mihaimasala
commited on
Commit
•
f2b05bc
1
Parent(s):
acea762
Update README.md
Browse files
README.md
CHANGED
@@ -3,452 +3,453 @@ license: llama2
|
|
3 |
language:
|
4 |
- ro
|
5 |
base_model: meta-llama/Llama-2-7b-hf
|
6 |
-
model-index:
|
7 |
-
- name: OpenLLM-Ro/RoLlama2-7b-Base-2024-05-14
|
8 |
-
results:
|
9 |
-
- task:
|
10 |
-
type: text-generation
|
11 |
-
dataset:
|
12 |
-
name: Romanian_Academic_Benchmarks
|
13 |
-
type: Romanian_Academic_Benchmarks
|
14 |
-
metrics:
|
15 |
-
- name: Average accuracy
|
16 |
-
type: accuracy
|
17 |
-
value: 38.03
|
18 |
-
- task:
|
19 |
-
type: text-generation
|
20 |
-
dataset:
|
21 |
-
name: OpenLLM-Ro/ro_arc_challenge
|
22 |
-
type: OpenLLM-Ro/ro_arc_challenge
|
23 |
-
metrics:
|
24 |
-
- name: Average accuracy
|
25 |
-
type: accuracy
|
26 |
-
value: 37.95
|
27 |
-
- task:
|
28 |
-
type: text-generation
|
29 |
-
dataset:
|
30 |
-
name: OpenLLM-Ro/ro_mmlu
|
31 |
-
type: OpenLLM-Ro/ro_mmlu
|
32 |
-
metrics:
|
33 |
-
- name: Average accuracy
|
34 |
-
type: accuracy
|
35 |
-
value: 27.22
|
36 |
-
- task:
|
37 |
-
type: text-generation
|
38 |
-
dataset:
|
39 |
-
name: OpenLLM-Ro/ro_winogrande
|
40 |
-
type: OpenLLM-Ro/ro_winogrande
|
41 |
-
metrics:
|
42 |
-
- name: Average accuracy
|
43 |
-
type: accuracy
|
44 |
-
value: 59.29
|
45 |
-
- task:
|
46 |
-
type: text-generation
|
47 |
-
dataset:
|
48 |
-
name: OpenLLM-Ro/ro_hellaswag
|
49 |
-
type: OpenLLM-Ro/ro_hellaswag
|
50 |
-
metrics:
|
51 |
-
- name: Average accuracy
|
52 |
-
type: accuracy
|
53 |
-
value: 57.22
|
54 |
-
- task:
|
55 |
-
type: text-generation
|
56 |
-
dataset:
|
57 |
-
name: OpenLLM-Ro/ro_gsm8k
|
58 |
-
type: OpenLLM-Ro/ro_gsm8k
|
59 |
-
metrics:
|
60 |
-
- name: Average accuracy
|
61 |
-
type: accuracy
|
62 |
-
value: 2.53
|
63 |
-
- task:
|
64 |
-
type: text-generation
|
65 |
-
dataset:
|
66 |
-
name: OpenLLM-Ro/ro_truthfulqa
|
67 |
-
type: OpenLLM-Ro/ro_truthfulqa
|
68 |
-
metrics:
|
69 |
-
- name: Average accuracy
|
70 |
-
type: accuracy
|
71 |
-
value: 44
|
72 |
-
- task:
|
73 |
-
type: text-generation
|
74 |
-
dataset:
|
75 |
-
name: LaRoSeDa_binary
|
76 |
-
type: LaRoSeDa_binary
|
77 |
-
metrics:
|
78 |
-
- name: Average macro-f1
|
79 |
-
type: macro-f1
|
80 |
-
value: 83.25
|
81 |
-
- task:
|
82 |
-
type: text-generation
|
83 |
-
dataset:
|
84 |
-
name: LaRoSeDa_multiclass
|
85 |
-
type: LaRoSeDa_multiclass
|
86 |
-
metrics:
|
87 |
-
- name: Average macro-f1
|
88 |
-
type: macro-f1
|
89 |
-
value: 61.04
|
90 |
-
- task:
|
91 |
-
type: text-generation
|
92 |
-
dataset:
|
93 |
-
name: LaRoSeDa_binary_finetuned
|
94 |
-
type: LaRoSeDa_binary_finetuned
|
95 |
-
metrics:
|
96 |
-
- name: Average macro-f1
|
97 |
-
type: macro-f1
|
98 |
-
value: 98.97
|
99 |
-
- task:
|
100 |
-
type: text-generation
|
101 |
-
dataset:
|
102 |
-
name: LaRoSeDa_multiclass_finetuned
|
103 |
-
type: LaRoSeDa_multiclass_finetuned
