Hugging Face's logo

tensorblock
/
bloom-3b-GGUF

Text Generation
GGUF
TensorBlock
GGUF
Eval Results
Inference Endpoints
Model card Files Community
morriszms commited on
Commit
ce494ec
•
1 Parent(s): 902f69d

Upload folder using huggingface_hub

Browse files
Files changed (14) hide show
  1. .gitattributes +12 -0
  2. README.md +1660 -0
  3. bloom-3b-Q2_K.gguf +3 -0
  4. bloom-3b-Q3_K_L.gguf +3 -0
  5. bloom-3b-Q3_K_M.gguf +3 -0
  6. bloom-3b-Q3_K_S.gguf +3 -0
  7. bloom-3b-Q4_0.gguf +3 -0
  8. bloom-3b-Q4_K_M.gguf +3 -0
  9. bloom-3b-Q4_K_S.gguf +3 -0
  10. bloom-3b-Q5_0.gguf +3 -0
  11. bloom-3b-Q5_K_M.gguf +3 -0
  12. bloom-3b-Q5_K_S.gguf +3 -0
  13. bloom-3b-Q6_K.gguf +3 -0
  14. bloom-3b-Q8_0.gguf +3 -0
.gitattributes CHANGED
@@ -33,3 +33,15 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
 
 
 
 
 
 
 
 
 
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ bloom-3b-Q2_K.gguf filter=lfs diff=lfs merge=lfs -text
37
+ bloom-3b-Q3_K_L.gguf filter=lfs diff=lfs merge=lfs -text
38
+ bloom-3b-Q3_K_M.gguf filter=lfs diff=lfs merge=lfs -text
39
+ bloom-3b-Q3_K_S.gguf filter=lfs diff=lfs merge=lfs -text
40
+ bloom-3b-Q4_0.gguf filter=lfs diff=lfs merge=lfs -text
41
+ bloom-3b-Q4_K_M.gguf filter=lfs diff=lfs merge=lfs -text
42
+ bloom-3b-Q4_K_S.gguf filter=lfs diff=lfs merge=lfs -text
43
+ bloom-3b-Q5_0.gguf filter=lfs diff=lfs merge=lfs -text
44
+ bloom-3b-Q5_K_M.gguf filter=lfs diff=lfs merge=lfs -text
45
+ bloom-3b-Q5_K_S.gguf filter=lfs diff=lfs merge=lfs -text
46
+ bloom-3b-Q6_K.gguf filter=lfs diff=lfs merge=lfs -text
47
+ bloom-3b-Q8_0.gguf filter=lfs diff=lfs merge=lfs -text
README.md ADDED
@@ -0,0 +1,1660 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: bigscience-bloom-rail-1.0
3
+ language:
4
+ - ak
5
+ - ar
6
+ - as
7
+ - bm
8
+ - bn
9
+ - ca
10
+ - code
11
+ - en
12
+ - es
13
+ - eu
14
+ - fon
15
+ - fr
16
+ - gu
17
+ - hi
18
+ - id
19
+ - ig
20
+ - ki
21
+ - kn
22
+ - lg
23
+ - ln
24
+ - ml
25
+ - mr
26
+ - ne
27
+ - nso
28
+ - ny
29
+ - or
30
+ - pa
31
+ - pt
32
+ - rn
33
+ - rw
34
+ - sn
35
+ - st
36
+ - sw
37
+ - ta
38
+ - te
39
+ - tn
40
+ - ts
41
+ - tum
42
+ - tw
43
+ - ur
44
+ - vi
45
+ - wo
46
+ - xh
47
+ - yo
48
+ - zh
49
+ - zhs
50
+ - zht
51
+ - zu
52
+ pipeline_tag: text-generation
53
+ tags:
54
+ - TensorBlock
55
+ - GGUF
56
+ base_model: bigscience/bloom-3b
57
+ model-index:
58
+ - name: bloom
59
+ results:
60
+ - task:
61
+ type: text-generation
62
+ name: text generation
63
+ dataset:
64
+ name: arc_challenge
65
+ type: arc_challenge
66
+ metrics:
67
+ - type: acc
68
+ value: 0.27986348122866894
69
+ name: acc
70
+ verified: false
71
+ - task:
72
+ type: text-generation
73
+ name: text generation
74
+ dataset:
75
+ name: arc_easy
76
+ type: arc_easy
77
+ metrics:
78
+ - type: acc
79
+ value: 0.5946969696969697
80
+ name: acc
81
+ verified: false
82
+ - task:
83
+ type: text-generation
84
+ name: text generation
85
+ dataset:
86
+ name: axb
87
+ type: axb
88
+ metrics:
89
+ - type: acc
90
+ value: 0.4433876811594203
91
+ name: acc
92
+ verified: false
93
+ - task:
94
+ type: text-generation
95
+ name: text generation
96
+ dataset:
97
+ name: axg
98
+ type: axg
99
+ metrics:
100
+ - type: acc
101
+ value: 0.5
102
+ name: acc
103
+ verified: false
104
+ - task:
105
+ type: text-generation
106
+ name: text generation
107
+ dataset:
108
+ name: boolq
109
+ type: boolq
110
+ metrics:
111
+ - type: acc
112
+ value: 0.6165137614678899
113
+ name: acc
114
+ verified: false
115
+ - task:
116
+ type: text-generation
117
+ name: text generation
118
+ dataset:
119
+ name: cb
120
+ type: cb
121
+ metrics:
122
+ - type: acc
123
+ value: 0.30357142857142855
124
+ name: acc
125
+ verified: false
126
+ - task:
127
+ type: text-generation
128
+ name: text generation
129
+ dataset:
130
+ name: cola
131
+ type: cola
132
+ metrics:
133
+ - type: acc
134
+ value: 0.610738255033557
135
+ name: acc
136
+ verified: false
137
+ - task:
138
+ type: text-generation
139
+ name: text generation
140
+ dataset:
141
+ name: copa
142
+ type: copa
143
+ metrics:
144
+ - type: acc
145
+ value: 0.63
146
+ name: acc
147
+ verified: false
148
+ - task:
149
+ type: text-generation
150
+ name: text generation
151
+ dataset:
152
+ name: crows_pairs_english
153
+ type: crows_pairs_english
154
+ metrics:
155
+ - type: acc
156
+ value: 0.4973166368515206
157
+ name: acc
158
+ verified: false
159
+ - task:
160
+ type: text-generation
161
+ name: text generation
162
+ dataset:
163
+ name: crows_pairs_french
164
+ type: crows_pairs_french
165
+ metrics:
166
+ - type: acc
167
+ value: 0.5032796660703638
168
+ name: acc
169
+ verified: false
170
+ - task:
171
+ type: text-generation
172
+ name: text generation
173
+ dataset:
174
+ name: diabla
175
+ type: diabla
176
+ metrics:
177
+ - type: acc
178
+ value: 0.28888308977035493
179
+ name: acc
180
+ verified: false
181
+ - task:
182
+ type: text-generation
183
+ name: text generation
184
+ dataset:
185
+ name: gsarti/flores_101_afr
186
+ type: gsarti/flores_101_afr
187
+ metrics:
188
+ - type: byte_perplexity
189
+ value: 6.500798737976343
190
+ name: byte_perplexity
191
+ verified: false
192
+ - task:
193
+ type: text-generation
194
+ name: text generation
195
+ dataset:
196
+ name: gsarti/flores_101_amh
197
+ type: gsarti/flores_101_amh
198
+ metrics:
199
+ - type: byte_perplexity
200
+ value: 3.9726863338897145
201
+ name: byte_perplexity
202
+ verified: false
203
+ - task:
204
+ type: text-generation
205
+ name: text generation
206
+ dataset:
207
+ name: gsarti/flores_101_ara
208
+ type: gsarti/flores_101_ara
209
+ metrics:
210
+ - type: byte_perplexity
211
+ value: 1.8083841089875814
212
+ name: byte_perplexity
213
+ verified: false
214
+ - task:
215
+ type: text-generation
216
+ name: text generation
217
+ dataset:
218
+ name: gsarti/flores_101_asm
219
+ type: gsarti/flores_101_asm
