
+ <output>
31
+ <port id="0" precision="U8">
32
+ <dim>895419</dim>
33
+ </port>
34
+ </output>
35
+ </layer>
36
+ <layer id="3" name="StringTensorUnpack_1310372" type="StringTensorUnpack" version="extension">
37
+ <data mode="begins_ends" />
38
+ <input>
39
+ <port id="0" precision="U8">
40
+ <dim>895419</dim>
41
+ </port>
42
+ </input>
43
+ <output>
44
+ <port id="1" precision="I32">
45
+ <dim>-1</dim>
46
+ </port>
47
+ <port id="2" precision="I32">
48
+ <dim>-1</dim>
49
+ </port>
50
+ <port id="3" precision="U8">
51
+ <dim>-1</dim>
52
+ </port>
53
+ </output>
54
+ </layer>
55
+ <layer id="4" name="Constant_1310405" type="Const" version="opset1">
56
+ <data element_type="i32" shape="3" offset="895419" size="12" />
57
  <output>
58
  <port id="0" precision="I32">
59
+ <dim>3</dim>
60
  </port>
61
  </output>
62
  </layer>
63
+ <layer id="5" name="Constant_1310403" type="Const" version="opset1">
64
+ <data element_type="i32" shape="1" offset="895431" size="4" />
65
  <output>
66
  <port id="0" precision="I32">
67
+ <dim>1</dim>
68
  </port>
69
  </output>
70
  </layer>
71
+ <layer id="6" name="Constant_1310402" type="Const" version="opset1">
72
+ <data element_type="i32" shape="1" offset="895435" size="4" />
73
  <output>
74
+ <port id="0" precision="I32">
75
+ <dim>1</dim>
76
  </port>
77
  </output>
78
  </layer>
79
+ <layer id="7" name="Constant_1310404" type="Const" version="opset1">
80
+ <data element_type="i32" shape="1" offset="895439" size="4" />
81
  <output>
82
  <port id="0" precision="I32">
83
+ <dim>1</dim>
84
+ </port>
85
+ </output>
86
+ </layer>
87
+ <layer id="8" name="Constant_1310407" type="Const" version="opset1">
88
+ <data element_type="i64" shape="1" offset="895443" size="8" />
89
+ <output>
90
+ <port id="0" precision="I64">
91
+ <dim>1</dim>
92
+ </port>
93
+ </output>
94
+ </layer>
95
+ <layer id="9" name="Slice_1310406" type="Slice" version="opset8">
96
+ <input>
97
+ <port id="0" precision="I32">
98
+ <dim>3</dim>
99
+ </port>
100
+ <port id="1" precision="I32">
101
+ <dim>1</dim>
102
+ </port>
103
+ <port id="2" precision="I32">
104
+ <dim>1</dim>
105
+ </port>
106
+ <port id="3" precision="I32">
107
+ <dim>1</dim>
108
+ </port>
109
+ <port id="4" precision="I64">
110
+ <dim>1</dim>
111
+ </port>
112
+ </input>
113
+ <output>
114
+ <port id="5" precision="I32">
115
  <dim>3</dim>
116
  </port>
117
  </output>
118
  </layer>
119
+ <layer id="10" name="VocabDecoder_1310408" type="VocabDecoder" version="extension">
120
  <data skip_tokens="" />
121
  <input>
122
  <port id="0" precision="I32">
 
124
  <dim>-1</dim>
125
  </port>
126
  <port id="1" precision="I32">
127
+ <dim>-1</dim>
128
  </port>
129
  <port id="2" precision="I32">
130
+ <dim>-1</dim>
131
  </port>
132
  <port id="3" precision="U8">
133
+ <dim>-1</dim>
134
  </port>
135
  <port id="4" precision="I32">
136
  <dim>3</dim>
 
154
  </port>
155
  </output>
156
  </layer>
157
+ <layer id="11" name="Constant_1310410" type="Const" version="opset1">
158
+ <data element_type="u8" shape="3" offset="895451" size="3" />
159
  <output>
160
  <port id="0" precision="U8">
161
  <dim>3</dim>
162
  </port>
163
  </output>
164
  </layer>
165
+ <layer id="12" name="Constant_1310412" type="Const" version="opset1">
166
+ <data element_type="u8" shape="1" offset="895454" size="1" />
167
  <output>
168
  <port id="0" precision="U8">
169
  <dim>1</dim>
170
  </port>
171
  </output>
172
  </layer>
173
+ <layer id="13" name="RegexNormalization_1310413" type="RegexNormalization" version="extension">
174
  <data global_replace="true" />
175
  <input>
176
  <port id="0" precision="I32">
 
201
  </port>
202
  </output>
203
  </layer>
204
+ <layer id="14" name="ByteFallback_1310414" type="ByteFallback" version="extension">
205
  <input>
206
  <port id="0" precision="I32">
207
  <dim>-1</dim>
 
225
  </port>
226
  </output>
227
  </layer>
228
+ <layer id="15" name="FuzeRagged_1310415" type="FuzeRagged" version="extension">
229
  <input>
230
  <port id="0" precision="I32">
231
  <dim>-1</dim>
 
249
  </port>
250
  </output>
251
  </layer>
252
+ <layer id="16" name="Constant_1310417" type="Const" version="opset1">
253
+ <data element_type="u8" shape="2" offset="895455" size="2" />
254
  <output>
255
  <port id="0" precision="U8">
256
  <dim>2</dim>
257
  </port>
258
  </output>
259
  </layer>
260
+ <layer id="17" name="Constant_1310419" type="Const" version="opset1">
261
+ <data element_type="u8" shape="0" offset="895457" size="1" />
262
  <output>
263
  <port id="0" precision="U8">
264
  <dim>0</dim>
265
  </port>
266
  </output>
267
  </layer>
268
+ <layer id="18" name="RegexNormalization_1310420" type="RegexNormalization" version="extension">
269
  <data global_replace="true" />
270
  <input>
271
  <port id="0" precision="I32">
 
296
  </port>
297
  </output>
298
  </layer>
299
+ <layer id="19" name="UTF8Validate_1310421" type="UTF8Validate" version="extension">
300
  <data replace_mode="true" />
301
  <input>
302
  <port id="0" precision="I32">
 
321
  </port>
322
  </output>
323
  </layer>
324
+ <layer id="20" name="StringTensorPack_1310422" type="StringTensorPack" version="extension">
325
+ <data mode="begins_ends" />
326
  <input>
327
  <port id="0" precision="I32">
328
  <dim>-1</dim>
 
335
  </port>
336
  </input>
337
  <output>
338
+ <port id="3" precision="STRING" names="string_output">
339
  <dim>-1</dim>
340
  </port>
341
  </output>
342
  </layer>
343
+ <layer id="21" name="Result_1310423" type="Result" version="opset1">
344
  <input>
345
  <port id="0" precision="STRING">
346
  <dim>-1</dim>
 