|
104 |
-
metrics:
|
105 |
-
- name: Average macro-f1
|
106 |
-
type: macro-f1
|
107 |
-
value: 87.72
|
108 |
-
- task:
|
109 |
-
type: text-generation
|
110 |
-
dataset:
|
111 |
-
name: WMT_EN-RO
|
112 |
-
type: WMT_EN-RO
|
113 |
-
metrics:
|
114 |
-
- name: Average bleu
|
115 |
-
type: bleu
|
116 |
-
value: 10.01
|
117 |
-
- task:
|
118 |
-
type: text-generation
|
119 |
-
dataset:
|
120 |
-
name: WMT_RO-EN
|
121 |
-
type: WMT_RO-EN
|
122 |
-
metrics:
|
123 |
-
- name: Average bleu
|
124 |
-
type: bleu
|
125 |
-
value: 13.03
|
126 |
-
- task:
|
127 |
-
type: text-generation
|
128 |
-
dataset:
|
129 |
-
name: WMT_EN-RO_finetuned
|
130 |
-
type: WMT_EN-RO_finetuned
|
131 |
-
metrics:
|
132 |
-
- name: Average bleu
|
133 |
-
type: bleu
|
134 |
-
value: 27.85
|
135 |
-
- task:
|
136 |
-
type: text-generation
|
137 |
-
dataset:
|
138 |
-
name: WMT_RO-EN_finetuned
|
139 |
-
type: WMT_RO-EN_finetuned
|
140 |
-
metrics:
|
141 |
-
- name: Average bleu
|
142 |
-
type: bleu
|
143 |
-
value: 39.3
|
144 |
-
- task:
|
145 |
-
type: text-generation
|
146 |
-
dataset:
|
147 |
-
name: XQuAD
|
148 |
-
type: XQuAD
|
149 |
-
metrics:
|
150 |
-
- name: Average exact_match
|
151 |
-
type: exact_match
|
152 |
-
value: 30.15
|
153 |
-
- task:
|
154 |
-
type: text-generation
|
155 |
-
dataset:
|
156 |
-
name: XQuAD
|
157 |
-
type: XQuAD
|
158 |
-
metrics:
|
159 |
-
- name: Average f1
|
160 |
-
type: f1
|
161 |
-
value: 47.03
|
162 |
-
- task:
|
163 |
-
type: text-generation
|
164 |
-
dataset:
|
165 |
-
name: XQuAD_finetuned
|
166 |
-
type: XQuAD_finetuned
|
167 |
-
metrics:
|
168 |
-
- name: Average exact_match
|
169 |
-
type: exact_match
|
170 |
-
value: 67.06
|
171 |
-
- task:
|
172 |
-
type: text-generation
|
173 |
-
dataset:
|
174 |
-
name: XQuAD_finetuned
|
175 |
-
type: XQuAD_finetuned
|
176 |
-
metrics:
|
177 |
-
- name: Average f1
|
178 |
-
type: f1
|
179 |
-
value: 79.96
|
180 |
-
- task:
|
181 |
-
type: text-generation
|
182 |
-
dataset:
|
183 |
-
name: STS
|
184 |
-
type: STS
|
185 |
-
metrics:
|
186 |
-
- name: Average spearman
|
187 |
-
type: spearman
|
188 |
-
value: 7.89
|
189 |
-
- task:
|
190 |
-
type: text-generation
|
191 |
-
dataset:
|
192 |
-
name: STS
|
193 |
-
type: STS
|
194 |
-
metrics:
|
195 |
-
- name: Average pearson
|
196 |
-
type: pearson
|
197 |
-
value: 7.98
|
198 |
-
- task:
|
199 |
-
type: text-generation
|
200 |
-
dataset:
|
201 |
-
name: STS_finetuned
|
202 |
-
type: STS_finetuned
|
203 |
-
metrics:
|
204 |
-
- name: Average spearman
|
205 |
-
type: spearman
|
206 |
-
value: 71.75
|
207 |
-
- task:
|
208 |
-
type: text-generation
|
209 |
-
dataset:
|
210 |
-
name: STS_finetuned
|
211 |
-
type: STS_finetuned
|
212 |
-
metrics:
|
213 |
-
- name: Average pearson
|
214 |
-
type: pearson
|
215 |
-
value: 71.99
|
216 |
-
- task:
|
217 |
-
type: text-generation
|
218 |
-
dataset:
|
219 |
-