220
+ metrics:
221
+ - type: byte_perplexity
222
+ value: 5.699102962086425
223
+ name: byte_perplexity
224
+ verified: false
225
+ - task:
226
+ type: text-generation
227
+ name: text generation
228
+ dataset:
229
+ name: gsarti/flores_101_ast
230
+ type: gsarti/flores_101_ast
231
+ metrics:
232
+ - type: byte_perplexity
233
+ value: 3.9252047073429384
234
+ name: byte_perplexity
235
+ verified: false
236
+ - task:
237
+ type: text-generation
238
+ name: text generation
239
+ dataset:
240
+ name: gsarti/flores_101_azj
241
+ type: gsarti/flores_101_azj
242
+ metrics:
243
+ - type: byte_perplexity
244
+ value: 6.942805054270002
245
+ name: byte_perplexity
246
+ verified: false
247
+ - task:
248
+ type: text-generation
249
+ name: text generation
250
+ dataset:
251
+ name: gsarti/flores_101_bel
252
+ type: gsarti/flores_101_bel
253
+ metrics:
254
+ - type: byte_perplexity
255
+ value: 3.614136245847082
256
+ name: byte_perplexity
257
+ verified: false
258
+ - task:
259
+ type: text-generation
260
+ name: text generation
261
+ dataset:
262
+ name: gsarti/flores_101_ben
263
+ type: gsarti/flores_101_ben
264
+ metrics:
265
+ - type: byte_perplexity
266
+ value: 5.121491534300969
267
+ name: byte_perplexity
268
+ verified: false
269
+ - task:
270
+ type: text-generation
271
+ name: text generation
272
+ dataset:
273
+ name: gsarti/flores_101_bos
274
+ type: gsarti/flores_101_bos
275
+ metrics:
276
+ - type: byte_perplexity
277
+ value: 5.653353469118798
278
+ name: byte_perplexity
279
+ verified: false
280
+ - task:
281
+ type: text-generation
282
+ name: text generation
283
+ dataset:
284
+ name: gsarti/flores_101_bul
285
+ type: gsarti/flores_101_bul
286
+ metrics:
287
+ - type: byte_perplexity
288
+ value: 2.7014693938055068
289
+ name: byte_perplexity
290
+ verified: false
291
+ - task:
292
+ type: text-generation
293
+ name: text generation
294
+ dataset:
295
+ name: gsarti/flores_101_cat
296
+ type: gsarti/flores_101_cat
297
+ metrics:
298
+ - type: byte_perplexity
299
+ value: 2.305190041967345
300
+ name: byte_perplexity
301
+ verified: false
302
+ - task:
303
+ type: text-generation
304
+ name: text generation
305
+ dataset:
306
+ name: gsarti/flores_101_ceb
307
+ type: gsarti/flores_101_ceb
308
+ metrics:
309
+ - type: byte_perplexity
310
+ value: 6.291000321323428
311
+ name: byte_perplexity
312
+ verified: false
313
+ - task:
314
+ type: text-generation
315
+ name: text generation
316
+ dataset:
317
+ name: gsarti/flores_101_ces
318
+ type: gsarti/flores_101_ces
319
+ metrics:
320
+ - type: byte_perplexity
321
+ value: 5.447322753586386
322
+ name: byte_perplexity
323
+ verified: false
324
+ - task:
325
+ type: text-generation
326
+ name: text generation
327
+ dataset:
328
+ name: gsarti/flores_101_ckb
329
+ type: gsarti/flores_101_ckb
330
+ metrics:
331
+ - type: byte_perplexity
332
+ value: 3.7255124939234765
333
+ name: byte_perplexity
334
+ verified: false
335
+ - task:
336
+ type: text-generation
337
+ name: text generation
338
+ dataset:
339
+ name: gsarti/flores_101_cym
340
+ type: gsarti/flores_101_cym
341
+ metrics:
342
+ - type: byte_perplexity
343
+ value: 12.539424151448149
344
+ name: byte_perplexity
345
+ verified: false
346
+ - task:
347
+ type: text-generation
348
+ name: text generation
349
+ dataset:
350
+ name: gsarti/flores_101_dan
351
+ type: gsarti/flores_101_dan
352
+ metrics:
353
+ - type: byte_perplexity
354
+ value: 5.183309001005672
355
+ name: byte_perplexity
356
+ verified: false
357
+ - task:
358
+ type: text-generation
359
+ name: text generation
360
+ dataset:
361
+ name: gsarti/flores_101_deu
362
+ type: gsarti/flores_101_deu
363
+ metrics:
364
+ - type: byte_perplexity
365
+ value: 3.1180422286591347
366
+ name: byte_perplexity
367
+ verified: false
368
+ - task:
369
+ type: text-generation
370
+ name: text generation
371
+ dataset:
372
+ name: gsarti/flores_101_ell
373
+ type: gsarti/flores_101_ell
374
+ metrics:
375
+ - type: byte_perplexity
376
+ value: 2.467943456164706
377
+ name: byte_perplexity
378
+ verified: false
379
+ - task:
380
+ type: text-generation
381
+ name: text generation
382
+ dataset:
383
+ name: gsarti/flores_101_eng
384
+ type: gsarti/flores_101_eng
385
+ metrics:
386
+ - type: byte_perplexity
387
+ value: 2.018740628193298
388
+ name: byte_perplexity
389
+ verified: false
390
+ - task:
391
+ type: text-generation
392
+ name: text generation
393
+ dataset:
394
+ name: gsarti/flores_101_est
395
+ type: gsarti/flores_101_est
396
+ metrics:
397
+ - type: byte_perplexity
398
+ value: 9.11654425176368
399
+ name: byte_perplexity
400
+ verified: false
401
+ - task:
402
+ type: text-generation
403
+ name: text generation
404
+ dataset:
405
+ name: gsarti/flores_101_fas
406
+ type: gsarti/flores_101_fas
407
+ metrics:
408
+ - type: byte_perplexity
409
+ value: 3.058009097116482
410
+ name: byte_perplexity
411
+ verified: false
412
+ - task:
413
+ type: text-generation
414
+ name: text generation
415
+ dataset:
416
+ name: gsarti/flores_101_fin
417
+ type: gsarti/flores_101_fin
418
+ metrics:
419
+ - type: byte_perplexity
420
+ value: 6.847047959628553
421
+ name: byte_perplexity
422
+ verified: false
423
+ - task:
424
+ type: text-generation
425
+ name: text generation
426
+ dataset:
427
+ name: gsarti/flores_101_fra
428
+ type: gsarti/flores_101_fra
429
+ metrics:
430
+ - type: byte_perplexity
431
+ value: 1.9975177011840075
432
+ name: byte_perplexity
433
+ verified: false
434
+ - task:
435
+ type: text-generation
436
+ name: text generation
437
+ dataset:
438
+ name: gsarti/flores_101_ful
439
+ type: gsarti/flores_101_ful
440
+ metrics:
441
+ - type: byte_perplexity
442
+ value: 11.465912731488828
443
+ name: byte_perplexity
444
+ verified: false
445
+ - task:
446
+ type: text-generation
447
+ name: text generation
448
+ dataset:
449
+ name: gsarti/flores_101_gle
450
