Llama-Guard-3-8B-GGUF / scores /Llama-Guard-3-8B-iq3_m.tqa
eaddario's picture
Generate Perplexity, KLD, ARC, HellaSwag, MMLU, Truthful QA and WinoGrande scores
6de6927 verified
common_init_from_params: setting dry_penalty_last_n to ctx_size = 768
common_init_from_params: warming up the model with an empty run - please wait ... (--no-warmup to disable)
system_info: n_threads = 6 (n_threads_batch = 6) / 12 | Metal : EMBED_LIBRARY = 1 | CPU : NEON = 1 | ARM_FMA = 1 | FP16_VA = 1 | DOTPROD = 1 | LLAMAFILE = 1 | ACCELERATE = 1 | AARCH64_REPACK = 1 |
multiple_choice_score: there are 817 tasks in prompt
multiple_choice_score: selecting 750 random tasks from 817 tasks available
multiple_choice_score: preparing task data...done
multiple_choice_score : calculating TruthfulQA score over 750 tasks.
task acc_norm
1 100.00000000
2 50.00000000
3 33.33333333
4 25.00000000
5 40.00000000
6 33.33333333
7 28.57142857
8 37.50000000
9 33.33333333
10 30.00000000
11 27.27272727
12 25.00000000
13 30.76923077
14 28.57142857
15 26.66666667
16 25.00000000
17 23.52941176
18 22.22222222
19 21.05263158
20 20.00000000
21 19.04761905
22 18.18181818
23 17.39130435
24 20.83333333
25 20.00000000
26 19.23076923
27 18.51851852
28 17.85714286
29 20.68965517
30 23.33333333
31 25.80645161
32 28.12500000
33 30.30303030
34 29.41176471
35 28.57142857
36 27.77777778
37 29.72972973
38 28.94736842
39 30.76923077
40 30.00000000
41 29.26829268
42 28.57142857
43 30.23255814
44 31.81818182
45 33.33333333
46 32.60869565
47 31.91489362
48 33.33333333
49 32.65306122
50 34.00000000
51 33.33333333
52 32.69230769
53 32.07547170
54 31.48148148
55 32.72727273
56 32.14285714
57 31.57894737
58 32.75862069
59 33.89830508
60 35.00000000
61 34.42622951
62 35.48387097
63 34.92063492
64 34.37500000
65 33.84615385
66 33.33333333
67 32.83582090
68 32.35294118
69 33.33333333
70 32.85714286
71 32.39436620
72 33.33333333
73 34.24657534
74 33.78378378
75 33.33333333
76 34.21052632
77 33.76623377
78 34.61538462
79 34.17721519
80 35.00000000
81 34.56790123
82 35.36585366
83 34.93975904
84 34.52380952
85 34.11764706
86 33.72093023
87 33.33333333
88 32.95454545
89 33.70786517
90 33.33333333
91 32.96703297
92 32.60869565
93 32.25806452
94 31.91489362
95 31.57894737
96 32.29166667
97 32.98969072
98 32.65306122
99 33.33333333
100 33.00000000
101 33.66336634
102 33.33333333
103 33.00970874
104 33.65384615
105 33.33333333
106 33.96226415
107 34.57943925
108 34.25925926
109 33.94495413
110 34.54545455
111 34.23423423
112 34.82142857
113 35.39823009
114 35.96491228
115 35.65217391
116 35.34482759
117 35.04273504
118 34.74576271
119 34.45378151
120 35.00000000
121 35.53719008
122 35.24590164
123 35.77235772
124 36.29032258
125 36.80000000
126 37.30158730
127 37.00787402
128 36.71875000
129 37.20930233
130 37.69230769
131 37.40458015
132 37.12121212
133 36.84210526
134 36.56716418
135 37.03703704
136 36.76470588
137 37.22627737
138 37.68115942
139 38.12949640
140 37.85714286
141 37.58865248
142 37.32394366
143 37.06293706
144 37.50000000
145 37.24137931
146 36.98630137
147 36.73469388
148 36.48648649
149 36.24161074
150 36.00000000
151 36.42384106
152 36.18421053
153 35.94771242
154 35.71428571
155 35.48387097
