Hugging Face's logo

OpenVINO
/
falcon-7b-instruct-int8-ov

Text Generation
Transformers
OpenVINO
falcon
custom_code
text-generation-inference
Inference Endpoints
Model card Files Community
openvino-ci commited on
Commit
f44ba7f
•
1 Parent(s): 1c43de7

Upload folder using huggingface_hub

Browse files
Files changed (8) hide show
  1. README.md +1 -1
  2. generation_config.json +2 -2
  3. openvino_detokenizer.bin +3 -0
  4. openvino_detokenizer.xml +219 -0
  5. openvino_model.xml +4 -4
  6. openvino_tokenizer.bin +3 -0
  7. openvino_tokenizer.xml +726 -0
  8. tokenizer_config.json +4 -0
README.md CHANGED
@@ -14,7 +14,7 @@ This is [falcon-7b-instruct](https://huggingface.co/tiiuae/falcon-7b-instruct) m
14
  Weight compression was performed using `nncf.compress_weights` with the following parameters:
15
 
16
  * mode: **int8_asym**
17
- * ratio: **1**
18
 
19
  For more information on quantization, check the [OpenVINO model optimization guide](https://docs.openvino.ai/2024/openvino-workflow/model-optimization-guide/weight-compression.html).
20
 
 
14
  Weight compression was performed using `nncf.compress_weights` with the following parameters:
15
 
16
  * mode: **int8_asym**
17
+ * ratio: **1.0**
18
 
19
  For more information on quantization, check the [OpenVINO model optimization guide](https://docs.openvino.ai/2024/openvino-workflow/model-optimization-guide/weight-compression.html).
20
 
generation_config.json CHANGED
@@ -1,6 +1,6 @@
1
  {
2
  "_from_model_config": true,
3
- "bos_token_id": 1,
4
- "eos_token_id": 2,
5
  "transformers_version": "4.41.2"
6
  }
 
