Hugging Face's logo

OpenVINO
/
open_llama_7b_v2-int8-ov

Text Generation
Transformers
OpenVINO
llama
Inference Endpoints
Model card Files Community
openvino-ci commited on
Commit
7d006b3
•
1 Parent(s): 2db541a

Upload folder using huggingface_hub

Browse files
Files changed (6) hide show
  1. config.json +1 -1
  2. openvino_detokenizer.bin +3 -0
  3. openvino_detokenizer.xml +97 -0
  4. openvino_model.xml +4 -4
  5. openvino_tokenizer.bin +3 -0
  6. openvino_tokenizer.xml +231 -0
config.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "_name_or_path": "openlm-research/open_llama_7b_v2",
3
  "architectures": [
4
  "LlamaForCausalLM"
5
  ],
 
1
  {
2
+ "_name_or_path": "OpenVINO/open_llama_7b_v2-int8-ov",
3
  "architectures": [
4
  "LlamaForCausalLM"
5
  ],
openvino_detokenizer.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9200b58939b5f15eda267b46cd367416c6ffec891653b50930b2cb0944c0f112
3
+ size 511574
openvino_detokenizer.xml ADDED
@@ -0,0 +1,97 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ <?xml version="1.0"?>
2
+ <net name="detokenizer" version="11">
3
+ <layers>
4
+ <layer id="0" name="Parameter_175799" type="Parameter" version="opset1">
5
+ <data shape="?,?" element_type="i64" />
6
+ <output>
7
+ <port id="0" precision="I64" names="Parameter_175799">
8
+ <dim>-1</dim>
9
+ <dim>-1</dim>
10
+ </port>
11
+ </output>
12
+ </layer>
13
+ <layer id="1" name="Constant_175779" type="Const" version="opset1">
14
+ <data element_type="u8" shape="511574" offset="0" size="511574" />
15
+ <output>
16
+ <port id="0" precision="U8">
17
+ <dim>511574</dim>
18
+ </port>
19
+ </output>
20
+ </layer>
21
+ <layer id="2" name="Convert_175809" type="Convert" version="opset1">
22
+ <data destination_type="i32" />
23
+ <input>
24
+ <port id="0" precision="I64">
25
+ <dim>-1</dim>
26
+ <dim>-1</dim>
27
+ </port>
28
+ </input>
29
+ <output>
30
+ <port id="1" precision="I32">
31
+ <dim>-1</dim>
32
+ <dim>-1</dim>
33
+ </port>
34
+ </output>
35
+ </layer>
36
+ <layer id="3" name="SentencepieceDetokenizer_175800" type="SentencepieceDetokenizer" version="extension">
37
+ <input>
38
+ <port id="0" precision="U8">
39
+ <dim>511574</dim>
40
+ </port>
41
+ <port id="1" precision="I32">
42
+ <dim>-1</dim>
43
+ <dim>-1</dim>
44
+ </port>
45
+ </input>
46
+ <output>
47
+ <port id="2" precision="I32">
48
+ <dim>-1</dim>
49
+ </port>
50
+ <port id="3" precision="I32">
51
+ <dim>-1</dim>
52
+ </port>
53
+ <port id="4" precision="U8">
54
+ <dim>-1</dim>
55
+ </port>
56
+ </output>
57
+ </layer>
58
+ <layer id="4" name="StringTensorPack_175801" type="StringTensorPack" version="extension">
59
+ <data mode="begins_ends" />
60
+ <input>
61
+ <port id="0" precision="I32">
62
+ <dim>-1</dim>
63
+ </port>
64
+ <port id="1" precision="I32">
65
+ <dim>-1</dim>
66
+ </port>
67
+ <port id="2" precision="U8">
68
+ <dim>-1</dim>
69
+ </port>
70
+ </input>
71
+ <output>
72
+ <port id="3" precision="STRING" names="string_output">