350
  </layers>
351
  <edges>
352
  <edge from-layer="0" from-port="0" to-layer="1" to-port="0" />
353
+ <edge from-layer="1" from-port="1" to-layer="10" to-port="0" />
354
+ <edge from-layer="2" from-port="0" to-layer="3" to-port="0" />
355
+ <edge from-layer="3" from-port="3" to-layer="10" to-port="3" />
356
+ <edge from-layer="3" from-port="2" to-layer="10" to-port="2" />
357
+ <edge from-layer="3" from-port="1" to-layer="10" to-port="1" />
358
+ <edge from-layer="4" from-port="0" to-layer="9" to-port="0" />
359
+ <edge from-layer="5" from-port="0" to-layer="9" to-port="1" />
360
+ <edge from-layer="6" from-port="0" to-layer="9" to-port="2" />
 
 
361
  <edge from-layer="7" from-port="0" to-layer="9" to-port="3" />
362
  <edge from-layer="8" from-port="0" to-layer="9" to-port="4" />
363
+ <edge from-layer="9" from-port="5" to-layer="10" to-port="4" />
364
+ <edge from-layer="10" from-port="7" to-layer="13" to-port="0" />
365
+ <edge from-layer="10" from-port="8" to-layer="13" to-port="1" />
366
+ <edge from-layer="10" from-port="9" to-layer="13" to-port="2" />
367
+ <edge from-layer="10" from-port="6" to-layer="15" to-port="1" />
368
+ <edge from-layer="10" from-port="5" to-layer="15" to-port="0" />
369
+ <edge from-layer="11" from-port="0" to-layer="13" to-port="3" />
370
+ <edge from-layer="12" from-port="0" to-layer="13" to-port="4" />
371
+ <edge from-layer="13" from-port="6" to-layer="14" to-port="1" />
372
+ <edge from-layer="13" from-port="7" to-layer="14" to-port="2" />
373
+ <edge from-layer="13" from-port="5" to-layer="14" to-port="0" />
374
+ <edge from-layer="14" from-port="3" to-layer="15" to-port="2" />
375
+ <edge from-layer="14" from-port="4" to-layer="15" to-port="3" />
376
+ <edge from-layer="14" from-port="5" to-layer="18" to-port="2" />
377
+ <edge from-layer="15" from-port="4" to-layer="18" to-port="0" />
378
+ <edge from-layer="15" from-port="5" to-layer="18" to-port="1" />
379
+ <edge from-layer="16" from-port="0" to-layer="18" to-port="3" />
380
+ <edge from-layer="17" from-port="0" to-layer="18" to-port="4" />
381
+ <edge from-layer="18" from-port="5" to-layer="19" to-port="0" />
382
+ <edge from-layer="18" from-port="6" to-layer="19" to-port="1" />
383
+ <edge from-layer="18" from-port="7" to-layer="19" to-port="2" />
384
+ <edge from-layer="19" from-port="3" to-layer="20" to-port="0" />
385
+ <edge from-layer="19" from-port="4" to-layer="20" to-port="1" />
386
+ <edge from-layer="19" from-port="5" to-layer="20" to-port="2" />
387
+ <edge from-layer="20" from-port="3" to-layer="21" to-port="0" />
388
  </edges>
389
  <rt_info>
390
  <add_attention_mask value="True" />
 
396
  <detokenizer_input_type value="i64" />
397
  <eos_token_id value="2" />
398
  <handle_special_tokens_with_re />
 
399
  <number_of_inputs value="1" />
400
+ <openvino_tokenizers_version value="2025.0.0.0" />
401
+ <openvino_version value="2025.0.0" />
402
  <original_tokenizer_class value="&lt;class 'transformers.models.llama.tokenization_llama_fast.LlamaTokenizerFast'>" />
403
  <sentencepiece_version value="0.2.0" />
404
  <skip_special_tokens value="True" />
openvino_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b249dcb93f112be9e63ea0f608a8391df44cd880e24ec5c4bb731a963dc6115e
3
- size 3864927748
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1559693fc642bface3d0c1cf79f301304f99bb88915d3e502a44736e8b371750
3
+ size 3766459860
openvino_model.xml CHANGED
The diff for this file is too large to render. See raw diff
 
openvino_tokenizer.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e514763f2e2d8519065bd3ae90b5586617fe2dd6bee8424902848da2c404a5eb
3
- size 4689273
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f9dbb875f667e8e4f3259d6328c0712653fa0e843ae61b19a44519715252aeb5
3
+ size 3350481
openvino_tokenizer.xml CHANGED
@@ -1,21 +1,22 @@
1
  <?xml version="1.0"?>
2
  <net name="tokenizer" version="11">
3
  <layers>
4
- <layer id="0" name="Parameter_1339145" type="Parameter" version="opset1">
5
  <data shape="?" element_type="string" />
6
  <output>
7
- <port id="0" precision="STRING" names="Parameter_1339145">
8
  <dim>-1</dim>
9
  </port>
10
  </output>
11
  </layer>
12
- <layer id="1" name="Constant_1339151" type="Const" version="opset1">
13
  <data element_type="i64" shape="" offset="0" size="8" />
14
  <output>
15
  <port id="0" precision="I64" />
16
  </output>
17
  </layer>
18
- <layer id="2" name="StringTensorUnpack_1339146" type="StringTensorUnpack" version="opset15">
 