+ type: gsarti/flores_101_gle
451
+ metrics:
452
+ - type: byte_perplexity
453
+ value: 8.681491663539422
454
+ name: byte_perplexity
455
+ verified: false
456
+ - task:
457
+ type: text-generation
458
+ name: text generation
459
+ dataset:
460
+ name: gsarti/flores_101_glg
461
+ type: gsarti/flores_101_glg
462
+ metrics:
463
+ - type: byte_perplexity
464
+ value: 3.029991089015508
465
+ name: byte_perplexity
466
+ verified: false
467
+ - task:
468
+ type: text-generation
469
+ name: text generation
470
+ dataset:
471
+ name: gsarti/flores_101_guj
472
+ type: gsarti/flores_101_guj
473
+ metrics:
474
+ - type: byte_perplexity
475
+ value: 4.955224230286231
476
+ name: byte_perplexity
477
+ verified: false
478
+ - task:
479
+ type: text-generation
480
+ name: text generation
481
+ dataset:
482
+ name: gsarti/flores_101_hau
483
+ type: gsarti/flores_101_hau
484
+ metrics:
485
+ - type: byte_perplexity
486
+ value: 10.758347356372159
487
+ name: byte_perplexity
488
+ verified: false
489
+ - task:
490
+ type: text-generation
491
+ name: text generation
492
+ dataset:
493
+ name: gsarti/flores_101_heb
494
+ type: gsarti/flores_101_heb
495
+ metrics:
496
+ - type: byte_perplexity
497
+ value: 3.6004478129801667
498
+ name: byte_perplexity
499
+ verified: false
500
+ - task:
501
+ type: text-generation
502
+ name: text generation
503
+ dataset:
504
+ name: gsarti/flores_101_hin
505
+ type: gsarti/flores_101_hin
506
+ metrics:
507
+ - type: byte_perplexity
508
+ value: 4.712530650588064
509
+ name: byte_perplexity
510
+ verified: false
511
+ - task:
512
+ type: text-generation
513
+ name: text generation
514
+ dataset:
515
+ name: gsarti/flores_101_hrv
516
+ type: gsarti/flores_101_hrv
517
+ metrics:
518
+ - type: byte_perplexity
519
+ value: 5.822418943372185
520
+ name: byte_perplexity
521
+ verified: false
522
+ - task:
523
+ type: text-generation
524
+ name: text generation
525
+ dataset:
526
+ name: gsarti/flores_101_hun
527
+ type: gsarti/flores_101_hun
528
+ metrics:
529
+ - type: byte_perplexity
530
+ value: 6.440482646965992
531
+ name: byte_perplexity
532
+ verified: false
533
+ - task:
534
+ type: text-generation
535
+ name: text generation
536
+ dataset:
537
+ name: gsarti/flores_101_hye
538
+ type: gsarti/flores_101_hye
539
+ metrics:
540
+ - type: byte_perplexity
541
+ value: 3.657718918347166
542
+ name: byte_perplexity
543
+ verified: false
544
+ - task:
545
+ type: text-generation
546
+ name: text generation
547
+ dataset:
548
+ name: gsarti/flores_101_ibo
549
+ type: gsarti/flores_101_ibo
550
+ metrics:
551
+ - type: byte_perplexity
552
+ value: 5.564814003872672
553
+ name: byte_perplexity
554
+ verified: false
555
+ - task:
556
+ type: text-generation
557
+ name: text generation
558
+ dataset:
559
+ name: gsarti/flores_101_ind
560
+ type: gsarti/flores_101_ind
561
+ metrics:
562
+ - type: byte_perplexity
563
+ value: 2.1597101468869373
564
+ name: byte_perplexity
565
+ verified: false
566
+ - task:
567
+ type: text-generation
568
+ name: text generation
569
+ dataset:
570
+ name: gsarti/flores_101_isl
571
+ type: gsarti/flores_101_isl
572
+ metrics:
573
+ - type: byte_perplexity
574
+ value: 8.082349269518136
575
+ name: byte_perplexity
576
+ verified: false
577
+ - task:
578
+ type: text-generation
579
+ name: text generation
580
+ dataset:
581
+ name: gsarti/flores_101_ita
582
+ type: gsarti/flores_101_ita
583
+ metrics:
584
+ - type: byte_perplexity
585
+ value: 2.9687591414176207
586
+ name: byte_perplexity
587
+ verified: false
588
+ - task:
589
+ type: text-generation
590
+ name: text generation
591
+ dataset:
592
+ name: gsarti/flores_101_jav
593
+ type: gsarti/flores_101_jav
594
+ metrics:
595
+ - type: byte_perplexity
596
+ value: 7.0573805415708994
597
+ name: byte_perplexity
598
+ verified: false
599
+ - task:
600
+ type: text-generation
601
+ name: text generation
602
+ dataset:
603
+ name: gsarti/flores_101_jpn
604
+ type: gsarti/flores_101_jpn
605
+ metrics:
606
+ - type: byte_perplexity
607
+ value: 2.7758864197116933
608
+ name: byte_perplexity
609
+ verified: false
610
+ - task:
611
+ type: text-generation
612
+ name: text generation
613
+ dataset:
614
+ name: gsarti/flores_101_kam
615
+ type: gsarti/flores_101_kam
616
+ metrics:
617
+ - type: byte_perplexity
618
+ value: 11.072949642861332
619
+ name: byte_perplexity
620
+ verified: false
621
+ - task:
622
+ type: text-generation
623
+ name: text generation
624
+ dataset:
625
+ name: gsarti/flores_101_kan
626
+ type: gsarti/flores_101_kan
627
+ metrics:
628
+ - type: byte_perplexity
629
+ value: 5.551730651007082
630
+ name: byte_perplexity
631
+ verified: false
632
+ - task:
633
+ type: text-generation
634
+ name: text generation
635
+ dataset:
636
+ name: gsarti/flores_101_kat
637
+ type: gsarti/flores_101_kat
638
+ metrics:
639
+ - type: byte_perplexity
640
+ value: 2.522630524283745
641
+ name: byte_perplexity
642
+ verified: false
643
+ - task:
644
+ type: text-generation
645
+ name: text generation
646
+ dataset:
647
+ name: gsarti/flores_101_kaz
648
+ type: gsarti/flores_101_kaz
649
+ metrics:
650
+ - type: byte_perplexity
651
+ value: 3.3901748516975574
652
+ name: byte_perplexity
653
+ verified: false
654
+ - task:
655
+ type: text-generation
656
+ name: text generation
657
+ dataset:
658
+ name: gsarti/flores_101_kea
659
+ type: gsarti/flores_101_kea
660
+ metrics:
661
+ - type: byte_perplexity
662
+ value: 8.918534182590863
663
+ name: byte_perplexity
664
+ verified: false
665
+ - task:
666
+ type: text-generation
667
+ name: text generation
668
+ dataset:
669
+ name: gsarti/flores_101_kir
670
+ type: gsarti/flores_101_kir
671
+ metrics:
672
+ - type: byte_perplexity
673
+ value: 3.729278369847201
674
+ name: byte_perplexity
675
+ verified: false
676
+ - task:
677
+ type: text-generation
678
+ name: text generation
679
+ dataset:
680