name: OpenLLM-Ro/ro_arc_challenge
|
220 |
-
type: OpenLLM-Ro/ro_arc_challenge
|
221 |
-
metrics:
|
222 |
-
- name: 0-shot
|
223 |
-
type: accuracy
|
224 |
-
value: 35.56
|
225 |
-
- name: 1-shot
|
226 |
-
type: accuracy
|
227 |
-
value: 36.42
|
228 |
-
- name: 3-shot
|
229 |
-
type: accuracy
|
230 |
-
value: 38.56
|
231 |
-
- name: 5-shot
|
232 |
-
type: accuracy
|
233 |
-
value: 38.39
|
234 |
-
- name: 10-shot
|
235 |
-
type: accuracy
|
236 |
-
value: 39.07
|
237 |
-
- name: 25-shot
|
238 |
-
type: accuracy
|
239 |
-
value: 39.67
|
240 |
-
- task:
|
241 |
-
type: text-generation
|
242 |
-
dataset:
|
243 |
-
name: OpenLLM-Ro/ro_mmlu
|
244 |
-
type: OpenLLM-Ro/ro_mmlu
|
245 |
-
metrics:
|
246 |
-
- name: 0-shot
|
247 |
-
type: accuracy
|
248 |
-
value: 25.82
|
249 |
-
- name: 1-shot
|
250 |
-
type: accuracy
|
251 |
-
value: 25.48
|
252 |
-
- name: 3-shot
|
253 |
-
type: accuracy
|
254 |
-
value: 27.61
|
255 |
-
- name: 5-shot
|
256 |
-
type: accuracy
|
257 |
-
value: 29.96
|
258 |
-
- task:
|
259 |
-
type: text-generation
|
260 |
-
dataset:
|
261 |
-
name: OpenLLM-Ro/ro_winogrande
|
262 |
-
type: OpenLLM-Ro/ro_winogrande
|
263 |
-
metrics:
|
264 |
-
- name: 0-shot
|
265 |
-
type: accuracy
|
266 |
-
value: 58.72
|
267 |
-
- name: 1-shot
|
268 |
-
type: accuracy
|
269 |
-
value: 58.88
|
270 |
-
- name: 3-shot
|
271 |
-
type: accuracy
|
272 |
-
value: 60.38
|
273 |
-
- name: 5-shot
|
274 |
-
type: accuracy
|
275 |
-
value: 59.19
|
276 |
-
- task:
|
277 |
-
type: text-generation
|
278 |
-
dataset:
|
279 |
-
name: OpenLLM-Ro/ro_hellaswag
|
280 |
-
type: OpenLLM-Ro/ro_hellaswag
|
281 |
-
metrics:
|
282 |
-
- name: 0-shot
|
283 |
-
type: accuracy
|
284 |
-
value: 55.85
|
285 |
-
- name: 1-shot
|
286 |
-
type: accuracy
|
287 |
-
value: 57.06
|
288 |
-
- name: 3-shot
|
289 |
-
type: accuracy
|
290 |
-
value: 57.52
|
291 |
-
- name: 5-shot
|
292 |
-
type: accuracy
|
293 |
-
value: 57.89
|
294 |
-
- name: 10-shot
|
295 |
-
type: accuracy
|
296 |
-
value: 57.79
|
297 |
-
- task:
|
298 |
-
type: text-generation
|
299 |
-
dataset:
|
300 |
-
name: OpenLLM-Ro/ro_gsm8k
|
301 |
-
type: OpenLLM-Ro/ro_gsm8k
|
302 |
-
metrics:
|
303 |
-
- name: 0-shot
|
304 |
-
type: accuracy
|
305 |
-
value: 0
|
306 |
-
- name: 1-shot
|
307 |
-
type: accuracy
|
308 |
-
value: 2.96
|
309 |
-
- name: 3-shot
|
310 |
-
type: accuracy
|
311 |
-
value: 4.62
|
312 |
-
- task:
|
313 |
-
type: text-generation
|
314 |
-
dataset:
|
315 |
-
name: LaRoSeDa_binary
|
316 |
-
type: LaRoSeDa_binary
|
317 |
-
metrics:
|
318 |
-
- name: 0-shot
|
319 |
-
type: macro-f1
|
320 |
-
value: 42.78
|
321 |
-
- name: 1-shot
|
322 |
-
type: macro-f1
|
323 |
-
value: 98
|
324 |
-
- name: 3-shot
|
325 |
-
type: macro-f1
|
326 |
-
value: 95.13
|
327 |
-
- name: 5-shot
|
328 |
-
type: macro-f1
|
329 |
-
value: 97.07
|
330 |
-
- task:
|
331 |
-
type: text-generation
|
332 |
-