73
+ <dim>-1</dim>
74
+ </port>
75
+ </output>
76
+ </layer>
77
+ <layer id="5" name="Result_175802" type="Result" version="opset1">
78
+ <input>
79
+ <port id="0" precision="STRING">
80
+ <dim>-1</dim>
81
+ </port>
82
+ </input>
83
+ </layer>
84
+ </layers>
85
+ <edges>
86
+ <edge from-layer="0" from-port="0" to-layer="2" to-port="0" />
87
+ <edge from-layer="1" from-port="0" to-layer="3" to-port="0" />
88
+ <edge from-layer="2" from-port="1" to-layer="3" to-port="1" />
89
+ <edge from-layer="3" from-port="2" to-layer="4" to-port="0" />
90
+ <edge from-layer="3" from-port="3" to-layer="4" to-port="1" />
91
+ <edge from-layer="3" from-port="4" to-layer="4" to-port="2" />
92
+ <edge from-layer="4" from-port="3" to-layer="5" to-port="0" />
93
+ </edges>
94
+ <rt_info>
95
+ <eos_token_id value="2" />
96
+ </rt_info>
97
+ </net>
openvino_model.xml CHANGED
@@ -108600,10 +108600,10 @@
108600
  </weight_compression>
108601
  </nncf>
108602
  <optimum>
108603
- <nncf_version value="2.12.0.dev0+e8ea2521" />
108604
- <optimum_intel_version value="1.18.0.dev0+0a6075b" />
108605
- <optimum_version value="1.20.0" />
108606
- <pytorch_version value="2.3.0" />
108607
  <transformers_version value="4.41.2" />
108608
  </optimum>
108609
  </rt_info>
 
108600
  </weight_compression>
108601
  </nncf>
108602
  <optimum>
108603
+ <nncf_version value="2.12.0.dev0+d033c6a8" />
108604
+ <optimum_intel_version value="1.19.0.dev0+480eea1" />
108605
+ <optimum_version value="1.21.2" />
108606
+ <pytorch_version value="2.3.1+cpu" />
108607
  <transformers_version value="4.41.2" />
108608
  </optimum>
108609
  </rt_info>
openvino_tokenizer.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b7b77cf2b305f5fcd17706d2242cec5e69576e6888853fee55fef0c734926605
3
+ size 511582
openvino_tokenizer.xml ADDED
@@ -0,0 +1,231 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ <?xml version="1.0"?>
2
+ <net name="tokenizer" version="11">
3
+ <layers>
4
+ <layer id="0" name="string_input" type="Parameter" version="opset1">
5
+ <data shape="?" element_type="string" />
6
+ <output>
7
+ <port id="0" precision="STRING" names="string_input">
8
+ <dim>-1</dim>
9
+ </port>
10
+ </output>
11
+ </layer>
12
+ <layer id="1" name="Constant_175785" type="Const" version="opset1">
13
+ <data element_type="i32" shape="" offset="0" size="4" />
14
+ <output>
15
+ <port id="0" precision="I32" />
16
+ </output>
17
+ </layer>
18
+ <layer id="2" name="Constant_175778" type="Const" version="opset1">
19
+ <data element_type="u8" shape="511574" offset="4" size="511574" />
20
+ <output>
21
+ <port id="0" precision="U8">
22
+ <dim>511574</dim>
23
+ </port>
24
+ </output>
25
+ </layer>
26
+ <layer id="3" name="SentencepieceTokenizer_175781" type="SentencepieceTokenizer" version="extension">
27
+ <data nbest_size="0" alpha="0" add_bos="true" add_eos="false" reverse="false" />
28
+ <input>
29
+ <port id="0" precision="U8">
30
+ <dim>511574</dim>