1
  {
2
  "_from_model_config": true,
3
+ "bos_token_id": 11,
4
+ "eos_token_id": 11,
5
  "transformers_version": "4.41.2"
6
  }
openvino_detokenizer.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1375ab9da6d6881d45e5c53338b44adb17bef2295913a62010e8ac99013c2b41
3
+ size 722194
openvino_detokenizer.xml ADDED
@@ -0,0 +1,219 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ <?xml version="1.0"?>
2
+ <net name="detokenizer" version="11">
3
+ <layers>
4
+ <layer id="0" name="Parameter_253359" type="Parameter" version="opset1">
5
+ <data shape="?,?" element_type="i64" />
6
+ <output>
7
+ <port id="0" precision="I64" names="Parameter_253359">
8
+ <dim>-1</dim>
9
+ <dim>-1</dim>
10
+ </port>
11
+ </output>
12
+ </layer>
13
+ <layer id="1" name="Convert_253375" type="Convert" version="opset1">
14
+ <data destination_type="i32" />
15
+ <input>
16
+ <port id="0" precision="I64">
17
+ <dim>-1</dim>
18
+ <dim>-1</dim>
19
+ </port>
20
+ </input>
21
+ <output>
22
+ <port id="1" precision="I32">
23
+ <dim>-1</dim>
24
+ <dim>-1</dim>
25
+ </port>
26
+ </output>
27
+ </layer>
28
+ <layer id="2" name="Constant_253341" type="Const" version="opset1">
29
+ <data element_type="u8" shape="722145" offset="0" size="722145" />
30
+ <output>
31
+ <port id="0" precision="U8">
32
+ <dim>722145</dim>
33
+ </port>
34
+ </output>
35
+ </layer>
36
+ <layer id="3" name="StringTensorUnpack_253342" type="StringTensorUnpack" version="extension">
37
+ <data mode="begins_ends" />
38
+ <input>
39
+ <port id="0" precision="U8">
40
+ <dim>722145</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="VocabDecoder_253360" type="VocabDecoder" version="extension">
56
+ <data skip_tokens="0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11" />
57
+ <input>
58
+ <port id="0" precision="I32">
59
+ <dim>-1</dim>
60
+ <dim>-1</dim>
61
+ </port>
62
+ <port id="1" precision="I32">
63
+ <dim>-1</dim>
64
+ </port>
65
+ <port id="2" precision="I32">
66
+ <dim>-1</dim>
67
+ </port>
68
+ <port id="3" precision="U8">
69
+ <dim>-1</dim>
70
+ </port>
71
+ </input>
72
+ <output>
73
+ <port id="4" precision="I32">
74
+ <dim>-1</dim>
75
+ </port>
76
+ <port id="5" precision="I32">
77
+ <dim>-1</dim>
78
+ </port>
79
+ <port id="6" precision="I32">
80
+ <dim>-1</dim>
81
+ </port>
82
+ <port id="7" precision="I32">
83
+ <dim>-1</dim>
84
+ </port>
85
+ <port id="8" precision="U8">
86
+ <dim>-1</dim>
87
+ </port>
88
+ </output>
89
+ </layer>
90
+ <layer id="5" name="CharsToBytes_253361" type="CharsToBytes" version="extension">
91
+ <input>
92
+ <port id="0" precision="I32">
93
+ <dim>-1</dim>
94
+ </port>
95
+ <port id="1" precision="I32">
96
+ <dim>-1</dim>
97
+ </port>
98
+ <port id="2" precision="I32">
99
+ <dim>-1</dim>
100
+ </port>
101
+ <port id="3" precision="I32">
102
+ <dim>-1</dim>
103
+ </port>
104
+ <port id="4" precision="U8">
105
+ <dim>-1</dim>
106
+ </port>
107
+ </input>
108
+ <output>
109
+ <port id="5" precision="I32">
110
+ <dim>-1</dim>
111
+ </port>
112
+ <port id="6" precision="I32">
113
+ <dim>-1</dim>
114
+ </port>
115
+ <port id="7" precision="U8">
116
+ <dim>-1</dim>
117
+ </port>
118
+ </output>
119
+ </layer>
120
+ <layer id="6" name="Constant_253363" type="Const" version="opset1">
121
+ <data element_type="u8" shape="47" offset="722145" size="47" />
122
+ <output>
123
+ <port id="0" precision="U8">
124
+ <dim>47</dim>
125
+ </port>
126
+ </output>
127
+ </layer>
128
+ <layer id="7" name="Constant_253365" type="Const" version="opset1">
129
+ <data element_type="u8" shape="2" offset="722192" size="2" />
130
+ <output>
131
+ <port id="0" precision="U8">
132
+ <dim>2</dim>
133
+ </port>
134
+ </output>
135
+ </layer>
136
+ <layer id="8" name="RegexNormalization_253366" type="RegexNormalization" version="extension">
137
+ <data global_replace="true" />
138
+ <input>
139
+ <port id="0" precision="I32">
140
+ <dim>-1</dim>
141
+ </port>
142
+ <port id="1" precision="I32">
143
+ <dim>-1</dim>
144
+ </port>
145
+ <port id="2" precision="U8">
146
+ <dim>-1</dim>
147
+ </port>
148
+ <port id="3" precision="U8">
149
+ <dim>47</dim>
150
+ </port>
151
+ <port id="4" precision="U8">
152
+ <dim>2</dim>
153
+ </port>
154
+ </input>
155
+ <output>
156
+ <port id="5" precision="I32">
157
+ <dim>-1</dim>
158
+ </port>
159
+ <port id="6" precision="I32">
160
+ <dim>-1</dim>
161
+ </port>
162
+ <port id="7" precision="U8">
163
+ <dim>-1</dim>
164
+ </port>
165
+ </output>
166
+ </layer>
167
+ <layer id="9" name="StringTensorPack_253367" type="StringTensorPack" version="extension">
168
+ <data mode="begins_ends" />
169
+ <input>
170
+ <port id="0" precision="I32">
171
+ <dim>-1</dim>
172
+ </port>
173
+ <port id="1" precision="I32">
174
+ <dim>-1</dim>
175
+ </port>
176
+ <port id="2" precision="U8">
177
+ <dim>-1</dim>
178
+ </port>
179
+ </input>
180
+ <output>
181
+ <port id="3" precision="STRING" names="string_output">
182
+ <dim>-1</dim>
183
+ </port>
184
+ </output>
185
+ </layer>
186
+ <layer id="10" name="Result_253368" type="Result" version="opset1">
187
+ <input>
188
+ <port id="0" precision="STRING">
189
+ <dim>-1</dim>
190
+ </port>
191
+ </input>
192
+ </layer>
193
+ </layers>
194
+ <edges>
195
+ <edge from-layer="0" from-port="0" to-layer="1" to-port="0" />
196
+ <edge from-layer="1" from-port="1" to-layer="4" to-port="0" />
197
+ <edge from-layer="2" from-port="0" to-layer="3" to-port="0" />
198
+ <edge from-layer="3" from-port="1" to-layer="4" to-port="1" />
199
+ <edge from-layer="3" from-port="2" to-layer="4" to-port="2" />
200
+ <edge from-layer="3" from-port="3" to-layer="4" to-port="3" />
201
+ <edge from-layer="4" from-port="8" to-layer="5" to-port="4" />
202
+ <edge from-layer="4" from-port="7" to-layer="5" to-port="3" />
203
+ <edge from-layer="4" from-port="6" to-layer="5" to-port="2" />
204
+ <edge from-layer="4" from-port="5" to-layer="5" to-port="1" />
205
+ <edge from-layer="4" from-port="4" to-layer="5" to-port="0" />
206
+ <edge from-layer="5" from-port="5" to-layer="8" to-port="0" />
207
+ <edge from-layer="5" from-port="6" to-layer="8" to-port="1" />
208
+ <edge from-layer="5" from-port="7" to-layer="8" to-port="2" />
209
+ <edge from-layer="6" from-port="0" to-layer="8" to-port="3" />
210
+ <edge from-layer="7" from-port="0" to-layer="8" to-port="4" />
211
+ <edge from-layer="8" from-port="5" to-layer="9" to-port="0" />
212
+ <edge from-layer="8" from-port="6" to-layer="9" to-port="1" />
213
+ <edge from-layer="8" from-port="7" to-layer="9" to-port="2" />
214
+ <edge from-layer="9" from-port="3" to-layer="10" to-port="0" />
215
+ </edges>
216
+ <rt_info>
217
+ <eos_token_id value="11" />
218
+ </rt_info>
219
+ </net>
openvino_model.xml CHANGED
@@ -88413,10 +88413,10 @@
88413
  </weight_compression>
88414
  </nncf>
88415
  <optimum>
88416
- <nncf_version value="2.12.0.dev0+fd0e33c9" />
88417
- <optimum_intel_version value="1.19.0.dev0+eeb1df0" />
88418
- <optimum_version value="1.21.1" />
88419
- <pytorch_version value="2.3.0" />
88420
  <transformers_version value="4.41.2" />
88421
  </optimum>
88422
  </rt_info>
 