dataset:
|
333 |
-
name: LaRoSeDa_multiclass
|
334 |
-
type: LaRoSeDa_multiclass
|
335 |
-
metrics:
|
336 |
-
- name: 0-shot
|
337 |
-
type: macro-f1
|
338 |
-
value: 46.41
|
339 |
-
- name: 1-shot
|
340 |
-
type: macro-f1
|
341 |
-
value: 67.36
|
342 |
-
- name: 3-shot
|
343 |
-
type: macro-f1
|
344 |
-
value: 65.16
|
345 |
-
- name: 5-shot
|
346 |
-
type: macro-f1
|
347 |
-
value: 65.23
|
348 |
-
- task:
|
349 |
-
type: text-generation
|
350 |
-
dataset:
|
351 |
-
name: WMT_EN-RO
|
352 |
-
type: WMT_EN-RO
|
353 |
-
metrics:
|
354 |
-
- name: 0-shot
|
355 |
-
type: bleu
|
356 |
-
value: 4.45
|
357 |
-
- name: 1-shot
|
358 |
-
type: bleu
|
359 |
-
value: 8.61
|
360 |
-
- name: 3-shot
|
361 |
-
type: bleu
|
362 |
-
value: 12.25
|
363 |
-
- name: 5-shot
|
364 |
-
type: bleu
|
365 |
-
value: 14.73
|
366 |
-
- task:
|
367 |
-
type: text-generation
|
368 |
-
dataset:
|
369 |
-
name: WMT_RO-EN
|
370 |
-
type: WMT_RO-EN
|
371 |
-
metrics:
|
372 |
-
- name: 0-shot
|
373 |
-
type: bleu
|
374 |
-
value: 1.29
|
375 |
-
- name: 1-shot
|
376 |
-
type: bleu
|
377 |
-
value: 10.78
|
378 |
-
- name: 3-shot
|
379 |
-
type: bleu
|
380 |
-
value: 16.82
|
381 |
-
- name: 5-shot
|
382 |
-
type: bleu
|
383 |
-
value: 23.24
|
384 |
-
- task:
|
385 |
-
type: text-generation
|
386 |
-
dataset:
|
387 |
-
name: XQuAD_EM
|
388 |
-
type: XQuAD_EM
|
389 |
-
metrics:
|
390 |
-
- name: 0-shot
|
391 |
-
type: exact_match
|
392 |
-
value: 5.29
|
393 |
-
- name: 1-shot
|
394 |
-
type: exact_match
|
395 |
-
value: 33.95
|
396 |
-
- name: 3-shot
|
397 |
-
type: exact_match
|
398 |
-
value: 39.24
|
399 |
-
- name: 5-shot
|
400 |
-
type: exact_match
|
401 |
-
value: 42.1
|
402 |
-
- task:
|
403 |
-
type: text-generation
|
404 |
-
dataset:
|
405 |
-
name: XQuAD_F1
|
406 |
-
type: XQuAD_F1
|
407 |
-
metrics:
|
408 |
-
- name: 0-shot
|
409 |
-
type: f1
|
410 |
-
value: 16.17
|
411 |
-
- name: 1-shot
|
412 |
-
type: f1
|
413 |
-
value: 51.84
|
414 |
-
- name: 3-shot
|
415 |
-
type: f1
|
416 |
-
value: 58.82
|
417 |
-
- name: 5-shot
|
418 |
-
type: f1
|
419 |
-
value: 61.29
|
420 |
-
- task:
|
421 |
-
type: text-generation
|
422 |
-
dataset:
|
423 |
-
name: STS
|
424 |
-
type: STS
|
425 |
-
metrics:
|
426 |
-
- name: 0-shot
|
427 |
-
type: spearman
|
428 |
-
value: -1.74
|
429 |
-
- name: 1-shot
|
430 |
-
type: spearman
|
431 |
-
value: 15.47
|
432 |
-
- name: 3-shot
|
433 |
-
type: spearman
|
434 |
-
value: 9.93
|
435 |
-
- task:
|
436 |
-
type: text-generation
|
437 |
-
dataset:
|
438 |
-
name: STS
|
439 |
-
type: STS
|
440 |
-
metrics:
|
441 |
-
- name: 0-shot
|
442 |
-
type: pearson
|
443 |
-
value: -1.4
|
444 |
-
- name: 1-shot
|
445 |
-
type: pearson
|
446 |
-
value: 15
|
447 |
-
- name: 3-shot
|
448 |
-
type: pearson
|
449 |
-
value: 10.33
|
450 |
datasets:
|
451 |
- uonlp/CulturaX
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
452 |
---
