Hugging Face's logo

sgao42
/
Phi-3.5-mini-instruct-OV

OpenVINO
phi3
custom_code
Model card Files Community
sgao42 commited on
Commit
29587dc
·
verified ·
1 Parent(s): 3e5e003

Upload 12 files

Browse files
Files changed (8) hide show
  1. config.json +105 -3
  2. openvino_detokenizer.bin +3 -0
  3. openvino_detokenizer.xml +121 -0
  4. openvino_model.bin +2 -2
  5. openvino_model.xml +0 -0
  6. openvino_tokenizer.bin +3 -0
  7. openvino_tokenizer.xml +311 -0
  8. tokenizer_config.json +0 -1
config.json CHANGED
@@ -16,7 +16,7 @@
16
  "hidden_size": 3072,
17
  "initializer_range": 0.02,
18
  "intermediate_size": 8192,
19
- "max_position_embeddings": 131072,
20
  "model_type": "phi3",
21
  "num_attention_heads": 32,
22
  "num_hidden_layers": 32,
@@ -25,12 +25,114 @@
25
  "pad_token_id": 32000,
26
  "resid_pdrop": 0.0,
27
  "rms_norm_eps": 1e-05,
28
- "rope_scaling": null,
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
29
  "rope_theta": 10000.0,
30
  "sliding_window": 262144,
31
  "tie_word_embeddings": false,
32
  "torch_dtype": "bfloat16",
33
- "transformers_version": "4.44.2",
34
  "use_cache": true,
35
  "vocab_size": 32064
36
  }
 
16
  "hidden_size": 3072,
17
  "initializer_range": 0.02,
18
  "intermediate_size": 8192,
19
+ "max_position_embeddings": 4096,
20
  "model_type": "phi3",
21
  "num_attention_heads": 32,
22
  "num_hidden_layers": 32,
 