88413
  </weight_compression>
88414
  </nncf>
88415
  <optimum>
88416
+ <nncf_version value="2.12.0.dev0+d033c6a8" />
88417
+ <optimum_intel_version value="1.19.0.dev0+480eea1" />
88418
+ <optimum_version value="1.21.2" />
88419
+ <pytorch_version value="2.3.1+cpu" />
88420
  <transformers_version value="4.41.2" />
88421
  </optimum>
88422
  </rt_info>
openvino_tokenizer.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:946eda63c6cc3d75508a042c41b4f18c86ea6894ab3b006608241669ad64f89d
3
+ size 1507746
openvino_tokenizer.xml ADDED
@@ -0,0 +1,726 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ <?xml version="1.0"?>
2
+ <net name="tokenizer" version="11">
3
+ <layers>
4
+ <layer id="0" name="Parameter_253251" type="Parameter" version="opset1">
5
+ <data shape="?" element_type="string" />
6
+ <output>
7
+ <port id="0" precision="STRING" names="Parameter_253251">
8
+ <dim>-1</dim>
9
+ </port>
10
+ </output>
11
+ </layer>
12
+ <layer id="1" name="Constant_253257" 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_253252" type="StringTensorUnpack" version="extension">
19
+ <data mode="begins_ends" />
20
+ <input>
21
+ <port id="0" precision="STRING">
22
+ <dim>-1</dim>
23
+ </port>
24
+ </input>
25
+ <output>
26
+ <port id="1" precision="I32">
27
+ <dim>-1</dim>
28
+ </port>
29
+ <port id="2" precision="I32">
30
+ <dim>-1</dim>
31
+ </port>
32
+ <port id="3" precision="U8">
33
+ <dim>-1</dim>
34
+ </port>
35
+ </output>
36
+ </layer>
37
+ <layer id="3" name="ShapeOf_253253" type="ShapeOf" version="opset3">
38
+ <data output_type="i64" />
39
+ <input>
40
+ <port id="0" precision="I32">
41
+ <dim>-1</dim>
42
+ </port>
43
+ </input>
44
+ <output>
45
+ <port id="1" precision="I64">
46
+ <dim>1</dim>
47
+ </port>
48
+ </output>
49
+ </layer>
50
+ <layer id="4" name="Constant_253254" 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_253255" 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_253256" type="Gather" version="opset8">
63
+ <data batch_dims="0" />
64
+ <input>
65
+ <port id="0" precision="I64">
66
+ <dim>1</dim>
67
+ </port>
68
+ <port id="1" precision="I64" />
69
+ <port id="2" precision="I64" />
70
+ </input>
71
+ <output>
72
+ <port id="3" precision="I64" />
73
+ </output>
74
+ </layer>
75
+ <layer id="7" name="Constant_253258" 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_253259" type="Range" version="opset4">
82
+ <data output_type="i32" />
83
+ <input>
84
+ <port id="0" precision="I64" />
85
+ <port id="1" precision="I64" />
86
+ <port id="2" precision="I64" />
87
+ </input>
88
+ <output>
89
+ <port id="3" precision="I32">
90
+ <dim>-1</dim>
91
+ </port>
92
+ </output>
93
+ </layer>
94
+ <layer id="9" name="Constant_253261" 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_253262" 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_253263" type="Add" version="opset1">
107
+ <data auto_broadcast="numpy" />
108
+ <input>
109
+ <port id="0" precision="I64" />
110
+ <port id="1" precision="I64" />
111
+ </input>
112
+ <output>
113
+ <port id="2" precision="I64" />
114
+ </output>
115
+ </layer>
116
+ <layer id="12" name="Constant_253264" 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_253265" type="Range" version="opset4">
123
+ <data output_type="i32" />
124
+ <input>
125
+ <port id="0" precision="I64" />
126
+ <port id="1" precision="I64" />
127
+ <port id="2" precision="I64" />
128
+ </input>
129
+ <output>
130
+ <port id="3" precision="I32">
131
+ <dim>-1</dim>
132
+ </port>
133
+ </output>
134
+ </layer>
135
+ <layer id="14" name="Constant_253328" type="Const" version="opset1">
136
+ <data element_type="u8" shape="5" offset="16" size="5" />
137
+ <output>
138
+ <port id="0" precision="U8">
139
+ <dim>5</dim>
140
+ </port>
141
+ </output>
142
+ </layer>
143
+ <layer id="15" name="RegexSplit_253329" type="RegexSplit" version="extension">