|
453 |
|
454 |
# Model Card for Model ID
|
|
|
3 |
language:
|
4 |
- ro
|
5 |
base_model: meta-llama/Llama-2-7b-hf
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
6 |
datasets:
|
7 |
- uonlp/CulturaX
|
8 |
+
model-index:
|
9 |
+
- name: OpenLLM-Ro/RoLlama2-7b-Base-2024-05-14
|
10 |
+
results:
|
11 |
+
- task:
|
12 |
+
type: text-generation
|
13 |
+
dataset:
|
14 |
+
name: Romanian_Academic_Benchmarks
|
15 |
+
type: Romanian_Academic_Benchmarks
|
16 |
+
metrics:
|
17 |
+
- name: Average accuracy
|
18 |
+
type: accuracy
|
19 |
+
value: 38.03
|
20 |
+
- task:
|
21 |
+
type: text-generation
|
22 |
+
dataset:
|
23 |
+
name: OpenLLM-Ro/ro_arc_challenge
|
24 |
+
type: OpenLLM-Ro/ro_arc_challenge
|
25 |
+
metrics:
|
26 |
+
- name: Average accuracy
|
27 |
+
type: accuracy
|
28 |
+
value: 37.95
|
29 |
+
- task:
|
30 |
+
type: text-generation
|
31 |
+
dataset:
|
32 |
+
name: OpenLLM-Ro/ro_mmlu
|
33 |
+
type: OpenLLM-Ro/ro_mmlu
|
34 |
+
metrics:
|
35 |
+
- name: Average accuracy
|
36 |
+
type: accuracy
|
37 |
+
value: 27.22
|
38 |
+
- task:
|
39 |
+
type: text-generation
|
40 |
+
dataset:
|
41 |
+
name: OpenLLM-Ro/ro_winogrande
|
42 |
+
type: OpenLLM-Ro/ro_winogrande
|
43 |
+
metrics:
|
44 |
+
- name: Average accuracy
|
45 |
+
type: accuracy
|
46 |
+
value: 59.29
|
47 |
+
- task:
|
48 |
+
type: text-generation
|
49 |
+
dataset:
|
50 |
+
name: OpenLLM-Ro/ro_hellaswag
|
51 |
+
type: OpenLLM-Ro/ro_hellaswag
|
52 |
+
metrics:
|
53 |
+
- name: Average accuracy
|
54 |
+
type: accuracy
|
55 |
+
value: 57.22
|
56 |
+
- task:
|
57 |
+
type: text-generation
|
58 |
+
dataset:
|
59 |
+
name: OpenLLM-Ro/ro_gsm8k
|
60 |
+
type: OpenLLM-Ro/ro_gsm8k
|
61 |
+
metrics:
|
62 |
+
- name: Average accuracy
|
63 |
+
type: accuracy
|
64 |
+
value: 2.53
|
65 |
+
- task:
|
66 |
+
type: text-generation
|
67 |
+
dataset:
|
68 |
+
name: OpenLLM-Ro/ro_truthfulqa
|
69 |
+
type: OpenLLM-Ro/ro_truthfulqa
|
70 |
+
metrics:
|
71 |
+
- name: Average accuracy
|
72 |
+
type: accuracy
|
73 |
+
value: 44.00
|
74 |
+
- task:
|
75 |
+
type: text-generation
|
76 |
+
dataset:
|
77 |
+
name: LaRoSeDa_binary
|
78 |
+
type: LaRoSeDa_binary
|
79 |
+
metrics:
|
80 |
+
- name: Average macro-f1
|
81 |
+
type: macro-f1
|
82 |
+
value: 83.25
|
83 |
+
- task:
|
84 |
+
type: text-generation
|
85 |
+
dataset:
|
86 |
+
name: LaRoSeDa_multiclass
|
87 |
+
type: LaRoSeDa_multiclass
|
88 |
+
metrics:
|
89 |
+
- name: Average macro-f1
|
90 |
+
type: macro-f1
|
91 |
+
value: 61.04
|
92 |
+
- task:
|
93 |
+
type: text-generation
|
94 |
+
dataset:
|
95 |
+
name: LaRoSeDa_binary_finetuned
|
96 |
+
type: LaRoSeDa_binary_finetuned
|
97 |
+
metrics:
|
98 |
+
- name: Average macro-f1
|
99 |
+
type: macro-f1
|
100 |
+
value: 98.97
|
101 |
+
- task:
|
102 |
+
type: text-generation
|
103 |
+
dataset:
|
104 |
+
name: LaRoSeDa_multiclass_finetuned
|
105 |
+
type: LaRoSeDa_multiclass_finetuned