25
  "pad_token_id": 32000,
26
  "resid_pdrop": 0.0,
27
  "rms_norm_eps": 1e-05,
28
+ "rope_scaling": {
29
+ "long_factor": [
30
+ 1.0800000429153442,
31
+ 1.1100000143051147,
32
+ 1.1399999856948853,
33
+ 1.340000033378601,
34
+ 1.5899999141693115,
35
+ 1.600000023841858,
36
+ 1.6200000047683716,
37
+ 2.620000123977661,
38
+ 3.2300000190734863,
39
+ 3.2300000190734863,
40
+ 4.789999961853027,
41
+ 7.400000095367432,
42
+ 7.700000286102295,
43
+ 9.09000015258789,
44
+ 12.199999809265137,
45
+ 17.670000076293945,
46
+ 24.46000099182129,
47
+ 28.57000160217285,
48
+ 30.420001983642578,
49
+ 30.840002059936523,
50
+ 32.590003967285156,
51
+ 32.93000411987305,
52
+ 42.320003509521484,
53
+ 44.96000289916992,
54
+ 50.340003967285156,
55
+ 50.45000457763672,
56
+ 57.55000305175781,
57
+ 57.93000411987305,
58
+ 58.21000289916992,
59
+ 60.1400032043457,
60
+ 62.61000442504883,
61
+ 62.62000274658203,
62
+ 62.71000289916992,
63
+ 63.1400032043457,
64
+ 63.1400032043457,
65
+ 63.77000427246094,
66
+ 63.93000411987305,
67
+ 63.96000289916992,
68
+ 63.970001220703125,
69
+ 64.02999877929688,
70
+ 64.06999969482422,
71
+ 64.08000183105469,
72
+ 64.12000274658203,
73
+ 64.41000366210938,
74
+ 64.4800033569336,
75
+ 64.51000213623047,
76
+ 64.52999877929688,
77
+ 64.83999633789062
78
+ ],
79
+ "short_factor": [
80
+ 1.0,
81
+ 1.0199999809265137,
82
+ 1.0299999713897705,
83
+ 1.0299999713897705,
84
+ 1.0499999523162842,
85
+ 1.0499999523162842,
86
+ 1.0499999523162842,
87
+ 1.0499999523162842,
88
+ 1.0499999523162842,
89
+ 1.0699999332427979,
90
+ 1.0999999046325684,
91
+ 1.1099998950958252,
92
+ 1.1599998474121094,
93
+ 1.1599998474121094,
94
+ 1.1699998378753662,
95
+ 1.2899998426437378,
96
+ 1.339999794960022,
97
+ 1.679999828338623,
98
+ 1.7899998426437378,
99
+ 1.8199998140335083,
100
+ 1.8499997854232788,
101
+ 1.8799997568130493,
102
+ 1.9099997282028198,
103
+ 1.9399996995925903,
104
+ 1.9899996519088745,
105
+ 2.0199997425079346,
106
+ 2.0199997425079346,
107
+ 2.0199997425079346,
108
+ 2.0199997425079346,
109
+ 2.0199997425079346,
110
+ 2.0199997425079346,
111
+ 2.0299997329711914,
112
+ 2.0299997329711914,
113
+ 2.0299997329711914,
114
+ 2.0299997329711914,
115
+ 2.0299997329711914,
116
+ 2.0299997329711914,
117
+ 2.0299997329711914,
118
+ 2.0299997329711914,
119
+ 2.0299997329711914,
120
+ 2.0799996852874756,
121
+ 2.0899996757507324,
122
+ 2.189999580383301,
123
+ 2.2199995517730713,
124
+ 2.5899994373321533,
125
+ 2.729999542236328,
126
+ 2.749999523162842,
127
+ 2.8399994373321533
128
+ ],
129
+ "type": "longrope"
130
+ },
131
  "rope_theta": 10000.0,
132
  "sliding_window": 262144,
133
  "tie_word_embeddings": false,
134
  "torch_dtype": "bfloat16",
135
+ "transformers_version": "4.40.2",
136
  "use_cache": true,
137
  "vocab_size": 32064
138
  }
openvino_detokenizer.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:34f54bcfdecf199ed7333d7748c8e357c9d98b26c0c5800c7e0809e48edd74d3
3
+ size 499991
openvino_detokenizer.xml ADDED