+ name: gsarti/flores_101_kor
681
+ type: gsarti/flores_101_kor
682
+ metrics:
683
+ - type: byte_perplexity
684
+ value: 3.932884847226212
685
+ name: byte_perplexity
686
+ verified: false
687
+ - task:
688
+ type: text-generation
689
+ name: text generation
690
+ dataset:
691
+ name: gsarti/flores_101_lao
692
+ type: gsarti/flores_101_lao
693
+ metrics:
694
+ - type: byte_perplexity
695
+ value: 2.9077314760849924
696
+ name: byte_perplexity
697
+ verified: false
698
+ - task:
699
+ type: text-generation
700
+ name: text generation
701
+ dataset:
702
+ name: gsarti/flores_101_lav
703
+ type: gsarti/flores_101_lav
704
+ metrics:
705
+ - type: byte_perplexity
706
+ value: 7.777221919194806
707
+ name: byte_perplexity
708
+ verified: false
709
+ - task:
710
+ type: text-generation
711
+ name: text generation
712
+ dataset:
713
+ name: gsarti/flores_101_lin
714
+ type: gsarti/flores_101_lin
715
+ metrics:
716
+ - type: byte_perplexity
717
+ value: 7.524842908050988
718
+ name: byte_perplexity
719
+ verified: false
720
+ - task:
721
+ type: text-generation
722
+ name: text generation
723
+ dataset:
724
+ name: gsarti/flores_101_lit
725
+ type: gsarti/flores_101_lit
726
+ metrics:
727
+ - type: byte_perplexity
728
+ value: 7.369179434621725
729
+ name: byte_perplexity
730
+ verified: false
731
+ - task:
732
+ type: text-generation
733
+ name: text generation
734
+ dataset:
735
+ name: gsarti/flores_101_ltz
736
+ type: gsarti/flores_101_ltz
737
+ metrics:
738
+ - type: byte_perplexity
739
+ value: 8.801059747949214
740
+ name: byte_perplexity
741
+ verified: false
742
+ - task:
743
+ type: text-generation
744
+ name: text generation
745
+ dataset:
746
+ name: gsarti/flores_101_lug
747
+ type: gsarti/flores_101_lug
748
+ metrics:
749
+ - type: byte_perplexity
750
+ value: 8.483203026364786
751
+ name: byte_perplexity
752
+ verified: false
753
+ - task:
754
+ type: text-generation
755
+ name: text generation
756
+ dataset:
757
+ name: gsarti/flores_101_luo
758
+ type: gsarti/flores_101_luo
759
+ metrics:
760
+ - type: byte_perplexity
761
+ value: 11.975963093623681
762
+ name: byte_perplexity
763
+ verified: false
764
+ - task:
765
+ type: text-generation
766
+ name: text generation
767
+ dataset:
768
+ name: gsarti/flores_101_mal
769
+ type: gsarti/flores_101_mal
770
+ metrics:
771
+ - type: byte_perplexity
772
+ value: 4.615948455160037
773
+ name: byte_perplexity
774
+ verified: false
775
+ - task:
776
+ type: text-generation
777
+ name: text generation
778
+ dataset:
779
+ name: gsarti/flores_101_mar
780
+ type: gsarti/flores_101_mar
781
+ metrics:
782
+ - type: byte_perplexity
783
+ value: 5.483253482821379
784
+ name: byte_perplexity
785
+ verified: false
786
+ - task:
787
+ type: text-generation
788
+ name: text generation
789
+ dataset:
790
+ name: gsarti/flores_101_mkd
791
+ type: gsarti/flores_101_mkd
792
+ metrics:
793
+ - type: byte_perplexity
794
+ value: 2.9656732291754087
795
+ name: byte_perplexity
796
+ verified: false
797
+ - task:
798
+ type: text-generation
799
+ name: text generation
800
+ dataset:
801
+ name: gsarti/flores_101_mlt
802
+ type: gsarti/flores_101_mlt
803
+ metrics:
804
+ - type: byte_perplexity
805
+ value: 15.004773437665275
806
+ name: byte_perplexity
807
+ verified: false
808
+ - task:
809
+ type: text-generation
810
+ name: text generation
811
+ dataset:
812
+ name: gsarti/flores_101_mon
813
+ type: gsarti/flores_101_mon
814
+ metrics:
815
+ - type: byte_perplexity
816
+ value: 3.410598542315402
817
+ name: byte_perplexity
818
+ verified: false
819
+ - task:
820
+ type: text-generation
821
+ name: text generation
822
+ dataset:
823
+ name: gsarti/flores_101_mri
824
+ type: gsarti/flores_101_mri
825
+ metrics:
826
+ - type: byte_perplexity
827
+ value: 7.474035895661322
828
+ name: byte_perplexity
829
+ verified: false
830
+ - task:
831
+ type: text-generation
832
+ name: text generation
833
+ dataset:
834
+ name: gsarti/flores_101_msa
835
+ type: gsarti/flores_101_msa
836
+ metrics:
837
+ - type: byte_perplexity
838
+ value: 2.5710001772665634
839
+ name: byte_perplexity
840
+ verified: false
841
+ - task:
842
+ type: text-generation
843
+ name: text generation
844
+ dataset:
845
+ name: gsarti/flores_101_mya
846
+ type: gsarti/flores_101_mya
847
+ metrics:
848
+ - type: byte_perplexity
849
+ value: 2.413577969878331
850
+ name: byte_perplexity
851
+ verified: false
852
+ - task:
853
+ type: text-generation
854
+ name: text generation
855
+ dataset:
856
+ name: gsarti/flores_101_nld
857
+ type: gsarti/flores_101_nld
858
+ metrics:
859
+ - type: byte_perplexity
860
+ value: 4.127831721885065
861
+ name: byte_perplexity
862
+ verified: false
863
+ - task:
864
+ type: text-generation
865
+ name: text generation
866
+ dataset:
867
+ name: gsarti/flores_101_nob
868
+ type: gsarti/flores_101_nob
869
+ metrics:
870
+ - type: byte_perplexity
871
+ value: 5.402763169129877
872
+ name: byte_perplexity
873
+ verified: false
874
+ - task:
875
+ type: text-generation
876
+ name: text generation
877
+ dataset:
878
+ name: gsarti/flores_101_npi
879
+ type: gsarti/flores_101_npi
880
+ metrics:
881
+ - type: byte_perplexity
882
+ value: 5.199342701937889
883
+ name: byte_perplexity
884
+ verified: false
885
+ - task:
886
+ type: text-generation
887
+ name: text generation
888
+ dataset:
889
+ name: gsarti/flores_101_nso
890
+ type: gsarti/flores_101_nso
891
+ metrics:
892
+ - type: byte_perplexity
893
+ value: 8.154626800955667
894
+ name: byte_perplexity
895
+ verified: false
896
+ - task:
897
+ type: text-generation
898
+ name: text generation
899
+ dataset:
900
+ name: gsarti/flores_101_nya
901
+ type: gsarti/flores_101_nya
902
+ metrics:
903
+ - type: byte_perplexity
904
+ value: 8.179860208369393
905
+ name: byte_perplexity
906
+ verified: false
907
+ - task:
908
+ type: text-generation
909
+ name: text generation