|
106 |
+
metrics:
|
107 |
+
- name: Average macro-f1
|
108 |
+
type: macro-f1
|
109 |
+
value: 87.72
|
110 |
+
- task:
|
111 |
+
type: text-generation
|
112 |
+
dataset:
|
113 |
+
name: WMT_EN-RO
|
114 |
+
type: WMT_EN-RO
|
115 |
+
metrics:
|
116 |
+
- name: Average bleu
|
117 |
+
type: bleu
|
118 |
+
value: 10.01
|
119 |
+
- task:
|
120 |
+
type: text-generation
|
121 |
+
dataset:
|
122 |
+
name: WMT_RO-EN
|
123 |
+
type: WMT_RO-EN
|
124 |
+
metrics:
|
125 |
+
- name: Average bleu
|
126 |
+
type: bleu
|
127 |
+
value: 13.03
|
128 |
+
- task:
|
129 |
+
type: text-generation
|
130 |
+
dataset:
|
131 |
+
name: WMT_EN-RO_finetuned
|
132 |
+
type: WMT_EN-RO_finetuned
|
133 |
+
metrics:
|
134 |
+
- name: Average bleu
|
135 |
+
type: bleu
|
136 |
+
value: 27.85
|
137 |
+
- task:
|
138 |
+
type: text-generation
|
139 |
+
dataset:
|
140 |
+
name: WMT_RO-EN_finetuned
|
141 |
+
type: WMT_RO-EN_finetuned
|
142 |
+
metrics:
|
143 |
+
- name: Average bleu
|
144 |
+
type: bleu
|
145 |
+
value: 39.30
|
146 |
+
- task:
|
147 |
+
type: text-generation
|
148 |
+
dataset:
|
149 |
+
name: XQuAD
|
150 |
+
type: XQuAD
|
151 |
+
metrics:
|
152 |
+
- name: Average exact_match
|
153 |
+
type: exact_match
|
154 |
+
value: 30.15
|
155 |
+
- task:
|
156 |
+
type: text-generation
|
157 |
+
dataset:
|
158 |
+
name: XQuAD
|
159 |
+
type: XQuAD
|
160 |
+
metrics:
|
161 |
+
- name: Average f1
|
162 |
+
type: f1
|
163 |
+
value: 47.03
|
164 |
+
- task:
|
165 |
+
type: text-generation
|
166 |
+
dataset:
|
167 |
+
name: XQuAD_finetuned
|
168 |
+
type: XQuAD_finetuned
|
169 |
+
metrics:
|
170 |
+
- name: Average exact_match
|
171 |
+
type: exact_match
|
172 |
+
value: 67.06
|
173 |
+
- task:
|
174 |
+
type: text-generation
|
175 |
+
dataset:
|
176 |
+
name: XQuAD_finetuned
|
177 |
+
type: XQuAD_finetuned
|
178 |
+
metrics:
|
179 |
+
- name: Average f1
|
180 |
+
type: f1
|
181 |
+
value: 79.96
|
182 |
+
- task:
|
183 |
+
type: text-generation
|
184 |
+
dataset:
|
185 |
+
name: STS
|
186 |
+
type: STS
|
187 |
+
metrics:
|
188 |
+
- name: Average spearman
|
189 |
+
type: spearman
|
190 |
+
value: 7.89
|
191 |
+
- task:
|
192 |
+
type: text-generation
|
193 |
+
dataset:
|
194 |
+
name: STS
|
195 |
+
type: STS
|
196 |
+
metrics:
|
197 |
+
- name: Average pearson
|
198 |
+
type: pearson
|
199 |
+
value: 7.98
|
200 |
+
- task:
|
201 |
+
type: text-generation
|
202 |
+
dataset:
|
203 |
+
name: STS_finetuned
|
204 |
+
type: STS_finetuned
|
205 |
+
metrics:
|
206 |
+
- name: Average spearman
|
207 |
+
type: spearman
|
208 |
+
value: 71.75
|
209 |
+
- task:
|
210 |
+
type: text-generation
|
211 |
+
dataset:
|
212 |
+
name: STS_finetuned
|
213 |
+
type: STS_finetuned
|
214 |
+
metrics:
|
215 |
+
- name: Average pearson
|
216 |
+
type: pearson
|
217 |
+
value: 71.99
|
218 |
+
- task:
|
219 |
+
type: text-generation
|
220 |
+
dataset:
|
221 |
+