@@ -0,0 +1,121 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ <?xml version="1.0"?>
2
+ <net name="detokenizer" version="11">
3
+ <layers>
4
+ <layer id="0" name="Parameter_24" type="Parameter" version="opset1">
5
+ <data shape="?,?" element_type="i64" />
6
+ <output>
7
+ <port id="0" precision="I64" names="Parameter_24">
8
+ <dim>-1</dim>
9
+ <dim>-1</dim>
10
+ </port>
11
+ </output>
12
+ </layer>
13
+ <layer id="1" name="Constant_2" type="Const" version="opset1">
14
+ <data element_type="u8" shape="499991" offset="0" size="499991" />
15
+ <output>
16
+ <port id="0" precision="U8">
17
+ <dim>499991</dim>
18
+ </port>
19
+ </output>
20
+ </layer>
21
+ <layer id="2" name="Convert_34" 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_25" type="SentencepieceDetokenizer" version="extension">
37
+ <input>
38
+ <port id="0" precision="U8">
39
+ <dim>499991</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_26" 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_27" 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
+ <add_attention_mask value="True" />
96
+ <add_prefix_space value="True" />
97
+ <add_special_tokens value="True" />
98
+ <bos_token_id value="1" />
99
+ <chat_template value="{% for message in messages %}{% if message['role'] == 'system' and message['content'] %}{{'&lt;|system|>&#10;' + message['content'] + '&lt;|end|>&#10;'}}{% elif message['role'] == 'user' %}{{'&lt;|user|>&#10;' + message['content'] + '&lt;|end|>&#10;'}}{% elif message['role'] == 'assistant' %}{{'&lt;|assistant|>&#10;' + message['content'] + '&lt;|end|>&#10;'}}{% endif %}{% endfor %}{% if add_generation_prompt %}{{ '&lt;|assistant|>&#10;' }}{% else %}{{ eos_token }}{% endif %}" />
100
+ <clean_up_tokenization_spaces value="False" />
101
+ <detokenizer_input_type value="i64" />
102
+ <eos_token_id value="32000" />
103
+ <handle_special_tokens_with_re value="False" />
104
+ <number_of_inputs value="1" />
105
+ <openvino_tokenizers_version value="2024.5.0.0" />
106
+ <openvino_version value="2024.5.0" />
107
+ <original_tokenizer_class value="&lt;class 'transformers.models.llama.tokenization_llama_fast.LlamaTokenizerFast'>" />
108
+ <pad_token_id value="32000" />
109
+ <sentencepiece_version value="0.2.0" />
110
+ <skip_special_tokens value="True" />
111
+ <streaming_detokenizer value="False" />
112
+ <tiktoken_version value="0.8.0" />
113
+ <tokenizer_output_type value="i64" />
114
+ <tokenizers_version value="0.19.1" />
115
+ <transformers_version value="4.40.2" />
116
+ <use_max_padding value="False" />
117
+ <use_sentencepiece_backend value="True" />
118
+ <utf8_replace_mode />
119
+ <with_detokenizer value="True" />
120
+ </rt_info>
121
+ </net>
openvino_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:796a512329183abddd3807c9ec13b821e47d734fb2a5fce421e646f2f8cf1797
3
- size 2066155732
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6a357cdfff637e5f6ed7b833046f591ae1dc5b03e65622b2fd02ab966d943869
3
+ size 2066155856
openvino_model.xml CHANGED
The diff for this file is too large to render. See raw diff
 