910
+ dataset:
911
+ name: gsarti/flores_101_oci
912
+ type: gsarti/flores_101_oci
913
+ metrics:
914
+ - type: byte_perplexity
915
+ value: 4.8617357393685845
916
+ name: byte_perplexity
917
+ verified: false
918
+ - task:
919
+ type: text-generation
920
+ name: text generation
921
+ dataset:
922
+ name: gsarti/flores_101_orm
923
+ type: gsarti/flores_101_orm
924
+ metrics:
925
+ - type: byte_perplexity
926
+ value: 12.911595421079408
927
+ name: byte_perplexity
928
+ verified: false
929
+ - task:
930
+ type: text-generation
931
+ name: text generation
932
+ dataset:
933
+ name: gsarti/flores_101_ory
934
+ type: gsarti/flores_101_ory
935
+ metrics:
936
+ - type: byte_perplexity
937
+ value: 5.189421861225964
938
+ name: byte_perplexity
939
+ verified: false
940
+ - task:
941
+ type: text-generation
942
+ name: text generation
943
+ dataset:
944
+ name: gsarti/flores_101_pan
945
+ type: gsarti/flores_101_pan
946
+ metrics:
947
+ - type: byte_perplexity
948
+ value: 4.698477289331806
949
+ name: byte_perplexity
950
+ verified: false
951
+ - task:
952
+ type: text-generation
953
+ name: text generation
954
+ dataset:
955
+ name: gsarti/flores_101_pol
956
+ type: gsarti/flores_101_pol
957
+ metrics:
958
+ - type: byte_perplexity
959
+ value: 4.625550458479643
960
+ name: byte_perplexity
961
+ verified: false
962
+ - task:
963
+ type: text-generation
964
+ name: text generation
965
+ dataset:
966
+ name: gsarti/flores_101_por
967
+ type: gsarti/flores_101_por
968
+ metrics:
969
+ - type: byte_perplexity
970
+ value: 1.9754515986213523
971
+ name: byte_perplexity
972
+ verified: false
973
+ - task:
974
+ type: text-generation
975
+ name: text generation
976
+ dataset:
977
+ name: gsarti/flores_101_pus
978
+ type: gsarti/flores_101_pus
979
+ metrics:
980
+ - type: byte_perplexity
981
+ value: 4.4963371422771585
982
+ name: byte_perplexity
983
+ verified: false
984
+ - task:
985
+ type: text-generation
986
+ name: text generation
987
+ dataset:
988
+ name: gsarti/flores_101_ron
989
+ type: gsarti/flores_101_ron
990
+ metrics:
991
+ - type: byte_perplexity
992
+ value: 4.965456830031304
993
+ name: byte_perplexity
994
+ verified: false
995
+ - task:
996
+ type: text-generation
997
+ name: text generation
998
+ dataset:
999
+ name: gsarti/flores_101_rus
1000
+ type: gsarti/flores_101_rus
1001
+ metrics:
1002
+ - type: byte_perplexity
1003
+ value: 2.0498020542445303
1004
+ name: byte_perplexity
1005
+ verified: false
1006
+ - task:
1007
+ type: text-generation
1008
+ name: text generation
1009
+ dataset:
1010
+ name: gsarti/flores_101_slk
1011
+ type: gsarti/flores_101_slk
1012
+ metrics:
1013
+ - type: byte_perplexity
1014
+ value: 6.450822127057479
1015
+ name: byte_perplexity
1016
+ verified: false
1017
+ - task:
1018
+ type: text-generation
1019
+ name: text generation
1020
+ dataset:
1021
+ name: gsarti/flores_101_slv
1022
+ type: gsarti/flores_101_slv
1023
+ metrics:
1024
+ - type: byte_perplexity
1025
+ value: 6.620252120186232
1026
+ name: byte_perplexity
1027
+ verified: false
1028
+ - task:
1029
+ type: text-generation
1030
+ name: text generation
1031
+ dataset:
1032
+ name: gsarti/flores_101_sna
1033
+ type: gsarti/flores_101_sna
1034
+ metrics:
1035
+ - type: byte_perplexity
1036
+ value: 8.462166771382726
1037
+ name: byte_perplexity
1038
+ verified: false
1039
+ - task:
1040
+ type: text-generation
1041
+ name: text generation
1042
+ dataset:
1043
+ name: gsarti/flores_101_snd
1044
+ type: gsarti/flores_101_snd
1045
+ metrics:
1046
+ - type: byte_perplexity
1047
+ value: 5.466066951221973
1048
+ name: byte_perplexity
1049
+ verified: false
1050
+ - task:
1051
+ type: text-generation
1052
+ name: text generation
1053
+ dataset:
1054
+ name: gsarti/flores_101_som
1055
+ type: gsarti/flores_101_som
1056
+ metrics:
1057
+ - type: byte_perplexity
1058
+ value: 11.95918054093392
1059
+ name: byte_perplexity
1060
+ verified: false
1061
+ - task:
1062
+ type: text-generation
1063
+ name: text generation
1064
+ dataset:
1065
+ name: gsarti/flores_101_spa
1066
+ type: gsarti/flores_101_spa
1067
+ metrics:
1068
+ - type: byte_perplexity
1069
+ value: 1.8965140104323535
1070
+ name: byte_perplexity
1071
+ verified: false
1072
+ - task:
1073
+ type: text-generation
1074
+ name: text generation
1075
+ dataset:
1076
+ name: gsarti/flores_101_srp
1077
+ type: gsarti/flores_101_srp
1078
+ metrics:
1079
+ - type: byte_perplexity
1080
+ value: 2.871214785885079
1081
+ name: byte_perplexity
1082
+ verified: false
1083
+ - task:
1084
+ type: text-generation
1085
+ name: text generation
1086
+ dataset:
1087
+ name: gsarti/flores_101_swe
1088
+ type: gsarti/flores_101_swe
1089
+ metrics:
1090
+ - type: byte_perplexity
1091
+ value: 5.054972008155866
1092
+ name: byte_perplexity
1093
+ verified: false
1094
+ - task:
1095
+ type: text-generation
1096
+ name: text generation
1097
+ dataset:
1098
+ name: gsarti/flores_101_swh
1099
+ type: gsarti/flores_101_swh
1100
+ metrics:
1101
+ - type: byte_perplexity
1102
+ value: 3.6973091886730676
1103
+ name: byte_perplexity
1104
+ verified: false
1105
+ - task:
1106
+ type: text-generation
1107
+ name: text generation
1108
+ dataset:
1109
+ name: gsarti/flores_101_tam
1110
+ type: gsarti/flores_101_tam
1111
+ metrics:
1112
+ - type: byte_perplexity
1113
+ value: 4.539493400469833
1114
+ name: byte_perplexity
1115
+ verified: false
1116
+ - task:
1117
+ type: text-generation
1118
+ name: text generation
1119
+ dataset:
1120
+ name: gsarti/flores_101_tel
1121
+ type: gsarti/flores_101_tel
1122
+ metrics:
1123
+ - type: byte_perplexity
1124
+ value: 5.807499987508966
1125
+ name: byte_perplexity
1126
+ verified: false
1127
+ - task:
1128
+ type: text-generation
1129
+ name: text generation
1130
+ dataset:
1131
+ name: gsarti/flores_101_tgk
1132
+ type: gsarti/flores_101_tgk
1133
+ metrics:
1134