156 35.89743590
157 35.66878981
158 35.44303797
159 35.22012579
160 35.00000000
161 35.40372671
162 35.18518519
163 34.96932515
164 34.75609756
165 35.15151515
166 35.54216867
167 35.32934132
168 35.71428571
169 35.50295858
170 35.29411765
171 35.08771930
172 34.88372093
173 34.68208092
174 35.05747126
175 34.85714286
176 35.22727273
177 35.02824859
178 34.83146067
179 34.63687151
180 34.44444444
181 34.25414365
182 34.06593407
183 33.87978142
184 33.69565217
185 33.51351351
186 33.87096774
187 33.68983957
188 33.51063830
189 33.33333333
190 33.15789474
191 32.98429319
192 32.81250000
193 32.64248705
194 32.98969072
195 32.82051282
196 32.65306122
197 32.48730964
198 32.32323232
199 32.16080402
200 32.50000000
201 32.33830846
202 32.67326733
203 32.51231527
204 32.35294118
205 32.68292683
206 32.52427184
207 32.85024155
208 32.69230769
209 32.53588517
210 32.38095238
211 32.22748815
212 32.07547170
213 31.92488263
214 31.77570093
215 31.62790698
216 31.94444444
217 31.79723502
218 32.11009174
219 31.96347032
220 31.81818182
221 31.67420814
222 31.53153153
223 31.39013453
224 31.69642857
225 31.55555556
226 31.41592920
227 31.71806167
228 31.57894737
229 31.44104803
230 31.30434783
231 31.16883117
232 31.03448276
233 30.90128755
234 31.19658120
235 31.06382979
236 31.35593220
237 31.22362869
238 31.51260504
239 31.38075314
240 31.25000000
241 31.12033195
242 30.99173554
243 30.86419753
244 30.73770492
245 30.61224490
246 30.48780488
247 30.36437247
248 30.24193548
249 30.52208835
250 30.40000000
251 30.67729084
252 30.95238095
253 30.83003953
254 30.70866142
255 30.98039216
256 30.85937500
257 30.73929961
258 30.62015504
259 30.88803089
260 30.76923077
261 31.03448276
262 30.91603053
263 30.79847909
264 31.06060606
265 31.32075472
266 31.20300752
267 31.08614232
268 30.97014925
269 30.85501859
270 30.74074074
271 30.62730627
272 30.51470588
273 30.40293040
274 30.29197080
275 30.18181818
276 30.07246377
277 29.96389892
278 29.85611511
279 29.74910394
280 30.00000000
281 30.24911032
282 30.14184397
283 30.38869258
284 30.28169014
285 30.17543860
286 30.41958042
287 30.66202091
288 30.90277778
289 30.79584775
290 30.68965517
291 30.92783505
292 31.16438356
293 31.39931741
294 31.29251701
295 31.52542373
296 31.75675676
297 31.98653199
298 31.87919463
299 31.77257525
300 31.66666667
301 31.56146179
302 31.45695364
303 31.35313531
304 31.25000000
305 31.14754098
306 31.04575163
307 30.94462541
308 31.16883117
309 31.06796117
310 30.96774194
311 30.86816720
312 31.08974359
313 30.99041534
314 30.89171975
315 30.79365079
316 30.69620253
317 30.91482650
318 30.81761006
319 30.72100313
320 30.62500000
321 30.52959502
322 30.74534161
323 30.65015480
324 30.55555556
325 30.46153846
326 30.36809816
327 30.58103976
328 30.79268293
329 30.69908815
330 30.90909091
331 30.81570997
332 31.02409639
333 31.23123123
334 31.13772455
335 31.34328358
336 31.54761905
337 31.45400593
338 31.36094675
339 31.56342183
340 31.47058824
341 31.37829912
342 31.57894737
343 31.48688047
344 31.68604651
345 31.88405797
346 31.79190751
347 31.98847262
348 31.89655172
349 31.80515759
350 31.71428571
351 31.62393162
352 31.81818182
353 31.72804533
354 31.63841808
355 31.83098592
