Initial commit
Browse files- README.md +285 -3
- config.json +223 -0
- merges.txt +0 -0
- model.safetensors +3 -0
- special_tokens_map.json +6 -0
- tokenizer.json +0 -0
- tokenizer_config.json +20 -0
- vocab.json +0 -0
README.md
CHANGED
@@ -1,3 +1,285 @@
|
|
1 |
-
---
|
2 |
-
license: mit
|
3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: mit
|
3 |
+
base_model: openai-community/gpt2
|
4 |
+
tags:
|
5 |
+
- generated_from_trainer
|
6 |
+
metrics:
|
7 |
+
- accuracy
|
8 |
+
model-index:
|
9 |
+
- name: gpt2-lang-ident
|
10 |
+
results: []
|
11 |
+
---
|
12 |
+
|
13 |
+
# gpt2-lang-ident
|
14 |
+
|
15 |
+
This model is a fine-tuned version of [openai-community/gpt2](https://huggingface.co/openai-community/gpt2) on sampled sentences from `stanford-oval/ccnews` and `qanastek/EMEA-V3` datasets.
|
16 |
+
It achieves the following results on the evaluation set:
|
17 |
+
- Loss: 0.1210
|
18 |
+
- Accuracy: 0.9721
|
19 |
+
|
20 |
+
## Model description
|
21 |
+
|
22 |
+
This model is trained to predict the language of the input text.
|
23 |
+
|
24 |
+
## Intended uses & limitations
|
25 |
+
|
26 |
+
The model can predict the following languages:
|
27 |
+
```
|
28 |
+
["af", "am", "ar", "as", "az", "ba", "be", "bg", "bn", "ca", "ceb", "ckb", "cs", "cy", "da", "de", "dv", "el", "en", "eo", "es", "et", "eu", "fa", "fi", "fr", "fy", "ga", "gd", "gl", "gu", "he", "hi", "hr", "hu", "hy", "id", "is", "it", "ja", "ka", "kk", "kn", "ku", "ky", "la", "lb", "lt", "lv", "mg", "mk", "ml", "mn", "mr", "mt", "my", "nds", "ne", "nl", "nn", "no", "or", "pa", "pl", "ps", "pt", "ro", "ru", "sah", "sd", "si", "sk", "sl", "sq", "sr", "sv", "sw", "ta", "te", "tg", "th", "tk", "tl", "tr", "tt", "ug", "uk", "ur", "vi", "yi"]
|
29 |
+
```
|
30 |
+
|
31 |
+
How to use:
|
32 |
+
|
33 |
+
```python
|
34 |
+
from transformers import (AutoModelForSequenceClassification, AutoTokenizer,
|
35 |
+
pipeline)
|
36 |
+
|
37 |
+
checkpoint = f"nie3e/gpt2-lang-ident"
|
38 |
+
model = AutoModelForSequenceClassification.from_pretrained(checkpoint)
|
39 |
+
tokenizer = AutoTokenizer.from_pretrained(checkpoint)
|
40 |
+
|
41 |
+
pipe = pipeline(
|
42 |
+
task="text-classification",
|
43 |
+
model=model,
|
44 |
+
tokenizer=tokenizer,
|
45 |
+
top_k=5
|
46 |
+
)
|
47 |
+
|
48 |
+
result = pipe("To jest model służący do identyfikacji języka!")