name: OpenLLM-Ro/ro_arc_challenge
|
222 |
+
type: OpenLLM-Ro/ro_arc_challenge
|
223 |
+
metrics:
|
224 |
+
- name: 0-shot
|
225 |
+
type: accuracy
|
226 |
+
value: 35.56
|
227 |
+
- name: 1-shot
|
228 |
+
type: accuracy
|
229 |
+
value: 36.42
|
230 |
+
- name: 3-shot
|
231 |
+
type: accuracy
|
232 |
+
value: 38.56
|
233 |
+
- name: 5-shot
|
234 |
+
type: accuracy
|
235 |
+
value: 38.39
|
236 |
+
- name: 10-shot
|
237 |
+
type: accuracy
|
238 |
+
value: 39.07
|
239 |
+
- name: 25-shot
|
240 |
+
type: accuracy
|
241 |
+
value: 39.67
|
242 |
+
- task:
|
243 |
+
type: text-generation
|
244 |
+
dataset:
|
245 |
+
name: OpenLLM-Ro/ro_mmlu
|
246 |
+
type: OpenLLM-Ro/ro_mmlu
|
247 |
+
metrics:
|
248 |
+
- name: 0-shot
|
249 |
+
type: accuracy
|
250 |
+
value: 25.82
|
251 |
+
- name: 1-shot
|
252 |
+
type: accuracy
|
253 |
+
value: 25.48
|
254 |
+
- name: 3-shot
|
255 |
+
type: accuracy
|
256 |
+
value: 27.61
|
257 |
+
- name: 5-shot
|
258 |
+
type: accuracy
|
259 |
+
value: 29.96
|
260 |
+
- task:
|
261 |
+
type: text-generation
|
262 |
+
dataset:
|
263 |
+
name: OpenLLM-Ro/ro_winogrande
|
264 |
+
type: OpenLLM-Ro/ro_winogrande
|
265 |
+
metrics:
|
266 |
+
- name: 0-shot
|
267 |
+
type: accuracy
|
268 |
+
value: 58.72
|
269 |
+
- name: 1-shot
|
270 |
+
type: accuracy
|
271 |
+
value: 58.88
|
272 |
+
- name: 3-shot
|
273 |
+
type: accuracy
|
274 |
+
value: 60.38
|
275 |
+
- name: 5-shot
|
276 |
+
type: accuracy
|
277 |
+
value: 59.19
|
278 |
+
- task:
|
279 |
+
type: text-generation
|
280 |
+
dataset:
|
281 |
+
name: OpenLLM-Ro/ro_hellaswag
|
282 |
+
type: OpenLLM-Ro/ro_hellaswag
|
283 |
+
metrics:
|
284 |
+
- name: 0-shot
|
285 |
+
type: accuracy
|
286 |
+
value: 55.85
|
287 |
+
- name: 1-shot
|
288 |
+
type: accuracy
|
289 |
+
value: 57.06
|
290 |
+
- name: 3-shot
|
291 |
+
type: accuracy
|
292 |
+
value: 57.52
|
293 |
+
- name: 5-shot
|
294 |
+
type: accuracy
|
295 |
+
value: 57.89
|
296 |
+
- name: 10-shot
|
297 |
+
type: accuracy
|
298 |
+
value: 57.79
|
299 |
+
- task:
|
300 |
+
type: text-generation
|
301 |
+
dataset:
|
302 |
+
name: OpenLLM-Ro/ro_gsm8k
|
303 |
+
type: OpenLLM-Ro/ro_gsm8k
|
304 |
+
metrics:
|
305 |
+
- name: 0-shot
|
306 |
+
type: accuracy
|
307 |
+
value: 0.00
|
308 |
+
- name: 1-shot
|
309 |
+
type: accuracy
|
310 |
+
value: 2.96
|
311 |
+
- name: 3-shot
|
312 |
+
type: accuracy
|
313 |
+
value: 4.62
|
314 |
+
- task:
|
315 |
+
type: text-generation
|
316 |
+
dataset:
|
317 |
+
name: LaRoSeDa_binary
|
318 |
+
type: LaRoSeDa_binary
|
319 |
+
metrics:
|
320 |
+
- name: 0-shot
|
321 |
+
type: macro-f1
|
322 |
+
value: 42.78
|
323 |
+
- name: 1-shot
|
324 |
+
type: macro-f1
|
325 |
+
value: 98.00
|
326 |
+
- name: 3-shot
|
327 |
+
type: macro-f1
|
328 |
+
value: 95.13
|
329 |
+
- name: 5-shot
|
330 |
+
type: macro-f1
|
331 |
+
value: 97.07
|
332 |
+
- task:
|
333 |
+
type: text-generation
|
334 |