356 31.74157303
357 31.93277311
358 31.84357542
359 31.75487465
360 31.66666667
361 31.85595568
362 31.76795580
363 31.95592287
364 32.14285714
365 32.05479452
366 31.96721311
367 32.15258856
368 32.33695652
369 32.24932249
370 32.16216216
371 32.07547170
372 32.25806452
373 32.17158177
374 32.35294118
375 32.26666667
376 32.18085106
377 32.36074271
378 32.27513228
379 32.18997361
380 32.36842105
381 32.28346457
382 32.46073298
383 32.37597911
384 32.29166667
385 32.20779221
386 32.12435233
387 32.04134367
388 32.21649485
389 32.13367609
390 32.05128205
391 32.22506394
392 32.14285714
393 32.31552163
394 32.48730964
395 32.40506329
396 32.32323232
397 32.49370277
398 32.41206030
399 32.33082707
400 32.25000000
401 32.16957606
402 32.33830846
403 32.25806452
404 32.17821782
405 32.09876543
406 32.01970443
407 31.94103194
408 31.86274510
409 31.78484108
410 31.70731707
411 31.63017032
412 31.79611650
413 31.71912833
414 31.88405797
415 32.04819277
416 31.97115385
417 32.13429257
418 32.05741627
419 31.98090692
420 31.90476190
421 32.06650831
422 31.99052133
423 32.15130024
424 32.31132075
425 32.23529412
426 32.15962441
427 32.08430913
428 32.00934579
429 31.93473193
430 31.86046512
431 31.78654292
432 31.94444444
433 32.10161663
434 32.25806452
435 32.18390805
436 32.33944954
437 32.26544622
438 32.42009132
439 32.34624146
440 32.50000000
441 32.42630385
442 32.57918552
443 32.50564334
444 32.65765766
445 32.58426966
446 32.51121076
447 32.43847875
448 32.36607143
449 32.51670379
450 32.66666667
451 32.59423503
452 32.52212389
453 32.67108168
454 32.59911894
455 32.52747253
456 32.45614035
457 32.38512035
458 32.31441048
459 32.24400871
460 32.17391304
461 32.10412148
462 32.25108225
463 32.39740821
464 32.54310345
465 32.68817204
466 32.61802575
467 32.54817987
468 32.69230769
469 32.62260128
470 32.55319149
471 32.48407643
472 32.62711864
473 32.55813953
474 32.48945148
475 32.63157895
476 32.56302521
477 32.49475891
478 32.42677824
479 32.35908142
480 32.50000000
481 32.43243243
482 32.57261411
483 32.50517598
484 32.64462810
485 32.78350515
486 32.71604938
487 32.85420945
488 32.78688525
489 32.92433538
490 32.85714286
491 32.99389002
492 33.13008130
493 33.06288032
494 33.19838057
495 33.13131313
496 33.06451613
497 32.99798793
498 32.93172691
499 32.86573146
500 32.80000000
501 32.73453094
502 32.66932271
503 32.60437376
504 32.73809524
505 32.67326733
506 32.60869565
507 32.54437870
508 32.67716535
509 32.80943026
510 32.74509804
511 32.68101761
512 32.61718750
513 32.55360624
514 32.49027237
515 32.62135922
516 32.75193798
517 32.88201161
518 32.81853282
519 32.75529865
520 32.69230769
521 32.62955854
522 32.56704981
523 32.50478011
524 32.44274809
525 32.38095238
526 32.31939163
527 32.44781784
528 32.38636364
529 32.32514178
530 32.26415094
531 32.20338983
532 32.33082707
533 32.45778612
534 32.39700375
535 32.33644860
536 32.27611940
537 32.40223464
538 32.34200743
539 32.28200371
540 32.22222222
541 32.16266174
542 32.10332103
543 32.04419890
544 31.98529412
545 31.92660550
546 32.05128205
547 32.17550274
548 32.29927007
549 32.24043716
550 32.18181818
551 32.12341198
552 32.24637681
553 32.36889693
554 32.31046931
555 32.25225225