19
  <input>
20
  <port id="0" precision="STRING">
21
  <dim>-1</dim>
@@ -33,7 +34,7 @@
33
  </port>
34
  </output>
35
  </layer>
36
- <layer id="3" name="ShapeOf_1339147" type="ShapeOf" version="opset3">
37
  <data output_type="i64" />
38
  <input>
39
  <port id="0" precision="I32">
@@ -46,19 +47,19 @@
46
  </port>
47
  </output>
48
  </layer>
49
- <layer id="4" name="Constant_1339148" type="Const" version="opset1">
50
  <data element_type="i64" shape="" offset="0" size="8" />
51
  <output>
52
  <port id="0" precision="I64" />
53
  </output>
54
  </layer>
55
- <layer id="5" name="Constant_1339149" type="Const" version="opset1">
56
  <data element_type="i64" shape="" offset="0" size="8" />
57
  <output>
58
  <port id="0" precision="I64" />
59
  </output>
60
  </layer>
61
- <layer id="6" name="Gather_1339150" type="Gather" version="opset8">
62
  <data batch_dims="0" />
63
  <input>
64
  <port id="0" precision="I64">
@@ -71,13 +72,13 @@
71
  <port id="3" precision="I64" />
72
  </output>
73
  </layer>
74
- <layer id="7" name="Constant_1339152" type="Const" version="opset1">
75
  <data element_type="i64" shape="" offset="8" size="8" />
76
  <output>
77
  <port id="0" precision="I64" />
78
  </output>
79
  </layer>
80
- <layer id="8" name="Range_1339153" type="Range" version="opset4">
81
  <data output_type="i32" />
82
  <input>
83
  <port id="0" precision="I64" />
@@ -90,19 +91,19 @@
90
  </port>
91
  </output>
92
  </layer>
93
- <layer id="9" name="Constant_1339154" type="Const" version="opset1">
94
  <data element_type="i64" shape="" offset="8" size="8" />
95
  <output>
96
  <port id="0" precision="I64" />
97
  </output>
98
  </layer>
99
- <layer id="10" name="Constant_1339155" type="Const" version="opset1">
100
  <data element_type="i64" shape="" offset="8" size="8" />
101
  <output>
102
  <port id="0" precision="I64" />
103
  </output>
104
  </layer>
105
- <layer id="11" name="Add_1339156" type="Add" version="opset1">
106
  <data auto_broadcast="numpy" />
107
  <input>
108
  <port id="0" precision="I64" />
@@ -112,13 +113,13 @@
112
  <port id="2" precision="I64" />
113
  </output>
114
  </layer>
115
- <layer id="12" name="Constant_1339157" type="Const" version="opset1">
116
  <data element_type="i64" shape="" offset="8" size="8" />
117
  <output>
118
  <port id="0" precision="I64" />
119
  </output>
120
  </layer>
121
- <layer id="13" name="Range_1339158" type="Range" version="opset4">
122
  <data output_type="i32" />
123
  <input>
124
  <port id="0" precision="I64" />
@@ -131,7 +132,7 @@
131
  </port>
132
  </output>
133
  </layer>
134
- <layer id="14" name="Constant_1339220" type="Const" version="opset1">
135
  <data element_type="u8" shape="117" offset="16" size="117" />
136
  <output>
137
  <port id="0" precision="U8">
@@ -139,7 +140,7 @@
139
  </port>
140
  </output>
141
  </layer>
142
- <layer id="15" name="SpecialTokensSplit_1339221" type="SpecialTokensSplit" version="extension">
143
  <input>
144
  <port id="0" precision="I32">
145
  <dim>-1</dim>
@@ -181,7 +182,7 @@
181
  </port>
182
  </output>
183
  </layer>
184
- <layer id="16" name="Constant_1339223" type="Const" version="opset1">
185
  <data element_type="u8" shape="1" offset="133" size="1" />
186
  <output>
187
  <port id="0" precision="U8">
@@ -189,7 +190,7 @@
189
  </port>
190
  </output>
191
  </layer>
192
- <layer id="17" name="Constant_1339225" type="Const" version="opset1">
193
  <data element_type="u8" shape="3" offset="134" size="3" />
194
  <output>
195
  <port id="0" precision="U8">
@@ -197,7 +198,7 @@
197
  </port>
198
  </output>
199
  </layer>
200
- <layer id="18" name="RegexNormalization_1339226" type="RegexNormalization" version="extension">
201
  <data global_replace="true" />
202
  <input>
203
  <port id="0" precision="I32">
@@ -234,7 +235,7 @@
234
  </port>
235
  </output>
236
  </layer>
237
- <layer id="19" name="Constant_1339228" type="Const" version="opset1">
238
  <data element_type="u8" shape="11" offset="137" size="11" />
239
  <output>
240
  <port id="0" precision="U8">
@@ -242,7 +243,7 @@
242
  </port>
243
  </output>
244
  </layer>
245
- <layer id="20" name="Constant_1339230" type="Const" version="opset1">
246
  <data element_type="u8" shape="5" offset="148" size="5" />
247
  <output>
248
  <port id="0" precision="U8">
@@ -250,7 +251,7 @@
250
  </port>
251
  </output>
252
  </layer>
253
- <layer id="21" name="RegexNormalization_1339231" type="RegexNormalization" version="extension">
254
  <data global_replace="true" />
255
  <input>
256
  <port id="0" precision="I32">
@@ -287,111 +288,123 @@
287
  </port>
288
  </output>
289
  </layer>
290
- <layer id="22" name="Constant_1339233" type="Const" version="opset1">
291
- <data element_type="i32" shape="64000" offset="153" size="256000" />
292
  <output>
293
- <port id="0" precision="I32">
294
- <dim>64000</dim>
295
  </port>
296
  </output>
297
  </layer>
298
- <layer id="23" name="Constant_1339235" type="Const" version="opset1">
299
- <data element_type="i32" shape="64000" offset="256153" size="256000" />
 
 
 
 
 
300
  <output>
301
- <port id="0" precision="I32">
302
- <dim>64000</dim>
 
 
 
 
 
 
303
  </port>
304
  </output>
305
  </layer>
306
- <layer id="24" name="Constant_1339237" type="Const" version="opset1">
307
- <data element_type="u8" shape="639411" offset="512153" size="639411" />
308
  <output>
309
  <port id="0" precision="U8">
310
- <dim>639411</dim>
311
  </port>
312
  </output>
313
  </layer>
314
- <layer id="25" name="Constant_1339245" type="Const" version="opset1">
315
- <data element_type="i32" shape="135342" offset="1151564" size="541368" />
316
- <output>
317
- <port id="0" precision="I32">
318
- <dim>135342</dim>
319
  </port>
320
- </output>
321
- </layer>
322
- <layer id="26" name="Constant_1339247" type="Const" version="opset1">
323
- <data element_type="i32" shape="135342" offset="1692932" size="541368" />
324
  <output>
325
- <port id="0" precision="I32">
326
- <dim>135342</dim>
 
 
 
 
 
 
327
  </port>
328
  </output>
329
  </layer>
330
- <layer id="27" name="Constant_1339249" type="Const" version="opset1">
331
- <data element_type="u8" shape="852341" offset="2234300" size="852341" />
332
  <output>
333
  <port id="0" precision="U8">
334
- <dim>852341</dim>
335
  </port>
336
  </output>
337
  </layer>
338
- <layer id="28" name="Constant_1339251" type="Const" version="opset1">
339
- <data element_type="i32" shape="135342" offset="3086641" size="541368" />
340
- <output>
341
- <port id="0" precision="I32">
342
- <dim>135342</dim>
343
  </port>
344
- </output>
345
- </layer>
346
- <layer id="29" name="Constant_1339253" type="Const" version="opset1">
347
- <data element_type="i32" shape="135342" offset="3628009" size="541368" />
348
  <output>
349
- <port id="0" precision="I32">
350
- <dim>135342</dim>
 
 
 
 
 