144
+ <data behaviour="contiguous" invert="false" max_splits="-1" />
145
+ <input>
146
+ <port id="0" precision="I32">
147
+ <dim>-1</dim>
148
+ </port>
149
+ <port id="1" precision="I32">
150
+ <dim>-1</dim>
151
+ </port>
152
+ <port id="2" precision="I32">
153
+ <dim>-1</dim>
154
+ </port>
155
+ <port id="3" precision="I32">
156
+ <dim>-1</dim>
157
+ </port>
158
+ <port id="4" precision="U8">
159
+ <dim>-1</dim>
160
+ </port>
161
+ <port id="5" precision="U8">
162
+ <dim>5</dim>
163
+ </port>
164
+ </input>
165
+ <output>
166
+ <port id="6" precision="I32">
167
+ <dim>-1</dim>
168
+ </port>
169
+ <port id="7" precision="I32">
170
+ <dim>-1</dim>
171
+ </port>
172
+ <port id="8" precision="I32">
173
+ <dim>-1</dim>
174
+ </port>
175
+ <port id="9" precision="I32">
176
+ <dim>-1</dim>
177
+ </port>
178
+ <port id="10" precision="U8">
179
+ <dim>-1</dim>
180
+ </port>
181
+ </output>
182
+ </layer>
183
+ <layer id="16" name="Constant_253331" type="Const" version="opset1">
184
+ <data element_type="u8" shape="64" offset="21" size="64" />
185
+ <output>
186
+ <port id="0" precision="U8">
187
+ <dim>64</dim>
188
+ </port>
189
+ </output>
190
+ </layer>
191
+ <layer id="17" name="RegexSplit_253332" type="RegexSplit" version="extension">
192
+ <data behaviour="isolate" invert="false" max_splits="-1" />
193
+ <input>
194
+ <port id="0" precision="I32">
195
+ <dim>-1</dim>
196
+ </port>
197
+ <port id="1" precision="I32">
198
+ <dim>-1</dim>
199
+ </port>
200
+ <port id="2" precision="I32">
201
+ <dim>-1</dim>
202
+ </port>
203
+ <port id="3" precision="I32">
204
+ <dim>-1</dim>
205
+ </port>
206
+ <port id="4" precision="U8">
207
+ <dim>-1</dim>
208
+ </port>
209
+ <port id="5" precision="U8">
210
+ <dim>64</dim>
211
+ </port>
212
+ </input>
213
+ <output>
214
+ <port id="6" precision="I32">
215
+ <dim>-1</dim>
216
+ </port>
217
+ <port id="7" precision="I32">
218
+ <dim>-1</dim>
219
+ </port>
220
+ <port id="8" precision="I32">
221
+ <dim>-1</dim>
222
+ </port>
223
+ <port id="9" precision="I32">
224
+ <dim>-1</dim>
225
+ </port>
226
+ <port id="10" precision="U8">
227
+ <dim>-1</dim>
228
+ </port>
229
+ </output>
230
+ </layer>
231
+ <layer id="18" name="BytesToChars_253333" type="BytesToChars" version="extension">
232
+ <input>
233
+ <port id="0" precision="I32">
234
+ <dim>-1</dim>
235
+ </port>
236
+ <port id="1" precision="I32">
237
+ <dim>-1</dim>
238
+ </port>
239
+ <port id="2" precision="I32">
240
+ <dim>-1</dim>
241
+ </port>
242
+ <port id="3" precision="I32">
243
+ <dim>-1</dim>
244
+ </port>
245
+ <port id="4" precision="U8">
246
+ <dim>-1</dim>
247
+ </port>
248
+ </input>
249
+ <output>
250
+ <port id="5" precision="I32">
251
+ <dim>-1</dim>
252
+ </port>
253
+ <port id="6" precision="I32">
254
+ <dim>-1</dim>
255
+ </port>
256
+ <port id="7" precision="I32">
257
+ <dim>-1</dim>
258
+ </port>
259
+ <port id="8" precision="I32">
260
+ <dim>-1</dim>
261
+ </port>
262
+ <port id="9" precision="U8">
263
+ <dim>-1</dim>
264
+ </port>
265
+ </output>
266
+ </layer>
267
+ <layer id="19" name="Constant_253335" type="Const" version="opset1">
268
+ <data element_type="u8" shape="20" offset="85" size="20" />
269
+ <output>
270
+ <port id="0" precision="U8">
271
+ <dim>20</dim>
272
+ </port>
273
+ </output>
274
+ </layer>
275
+ <layer id="20" name="RegexSplit_253336" type="RegexSplit" version="extension">
276
+ <data behaviour="contiguous" invert="false" max_splits="-1" />
277
+ <input>
278
+ <port id="0" precision="I32">
279
+ <dim>-1</dim>
280
+ </port>
281
+ <port id="1" precision="I32">
282
+ <dim>-1</dim>
283
+ </port>
284
+ <port id="2" precision="I32">
285
+ <dim>-1</dim>
286
+ </port>
287
+ <port id="3" precision="I32">
288
+ <dim>-1</dim>
289
+ </port>
290
+ <port id="4" precision="U8">
291
+ <dim>-1</dim>
292