openvino_tokenizer.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:27a382f4cd1318883609d060c54fd98f5d98c02b94062e129e901a37595eedad
3
+ size 499989
openvino_tokenizer.xml ADDED
@@ -0,0 +1,311 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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_5" 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_1" type="Const" version="opset1">
19
+ <data element_type="u8" shape="499969" offset="4" size="499969" />
20
+ <output>
21
+ <port id="0" precision="U8">
22
+ <dim>499969</dim>
23
+ </port>
24
+ </output>
25
+ </layer>
26
+ <layer id="3" name="SentencepieceTokenizer_4" type="SentencepieceTokenizer" version="extension">
27
+ <data nbest_size="0" alpha="0" add_bos="false" add_eos="false" reverse="true" />
28
+ <input>
29
+ <port id="0" precision="U8">
30
+ <dim>499969</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_6" 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_7" type="Const" version="opset1">
65
+ <data element_type="i32" shape="" offset="499973" size="4" />
66
+ <output>
67
+ <port id="0" precision="I32" />
68
+ </output>
69
+ </layer>
70
+ <layer id="6" name="ShapeOf_8" 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_9" 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_12" 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="Constant_16" type="Const" version="opset1">
119
+ <data element_type="i64" shape="1" offset="499977" size="8" />
120
+ <output>
121
+ <port id="0" precision="I64">
122
+ <dim>1</dim>
123
+ </port>
124
+ </output>
125
+ </layer>
126
+ <layer id="10" name="Reverse_17" type="Reverse" version="opset1">
127
+ <data mode="index" />
128
+ <input>
129
+ <port id="0" precision="I32">
130
+ <dim>-1</dim>
131
+ <dim>-1</dim>
132
+ </port>
133
+ <port id="1" precision="I64">
134
+ <dim>1</dim>
135
+ </port>
136
+ </input>
137
+ <output>
138
+ <port id="2" precision="I32">
139
+ <dim>-1</dim>
140
+ <dim>-1</dim>
141
+ </port>
142
+ </output>
143
+ </layer>
144
+ <layer id="11" name="Reverse_17" type="Convert" version="opset1">
145
+ <data destination_type="i64" />
146
+ <input>
147
+ <port id="0" precision="I32">
148
+ <dim>-1</dim>
149
+ <dim>-1</dim>
150
+ </port>
151
+ </input>
152
+ <output>
153
+ <port id="1" precision="I64" names="attention_mask">
154
+ <dim>-1</dim>
155
+ <dim>-1</dim>
156
+ </port>
157
+ </output>
158
+ </layer>
159
+ <layer id="13" name="Constant_13" type="Const" version="opset1">
160
+ <data element_type="i32" shape="" offset="499985" size="4" />
161
+ <output>
162
+ <port id="0" precision="I32" />
163
+ </output>
164
+ </layer>
165
+ <layer id="14" name="Broadcast_14" type="Broadcast" version="opset3">
166
+ <data mode="bidirectional" />
167
+ <input>
168
+ <port id="0" precision="I32" />
169
+ <port id="1" precision="I64">
170
+ <dim>2</dim>
171
+ </port>
172
+ </input>
173
+ <output>
174
+ <port id="2" precision="I32">
175
+ <dim>-1</dim>
176
+ <dim>-1</dim>
177
+ </port>
178
+ </output>
179
+ </layer>
180
+ <layer id="15" name="ScatterNDUpdate_15" type="ScatterNDUpdate" version="opset4">
181
+ <input>
182
+ <port id="0" precision="I32">
183
+ <dim>-1</dim>
184
+ <dim>-1</dim>
185
+ </port>
186
+ <port id="1" precision="I64">
187
+ <dim>-1</dim>
188
+ <dim>2</dim>
189
+ </port>
190
+ <port id="2" precision="I32">
191
+ <dim>-1</dim>
192
+ </port>
193
+ </input>
194
+ <output>
195
+ <port id="3" precision="I32">
196
+ <dim>-1</dim>
197
+ <dim>-1</dim>
198
+ </port>
199
+ </output>
200
+ </layer>
201
+ <layer id="16" name="Constant_18" type="Const" version="opset1">
202
+ <data element_type="i64" shape="1" offset="499977" size="8" />
203
+ <output>
204
+ <port id="0" precision="I64">
205
+ <dim>1</dim>
206
+ </port>
207
+ </output>
208
+ </layer>
209
+ <layer id="17" name="Reverse_19" type="Reverse" version="opset1">
210
+ <data mode="index" />
211
+ <input>
212
+ <port id="0" precision="I32">
213
+ <dim>-1</dim>
214
+ <dim>-1</dim>
215
+ </port>
216
+ <port id="1" precision="I64">
217
+ <dim>1</dim>
218
+ </port>
219
+ </input>
220
+ <output>
221
+ <port id="2" precision="I32">
222
+ <dim>-1</dim>
223
+ <dim>-1</dim>
224
+ </port>
225
+ </output>
226
+ </layer>
227
+ <layer id="18" name="Reverse_19" type="Convert" version="opset1">
228
+ <data destination_type="i64" />
229
+ <input>
230