 
351
  </port>
352
  </output>
353
  </layer>
354
- <layer id="30" name="Constant_1339255" type="Const" version="opset1">
355
- <data element_type="u8" shape="519587" offset="4169377" size="519587" />
356
  <output>
357
  <port id="0" precision="U8">
358
- <dim>519587</dim>
359
  </port>
360
  </output>
361
  </layer>
362
- <layer id="31" name="Constant_1339239" type="Const" version="opset1">
363
- <data element_type="i32" shape="22" offset="4688964" size="88" />
364
- <output>
365
- <port id="0" precision="I32">
366
- <dim>22</dim>
367
  </port>
368
- </output>
369
- </layer>
370
- <layer id="32" name="Constant_1339241" type="Const" version="opset1">
371
- <data element_type="i32" shape="22" offset="4689052" size="88" />
372
  <output>
373
- <port id="0" precision="I32">
374
- <dim>22</dim>
375
  </port>
376
- </output>
377
- </layer>
378
- <layer id="33" name="Constant_1339243" type="Const" version="opset1">
379
- <data element_type="u8" shape="37" offset="4689140" size="37" />
380
- <output>
381
- <port id="0" precision="U8">
382
- <dim>37</dim>
383
  </port>
384
  </output>
385
  </layer>
386
- <layer id="34" name="Constant_1339256" type="Const" version="opset1">
387
- <data element_type="i32" shape="22" offset="4689177" size="88" />
388
  <output>
389
  <port id="0" precision="I32">
390
  <dim>22</dim>
391
  </port>
392
  </output>
393
  </layer>
394
- <layer id="35" name="BPETokenizer_1339257" type="BPETokenizer" version="extension">
395
  <data unk_token="&lt;unk>" fuse_unk="true" suffix_indicator="" end_suffix="" byte_fallback="true" cache_capacity="20000" />
396
  <input>
397
  <port id="0" precision="I32">
@@ -410,40 +423,40 @@
410
  <dim>-1</dim>
411
  </port>
412
  <port id="5" precision="I32">
413
- <dim>64000</dim>
414
  </port>
415
  <port id="6" precision="I32">
416
- <dim>64000</dim>
417
  </port>
418
  <port id="7" precision="U8">
419
- <dim>639411</dim>
420
  </port>
421
  <port id="8" precision="I32">
422
- <dim>135342</dim>
423
  </port>
424
  <port id="9" precision="I32">
425
- <dim>135342</dim>
426
  </port>
427
  <port id="10" precision="U8">
428
- <dim>852341</dim>
429
  </port>
430
  <port id="11" precision="I32">
431
- <dim>135342</dim>
432
  </port>
433
  <port id="12" precision="I32">
434
- <dim>135342</dim>
435
  </port>
436
  <port id="13" precision="U8">
437
- <dim>519587</dim>
438
  </port>
439
  <port id="14" precision="I32">
440
- <dim>22</dim>
441
  </port>
442
  <port id="15" precision="I32">
443
- <dim>22</dim>
444
  </port>
445
  <port id="16" precision="U8">
446
- <dim>37</dim>
447
  </port>
448
  <port id="17" precision="I32">
449
  <dim>22</dim>
@@ -461,7 +474,7 @@
461
  </port>
462
  </output>
463
  </layer>
464
- <layer id="36" name="Subtract_1339258" type="Subtract" version="opset1">
465
  <data auto_broadcast="numpy" />
466
  <input>
467
  <port id="0" precision="I32">
@@ -477,13 +490,13 @@
477
  </port>
478
  </output>
479
  </layer>
480
- <layer id="37" name="Constant_1339259" type="Const" version="opset1">
481
- <data element_type="i32" shape="" offset="4689265" size="4" />
482
  <output>
483
  <port id="0" precision="I32" />
484
  </output>
485
  </layer>
486
- <layer id="38" name="Minimum_1339260" type="Minimum" version="opset1">
487
  <data auto_broadcast="numpy" />
488
  <input>
489
  <port id="0" precision="I32">
@@ -497,7 +510,7 @@
497
  </port>
498
  </output>
499
  </layer>
500
- <layer id="39" name="Subtract_1339261" type="Subtract" version="opset1">
501
  <data auto_broadcast="numpy" />
502
  <input>
503
  <port id="0" precision="I32">
@@ -513,15 +526,15 @@
513
  </port>
514
  </output>
515
  </layer>
516
- <layer id="40" name="Constant_1339262" type="Const" version="opset1">
517
- <data element_type="i32" shape="1" offset="4689269" size="4" />
518
  <output>
519
  <port id="0" precision="I32">
520
  <dim>1</dim>
521
  </port>
522
  </output>
523
  </layer>
524
- <layer id="41" name="CombineSegments_1339263" type="CombineSegments" version="extension">
525
  <input>
526
  <port id="0" precision="I32">
527
  <dim>-1</dim>
@@ -557,7 +570,7 @@
557
  </port>
558
  </output>
559
  </layer>
560
- <layer id="42" name="Subtract_1339264" type="Subtract" version="opset1">
561
  <data auto_broadcast="numpy" />
562
  <input>
563
  <port id="0" precision="I32">
@@ -573,13 +586,13 @@
573
  </port>
574
  </output>
575
  </layer>
576
- <layer id="43" name="Constant_1339265" type="Const" version="opset1">
577
- <data element_type="i32" shape="" offset="4689269" size="4" />
578
  <output>
579
  <port id="0" precision="I32" />
580
  </output>
581
  </layer>
582
- <layer id="44" name="ReduceMax_1339266" type="ReduceMax" version="opset1">
583
  <data keep_dims="false" />
584
  <input>
585
  <port id="0" precision="I32">
@@ -591,14 +604,14 @@
591
  <port id="2" precision="I32" />
592
  </output>
593
  </layer>
594
- <layer id="45" name="Constant_1339267" type="Const" version="opset1">
595
- <data element_type="i32" shape="" offset="4689269" size="4" />
596
  <output>
597
  <port id="0" precision="I32" />
598
  </output>
599
  </layer>
600
- <layer id="46" name="RaggedToDense_1339268" type="RaggedToDense" version="extension">
601
- <data pad_right="false" m_pad_max_length="false" />
602
  <input>
603
  <port id="0" precision="I32">
604
  <dim>-1</dim>
@@ -623,7 +636,7 @@
623
  </port>
624
  </output>
625
  </layer>
626
- <layer id="47" name="Convert_1339269" type="Convert" version="opset1">
627
  <data destination_type="i32" />
628
  <input>
629
  <port id="0" precision="BOOL">
@@ -638,7 +651,7 @@
638
  </port>
639
  </output>
640
  </layer>
641
- <layer id="48" name="Convert_1339269.0" type="Convert" version="opset1">
642
  <data destination_type="i64" />
643
  <input>
644
  <port id="0" precision="I32">
@@ -653,7 +666,7 @@
653
  </port>
654
  </output>
655
  </layer>
656
- <layer id="50" name="RaggedToDense_1339268.0" type="Convert" version="opset1">
657
  <data destination_type="i64" />
658
  <input>
659
  <port id="0" precision="I32">
@@ -668,7 +681,7 @@
668
  </port>
669
  </output>
670
  </layer>
671
- <layer id="51" name="Result_1339272" type="Result" version="opset1" output_names="input_ids">
672
  <input>
673
  <port id="0" precision="I64">
674
  <dim>-1</dim>
@@ -676,7 +689,7 @@
676
  </port>
677
  </input>
678
  </layer>
679
- <layer id="49" name="Result_1339274" type="Result" version="opset1" output_names="attention_mask">
680
  <input>
681
  <port id="0" precision="I64">
682
  <dim>-1</dim>
@@ -695,8 +708,8 @@
695
  <edge from-layer="3" from-port="1" to-layer="6" to-port="0" />
696
  <edge from-layer="4" from-port="0" to-layer="6" to-port="1" />
697
  <edge from-layer="5" from-port="0" to-layer="6" to-port="2" />
698
- <edge from-layer="6" from-port="3" to-layer="11" to-port="0" />
699
  <edge from-layer="6" from-port="3" to-layer="8" to-port="1" />
 