+ </port>
293
+ <port id="5" precision="U8">
294
+ <dim>20</dim>
295
+ </port>
296
+ </input>
297
+ <output>
298
+ <port id="6" precision="I32">
299
+ <dim>-1</dim>
300
+ </port>
301
+ <port id="7" precision="I32">
302
+ <dim>-1</dim>
303
+ </port>
304
+ <port id="8" precision="I32">
305
+ <dim>-1</dim>
306
+ </port>
307
+ <port id="9" precision="I32">
308
+ <dim>-1</dim>
309
+ </port>
310
+ <port id="10" precision="U8">
311
+ <dim>-1</dim>
312
+ </port>
313
+ </output>
314
+ </layer>
315
+ <layer id="21" name="Constant_253338" type="Const" version="opset1">
316
+ <data element_type="u8" shape="15" offset="105" size="15" />
317
+ <output>
318
+ <port id="0" precision="U8">
319
+ <dim>15</dim>
320
+ </port>
321
+ </output>
322
+ </layer>
323
+ <layer id="22" name="RegexSplit_253339" type="RegexSplit" version="extension">
324
+ <data behaviour="isolate" invert="false" max_splits="-1" />
325
+ <input>
326
+ <port id="0" precision="I32">
327
+ <dim>-1</dim>
328
+ </port>
329
+ <port id="1" precision="I32">
330
+ <dim>-1</dim>
331
+ </port>
332
+ <port id="2" precision="I32">
333
+ <dim>-1</dim>
334
+ </port>
335
+ <port id="3" precision="I32">
336
+ <dim>-1</dim>
337
+ </port>
338
+ <port id="4" precision="U8">
339
+ <dim>-1</dim>
340
+ </port>
341
+ <port id="5" precision="U8">
342
+ <dim>15</dim>
343
+ </port>
344
+ </input>
345
+ <output>
346
+ <port id="6" precision="I32">
347
+ <dim>-1</dim>
348
+ </port>
349
+ <port id="7" precision="I32">
350
+ <dim>-1</dim>
351
+ </port>
352
+ <port id="8" precision="I32">
353
+ <dim>-1</dim>
354
+ </port>
355
+ <port id="9" precision="I32">
356
+ <dim>-1</dim>
357
+ </port>
358
+ <port id="10" precision="U8">
359
+ <dim>-1</dim>
360
+ </port>
361
+ </output>
362
+ </layer>
363
+ <layer id="23" name="Constant_253341" type="Const" version="opset1">
364
+ <data element_type="u8" shape="722145" offset="120" size="722145" />
365
+ <output>
366
+ <port id="0" precision="U8">
367
+ <dim>722145</dim>
368
+ </port>
369
+ </output>
370
+ </layer>
371
+ <layer id="24" name="StringTensorUnpack_253342" type="StringTensorUnpack" version="extension">
372
+ <data mode="begins_ends" />
373
+ <input>
374
+ <port id="0" precision="U8">
375
+ <dim>722145</dim>
376
+ </port>
377
+ </input>
378
+ <output>
379
+ <port id="1" precision="I32">
380
+ <dim>-1</dim>
381
+ </port>
382
+ <port id="2" precision="I32">
383
+ <dim>-1</dim>
384
+ </port>
385
+ <port id="3" precision="U8">
386
+ <dim>-1</dim>
387
+ </port>
388
+ </output>
389
+ </layer>
390
+ <layer id="25" name="Constant_253344" type="Const" version="opset1">
391
+ <data element_type="u8" shape="785473" offset="722265" size="785473" />
392
+ <output>
393
+ <port id="0" precision="U8">
394
+ <dim>785473</dim>
395
+ </port>
396
+ </output>
397
+ </layer>
398
+ <layer id="26" name="StringTensorUnpack_253345" type="StringTensorUnpack" version="extension">
399
+ <data mode="begins_ends" />
400
+ <input>
401
+ <port id="0" precision="U8">
402
+ <dim>785473</dim>
403
+ </port>
404
+ </input>
405
+ <output>
406
+ <port id="1" precision="I32">
407
+ <dim>-1</dim>
408
+ </port>
409
+ <port id="2" precision="I32">
410
+ <dim>-1</dim>
411
+ </port>
412
+ <port id="3" precision="U8">
413
+ <dim>-1</dim>
414
+ </port>
415
+ </output>
416
+ </layer>
417
+ <layer id="27" name="BPETokenizer_253346" type="BPETokenizer" version="extension">
418
+ <data unk_token="" fuse_unk="false" suffix_indicator="" end_suffix="" byte_fallback="false" />
419
+ <input>
420
+ <port id="0" precision="I32">
421
+ <dim>-1</dim>
422
+ </port>
423
+ <port id="1" precision="I32">
424
+ <dim>-1</dim>
425
+ </port>
426
+ <port id="2" precision="I32">
427
+ <dim>-1</dim>
428
+ </port>
429
+ <port id="3" precision="I32">
430
+ <dim>-1</dim>
431
+ </port>
432
+ <port id="4" precision="U8">
433
+ <dim>-1</dim>
434
+ </port>