+ - type: byte_perplexity
1135
+ value: 3.5994818827380426
1136
+ name: byte_perplexity
1137
+ verified: false
1138
+ - task:
1139
+ type: text-generation
1140
+ name: text generation
1141
+ dataset:
1142
+ name: gsarti/flores_101_tgl
1143
+ type: gsarti/flores_101_tgl
1144
+ metrics:
1145
+ - type: byte_perplexity
1146
+ value: 5.667053833119858
1147
+ name: byte_perplexity
1148
+ verified: false
1149
+ - task:
1150
+ type: text-generation
1151
+ name: text generation
1152
+ dataset:
1153
+ name: gsarti/flores_101_tha
1154
+ type: gsarti/flores_101_tha
1155
+ metrics:
1156
+ - type: byte_perplexity
1157
+ value: 2.365940201944242
1158
+ name: byte_perplexity
1159
+ verified: false
1160
+ - task:
1161
+ type: text-generation
1162
+ name: text generation
1163
+ dataset:
1164
+ name: gsarti/flores_101_tur
1165
+ type: gsarti/flores_101_tur
1166
+ metrics:
1167
+ - type: byte_perplexity
1168
+ value: 4.885014749844601
1169
+ name: byte_perplexity
1170
+ verified: false
1171
+ - task:
1172
+ type: text-generation
1173
+ name: text generation
1174
+ dataset:
1175
+ name: gsarti/flores_101_ukr
1176
+ type: gsarti/flores_101_ukr
1177
+ metrics:
1178
+ - type: byte_perplexity
1179
+ value: 2.7240934990288483
1180
+ name: byte_perplexity
1181
+ verified: false
1182
+ - task:
1183
+ type: text-generation
1184
+ name: text generation
1185
+ dataset:
1186
+ name: gsarti/flores_101_umb
1187
+ type: gsarti/flores_101_umb
1188
+ metrics:
1189
+ - type: byte_perplexity
1190
+ value: 12.766915508610673
1191
+ name: byte_perplexity
1192
+ verified: false
1193
+ - task:
1194
+ type: text-generation
1195
+ name: text generation
1196
+ dataset:
1197
+ name: gsarti/flores_101_urd
1198
+ type: gsarti/flores_101_urd
1199
+ metrics:
1200
+ - type: byte_perplexity
1201
+ value: 1.9797467071381232
1202
+ name: byte_perplexity
1203
+ verified: false
1204
+ - task:
1205
+ type: text-generation
1206
+ name: text generation
1207
+ dataset:
1208
+ name: gsarti/flores_101_uzb
1209
+ type: gsarti/flores_101_uzb
1210
+ metrics:
1211
+ - type: byte_perplexity
1212
+ value: 12.002337637722146
1213
+ name: byte_perplexity
1214
+ verified: false
1215
+ - task:
1216
+ type: text-generation
1217
+ name: text generation
1218
+ dataset:
1219
+ name: gsarti/flores_101_vie
1220
+ type: gsarti/flores_101_vie
1221
+ metrics:
1222
+ - type: byte_perplexity
1223
+ value: 1.76578415476397
1224
+ name: byte_perplexity
1225
+ verified: false
1226
+ - task:
1227
+ type: text-generation
1228
+ name: text generation
1229
+ dataset:
1230
+ name: gsarti/flores_101_wol
1231
+ type: gsarti/flores_101_wol
1232
+ metrics:
1233
+ - type: byte_perplexity
1234
+ value: 9.144285650306488
1235
+ name: byte_perplexity
1236
+ verified: false
1237
+ - task:
1238
+ type: text-generation
1239
+ name: text generation
1240
+ dataset:
1241
+ name: gsarti/flores_101_xho
1242
+ type: gsarti/flores_101_xho
1243
+ metrics:
1244
+ - type: byte_perplexity
1245
+ value: 7.403240538286952
1246
+ name: byte_perplexity
1247
+ verified: false
1248
+ - task:
1249
+ type: text-generation
1250
+ name: text generation
1251
+ dataset:
1252
+ name: gsarti/flores_101_yor
1253
+ type: gsarti/flores_101_yor
1254
+ metrics:
1255
+ - type: byte_perplexity
1256
+ value: 5.91272037551173
1257
+ name: byte_perplexity
1258
+ verified: false
1259
+ - task:
1260
+ type: text-generation
1261
+ name: text generation
1262
+ dataset:
1263
+ name: gsarti/flores_101_zho_simpl
1264
+ type: gsarti/flores_101_zho_simpl
1265
+ metrics:
1266
+ - type: byte_perplexity
1267
+ value: 2.2769070822768533
1268
+ name: byte_perplexity
1269
+ verified: false
1270
+ - task:
1271
+ type: text-generation
1272
+ name: text generation
1273
+ dataset:
1274
+ name: gsarti/flores_101_zho_trad
1275
+ type: gsarti/flores_101_zho_trad
1276
+ metrics:
1277
+ - type: byte_perplexity
1278
+ value: 2.5180582198242383
1279
+ name: byte_perplexity
1280
+ verified: false
1281
+ - task:
1282
+ type: text-generation
1283
+ name: text generation
1284
+ dataset:
1285
+ name: gsarti/flores_101_zul
1286
+ type: gsarti/flores_101_zul
1287
+ metrics:
1288
+ - type: byte_perplexity
1289
+ value: 8.53353320693145
1290
+ name: byte_perplexity
1291
+ verified: false
1292
+ - task:
1293
+ type: text-generation
1294
+ name: text generation
1295
+ dataset:
1296
+ name: headqa
1297
+ type: headqa
1298
+ metrics:
1299
+ - type: acc
1300
+ value: 0.26440554339897887
1301
+ name: acc
1302
+ verified: false
1303
+ - task:
1304
+ type: text-generation
1305
+ name: text generation
1306
+ dataset:
1307
+ name: hellaswag
1308
+ type: hellaswag
1309
+ metrics:
1310
+ - type: acc
1311
+ value: 0.41236805417247563
1312
+ name: acc
1313
+ verified: false
1314
+ - task:
1315
+ type: text-generation
1316
+ name: text generation
1317
+ dataset:
1318
+ name: logiqa
1319
+ type: logiqa
1320
+ metrics:
1321
+ - type: acc
1322
+ value: 0.2073732718894009
1323
+ name: acc
1324
+ verified: false
1325
+ - task:
1326
+ type: text-generation
1327
+ name: text generation
1328
+ dataset:
1329
+ name: mathqa
1330
+ type: mathqa
1331
+ metrics:
1332
+ - type: acc
1333
+ value: 0.24958123953098826
1334
+ name: acc
1335
+ verified: false
1336
+ - task:
1337
+ type: text-generation
1338
+ name: text generation
1339
+ dataset:
1340
+ name: mc_taco
1341
+ type: mc_taco
1342
+ metrics:
1343
+ - type: em
1344
+ value: 0.11936936936936937
1345
+ name: em
1346
+ verified: false
1347
+ - task:
1348
+ type: text-generation
1349
+ name: text generation
1350
+ dataset:
1351
+ name: mnli
1352
+ type: mnli
1353
+ metrics:
1354
+ - type: acc
1355
+ value: 0.35496688741721855
1356
+ name: acc
1357
+ verified: false
1358
+ - task:
1359
+ type: text-generation
1360
+ name: text generation
1361
+ dataset:
1362
+ name: mnli_mismatched
1363
+ type: mnli_mismatched
1364
+ metrics:
1365
+ - type: acc
1366
+ value: 0.35211554109031734