700
  <edge from-layer="7" from-port="0" to-layer="8" to-port="2" />
701
  <edge from-layer="8" from-port="3" to-layer="15" to-port="0" />
702
  <edge from-layer="9" from-port="0" to-layer="13" to-port="0" />
@@ -705,11 +718,11 @@
705
  <edge from-layer="12" from-port="0" to-layer="13" to-port="2" />
706
  <edge from-layer="13" from-port="3" to-layer="15" to-port="1" />
707
  <edge from-layer="14" from-port="0" to-layer="15" to-port="5" />
708
- <edge from-layer="15" from-port="10" to-layer="18" to-port="2" />
709
- <edge from-layer="15" from-port="7" to-layer="35" to-port="1" />
710
- <edge from-layer="15" from-port="6" to-layer="35" to-port="0" />
711
- <edge from-layer="15" from-port="11" to-layer="18" to-port="3" />
712
  <edge from-layer="15" from-port="9" to-layer="18" to-port="1" />
 
 
 
 
713
  <edge from-layer="15" from-port="8" to-layer="18" to-port="0" />
714
  <edge from-layer="16" from-port="0" to-layer="18" to-port="4" />
715
  <edge from-layer="17" from-port="0" to-layer="18" to-port="5" />
@@ -719,46 +732,50 @@
719
  <edge from-layer="18" from-port="9" to-layer="21" to-port="3" />
720
  <edge from-layer="19" from-port="0" to-layer="21" to-port="4" />
721
  <edge from-layer="20" from-port="0" to-layer="21" to-port="5" />
722
- <edge from-layer="21" from-port="6" to-layer="35" to-port="2" />
723
- <edge from-layer="21" from-port="7" to-layer="35" to-port="3" />
724
- <edge from-layer="21" from-port="8" to-layer="35" to-port="4" />
725
- <edge from-layer="22" from-port="0" to-layer="35" to-port="5" />
726
- <edge from-layer="23" from-port="0" to-layer="35" to-port="6" />
727
- <edge from-layer="24" from-port="0" to-layer="35" to-port="7" />
728
- <edge from-layer="25" from-port="0" to-layer="35" to-port="8" />
729
- <edge from-layer="26" from-port="0" to-layer="35" to-port="9" />
730
- <edge from-layer="27" from-port="0" to-layer="35" to-port="10" />
731
- <edge from-layer="28" from-port="0" to-layer="35" to-port="11" />
732
- <edge from-layer="29" from-port="0" to-layer="35" to-port="12" />
733
- <edge from-layer="30" from-port="0" to-layer="35" to-port="13" />
734
- <edge from-layer="31" from-port="0" to-layer="35" to-port="14" />
735
- <edge from-layer="32" from-port="0" to-layer="35" to-port="15" />
736
- <edge from-layer="33" from-port="0" to-layer="35" to-port="16" />
737
- <edge from-layer="34" from-port="0" to-layer="35" to-port="17" />
738
- <edge from-layer="35" from-port="19" to-layer="36" to-port="0" />
739
- <edge from-layer="35" from-port="18" to-layer="36" to-port="1" />
740
- <edge from-layer="35" from-port="19" to-layer="39" to-port="0" />
741
- <edge from-layer="35" from-port="19" to-layer="41" to-port="1" />
742
- <edge from-layer="35" from-port="20" to-layer="41" to-port="2" />
743
- <edge from-layer="36" from-port="2" to-layer="38" to-port="0" />
744
- <edge from-layer="37" from-port="0" to-layer="38" to-port="1" />
745
- <edge from-layer="38" from-port="2" to-layer="39" to-port="1" />
746
- <edge from-layer="39" from-port="2" to-layer="41" to-port="0" />
747
- <edge from-layer="40" from-port="0" to-layer="41" to-port="3" />
748
- <edge from-layer="41" from-port="5" to-layer="42" to-port="0" />
749
- <edge from-layer="41" from-port="4" to-layer="42" to-port="1" />
750
- <edge from-layer="41" from-port="4" to-layer="46" to-port="0" />
751
- <edge from-layer="41" from-port="5" to-layer="46" to-port="1" />
752
- <edge from-layer="41" from-port="6" to-layer="46" to-port="2" />
753
- <edge from-layer="42" from-port="2" to-layer="44" to-port="0" />
754
- <edge from-layer="43" from-port="0" to-layer="44" to-port="1" />
755
- <edge from-layer="44" from-port="2" to-layer="46" to-port="3" />
756
- <edge from-layer="45" from-port="0" to-layer="46" to-port="4" />
757
- <edge from-layer="46" from-port="6" to-layer="47" to-port="0" />
758
- <edge from-layer="46" from-port="5" to-layer="50" to-port="0" />
759
- <edge from-layer="47" from-port="1" to-layer="48" to-port="0" />
760
- <edge from-layer="48" from-port="1" to-layer="49" to-port="0" />
761
- <edge from-layer="50" from-port="1" to-layer="51" to-port="0" />
 
 
 
 
762
  </edges>
763
  <rt_info>
764
  <add_attention_mask value="True" />
@@ -770,10 +787,9 @@
770
  <detokenizer_input_type value="i64" />
771
  <eos_token_id value="2" />
772
  <handle_special_tokens_with_re />
773
- <max_length />
774
  <number_of_inputs value="1" />
775
- <openvino_tokenizers_version value="2025.1.0.0-513-7881453f285" />
776
- <openvino_version value="2025.1.0-18378-fcb9c9d5d6d" />
777
  <original_tokenizer_class value="&lt;class 'transformers.models.llama.tokenization_llama_fast.LlamaTokenizerFast'>" />
778
  <sentencepiece_version value="0.2.0" />
779
  <skip_special_tokens value="True" />
 
1
  <?xml version="1.0"?>
2
  <net name="tokenizer" version="11">
3
  <layers>
4
+ <layer id="0" name="Parameter_1310283" type="Parameter" version="opset1">
5
  <data shape="?" element_type="string" />
6
  <output>
7
+ <port id="0" precision="STRING" names="Parameter_1310283">
8
  <dim>-1</dim>
9
  </port>
10
  </output>
11
  </layer>
12
+ <layer id="1" name="Constant_1310289" type="Const" version="opset1">
13
  <data element_type="i64" shape="" offset="0" size="8" />
14
  <output>
15
  <port id="0" precision="I64" />
16
  </output>
17
  </layer>
18
+ <layer id="2" name="StringTensorUnpack_1310284" type="StringTensorUnpack" version="extension">
19
+ <data mode="begins_ends" />
20
  <input>
21
  <port id="0" precision="STRING">
22
  <dim>-1</dim>
 
34
  </port>
35
  </output>
36
  </layer>
37
+ <layer id="3" name="ShapeOf_1310285" type="ShapeOf" version="opset3">
38
  <data output_type="i64" />
39
  <input>
40
  <port id="0" precision="I32">
 