435
+ <port id="5" precision="I32">
436
+ <dim>-1</dim>
437
+ </port>
438
+ <port id="6" precision="I32">
439
+ <dim>-1</dim>
440
+ </port>
441
+ <port id="7" precision="U8">
442
+ <dim>-1</dim>
443
+ </port>
444
+ <port id="8" precision="I32">
445
+ <dim>-1</dim>
446
+ </port>
447
+ <port id="9" precision="I32">
448
+ <dim>-1</dim>
449
+ </port>
450
+ <port id="10" precision="U8">
451
+ <dim>-1</dim>
452
+ </port>
453
+ </input>
454
+ <output>
455
+ <port id="11" precision="I32">
456
+ <dim>-1</dim>
457
+ </port>
458
+ <port id="12" precision="I32">
459
+ <dim>-1</dim>
460
+ </port>
461
+ <port id="13" precision="I32">
462
+ <dim>-1</dim>
463
+ </port>
464
+ </output>
465
+ </layer>
466
+ <layer id="28" name="Subtract_253347" type="Subtract" version="opset1">
467
+ <data auto_broadcast="numpy" />
468
+ <input>
469
+ <port id="0" precision="I32">
470
+ <dim>-1</dim>
471
+ </port>
472
+ <port id="1" precision="I32">
473
+ <dim>-1</dim>
474
+ </port>
475
+ </input>
476
+ <output>
477
+ <port id="2" precision="I32">
478
+ <dim>-1</dim>
479
+ </port>
480
+ </output>
481
+ </layer>
482
+ <layer id="29" name="Constant_253348" type="Const" version="opset1">
483
+ <data element_type="i32" shape="" offset="1507738" size="4" />
484
+ <output>
485
+ <port id="0" precision="I32" />
486
+ </output>
487
+ </layer>
488
+ <layer id="30" name="Minimum_253349" type="Minimum" version="opset1">
489
+ <data auto_broadcast="numpy" />
490
+ <input>
491
+ <port id="0" precision="I32">
492
+ <dim>-1</dim>
493
+ </port>
494
+ <port id="1" precision="I32" />
495
+ </input>
496
+ <output>
497
+ <port id="2" precision="I32">
498
+ <dim>-1</dim>
499
+ </port>
500
+ </output>
501
+ </layer>
502
+ <layer id="31" name="Add_253350" type="Add" version="opset1">
503
+ <data auto_broadcast="numpy" />
504
+ <input>
505
+ <port id="0" precision="I32">
506
+ <dim>-1</dim>
507
+ </port>
508
+ <port id="1" precision="I32">
509
+ <dim>-1</dim>
510
+ </port>
511
+ </input>
512
+ <output>
513
+ <port id="2" precision="I32">
514
+ <dim>-1</dim>
515
+ </port>
516
+ </output>
517
+ </layer>
518
+ <layer id="32" name="Subtract_253351" type="Subtract" version="opset1">
519
+ <data auto_broadcast="numpy" />
520
+ <input>
521
+ <port id="0" precision="I32">
522
+ <dim>-1</dim>
523
+ </port>
524
+ <port id="1" precision="I32">
525
+ <dim>-1</dim>
526
+ </port>
527
+ </input>
528
+ <output>
529
+ <port id="2" precision="I32">
530
+ <dim>-1</dim>
531
+ </port>
532
+ </output>
533
+ </layer>
534
+ <layer id="33" name="Constant_253352" type="Const" version="opset1">
535
+ <data element_type="i32" shape="" offset="1507742" size="4" />
536
+ <output>
537
+ <port id="0" precision="I32" />
538
+ </output>
539
+ </layer>
540
+ <layer id="34" name="ReduceMax_253353" type="ReduceMax" version="opset1">
541
+ <data keep_dims="false" />
542
+ <input>
543
+ <port id="0" precision="I32">
544
+ <dim>-1</dim>
545
+ </port>
546
+ <port id="1" precision="I32" />
547
+ </input>
548
+ <output>
549
+ <port id="2" precision="I32" />
550
+ </output>
551
+ </layer>
552
+ <layer id="35" name="Constant_253354" type="Const" version="opset1">
553
+ <data element_type="i32" shape="" offset="1507742" size="4" />
554
+ <output>
555
+ <port id="0" precision="I32" />
556
+ </output>
557
+ </layer>
558
+ <layer id="36" name="RaggedToDense_253355" type="RaggedToDense" version="extension">
559
+ <data pad_right="true" />
560
+ <input>
561
+ <port id="0" precision="I32">
562
+ <dim>-1</dim>
563
+ </port>
564
+ <port id="1" precision="I32">
565
+ <dim>-1</dim>
566
+ </port>
567
+ <port id="2" precision="I32">
568
+ <dim>-1</dim>
569
+ </port>
570
+ <port id="3" precision="I32" />
571
+ <port id="4" precision="I32" />
572
+ </input>
573
+ <output>
574
+ <port id="5" precision="I32">
575
+ <dim>-1</dim>
576
+ <dim>-1</dim>
577
+ </port>