31
+ </port>
32
+ <port id="1" precision="STRING">
33
+ <dim>-1</dim>
34
+ </port>
35
+ </input>
36
+ <output>
37
+ <port id="2" precision="I64">
38
+ <dim>-1</dim>
39
+ <dim>2</dim>
40
+ </port>
41
+ <port id="3" precision="I32">
42
+ <dim>-1</dim>
43
+ </port>
44
+ <port id="4" precision="I64">
45
+ <dim>2</dim>
46
+ </port>
47
+ </output>
48
+ </layer>
49
+ <layer id="4" name="Broadcast_175786" type="Broadcast" version="opset3">
50
+ <data mode="numpy" />
51
+ <input>
52
+ <port id="0" precision="I32" />
53
+ <port id="1" precision="I64">
54
+ <dim>2</dim>
55
+ </port>
56
+ </input>
57
+ <output>
58
+ <port id="2" precision="I32">
59
+ <dim>-1</dim>
60
+ <dim>-1</dim>
61
+ </port>
62
+ </output>
63
+ </layer>
64
+ <layer id="5" name="Constant_175787" type="Const" version="opset1">
65
+ <data element_type="i32" shape="" offset="511578" size="4" />
66
+ <output>
67
+ <port id="0" precision="I32" />
68
+ </output>
69
+ </layer>
70
+ <layer id="6" name="ShapeOf_175788" type="ShapeOf" version="opset3">
71
+ <data output_type="i64" />
72
+ <input>
73
+ <port id="0" precision="I32">
74
+ <dim>-1</dim>
75
+ </port>
76
+ </input>
77
+ <output>
78
+ <port id="1" precision="I64">
79
+ <dim>1</dim>
80
+ </port>
81
+ </output>
82
+ </layer>
83
+ <layer id="7" name="Broadcast_175789" type="Broadcast" version="opset3">
84
+ <data mode="numpy" />
85
+ <input>
86
+ <port id="0" precision="I32" />
87
+ <port id="1" precision="I64">
88
+ <dim>1</dim>
89
+ </port>
90
+ </input>
91
+ <output>
92
+ <port id="2" precision="I32">
93
+ <dim>-1</dim>
94
+ </port>
95
+ </output>
96
+ </layer>
97
+ <layer id="8" name="ScatterNDUpdate_175793" type="ScatterNDUpdate" version="opset4">
98
+ <input>
99
+ <port id="0" precision="I32">
100
+ <dim>-1</dim>
101
+ <dim>-1</dim>
102
+ </port>
103
+ <port id="1" precision="I64">
104
+ <dim>-1</dim>
105
+ <dim>2</dim>
106
+ </port>
107
+ <port id="2" precision="I32">
108
+ <dim>-1</dim>
109
+ </port>
110
+ </input>
111
+ <output>
112
+ <port id="3" precision="I32">
113
+ <dim>-1</dim>
114
+ <dim>-1</dim>
115
+ </port>
116
+ </output>
117
+ </layer>
118
+ <layer id="9" name="ScatterNDUpdate_175793" type="Convert" version="opset1">
119
+ <data destination_type="i64" />
120
+ <input>
121
+ <port id="0" precision="I32">
122
+ <dim>-1</dim>
123
+ <dim>-1</dim>
124
+ </port>
125
+ </input>
126
+ <output>
127
+ <port id="1" precision="I64" names="attention_mask">
128
+ <dim>-1</dim>
129
+ <dim>-1</dim>
130
+ </port>
131
+ </output>
132
+ </layer>
133
+ <layer id="11" name="Constant_175782" type="Const" version="opset1">
134
+ <data element_type="i32" shape="" offset="0" size="4" />
135
+ <output>
136
+ <port id="0" precision="I32" />
137
+ </output>
138
+ </layer>
139
+ <layer id="12" name="Broadcast_175783" type="Broadcast" version="opset3">
140
+ <data mode="numpy" />
141
+ <input>
142
+ <port id="0" precision="I32" />
143
+ <port id="1" precision="I64">
144