47
  </port>
48
  </output>
49
  </layer>
50
+ <layer id="4" name="Constant_1310286" type="Const" version="opset1">
51
  <data element_type="i64" shape="" offset="0" size="8" />
52
  <output>
53
  <port id="0" precision="I64" />
54
  </output>
55
  </layer>
56
+ <layer id="5" name="Constant_1310287" type="Const" version="opset1">
57
  <data element_type="i64" shape="" offset="0" size="8" />
58
  <output>
59
  <port id="0" precision="I64" />
60
  </output>
61
  </layer>
62
+ <layer id="6" name="Gather_1310288" type="Gather" version="opset8">
63
  <data batch_dims="0" />
64
  <input>
65
  <port id="0" precision="I64">
 
72
  <port id="3" precision="I64" />
73
  </output>
74
  </layer>
75
+ <layer id="7" name="Constant_1310290" type="Const" version="opset1">
76
  <data element_type="i64" shape="" offset="8" size="8" />
77
  <output>
78
  <port id="0" precision="I64" />
79
  </output>
80
  </layer>
81
+ <layer id="8" name="Range_1310291" type="Range" version="opset4">
82
  <data output_type="i32" />
83
  <input>
84
  <port id="0" precision="I64" />
 
91
  </port>
92
  </output>
93
  </layer>
94
+ <layer id="9" name="Constant_1310292" type="Const" version="opset1">
95
  <data element_type="i64" shape="" offset="8" size="8" />
96
  <output>
97
  <port id="0" precision="I64" />
98
  </output>
99
  </layer>
100
+ <layer id="10" name="Constant_1310293" type="Const" version="opset1">
101
  <data element_type="i64" shape="" offset="8" size="8" />
102
  <output>
103
  <port id="0" precision="I64" />
104
  </output>
105
  </layer>
106
+ <layer id="11" name="Add_1310294" type="Add" version="opset1">
107
  <data auto_broadcast="numpy" />
108
  <input>
109
  <port id="0" precision="I64" />
 
113
  <port id="2" precision="I64" />
114
  </output>
115
  </layer>
116
+ <layer id="12" name="Constant_1310295" type="Const" version="opset1">
117
  <data element_type="i64" shape="" offset="8" size="8" />
118
  <output>
119
  <port id="0" precision="I64" />
120
  </output>
121
  </layer>
122
+ <layer id="13" name="Range_1310296" type="Range" version="opset4">
123
  <data output_type="i32" />
124
  <input>
125
  <port id="0" precision="I64" />
 
132
  </port>
133
  </output>
134
  </layer>
135
+ <layer id="14" name="Constant_1310358" type="Const" version="opset1">
136
  <data element_type="u8" shape="117" offset="16" size="117" />
137
  <output>
138
  <port id="0" precision="U8">
 
140
  </port>
141
  </output>
142
  </layer>
143
+ <layer id="15" name="SpecialTokensSplit_1310359" type="SpecialTokensSplit" version="extension">
144
  <input>
145
  <port id="0" precision="I32">
146
  <dim>-1</dim>
 
182
  </port>
183
  </output>
184
  </layer>
185
+ <layer id="16" name="Constant_1310361" type="Const" version="opset1">
186
  <data element_type="u8" shape="1" offset="133" size="1" />
187
  <output>
188
  <port id="0" precision="U8">
 
190
  </port>
191
  </output>
192
  </layer>
193
+ <layer id="17" name="Constant_1310363" type="Const" version="opset1">
194
  <data element_type="u8" shape="3" offset="134" size="3" />
195
  <output>
196
  <port id="0" precision="U8">
 
198
  </port>
199
  </output>
200
  </layer>
201
+ <layer id="18" name="RegexNormalization_1310364" type="RegexNormalization" version="extension">
202
  <data global_replace="true" />
203
  <input>
204
  <port id="0" precision="I32">
 
235
  </port>
236
  </output>
237
  </layer>
238
+ <layer id="19" name="Constant_1310366" type="Const" version="opset1">
239
  <data element_type="u8" shape="11" offset="137" size="11" />
240
  <output>
241
  <port id="0" precision="U8">
 
243
  </port>
244
  </output>
245
  </layer>
246
+ <layer id="20" name="Constant_1310368" type="Const" version="opset1">
247
  <data element_type="u8" shape="5" offset="148" size="5" />
248
  <output>
249
  <port id="0" precision="U8">
 
251
  </port>
252
  </output>
253
  </layer>
254
+ <layer id="21" name="RegexNormalization_1310369" type="RegexNormalization" version="extension">
255
  <data global_replace="true" />
256
  <input>
257
  <port id="0" precision="I32">
 
288
  </port>
289
  </output>
290
  </layer>
291
+ <layer id="22" name="Constant_1310371" type="Const" version="opset1">
292
+ <data element_type="u8" shape="895419" offset="153" size="895419" />
293
  <output>
294
+ <port id="0" precision="U8">
295
+ <dim>895419</dim>
296
  </port>
297
  </output>
298
  </layer>
299
+ <layer id="23" name="StringTensorUnpack_1310372" type="StringTensorUnpack" version="extension">
300
+ <data mode="begins_ends" />
301
+ <input>
302
+ <port id="0" precision="U8">
303
+ <dim>895419</dim>
304
+ </port>
305
+ </input>
306
  <output>
307
+ <port id="1" precision="I32">
308
+ <dim>-1</dim>
309
+ </port>
310
+ <port id="2" precision="I32">
311
+ <dim>-1</dim>
312
+ </port>
313
+ <port id="3" precision="U8">
314
+ <dim>-1</dim>
315
  </port>
316
  </output>
317
  </layer>
318
+ <layer id="24" name="Constant_1310377" type="Const" version="opset1">
319
+ <data element_type="u8" shape="1393717" offset="895572" size="1393717" />
320
  <output>
321
  <port id="0" precision="U8">
322
+ <dim>1393717</dim>
323
  </port>
324
  </output>
325
  </layer>
326
+ <layer id="25" name="StringTensorUnpack_1310378" type="StringTensorUnpack" version="extension">
327
+ <data mode="begins_ends" />
328
+ <input>
329
+ <port id="0" precision="U8">
330
+ <dim>1393717</dim>
331
  </port>
332
+ </input>
 
 
 
333
  <output>
334
+ <port id="1" precision="I32">
335
+ <dim>-1</dim>
336
+ </port>
337
+ <port id="2" precision="I32">
338
+ <dim>-1</dim>
339
+ </port>
340
+ <port id="3" precision="U8">
341
+ <dim>-1</dim>
342
  </port>
343
  </output>
344
  </layer>
345
+ <layer id="26" name="Constant_1310380" type="Const" version="opset1">
346
+ <data element_type="u8" shape="1060963" offset="2289289" size="1060963" />
347
  <output>
348
  <port id="0" precision="U8">
349
+ <dim>1060963</dim>
350
  </port>
351
  </output>
352
  </layer>
353
+ <layer id="27" name="StringTensorUnpack_1310381" type="StringTensorUnpack" version="extension">
354
+ <data mode="begins_ends" />
355
+ <input>
356
+ <port id="0" precision="U8">
357
+ <dim>1060963</dim>
358
  </port>
359
+ </input>
 