1367
+ name: acc
1368
+ verified: false
1369
+ - task:
1370
+ type: text-generation
1371
+ name: text generation
1372
+ dataset:
1373
+ name: mrpc
1374
+ type: mrpc
1375
+ metrics:
1376
+ - type: acc
1377
+ value: 0.5857843137254902
1378
+ name: acc
1379
+ verified: false
1380
+ - task:
1381
+ type: text-generation
1382
+ name: text generation
1383
+ dataset:
1384
+ name: multirc
1385
+ type: multirc
1386
+ metrics:
1387
+ - type: acc
1388
+ value: 0.5375412541254125
1389
+ name: acc
1390
+ verified: false
1391
+ - task:
1392
+ type: text-generation
1393
+ name: text generation
1394
+ dataset:
1395
+ name: openbookqa
1396
+ type: openbookqa
1397
+ metrics:
1398
+ - type: acc
1399
+ value: 0.216
1400
+ name: acc
1401
+ verified: false
1402
+ - task:
1403
+ type: text-generation
1404
+ name: text generation
1405
+ dataset:
1406
+ name: piqa
1407
+ type: piqa
1408
+ metrics:
1409
+ - type: acc
1410
+ value: 0.7078346028291621
1411
+ name: acc
1412
+ verified: false
1413
+ - task:
1414
+ type: text-generation
1415
+ name: text generation
1416
+ dataset:
1417
+ name: prost
1418
+ type: prost
1419
+ metrics:
1420
+ - type: acc
1421
+ value: 0.22683603757472245
1422
+ name: acc
1423
+ verified: false
1424
+ - task:
1425
+ type: text-generation
1426
+ name: text generation
1427
+ dataset:
1428
+ name: pubmedqa
1429
+ type: pubmedqa
1430
+ metrics:
1431
+ - type: acc
1432
+ value: 0.616
1433
+ name: acc
1434
+ verified: false
1435
+ - task:
1436
+ type: text-generation
1437
+ name: text generation
1438
+ dataset:
1439
+ name: qnli
1440
+ type: qnli
1441
+ metrics:
1442
+ - type: acc
1443
+ value: 0.5072304594545122
1444
+ name: acc
1445
+ verified: false
1446
+ - task:
1447
+ type: text-generation
1448
+ name: text generation
1449
+ dataset:
1450
+ name: qqp
1451
+ type: qqp
1452
+ metrics:
1453
+ - type: acc
1454
+ value: 0.3842443729903537
1455
+ name: acc
1456
+ verified: false
1457
+ - task:
1458
+ type: text-generation
1459
+ name: text generation
1460
+ dataset:
1461
+ name: race
1462
+ type: race
1463
+ metrics:
1464
+ - type: acc
1465
+ value: 0.3521531100478469
1466
+ name: acc
1467
+ verified: false
1468
+ - task:
1469
+ type: text-generation
1470
+ name: text generation
1471
+ dataset:
1472
+ name: rte
1473
+ type: rte
1474
+ metrics:
1475
+ - type: acc
1476
+ value: 0.47653429602888087
1477
+ name: acc
1478
+ verified: false
1479
+ - task:
1480
+ type: text-generation
1481
+ name: text generation
1482
+ dataset:
1483
+ name: sciq
1484
+ type: sciq
1485
+ metrics:
1486
+ - type: acc
1487
+ value: 0.892
1488
+ name: acc
1489
+ verified: false
1490
+ - task:
1491
+ type: text-generation
1492
+ name: text generation
1493
+ dataset:
1494
+ name: sst
1495
+ type: sst
1496
+ metrics:
1497
+ - type: acc
1498
+ value: 0.5177752293577982
1499
+ name: acc
1500
+ verified: false
1501
+ - task:
1502
+ type: text-generation
1503
+ name: text generation
1504
+ dataset:
1505
+ name: triviaqa
1506
+ type: triviaqa
1507
+ metrics:
1508
+ - type: acc
1509
+ value: 0.041633518960487934
1510
+ name: acc
1511
+ verified: false
1512
+ - task:
1513
+ type: text-generation
1514
+ name: text generation
1515
+ dataset:
1516
+ name: tydiqa_primary
1517
+ type: tydiqa_primary
1518
+ metrics:
1519
+ - type: acc
1520
+ value: 0.3011337608795236
1521
+ name: acc
1522
+ verified: false
1523
+ - task:
1524
+ type: text-generation
1525
+ name: text generation
1526
+ dataset:
1527
+ name: webqs
1528
+ type: webqs
1529
+ metrics:
1530
+ - type: acc
1531
+ value: 0.01673228346456693
1532
+ name: acc
1533
+ verified: false
1534
+ - task:
1535
+ type: text-generation
1536
+ name: text generation
1537
+ dataset:
1538
+ name: wic
1539
+ type: wic
1540
+ metrics:
1541
+ - type: acc
1542
+ value: 0.5015673981191222
1543
+ name: acc
1544
+ verified: false
1545
+ - task:
1546
+ type: text-generation
1547
+ name: text generation
1548
+ dataset:
1549
+ name: winogrande
1550
+ type: winogrande
1551
+ metrics:
1552
+ - type: acc
1553
+ value: 0.5864246250986582
1554
+ name: acc
1555
+ verified: false
1556
+ - task:
1557
+ type: text-generation
1558
+ name: text generation
1559
+ dataset:
1560
+ name: wnli
1561
+ type: wnli
1562
+ metrics:
1563
+ - type: acc
1564
+ value: 0.471830985915493
1565
+ name: acc
1566
+ verified: false
1567
+ - task:
1568
+ type: text-generation
1569
+ name: text generation
1570
+ dataset:
1571
+ name: wsc
1572
+ type: wsc
1573
+ metrics:
1574
+ - type: acc
1575
+ value: 0.4423076923076923
1576
+ name: acc
1577
+ verified: false
1578
+ - task:
1579
+ type: text-generation
1580
+ name: text generation
1581
+ dataset:
1582
+ name: humaneval
1583
+ type: humaneval
1584
+ metrics:
1585
+ - type: pass@1
1586
+ value: 0.15524390243902436
1587
+ name: pass@1
1588
+ verified: false
1589
+ - type: pass@10
1590
+ value: 0.3220367632383857
1591
+ name: pass@10
1592
+ verified: false
1593
+ - type: pass@100
1594
+ value: 0.5545431515723145
1595
+ name: pass@100
1596
+ verified: false
1597
+ ---
1598
+
1599
+ <div style="width: auto; margin-left: auto; margin-right: auto">
1600
+ <img src="https://i.imgur.com/jC7kdl8.jpeg" alt="TensorBlock" style="width: 100%; min-width: 400px; display: block; margin: auto;">
1601
+ </div>
1602
+ <div style="display: flex; justify-content: space-between; width: 100%;">
1603
+ <div style="display: flex; flex-direction: column; align-items: flex-start;">
1604
+ <p style="margin-top: 0.5em; margin-bottom: 0em;">
1605
+ Feedback and support: TensorBlock's <a href="https://x.com/tensorblock_aoi">Twitter/X</a>, <a href="https://t.me/TensorBlock">Telegram Group</a> and <a href="https://x.com/tensorblock_aoi">Discord server</a>
1606
+ </p>
1607
+ </div>
1608
+ </div>
1609
+
1610
+ ## bigscience/bloom-3b - GGUF
1611
+
1612
+ This repo contains GGUF format model files for [bigscience/bloom-3b](https://huggingface.co/bigscience/bloom-3b).