556 32.19424460
557 32.13644524
558 32.07885305
559 32.02146691
560 32.14285714
561 32.26381462
562 32.20640569
563 32.14920071
564 32.26950355
565 32.21238938
566 32.33215548
567 32.45149912
568 32.57042254
569 32.51318102
570 32.45614035
571 32.57443082
572 32.51748252
573 32.46073298
574 32.40418118
575 32.52173913
576 32.63888889
577 32.75563258
578 32.69896194
579 32.81519862
580 32.75862069
581 32.87435456
582 32.81786942
583 32.76157804
584 32.87671233
585 32.82051282
586 32.93515358
587 32.87904600
588 32.82312925
589 32.76740238
590 32.88135593
591 32.82571912
592 32.77027027
593 32.71500843
594 32.65993266
595 32.60504202
596 32.55033557
597 32.49581240
598 32.44147157
599 32.38731219
600 32.33333333
601 32.44592346
602 32.55813953
603 32.50414594
604 32.45033113
605 32.39669421
606 32.50825083
607 32.45469522
608 32.40131579
609 32.51231527
610 32.45901639
611 32.40589198
612 32.51633987
613 32.46329527
614 32.41042345
615 32.35772358
616 32.30519481
617 32.41491086
618 32.36245955
619 32.47172859
620 32.41935484
621 32.36714976
622 32.47588424
623 32.42375602
624 32.37179487
625 32.32000000
626 32.26837061
627 32.21690590
628 32.32484076
629 32.27344992
630 32.22222222
631 32.17115689
632 32.12025316
633 32.06951027
634 32.01892744
635 31.96850394
636 31.91823899
637 31.86813187
638 31.81818182
639 31.76838811
640 31.87500000
641 31.82527301
642 31.93146417
643 31.88180404
644 31.83229814
645 31.93798450
646 31.88854489
647 31.99381762
648 32.09876543
649 32.04930663
650 32.15384615
651 32.25806452
652 32.36196319
653 32.46554364
654 32.41590214
655 32.36641221
656 32.46951220
657 32.57229833
658 32.52279635
659 32.62518968
660 32.57575758
661 32.52647504
662 32.47734139
663 32.42835596
664 32.37951807
665 32.33082707
666 32.43243243
667 32.38380810
668 32.33532934
669 32.28699552
670 32.38805970
671 32.33979136
672 32.29166667
673 32.24368499
674 32.19584570
675 32.29629630
676 32.39644970
677 32.34859675
678 32.30088496
679 32.25331370
680 32.20588235
681 32.15859031
682 32.25806452
683 32.21083455
684 32.30994152
685 32.26277372
686 32.21574344
687 32.16885007
688 32.12209302
689 32.07547170
690 32.02898551
691 32.12735166
692 32.22543353
693 32.32323232
694 32.27665706
695 32.23021583
696 32.32758621
697 32.28120516
698 32.23495702
699 32.18884120
700 32.28571429
701 32.23965763
702 32.19373219
703 32.14793741
704 32.10227273
705 32.05673759
706 32.15297450
707 32.24893918
708 32.20338983
709 32.15796897
710 32.11267606
711 32.06751055
712 32.02247191
713 31.97755961
714 31.93277311
715 31.88811189
716 31.84357542
717 31.93863319
718 32.03342618
719 32.12795549
720 32.08333333
721 32.03883495
722 31.99445983
723 31.95020747
724 31.90607735
725 31.86206897
726 31.95592287
727 31.91196699
728 31.86813187
729 31.82441701
730 31.78082192
731 31.73734610
732 31.83060109
733 31.78717599
734 31.74386921
735 31.70068027
736 31.65760870
737 31.75033921
738 31.70731707
739 31.79972936
740 31.89189189
741 31.84885290
742 31.80592992
743 31.76312248
744 31.85483871
745 31.81208054
746 31.90348525
747 31.99464525
748 31.95187166
749 32.04272363
750 32.13333333
Final result: 32.1333 +/- 1.7063
Random chance: 19.8992 +/- 1.4588