578
+ <port id="6" precision="BOOL">
579
+ <dim>-1</dim>
580
+ <dim>-1</dim>
581
+ </port>
582
+ </output>
583
+ </layer>
584
+ <layer id="37" name="Convert_253356" type="Convert" version="opset1">
585
+ <data destination_type="i32" />
586
+ <input>
587
+ <port id="0" precision="BOOL">
588
+ <dim>-1</dim>
589
+ <dim>-1</dim>
590
+ </port>
591
+ </input>
592
+ <output>
593
+ <port id="1" precision="I32">
594
+ <dim>-1</dim>
595
+ <dim>-1</dim>
596
+ </port>
597
+ </output>
598
+ </layer>
599
+ <layer id="38" name="Convert_253356" type="Convert" version="opset1">
600
+ <data destination_type="i64" />
601
+ <input>
602
+ <port id="0" precision="I32">
603
+ <dim>-1</dim>
604
+ <dim>-1</dim>
605
+ </port>
606
+ </input>
607
+ <output>
608
+ <port id="1" precision="I64" names="attention_mask">
609
+ <dim>-1</dim>
610
+ <dim>-1</dim>
611
+ </port>
612
+ </output>
613
+ </layer>
614
+ <layer id="40" name="RaggedToDense_253355.0" type="Convert" version="opset1">
615
+ <data destination_type="i64" />
616
+ <input>
617
+ <port id="0" precision="I32">
618
+ <dim>-1</dim>
619
+ <dim>-1</dim>
620
+ </port>
621
+ </input>
622
+ <output>
623
+ <port id="1" precision="I64" names="input_ids">
624
+ <dim>-1</dim>
625
+ <dim>-1</dim>
626
+ </port>
627
+ </output>
628
+ </layer>
629
+ <layer id="41" name="Result_253357" type="Result" version="opset1">
630
+ <input>
631
+ <port id="0" precision="I64">
632
+ <dim>-1</dim>
633
+ <dim>-1</dim>
634
+ </port>
635
+ </input>
636
+ </layer>
637
+ <layer id="39" name="Result_253358" type="Result" version="opset1">
638
+ <input>
639
+ <port id="0" precision="I64">
640
+ <dim>-1</dim>
641
+ <dim>-1</dim>
642
+ </port>
643
+ </input>
644
+ </layer>
645
+ </layers>
646
+ <edges>
647
+ <edge from-layer="0" from-port="0" to-layer="2" to-port="0" />
648
+ <edge from-layer="1" from-port="0" to-layer="8" to-port="0" />
649
+ <edge from-layer="2" from-port="1" to-layer="3" to-port="0" />
650
+ <edge from-layer="2" from-port="3" to-layer="15" to-port="4" />
651
+ <edge from-layer="2" from-port="2" to-layer="15" to-port="3" />
652
+ <edge from-layer="2" from-port="1" to-layer="15" to-port="2" />
653
+ <edge from-layer="3" from-port="1" to-layer="6" to-port="0" />
654
+ <edge from-layer="4" from-port="0" to-layer="6" to-port="1" />
655
+ <edge from-layer="5" from-port="0" to-layer="6" to-port="2" />
656
+ <edge from-layer="6" from-port="3" to-layer="8" to-port="1" />
657
+ <edge from-layer="6" from-port="3" to-layer="11" to-port="0" />
658
+ <edge from-layer="7" from-port="0" to-layer="8" to-port="2" />
659
+ <edge from-layer="8" from-port="3" to-layer="15" to-port="0" />
660
+ <edge from-layer="9" from-port="0" to-layer="13" to-port="0" />
661
+ <edge from-layer="10" from-port="0" to-layer="11" to-port="1" />
662
+ <edge from-layer="11" from-port="2" to-layer="13" to-port="1" />
663
+ <edge from-layer="12" from-port="0" to-layer="13" to-port="2" />
664
+ <edge from-layer="13" from-port="3" to-layer="15" to-port="1" />
665
+ <edge from-layer="14" from-port="0" to-layer="15" to-port="5" />
666
+ <edge from-layer="15" from-port="6" to-layer="17" to-port="0" />
667
+ <edge from-layer="15" from-port="10" to-layer="17" to-port="4" />
668
+ <edge from-layer="15" from-port="9" to-layer="17" to-port="3" />
669
+ <edge from-layer="15" from-port="7" to-layer="17" to-port="1" />
670
+ <edge from-layer="15" from-port="8" to-layer="17" to-port="2" />
671
+ <edge from-layer="16" from-port="0" to-layer="17" to-port="5" />
672
+ <edge from-layer="17" from-port="6" to-layer="18" to-port="0" />
673
+ <edge from-layer="17" from-port="7" to-layer="18" to-port="1" />
674
+ <edge from-layer="17" from-port="8" to-layer="18" to-port="2" />
675
+ <edge from-layer="17" from-port="9" to-layer="18" to-port="3" />
676
+ <edge from-layer="17" from-port="10" to-layer="18" to-port="4" />