|
49 |
+
print(result)
|
50 |
+
```
|
51 |
+
```
|
52 |
+
[[{'label': 'pl', 'score': 0.9999653100967407}, {'label': 'sr', 'score': 1.5228776646836195e-05}, {'label': 'hr', 'score': 1.057955432770541e-05}, {'label': 'bn', 'score': 1.590750912328076e-06}, {'label': 'cs', 'score': 1.3942196801508544e-06}]]
|
53 |
+
```
|
54 |
+
|
55 |
+
## Training and evaluation data
|
56 |
+
|
57 |
+
<details><summary>Training data ([lang]: count)</summary>
|
58 |
+
|
59 |
+
[bn]: 1947
|
60 |
+
[ar]: 1947
|
61 |
+
[vi]: 1947
|
62 |
+
[uk]: 1947
|
63 |
+
[kn]: 1947
|
64 |
+
[mr]: 1947
|
65 |
+
[id]: 1947
|
66 |
+
[te]: 1947
|
67 |
+
[no]: 1947
|
68 |
+
[ru]: 1947
|
69 |
+
[he]: 1947
|
70 |
+
[az]: 1947
|
71 |
+
[ca]: 1946
|
72 |
+
[fa]: 1946
|
73 |
+
[hi]: 1946
|
74 |
+
[th]: 1946
|
75 |
+
[tr]: 1946
|
76 |
+
[mk]: 1946
|
77 |
+
[ta]: 1945
|
78 |
+
[sq]: 1945
|
79 |
+
[ur]: 1942
|
80 |
+
[gu]: 1939
|
81 |
+
[ml]: 1936
|
82 |
+
[is]: 1738
|
83 |
+
[de]: 1543
|
84 |
+
[da]: 1521
|
85 |
+
[fi]: 1461
|
86 |
+
[el]: 1431
|
87 |
+
[nl]: 1424
|
88 |
+
[fr]: 1408
|
89 |
+
[cs]: 1401
|
90 |
+
[es]: 1397
|
91 |
+
[en]: 1394
|
92 |
+
[lt]: 1392
|
93 |
+
[hu]: 1379
|
94 |
+
[pt]: 1375
|
95 |
+
[lv]: 1373
|
96 |
+
[it]: 1360
|
97 |
+
[pl]: 1355
|
98 |
+
[sk]: 1355
|
99 |
+
[et]: 1348
|
100 |
+
[sl]: 1328
|
101 |
+
[sv]: 1300
|
102 |
+
[bg]: 1278
|
103 |
+
[mt]: 1234
|
104 |
+
[ro]: 1218
|
105 |
+
[kk]: 1179
|
106 |
+
[hy]: 1176
|
107 |
+
[or]: 1112
|
108 |
+
[pa]: 780
|
109 |
+
[sr]: 744
|
110 |
+
[as]: 735
|
111 |
+
[hr]: 722
|
112 |
+
[ne]: 626
|
113 |
+
[gl]: 566
|
114 |
+
[ckb]: 563
|
115 |
+
[ka]: 560
|
116 |
+
[ug]: 485
|
117 |
+
[ky]: 453
|
118 |
+
[eu]: 351
|
119 |
+
[ps]: 311
|
120 |
+
[tl]: 307
|
121 |
+
[fy]: 290
|
122 |
+
[mn]: 289
|
123 |
+
[si]: 244
|
124 |
+
[cy]: 214
|
125 |
+
[nn]: 212
|
126 |
+
[ku]: 195
|
127 |
+
[tg]: 176
|
128 |
+
[am]: 141
|
129 |
+
[tt]: 121
|
130 |
+
[ja]: 104
|
131 |
+
[lb]: 93
|
132 |
+
[tk]: 72
|
133 |
+
[be]: 64
|
134 |
+
[sw]: 45
|
135 |
+
[af]: 44
|
136 |
+
[my]: 40
|
137 |
+
[ceb]: 35
|
138 |
+
[la]: 33
|
139 |
+
[dv]: 20
|
140 |
+
[ba]: 19
|
141 |
+
[ga]: 19
|
142 |
+
[eo]: 19
|
143 |
+
[gd]: 16
|
144 |
+
[mg]: 15
|
145 |
+
[yi]: 14
|
146 |
+
[sah]: 14
|
147 |
+
[sd]: 11
|
148 |
+
[nds]: 11
|
149 |
+
</details>
|
150 |
+
|
151 |
+
<details><summary>Eval data ([lang]: count)</summary>
|
152 |
+
|
153 |
+
[te]: 195
|
154 |