 
 
360
  <output>
361
+ <port id="1" precision="I32">
362
+ <dim>-1</dim>
363
+ </port>
364
+ <port id="2" precision="I32">
365
+ <dim>-1</dim>
366
+ </port>
367
+ <port id="3" precision="U8">
368
+ <dim>-1</dim>
369
  </port>
370
  </output>
371
  </layer>
372
+ <layer id="28" name="Constant_1310374" type="Const" version="opset1">
373
+ <data element_type="u8" shape="133" offset="3350252" size="133" />
374
  <output>
375
  <port id="0" precision="U8">
376
+ <dim>133</dim>
377
  </port>
378
  </output>
379
  </layer>
380
+ <layer id="29" name="StringTensorUnpack_1310375" type="StringTensorUnpack" version="extension">
381
+ <data mode="begins_ends" />
382
+ <input>
383
+ <port id="0" precision="U8">
384
+ <dim>133</dim>
385
  </port>
386
+ </input>
 
 
 
387
  <output>
388
+ <port id="1" precision="I32">
389
+ <dim>-1</dim>
390
  </port>
391
+ <port id="2" precision="I32">
392
+ <dim>-1</dim>
393
+ </port>
394
+ <port id="3" precision="U8">
395
+ <dim>-1</dim>
 
 
396
  </port>
397
  </output>
398
  </layer>
399
+ <layer id="30" name="Constant_1310382" type="Const" version="opset1">
400
+ <data element_type="i32" shape="22" offset="3350385" size="88" />
401
  <output>
402
  <port id="0" precision="I32">
403
  <dim>22</dim>
404
  </port>
405
  </output>
406
  </layer>
407
+ <layer id="31" name="BPETokenizer_1310383" type="BPETokenizer" version="extension">
408
  <data unk_token="&lt;unk>" fuse_unk="true" suffix_indicator="" end_suffix="" byte_fallback="true" cache_capacity="20000" />
409
  <input>
410
  <port id="0" precision="I32">
 
423
  <dim>-1</dim>
424
  </port>
425
  <port id="5" precision="I32">
426
+ <dim>-1</dim>
427
  </port>
428
  <port id="6" precision="I32">
429
+ <dim>-1</dim>
430
  </port>
431
  <port id="7" precision="U8">
432
+ <dim>-1</dim>
433
  </port>
434
  <port id="8" precision="I32">
435
+ <dim>-1</dim>
436
  </port>
437
  <port id="9" precision="I32">
438
+ <dim>-1</dim>
439
  </port>
440
  <port id="10" precision="U8">
441
+ <dim>-1</dim>
442
  </port>
443
  <port id="11" precision="I32">
444
+ <dim>-1</dim>
445
  </port>
446
  <port id="12" precision="I32">
447
+ <dim>-1</dim>
448
  </port>
449
  <port id="13" precision="U8">
450
+ <dim>-1</dim>
451
  </port>
452
  <port id="14" precision="I32">
453
+ <dim>-1</dim>
454
  </port>
455
  <port id="15" precision="I32">
456
+ <dim>-1</dim>
457
  </port>
458
  <port id="16" precision="U8">
459
+ <dim>-1</dim>
460
  </port>
461
  <port id="17" precision="I32">
462
  <dim>22</dim>
 
474
  </port>
475
  </output>
476
  </layer>
477
+ <layer id="32" name="Subtract_1310384" type="Subtract" version="opset1">
478
  <data auto_broadcast="numpy" />
479
  <input>
480
  <port id="0" precision="I32">
 
490
  </port>
491
  </output>
492
  </layer>
493
+ <layer id="33" name="Constant_1310385" type="Const" version="opset1">
494
+ <data element_type="i32" shape="" offset="3350473" size="4" />
495
  <output>
496
  <port id="0" precision="I32" />
497
  </output>
498
  </layer>
499
+ <layer id="34" name="Minimum_1310386" type="Minimum" version="opset1">
500
  <data auto_broadcast="numpy" />
501
  <input>
502
  <port id="0" precision="I32">
 
510
  </port>
511
  </output>
512
  </layer>
513
+ <layer id="35" name="Subtract_1310387" type="Subtract" version="opset1">
514
  <data auto_broadcast="numpy" />
515
  <input>
516
  <port id="0" precision="I32">
 
526
  </port>
527
  </output>
528
  </layer>
529
+ <layer id="36" name="Constant_1310388" type="Const" version="opset1">
530
+ <data element_type="i32" shape="1" offset="3350477" size="4" />
531
  <output>
532
  <port id="0" precision="I32">
533
  <dim>1</dim>
534
  </port>
535
  </output>
536
  </layer>
537
+ <layer id="37" name="CombineSegments_1310389" type="CombineSegments" version="extension">
538
  <input>
539
  <port id="0" precision="I32">
540
  <dim>-1</dim>
 
570
  </port>
571
  </output>
572
  </layer>
573
+ <layer id="38" name="Subtract_1310390" type="Subtract" version="opset1">
574
  <data auto_broadcast="numpy" />
575
  <input>
576
  <port id="0" precision="I32">
 
586
  </port>
587
  </output>
588
  </layer>
589
+ <layer id="39" name="Constant_1310391" type="Const" version="opset1">
590
+ <data element_type="i32" shape="" offset="3350477" size="4" />
591
  <output>
592
  <port id="0" precision="I32" />
593
  </output>
594
  </layer>
595
+ <layer id="40" name="ReduceMax_1310392" type="ReduceMax" version="opset1">
596
  <data keep_dims="false" />
597
  <input>
598
  <port id="0" precision="I32">
 
604
  <port id="2" precision="I32" />
605
  </output>
606
  </layer>
607
+ <layer id="41" name="Constant_1310393" type="Const" version="opset1">
608
+ <data element_type="i32" shape="" offset="3350477" size="4" />
609
  <output>
610
  <port id="0" precision="I32" />
611
  </output>
612
  </layer>
613
+ <layer id="42" name="RaggedToDense_1310394" type="RaggedToDense" version="extension">
614
+ <data pad_right="false" />
615
  <input>
616
  <port id="0" precision="I32">
617
  <dim>-1</dim>
 
636
  </port>
637
  </output>
638
  </layer>
639
+ <layer id="43" name="Convert_1310395" type="Convert" version="opset1">
640
  <data destination_type="i32" />
641
  <input>
642
  <port id="0" precision="BOOL">
 
651
  </port>
652
  </output>
653
  </layer>
654
+ <layer id="44" name="Convert_1310395.0" type="Convert" version="opset1">
655
  <data destination_type="i64" />
656
  <input>
657
  <port id="0" precision="I32">
 