677
+ <edge from-layer="18" from-port="5" to-layer="20" to-port="0" />
678
+ <edge from-layer="18" from-port="9" to-layer="20" to-port="4" />
679
+ <edge from-layer="18" from-port="8" to-layer="20" to-port="3" />
680
+ <edge from-layer="18" from-port="7" to-layer="20" to-port="2" />
681
+ <edge from-layer="18" from-port="6" to-layer="20" to-port="1" />
682
+ <edge from-layer="19" from-port="0" to-layer="20" to-port="5" />
683
+ <edge from-layer="20" from-port="10" to-layer="22" to-port="4" />
684
+ <edge from-layer="20" from-port="7" to-layer="22" to-port="1" />
685
+ <edge from-layer="20" from-port="9" to-layer="22" to-port="3" />
686
+ <edge from-layer="20" from-port="6" to-layer="22" to-port="0" />
687
+ <edge from-layer="20" from-port="8" to-layer="22" to-port="2" />
688
+ <edge from-layer="21" from-port="0" to-layer="22" to-port="5" />
689
+ <edge from-layer="22" from-port="6" to-layer="27" to-port="0" />
690
+ <edge from-layer="22" from-port="7" to-layer="27" to-port="1" />
691
+ <edge from-layer="22" from-port="8" to-layer="27" to-port="2" />
692
+ <edge from-layer="22" from-port="9" to-layer="27" to-port="3" />
693
+ <edge from-layer="22" from-port="10" to-layer="27" to-port="4" />
694
+ <edge from-layer="23" from-port="0" to-layer="24" to-port="0" />
695
+ <edge from-layer="24" from-port="2" to-layer="27" to-port="6" />
696
+ <edge from-layer="24" from-port="3" to-layer="27" to-port="7" />
697
+ <edge from-layer="24" from-port="1" to-layer="27" to-port="5" />
698
+ <edge from-layer="25" from-port="0" to-layer="26" to-port="0" />
699
+ <edge from-layer="26" from-port="1" to-layer="27" to-port="8" />
700
+ <edge from-layer="26" from-port="2" to-layer="27" to-port="9" />
701
+ <edge from-layer="26" from-port="3" to-layer="27" to-port="10" />
702
+ <edge from-layer="27" from-port="12" to-layer="28" to-port="0" />
703
+ <edge from-layer="27" from-port="11" to-layer="28" to-port="1" />
704
+ <edge from-layer="27" from-port="13" to-layer="36" to-port="2" />
705
+ <edge from-layer="27" from-port="11" to-layer="31" to-port="0" />
706
+ <edge from-layer="27" from-port="11" to-layer="36" to-port="0" />
707
+ <edge from-layer="27" from-port="11" to-layer="32" to-port="1" />
708
+ <edge from-layer="28" from-port="2" to-layer="30" to-port="0" />
709
+ <edge from-layer="29" from-port="0" to-layer="30" to-port="1" />
710
+ <edge from-layer="30" from-port="2" to-layer="31" to-port="1" />
711
+ <edge from-layer="31" from-port="2" to-layer="32" to-port="0" />
712
+ <edge from-layer="31" from-port="2" to-layer="36" to-port="1" />
713
+ <edge from-layer="32" from-port="2" to-layer="34" to-port="0" />
714
+ <edge from-layer="33" from-port="0" to-layer="34" to-port="1" />
715
+ <edge from-layer="34" from-port="2" to-layer="36" to-port="3" />
716
+ <edge from-layer="35" from-port="0" to-layer="36" to-port="4" />
717
+ <edge from-layer="36" from-port="6" to-layer="37" to-port="0" />
718
+ <edge from-layer="36" from-port="5" to-layer="40" to-port="0" />
719
+ <edge from-layer="37" from-port="1" to-layer="38" to-port="0" />
720
+ <edge from-layer="38" from-port="1" to-layer="39" to-port="0" />
721
+ <edge from-layer="40" from-port="1" to-layer="41" to-port="0" />
722
+ </edges>
723
+ <rt_info>
724
+ <eos_token_id value="11" />
725
+ </rt_info>
726
+ </net>
tokenizer_config.json CHANGED
@@ -113,6 +113,10 @@
113
  ],
114
  "clean_up_tokenization_spaces": true,
115
  "eos_token": "<|endoftext|>",
 
 
 
 
116
  "model_max_length": 2048,
117
  "tokenizer_class": "PreTrainedTokenizerFast"
118
  }
 
113
  ],
114
  "clean_up_tokenization_spaces": true,
115
  "eos_token": "<|endoftext|>",
116
+ "model_input_names": [
117
+ "input_ids",
118
+ "attention_mask"
119
+ ],
120
  "model_max_length": 2048,
121
  "tokenizer_class": "PreTrainedTokenizerFast"
122
  }