+
[mk]: 195
|
155 |
+
[bn]: 195
|
156 |
+
[uk]: 195
|
157 |
+
[hi]: 195
|
158 |
+
[ar]: 195
|
159 |
+
[sq]: 195
|
160 |
+
[kn]: 195
|
161 |
+
[tr]: 195
|
162 |
+
[ca]: 195
|
163 |
+
[az]: 195
|
164 |
+
[fa]: 195
|
165 |
+
[ru]: 195
|
166 |
+
[mr]: 195
|
167 |
+
[id]: 195
|
168 |
+
[no]: 195
|
169 |
+
[vi]: 195
|
170 |
+
[th]: 195
|
171 |
+
[he]: 195
|
172 |
+
[gu]: 194
|
173 |
+
[ml]: 194
|
174 |
+
[ta]: 194
|
175 |
+
[ur]: 194
|
176 |
+
[is]: 174
|
177 |
+
[de]: 154
|
178 |
+
[da]: 152
|
179 |
+
[fi]: 146
|
180 |
+
[el]: 143
|
181 |
+
[nl]: 142
|
182 |
+
[fr]: 141
|
183 |
+
[es]: 140
|
184 |
+
[cs]: 140
|
185 |
+
[en]: 139
|
186 |
+
[lt]: 139
|
187 |
+
[hu]: 138
|
188 |
+
[lv]: 137
|
189 |
+
[pt]: 137
|
190 |
+
[it]: 136
|
191 |
+
[et]: 135
|
192 |
+
[pl]: 135
|
193 |
+
[sk]: 135
|
194 |
+
[sl]: 133
|
195 |
+
[sv]: 130
|
196 |
+
[bg]: 128
|
197 |
+
[mt]: 123
|
198 |
+
[ro]: 122
|
199 |
+
[hy]: 118
|
200 |
+
[kk]: 118
|
201 |
+
[or]: 111
|
202 |
+
[pa]: 78
|
203 |
+
[sr]: 74
|
204 |
+
[as]: 74
|
205 |
+
[hr]: 72
|
206 |
+
[ne]: 63
|
207 |
+
[gl]: 57
|
208 |
+
[ckb]: 56
|
209 |
+
[ka]: 56
|
210 |
+
[ug]: 49
|
211 |
+
[ky]: 45
|
212 |
+
[eu]: 35
|
213 |
+
[ps]: 31
|
214 |
+
[tl]: 31
|
215 |
+
[mn]: 29
|
216 |
+
[fy]: 29
|
217 |
+
[si]: 24
|
218 |
+
[nn]: 21
|
219 |
+
[cy]: 21
|
220 |
+
[ku]: 19
|
221 |
+
[tg]: 18
|
222 |
+
[am]: 14
|
223 |
+
[tt]: 12
|
224 |
+
[ja]: 10
|
225 |
+
[lb]: 9
|
226 |
+
[tk]: 7
|
227 |
+
[be]: 6
|
228 |
+
[my]: 4
|
229 |
+
[sw]: 4
|
230 |
+
[af]: 4
|
231 |
+
[ceb]: 3
|
232 |
+
[la]: 3
|
233 |
+
[ba]: 2
|
234 |
+
[dv]: 2
|
235 |
+
[eo]: 2
|
236 |
+
[gd]: 2
|
237 |
+
[ga]: 2
|
238 |
+
[mg]: 1
|
239 |
+
[sd]: 1
|
240 |
+
[nds]: 1
|
241 |
+
[yi]: 1
|
242 |
+
[sah]: 1
|
243 |
+
</details>
|
244 |
+
|
245 |
+
### Training procedure
|
246 |
+
|
247 |
+
GPU: RTX 3090 \
|
248 |
+
Training time: 1h 53min
|
249 |
+
|
250 |
+
### Training hyperparameters
|
251 |
+
|
252 |
+
The following hyperparameters were used during training:
|
253 |
+
- learning_rate: 2e-05
|
254 |
+
- train_batch_size: 8
|
255 |
+
- eval_batch_size: 4
|
256 |
+
- seed: 42
|
257 |
+
- gradient_accumulation_steps: 4
|
258 |
+
- total_train_batch_size: 32
|
259 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
260 |
+
- lr_scheduler_type: linear
|
261 |
+
- num_epochs: 10
|
262 |
+
- mixed_precision_training: Native AMP
|
263 |
+
|
264 |
+