+ <dim>2</dim>
145
+ </port>
146
+ </input>
147
+ <output>
148
+ <port id="2" precision="I32">
149
+ <dim>-1</dim>
150
+ <dim>-1</dim>
151
+ </port>
152
+ </output>
153
+ </layer>
154
+ <layer id="13" name="ScatterNDUpdate_175784" type="ScatterNDUpdate" version="opset4">
155
+ <input>
156
+ <port id="0" precision="I32">
157
+ <dim>-1</dim>
158
+ <dim>-1</dim>
159
+ </port>
160
+ <port id="1" precision="I64">
161
+ <dim>-1</dim>
162
+ <dim>2</dim>
163
+ </port>
164
+ <port id="2" precision="I32">
165
+ <dim>-1</dim>
166
+ </port>
167
+ </input>
168
+ <output>
169
+ <port id="3" precision="I32">
170
+ <dim>-1</dim>
171
+ <dim>-1</dim>
172
+ </port>
173
+ </output>
174
+ </layer>
175
+ <layer id="14" name="ScatterNDUpdate_175784" type="Convert" version="opset1">
176
+ <data destination_type="i64" />
177
+ <input>
178
+ <port id="0" precision="I32">
179
+ <dim>-1</dim>
180
+ <dim>-1</dim>
181
+ </port>
182
+ </input>
183
+ <output>
184
+ <port id="1" precision="I64" names="input_ids">
185
+ <dim>-1</dim>
186
+ <dim>-1</dim>
187
+ </port>
188
+ </output>
189
+ </layer>
190
+ <layer id="15" name="Result_175794" type="Result" version="opset1">
191
+ <input>
192
+ <port id="0" precision="I64">
193
+ <dim>-1</dim>
194
+ <dim>-1</dim>
195
+ </port>
196
+ </input>
197
+ </layer>
198
+ <layer id="10" name="Result_175795" type="Result" version="opset1">
199
+ <input>
200
+ <port id="0" precision="I64">
201
+ <dim>-1</dim>
202
+ <dim>-1</dim>
203
+ </port>
204
+ </input>
205
+ </layer>
206
+ </layers>
207
+ <edges>
208
+ <edge from-layer="0" from-port="0" to-layer="3" to-port="1" />
209
+ <edge from-layer="1" from-port="0" to-layer="4" to-port="0" />
210
+ <edge from-layer="2" from-port="0" to-layer="3" to-port="0" />
211
+ <edge from-layer="3" from-port="4" to-layer="4" to-port="1" />
212
+ <edge from-layer="3" from-port="3" to-layer="6" to-port="0" />
213
+ <edge from-layer="3" from-port="2" to-layer="8" to-port="1" />
214
+ <edge from-layer="3" from-port="3" to-layer="13" to-port="2" />
215
+ <edge from-layer="3" from-port="2" to-layer="13" to-port="1" />
216
+ <edge from-layer="3" from-port="4" to-layer="12" to-port="1" />
217
+ <edge from-layer="4" from-port="2" to-layer="8" to-port="0" />
218
+ <edge from-layer="5" from-port="0" to-layer="7" to-port="0" />
219
+ <edge from-layer="6" from-port="1" to-layer="7" to-port="1" />
220
+ <edge from-layer="7" from-port="2" to-layer="8" to-port="2" />
221
+ <edge from-layer="8" from-port="3" to-layer="9" to-port="0" />
222
+ <edge from-layer="9" from-port="1" to-layer="10" to-port="0" />
223
+ <edge from-layer="11" from-port="0" to-layer="12" to-port="0" />
224
+ <edge from-layer="12" from-port="2" to-layer="13" to-port="0" />
225
+ <edge from-layer="13" from-port="3" to-layer="14" to-port="0" />
226
+ <edge from-layer="14" from-port="1" to-layer="15" to-port="0" />
227
+ </edges>
228
+ <rt_info>
229
+ <eos_token_id value="2" />
230
+ </rt_info>
231
+ </net>