1613
+
1614
+ The files were quantized using machines provided by [TensorBlock](https://tensorblock.co/), and they are compatible with llama.cpp as of [commit b4011](https://github.com/ggerganov/llama.cpp/commit/a6744e43e80f4be6398fc7733a01642c846dce1d).
1615
+
1616
+ ## Prompt template
1617
+
1618
+ ```
1619
+
1620
+ ```
1621
+
1622
+ ## Model file specification
1623
+
1624
+ | Filename | Quant type | File Size | Description |
1625
+ | -------- | ---------- | --------- | ----------- |
1626
+ | [bloom-3b-Q2_K.gguf](https://huggingface.co/tensorblock/bloom-3b-GGUF/tree/main/bloom-3b-Q2_K.gguf) | Q2_K | 1.516 GB | smallest, significant quality loss - not recommended for most purposes |
1627
+ | [bloom-3b-Q3_K_S.gguf](https://huggingface.co/tensorblock/bloom-3b-GGUF/tree/main/bloom-3b-Q3_K_S.gguf) | Q3_K_S | 1.707 GB | very small, high quality loss |
1628
+ | [bloom-3b-Q3_K_M.gguf](https://huggingface.co/tensorblock/bloom-3b-GGUF/tree/main/bloom-3b-Q3_K_M.gguf) | Q3_K_M | 1.905 GB | very small, high quality loss |
1629
+ | [bloom-3b-Q3_K_L.gguf](https://huggingface.co/tensorblock/bloom-3b-GGUF/tree/main/bloom-3b-Q3_K_L.gguf) | Q3_K_L | 2.016 GB | small, substantial quality loss |
1630
+ | [bloom-3b-Q4_0.gguf](https://huggingface.co/tensorblock/bloom-3b-GGUF/tree/main/bloom-3b-Q4_0.gguf) | Q4_0 | 2.079 GB | legacy; small, very high quality loss - prefer using Q3_K_M |
1631
+ | [bloom-3b-Q4_K_S.gguf](https://huggingface.co/tensorblock/bloom-3b-GGUF/tree/main/bloom-3b-Q4_K_S.gguf) | Q4_K_S | 2.088 GB | small, greater quality loss |
1632
+ | [bloom-3b-Q4_K_M.gguf](https://huggingface.co/tensorblock/bloom-3b-GGUF/tree/main/bloom-3b-Q4_K_M.gguf) | Q4_K_M | 2.235 GB | medium, balanced quality - recommended |
1633
+ | [bloom-3b-Q5_0.gguf](https://huggingface.co/tensorblock/bloom-3b-GGUF/tree/main/bloom-3b-Q5_0.gguf) | Q5_0 | 2.428 GB | legacy; medium, balanced quality - prefer using Q4_K_M |
1634
+ | [bloom-3b-Q5_K_S.gguf](https://huggingface.co/tensorblock/bloom-3b-GGUF/tree/main/bloom-3b-Q5_K_S.gguf) | Q5_K_S | 2.428 GB | large, low quality loss - recommended |
1635
+ | [bloom-3b-Q5_K_M.gguf](https://huggingface.co/tensorblock/bloom-3b-GGUF/tree/main/bloom-3b-Q5_K_M.gguf) | Q5_K_M | 2.546 GB | large, very low quality loss - recommended |
1636
+ | [bloom-3b-Q6_K.gguf](https://huggingface.co/tensorblock/bloom-3b-GGUF/tree/main/bloom-3b-Q6_K.gguf) | Q6_K | 2.799 GB | very large, extremely low quality loss |
1637
+ | [bloom-3b-Q8_0.gguf](https://huggingface.co/tensorblock/bloom-3b-GGUF/tree/main/bloom-3b-Q8_0.gguf) | Q8_0 | 3.621 GB | very large, extremely low quality loss - not recommended |
1638
+
1639
+
1640
+ ## Downloading instruction
1641
+
1642
+ ### Command line
1643
+
1644
+ Firstly, install Huggingface Client
1645
+
1646
+ ```shell
1647
+ pip install -U "huggingface_hub[cli]"
1648
+ ```
1649
+
1650
+ Then, downoad the individual model file the a local directory
1651
+
1652
+ ```shell
1653
+ huggingface-cli download tensorblock/bloom-3b-GGUF --include "bloom-3b-Q2_K.gguf" --local-dir MY_LOCAL_DIR
1654
+ ```
1655
+
1656
+ If you wanna download multiple model files with a pattern (e.g., `*Q4_K*gguf`), you can try:
1657
+
1658
+ ```shell
1659
+ huggingface-cli download tensorblock/bloom-3b-GGUF --local-dir MY_LOCAL_DIR --local-dir-use-symlinks False --include='*Q4_K*gguf'
1660
+ ```
bloom-3b-Q2_K.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7d5e17ed661fe73c50d52cfad8328ac966e70d4a23e925cb361e82e2280a6882
3
+ size 1628232512
bloom-3b-Q3_K_L.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d4a9d5efd6a34d0b4c9a342cf758198aec039b0f031b2ef2fed95814a842e5a7
3
+ size 2165013312
bloom-3b-Q3_K_M.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fe56d56a93fd4cf30f2ec5a0f92d7b28cf817d90eb7c0b7bcfab1ac70c61e4d0
3
+ size 2045410112
bloom-3b-Q3_K_S.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:260fe93cb4eef24b018291904493d61ead1c91210582361fc5015c775b5b83de
3
+ size 1833237312
bloom-3b-Q4_0.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ce16edcef3bd54202a14c9de97a1f67a34e6df6d80f59201d48baa9b2fa5fb9e
3
+ size 2231880512
bloom-3b-Q4_K_M.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7984af5ac90f59deaff474873c222e6746e05f89f31173470750d7267adbfc6f
3
+ size 2400226112
bloom-3b-Q4_K_S.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:477f3e33b42a9cca5cf33faa86962a806e22404ee1d267d2d3ea974f9d71ab76
3
+ size 2241710912
bloom-3b-Q5_0.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:27b8312cf7ae9daff2d1a9c4177016c0365383c852426ec50eb7d92685e8735a
3
+ size 2607074112
bloom-3b-Q5_K_M.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:203d390c1a295900758dfb9130995d2d31ddc5481383758d86a91862fe79f9d8
3
+ size 2734152512
bloom-3b-Q5_K_S.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:89eab193c61f0d9dde583dd23dbb0ef16b1612f2d6dec73d81b3d3d5070bc925
3
+ size 2607074112
bloom-3b-Q6_K.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:593fc1acfd384c8c8cec28ec2d6a21f8171667cfa194161f7ac32a18997b0fd2
3
+ size 3005717312
bloom-3b-Q8_0.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c8f0b31d3aa6d77ccd886884edf544f8421787ca7e02a7b9f8e4ecf4460bfd97
3
+ size 3888200512