### Training results
|
265 |
+
|
266 |
+
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
267 |
+
|:-------------:|:-----:|:-----:|:---------------:|:--------:|
|
268 |
+
| 0.2833 | 1.0 | 2812 | 0.2004 | 0.94 |
|
269 |
+
| 0.168 | 2.0 | 5625 | 0.1567 | 0.954 |
|
270 |
+
| 0.1131 | 3.0 | 8437 | 0.1429 | 0.9586 |
|
271 |
+
| 0.0832 | 4.0 | 11250 | 0.1257 | 0.967 |
|
272 |
+
| 0.0635 | 5.0 | 14062 | 0.1222 | 0.9682 |
|
273 |
+
| 0.0479 | 6.0 | 16875 | 0.1214 | 0.9704 |
|
274 |
+
| 0.0361 | 7.0 | 19687 | 0.1255 | 0.9712 |
|
275 |
+
| 0.0258 | 8.0 | 22500 | 0.1178 | 0.9712 |
|
276 |
+
| 0.0243 | 9.0 | 25312 | 0.1223 | 0.9724 |
|
277 |
+
| 0.0171 | 10.0 | 28120 | 0.1210 | 0.9721 |
|
278 |
+
|
279 |
+
|
280 |
+
### Framework versions
|
281 |
+
|
282 |
+
- Transformers 4.36.2
|
283 |
+
- Pytorch 2.1.2+cu121
|
284 |
+
- Datasets 2.16.1
|
285 |
+
- Tokenizers 0.15.0
|
config.json
ADDED
@@ -0,0 +1,223 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "openai-community/gpt2",
|
3 |
+
"activation_function": "gelu_new",
|
4 |
+
"architectures": [
|
5 |
+
"GPT2ForSequenceClassification"
|
6 |
+
],
|
7 |
+
"attn_pdrop": 0.1,
|
8 |
+
"bos_token_id": 50256,
|
9 |
+
"embd_pdrop": 0.1,
|
10 |
+
"eos_token_id": 50256,
|
11 |
+
"id2label": {"0": "af",
|
12 |
+
"1": "am",
|
13 |
+
"2": "ar",
|
14 |
+
"3": "as",
|
15 |
+
"4": "az",
|
16 |
+
"5": "ba",
|
17 |
+
"6": "be",
|
18 |
+
"7": "bg",
|
19 |
+
"8": "bn",
|
20 |
+
"9": "ca",
|
21 |
+
"10": "ceb",
|
22 |
+
"11": "ckb",
|
23 |
+
"12": "cs",
|
24 |
+
"13": "cy",
|
25 |
+
"14": "da",
|
26 |
+
"15": "de",
|
27 |
+
"16": "dv",
|
28 |
+
"17": "el",
|
29 |
+
"18": "en",
|
30 |
+
"19": "eo",
|
31 |
+
"20": "es",
|
32 |
+
"21": "et",
|
33 |
+
"22": "eu",
|
34 |
+
"23": "fa",
|
35 |
+
"24": "fi",
|
36 |
+
"25": "fr",
|
37 |
+
"26": "fy",
|
38 |
+
"27": "ga",
|
39 |
+
"28": "gd",
|
40 |
+
"29": "gl",
|
41 |
+
"30": "gu",
|
42 |
+
"31": "he",
|
43 |
+
"32": "hi",
|
44 |
+
"33": "hr",
|
45 |
+
"34": "hu",
|
46 |
+
"35": "hy",
|
47 |
+
"36": "id",
|
48 |
+
"37": "is",
|
49 |
+
"38": "it",
|
50 |
+
"39": "ja",
|
51 |
+
"40": "ka",
|
52 |
+
"41": "kk",
|
53 |
+
"42": "kn",
|
54 |
+
"43": "ku",
|
55 |
+
"44": "ky",
|
56 |
+
"45": "la",
|
57 |
+
"46": "lb",
|
58 |
+
"47": "lt",
|
59 |
+
"48": "lv",
|
60 |
+
"49": "mg",
|
61 |
+
"50": "mk",
|
62 |
+
"51": "ml",
|
63 |
+
"52": "mn",
|
64 |
+
"53": "mr",
|
65 |
+
"54": "mt",
|