+ <port id="0" precision="I32">
231
+ <dim>-1</dim>
232
+ <dim>-1</dim>
233
+ </port>
234
+ </input>
235
+ <output>
236
+ <port id="1" precision="I64" names="input_ids">
237
+ <dim>-1</dim>
238
+ <dim>-1</dim>
239
+ </port>
240
+ </output>
241
+ </layer>
242
+ <layer id="19" name="Result_20" type="Result" version="opset1">
243
+ <input>
244
+ <port id="0" precision="I64">
245
+ <dim>-1</dim>
246
+ <dim>-1</dim>
247
+ </port>
248
+ </input>
249
+ </layer>
250
+ <layer id="12" name="Result_21" type="Result" version="opset1">
251
+ <input>
252
+ <port id="0" precision="I64">
253
+ <dim>-1</dim>
254
+ <dim>-1</dim>
255
+ </port>
256
+ </input>
257
+ </layer>
258
+ </layers>
259
+ <edges>
260
+ <edge from-layer="0" from-port="0" to-layer="3" to-port="1" />
261
+ <edge from-layer="1" from-port="0" to-layer="4" to-port="0" />
262
+ <edge from-layer="2" from-port="0" to-layer="3" to-port="0" />
263
+ <edge from-layer="3" from-port="4" to-layer="4" to-port="1" />
264
+ <edge from-layer="3" from-port="3" to-layer="6" to-port="0" />
265
+ <edge from-layer="3" from-port="2" to-layer="8" to-port="1" />
266
+ <edge from-layer="3" from-port="4" to-layer="14" to-port="1" />
267
+ <edge from-layer="3" from-port="2" to-layer="15" to-port="1" />
268
+ <edge from-layer="3" from-port="3" to-layer="15" to-port="2" />
269
+ <edge from-layer="4" from-port="2" to-layer="8" to-port="0" />
270
+ <edge from-layer="5" from-port="0" to-layer="7" to-port="0" />
271
+ <edge from-layer="6" from-port="1" to-layer="7" to-port="1" />
272
+ <edge from-layer="7" from-port="2" to-layer="8" to-port="2" />
273
+ <edge from-layer="8" from-port="3" to-layer="10" to-port="0" />
274
+ <edge from-layer="9" from-port="0" to-layer="10" to-port="1" />
275
+ <edge from-layer="10" from-port="2" to-layer="11" to-port="0" />
276
+ <edge from-layer="11" from-port="1" to-layer="12" to-port="0" />
277
+ <edge from-layer="13" from-port="0" to-layer="14" to-port="0" />
278
+ <edge from-layer="14" from-port="2" to-layer="15" to-port="0" />
279
+ <edge from-layer="15" from-port="3" to-layer="17" to-port="0" />
280
+ <edge from-layer="16" from-port="0" to-layer="17" to-port="1" />
281
+ <edge from-layer="17" from-port="2" to-layer="18" to-port="0" />
282
+ <edge from-layer="18" from-port="1" to-layer="19" to-port="0" />
283
+ </edges>
284
+ <rt_info>
285
+ <add_attention_mask value="True" />
286
+ <add_prefix_space value="True" />
287
+ <add_special_tokens value="True" />
288
+ <bos_token_id value="1" />
289
+ <chat_template value="{% for message in messages %}{% if message['role'] == 'system' and message['content'] %}{{'&lt;|system|>&#10;' + message['content'] + '&lt;|end|>&#10;'}}{% elif message['role'] == 'user' %}{{'&lt;|user|>&#10;' + message['content'] + '&lt;|end|>&#10;'}}{% elif message['role'] == 'assistant' %}{{'&lt;|assistant|>&#10;' + message['content'] + '&lt;|end|>&#10;'}}{% endif %}{% endfor %}{% if add_generation_prompt %}{{ '&lt;|assistant|>&#10;' }}{% else %}{{ eos_token }}{% endif %}" />
290
+ <clean_up_tokenization_spaces value="False" />
291
+ <detokenizer_input_type value="i64" />
292
+ <eos_token_id value="32000" />
293
+ <handle_special_tokens_with_re value="False" />
294
+ <number_of_inputs value="1" />
295
+ <openvino_tokenizers_version value="2024.5.0.0" />
296
+ <openvino_version value="2024.5.0" />
297
+ <original_tokenizer_class value="&lt;class 'transformers.models.llama.tokenization_llama_fast.LlamaTokenizerFast'>" />
298
+ <pad_token_id value="32000" />
299
+ <sentencepiece_version value="0.2.0" />
300
+ <skip_special_tokens value="True" />
301
+ <streaming_detokenizer value="False" />
302
+ <tiktoken_version value="0.8.0" />
303
+ <tokenizer_output_type value="i64" />
304
+ <tokenizers_version value="0.19.1" />
305
+ <transformers_version value="4.40.2" />
306
+ <use_max_padding value="False" />
307
+ <use_sentencepiece_backend value="True" />
308
+ <utf8_replace_mode />
309
+ <with_detokenizer value="True" />
310
+ </rt_info>
311
+ </net>
tokenizer_config.json CHANGED
@@ -1,7 +1,6 @@
1
  {
2
  "add_bos_token": false,
3
  "add_eos_token": false,
4
- "add_prefix_space": null,
5
  "added_tokens_decoder": {
6
  "0": {
7
  "content": "<unk>",
 
1
  {
2
  "add_bos_token": false,
3
  "add_eos_token": false,
 
4
  "added_tokens_decoder": {
5
  "0": {
6
  "content": "<unk>",