666
  </port>
667
  </output>
668
  </layer>
669
+ <layer id="46" name="RaggedToDense_1310394.0" type="Convert" version="opset1">
670
  <data destination_type="i64" />
671
  <input>
672
  <port id="0" precision="I32">
 
681
  </port>
682
  </output>
683
  </layer>
684
+ <layer id="47" name="Result_1310398" type="Result" version="opset1">
685
  <input>
686
  <port id="0" precision="I64">
687
  <dim>-1</dim>
 
689
  </port>
690
  </input>
691
  </layer>
692
+ <layer id="45" name="Result_1310400" type="Result" version="opset1">
693
  <input>
694
  <port id="0" precision="I64">
695
  <dim>-1</dim>
 
708
  <edge from-layer="3" from-port="1" to-layer="6" to-port="0" />
709
  <edge from-layer="4" from-port="0" to-layer="6" to-port="1" />
710
  <edge from-layer="5" from-port="0" to-layer="6" to-port="2" />
 
711
  <edge from-layer="6" from-port="3" to-layer="8" to-port="1" />
712
+ <edge from-layer="6" from-port="3" to-layer="11" to-port="0" />
713
  <edge from-layer="7" from-port="0" to-layer="8" to-port="2" />
714
  <edge from-layer="8" from-port="3" to-layer="15" to-port="0" />
715
  <edge from-layer="9" from-port="0" to-layer="13" to-port="0" />
 
718
  <edge from-layer="12" from-port="0" to-layer="13" to-port="2" />
719
  <edge from-layer="13" from-port="3" to-layer="15" to-port="1" />
720
  <edge from-layer="14" from-port="0" to-layer="15" to-port="5" />
 
 
 
 
721
  <edge from-layer="15" from-port="9" to-layer="18" to-port="1" />
722
+ <edge from-layer="15" from-port="6" to-layer="31" to-port="0" />
723
+ <edge from-layer="15" from-port="7" to-layer="31" to-port="1" />
724
+ <edge from-layer="15" from-port="11" to-layer="18" to-port="3" />
725
+ <edge from-layer="15" from-port="10" to-layer="18" to-port="2" />
726
  <edge from-layer="15" from-port="8" to-layer="18" to-port="0" />
727
  <edge from-layer="16" from-port="0" to-layer="18" to-port="4" />
728
  <edge from-layer="17" from-port="0" to-layer="18" to-port="5" />
 
732
  <edge from-layer="18" from-port="9" to-layer="21" to-port="3" />
733
  <edge from-layer="19" from-port="0" to-layer="21" to-port="4" />
734
  <edge from-layer="20" from-port="0" to-layer="21" to-port="5" />
735
+ <edge from-layer="21" from-port="8" to-layer="31" to-port="4" />
736
+ <edge from-layer="21" from-port="7" to-layer="31" to-port="3" />
737
+ <edge from-layer="21" from-port="6" to-layer="31" to-port="2" />
738
+ <edge from-layer="22" from-port="0" to-layer="23" to-port="0" />
739
+ <edge from-layer="23" from-port="1" to-layer="31" to-port="5" />
740
+ <edge from-layer="23" from-port="2" to-layer="31" to-port="6" />
741
+ <edge from-layer="23" from-port="3" to-layer="31" to-port="7" />
742
+ <edge from-layer="24" from-port="0" to-layer="25" to-port="0" />
743
+ <edge from-layer="25" from-port="1" to-layer="31" to-port="8" />
744
+ <edge from-layer="25" from-port="2" to-layer="31" to-port="9" />
745
+ <edge from-layer="25" from-port="3" to-layer="31" to-port="10" />
746
+ <edge from-layer="26" from-port="0" to-layer="27" to-port="0" />
747
+ <edge from-layer="27" from-port="3" to-layer="31" to-port="13" />
748
+ <edge from-layer="27" from-port="2" to-layer="31" to-port="12" />
749
+ <edge from-layer="27" from-port="1" to-layer="31" to-port="11" />
750
+ <edge from-layer="28" from-port="0" to-layer="29" to-port="0" />
751
+ <edge from-layer="29" from-port="1" to-layer="31" to-port="14" />
752
+ <edge from-layer="29" from-port="2" to-layer="31" to-port="15" />
753
+ <edge from-layer="29" from-port="3" to-layer="31" to-port="16" />
754
+ <edge from-layer="30" from-port="0" to-layer="31" to-port="17" />
755
+ <edge from-layer="31" from-port="19" to-layer="32" to-port="0" />
756
+ <edge from-layer="31" from-port="18" to-layer="32" to-port="1" />
757
+ <edge from-layer="31" from-port="19" to-layer="35" to-port="0" />
758
+ <edge from-layer="31" from-port="20" to-layer="37" to-port="2" />
759
+ <edge from-layer="31" from-port="19" to-layer="37" to-port="1" />
760
+ <edge from-layer="32" from-port="2" to-layer="34" to-port="0" />
761
+ <edge from-layer="33" from-port="0" to-layer="34" to-port="1" />
762
+ <edge from-layer="34" from-port="2" to-layer="35" to-port="1" />
763
+ <edge from-layer="35" from-port="2" to-layer="37" to-port="0" />
764
+ <edge from-layer="36" from-port="0" to-layer="37" to-port="3" />
765
+ <edge from-layer="37" from-port="5" to-layer="42" to-port="1" />
766
+ <edge from-layer="37" from-port="6" to-layer="42" to-port="2" />
767
+ <edge from-layer="37" from-port="4" to-layer="42" to-port="0" />
768
+ <edge from-layer="37" from-port="4" to-layer="38" to-port="1" />
769
+ <edge from-layer="37" from-port="5" to-layer="38" to-port="0" />
770
+ <edge from-layer="38" from-port="2" to-layer="40" to-port="0" />
771
+ <edge from-layer="39" from-port="0" to-layer="40" to-port="1" />
772
+ <edge from-layer="40" from-port="2" to-layer="42" to-port="3" />
773
+ <edge from-layer="41" from-port="0" to-layer="42" to-port="4" />
774
+ <edge from-layer="42" from-port="6" to-layer="43" to-port="0" />
775
+ <edge from-layer="42" from-port="5" to-layer="46" to-port="0" />
776
+ <edge from-layer="43" from-port="1" to-layer="44" to-port="0" />
777
+ <edge from-layer="44" from-port="1" to-layer="45" to-port="0" />
778
+ <edge from-layer="46" from-port="1" to-layer="47" to-port="0" />
779
  </edges>
780
  <rt_info>
781
  <add_attention_mask value="True" />
 
787
  <detokenizer_input_type value="i64" />
788
  <eos_token_id value="2" />
789
  <handle_special_tokens_with_re />
 
790
  <number_of_inputs value="1" />
791
+ <openvino_tokenizers_version value="2025.0.0.0" />
792
+ <openvino_version value="2025.0.0" />
793
  <original_tokenizer_class value="&lt;class 'transformers.models.llama.tokenization_llama_fast.LlamaTokenizerFast'>" />
794
  <sentencepiece_version value="0.2.0" />
795
  <skip_special_tokens value="True" />