66 |
+
"55": "my",
|
67 |
+
"56": "nds",
|
68 |
+
"57": "ne",
|
69 |
+
"58": "nl",
|
70 |
+
"59": "nn",
|
71 |
+
"60": "no",
|
72 |
+
"61": "or",
|
73 |
+
"62": "pa",
|
74 |
+
"63": "pl",
|
75 |
+
"64": "ps",
|
76 |
+
"65": "pt",
|
77 |
+
"66": "ro",
|
78 |
+
"67": "ru",
|
79 |
+
"68": "sah",
|
80 |
+
"69": "sd",
|
81 |
+
"70": "si",
|
82 |
+
"71": "sk",
|
83 |
+
"72": "sl",
|
84 |
+
"73": "sq",
|
85 |
+
"74": "sr",
|
86 |
+
"75": "sv",
|
87 |
+
"76": "sw",
|
88 |
+
"77": "ta",
|
89 |
+
"78": "te",
|
90 |
+
"79": "tg",
|
91 |
+
"80": "th",
|
92 |
+
"81": "tk",
|
93 |
+
"82": "tl",
|
94 |
+
"83": "tr",
|
95 |
+
"84": "tt",
|
96 |
+
"85": "ug",
|
97 |
+
"86": "uk",
|
98 |
+
"87": "ur",
|
99 |
+
"88": "vi",
|
100 |
+
"89": "yi"},
|
101 |
+
"initializer_range": 0.02,
|
102 |
+
"label2id": {
|
103 |
+
"af": 0,
|
104 |
+
"am": 1,
|
105 |
+
"ar": 2,
|
106 |
+
"as": 3,
|
107 |
+
"az": 4,
|
108 |
+
"ba": 5,
|
109 |
+
"be": 6,
|
110 |
+
"bg": 7,
|
111 |
+
"bn": 8,
|
112 |
+
"ca": 9,
|
113 |
+
"ceb": 10,
|
114 |
+
"ckb": 11,
|
115 |
+
"cs": 12,
|
116 |
+
"cy": 13,
|
117 |
+
"da": 14,
|
118 |
+
"de": 15,
|
119 |
+
"dv": 16,
|
120 |
+
"el": 17,
|
121 |
+
"en": 18,
|
122 |
+
"eo": 19,
|
123 |
+
"es": 20,
|
124 |
+
"et": 21,
|
125 |
+
"eu": 22,
|
126 |
+
"fa": 23,
|
127 |
+
"fi": 24,
|
128 |
+
"fr": 25,
|
129 |
+
"fy": 26,
|
130 |
+
"ga": 27,
|
131 |
+
"gd": 28,
|
132 |
+
"gl": 29,
|
133 |
+
"gu": 30,
|
134 |
+
"he": 31,
|
135 |
+
"hi": 32,
|
136 |
+
"hr": 33,
|
137 |
+
"hu": 34,
|
138 |
+
"hy": 35,
|
139 |
+
"id": 36,
|
140 |
+
"is": 37,
|
141 |
+
"it": 38,
|
142 |
+
"ja": 39,
|
143 |
+
"ka": 40,
|
144 |
+
"kk": 41,
|
145 |
+
"kn": 42,
|
146 |
+
"ku": 43,
|
147 |
+
"ky": 44,
|
148 |
+
"la": 45,
|
149 |
+
"lb": 46,
|
150 |
+
"lt": 47,
|
151 |
+
"lv": 48,
|
152 |
+
"mg": 49,
|
153 |
+
"mk": 50,
|
154 |
+
"ml": 51,
|
155 |
+
"mn": 52,
|
156 |
+
"mr": 53,
|
157 |
+
"mt": 54,
|
158 |
+
"my": 55,
|
159 |
+
"nds": 56,
|
160 |
+
"ne": 57,
|
161 |
+
"nl": 58,
|
162 |
+
"nn": 59,
|
163 |
+
"no": 60,
|
164 |
+
"or": 61,
|
165 |
+
"pa": 62,
|
166 |
+
"pl": 63,
|
167 |
+
"ps": 64,
|
168 |
+
"pt": 65,
|
169 |
+
"ro": 66,
|
170 |
+
"ru": 67,
|
171 |
+
"sah": 68,
|
172 |
+
"sd": 69,
|
173 |
+
"si": 70,
|
174 |
+
"sk": 71,
|
175 |
+
"sl": 72,
|
176 |
+
"sq": 73,
|
177 |
+
"sr": 74,
|
178 |
+
"sv": 75,
|
179 |
+
"sw": 76,
|
180 |
+
"ta": 77,
|
181 |
+
"te": 78,
|
182 |