+
dataset:
|
335 |
+
name: LaRoSeDa_multiclass
|
336 |
+
type: LaRoSeDa_multiclass
|
337 |
+
metrics:
|
338 |
+
- name: 0-shot
|
339 |
+
type: macro-f1
|
340 |
+
value: 46.41
|
341 |
+
- name: 1-shot
|
342 |
+
type: macro-f1
|
343 |
+
value: 67.36
|
344 |
+
- name: 3-shot
|
345 |
+
type: macro-f1
|
346 |
+
value: 65.16
|
347 |
+
- name: 5-shot
|
348 |
+
type: macro-f1
|
349 |
+
value: 65.23
|
350 |
+
- task:
|
351 |
+
type: text-generation
|
352 |
+
dataset:
|
353 |
+
name: WMT_EN-RO
|
354 |
+
type: WMT_EN-RO
|
355 |
+
metrics:
|
356 |
+
- name: 0-shot
|
357 |
+
type: bleu
|
358 |
+
value: 4.45
|
359 |
+
- name: 1-shot
|
360 |
+
type: bleu
|
361 |
+
value: 8.61
|
362 |
+
- name: 3-shot
|
363 |
+
type: bleu
|
364 |
+
value: 12.25
|
365 |
+
- name: 5-shot
|
366 |
+
type: bleu
|
367 |
+
value: 14.73
|
368 |
+
- task:
|
369 |
+
type: text-generation
|
370 |
+
dataset:
|
371 |
+
name: WMT_RO-EN
|
372 |
+
type: WMT_RO-EN
|
373 |
+
metrics:
|
374 |
+
- name: 0-shot
|
375 |
+
type: bleu
|
376 |
+
value: 1.29
|
377 |
+
- name: 1-shot
|
378 |
+
type: bleu
|
379 |
+
value: 10.78
|
380 |
+
- name: 3-shot
|
381 |
+
type: bleu
|
382 |
+
value: 16.82
|
383 |
+
- name: 5-shot
|
384 |
+
type: bleu
|
385 |
+
value: 23.24
|
386 |
+
- task:
|
387 |
+
type: text-generation
|
388 |
+
dataset:
|
389 |
+
name: XQuAD_EM
|
390 |
+
type: XQuAD_EM
|
391 |
+
metrics:
|
392 |
+
- name: 0-shot
|
393 |
+
type: exact_match
|
394 |
+
value: 5.29
|
395 |
+
- name: 1-shot
|
396 |
+
type: exact_match
|
397 |
+
value: 33.95
|
398 |
+
- name: 3-shot
|
399 |
+
type: exact_match
|
400 |
+
value: 39.24
|
401 |
+
- name: 5-shot
|
402 |
+
type: exact_match
|
403 |
+
value: 42.10
|
404 |
+
- task:
|
405 |
+
type: text-generation
|
406 |
+
dataset:
|
407 |
+
name: XQuAD_F1
|
408 |
+
type: XQuAD_F1
|
409 |
+
metrics:
|
410 |
+
- name: 0-shot
|
411 |
+
type: f1
|
412 |
+
value: 16.17
|
413 |
+
- name: 1-shot
|
414 |
+
type: f1
|
415 |
+
value: 51.84
|
416 |
+
- name: 3-shot
|
417 |
+
type: f1
|
418 |
+
value: 58.82
|
419 |
+
- name: 5-shot
|
420 |
+
type: f1
|
421 |
+
value: 61.29
|
422 |
+
- task:
|
423 |
+
type: text-generation
|
424 |
+
dataset:
|
425 |
+
name: STS_Spearman
|
426 |
+
type: STS_Spearman
|
427 |
+
metrics:
|
428 |
+
- name: 1-shot
|
429 |
+
type: spearman
|
430 |
+
value: -1.74
|
431 |
+
- name: 3-shot
|
432 |
+
type: spearman
|
433 |
+
value: 15.47
|
434 |
+
- name: 5-shot
|
435 |
+
type: spearman
|
436 |
+
value: 9.93
|
437 |
+
- task:
|
438 |
+
type: text-generation
|
439 |
+
dataset:
|
440 |
+
name: STS_Pearson
|
441 |
+
type: STS_Pearson
|
442 |
+
metrics:
|
443 |
+
- name: 1-shot
|
444 |
+
type: pearson
|
445 |
+
value: -1.40
|
446 |
+
- name: 3-shot
|
447 |
+
type: pearson
|
448 |
+
value: 15.00
|
449 |
+
- name: 5-shot
|
450 |
+
type: pearson
|
451 |
+
value: 10.33
|
452 |
+
|
453 |
---
|
454 |
|
455 |
# Model Card for Model ID
|