+
"tg": 79,
|
183 |
+
"th": 80,
|
184 |
+
"tk": 81,
|
185 |
+
"tl": 82,
|
186 |
+
"tr": 83,
|
187 |
+
"tt": 84,
|
188 |
+
"ug": 85,
|
189 |
+
"uk": 86,
|
190 |
+
"ur": 87,
|
191 |
+
"vi": 88,
|
192 |
+
"yi": 89
|
193 |
+
},
|
194 |
+
"layer_norm_epsilon": 1e-05,
|
195 |
+
"model_type": "gpt2",
|
196 |
+
"n_ctx": 1024,
|
197 |
+
"n_embd": 768,
|
198 |
+
"n_head": 12,
|
199 |
+
"n_inner": null,
|
200 |
+
"n_layer": 12,
|
201 |
+
"n_positions": 1024,
|
202 |
+
"pad_token_id": 50256,
|
203 |
+
"problem_type": "single_label_classification",
|
204 |
+
"reorder_and_upcast_attn": false,
|
205 |
+
"resid_pdrop": 0.1,
|
206 |
+
"scale_attn_by_inverse_layer_idx": false,
|
207 |
+
"scale_attn_weights": true,
|
208 |
+
"summary_activation": null,
|
209 |
+
"summary_first_dropout": 0.1,
|
210 |
+
"summary_proj_to_labels": true,
|
211 |
+
"summary_type": "cls_index",
|
212 |
+
"summary_use_proj": true,
|
213 |
+
"task_specific_params": {
|
214 |
+
"text-generation": {
|
215 |
+
"do_sample": true,
|
216 |
+
"max_length": 50
|
217 |
+
}
|
218 |
+
},
|
219 |
+
"torch_dtype": "float32",
|
220 |
+
"transformers_version": "4.36.2",
|
221 |
+
"use_cache": true,
|
222 |
+
"vocab_size": 50257
|
223 |
+
}
|
merges.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:041d6a825692c31bad97df5b268b6f8ab6976cd971c33366fbf0c738fa61546a
|
3 |
+
size 498050768
|
special_tokens_map.json
ADDED
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"bos_token": "<|endoftext|>",
|
3 |
+
"eos_token": "<|endoftext|>",
|
4 |
+
"pad_token": "<|endoftext|>",
|
5 |
+
"unk_token": "<|endoftext|>"
|
6 |
+
}
|
tokenizer.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
tokenizer_config.json
ADDED
@@ -0,0 +1,20 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"add_prefix_space": false,
|
3 |
+
"added_tokens_decoder": {
|
4 |
+
"50256": {
|
5 |
+
"content": "<|endoftext|>",
|
6 |
+
"lstrip": false,
|
7 |
+
"normalized": true,
|
8 |
+
"rstrip": false,
|
9 |
+
"single_word": false,
|
10 |
+
"special": true
|
11 |
+
}
|
12 |
+
},
|
13 |
+
"bos_token": "<|endoftext|>",
|
14 |
+
"clean_up_tokenization_spaces": true,
|
15 |
+
"eos_token": "<|endoftext|>",
|
16 |
+
"model_max_length": 1024,
|
17 |
+
"pad_token": "<|endoftext|>",
|
18 |
+
"tokenizer_class": "GPT2Tokenizer",
|
19 |
+
"unk_token": "<|endoftext|>"
|
20 |
+
}
|
vocab.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|