Llama-Guard-3-8B-GGUF / scores /Llama-Guard-3-8B-Q4_K_M-naive.arc
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 869 tasks in prompt
multiple_choice_score: selecting 750 random tasks from 869 tasks available
multiple_choice_score: preparing task data...done
multiple_choice_score : calculating TruthfulQA score over 750 tasks.
task acc_norm
1 100.00000000
2 100.00000000
3 100.00000000
4 75.00000000
5 80.00000000
6 83.33333333
7 85.71428571
8 87.50000000
9 77.77777778
10 80.00000000
11 72.72727273
12 75.00000000
13 76.92307692
14 78.57142857
15 80.00000000
16 81.25000000
17 82.35294118
18 77.77777778
19 78.94736842
20 75.00000000
21 76.19047619
22 77.27272727
23 73.91304348
24 75.00000000
25 72.00000000
26 69.23076923
27 70.37037037
28 67.85714286
29 68.96551724
30 70.00000000
31 67.74193548
32 65.62500000
33 63.63636364
34 61.76470588
35 60.00000000
36 58.33333333
37 59.45945946
38 57.89473684
39 58.97435897
40 60.00000000
41 60.97560976
42 59.52380952
43 58.13953488
44 56.81818182
45 57.77777778
46 58.69565217
47 59.57446809
48 60.41666667
49 61.22448980
50 60.00000000
51 60.78431373
52 61.53846154
53 62.26415094
54 61.11111111
55 60.00000000
56 60.71428571
57 61.40350877
58 62.06896552
59 62.71186441
60 63.33333333
61 62.29508197
62 61.29032258
63 61.90476190
64 62.50000000
65 61.53846154
66 62.12121212
67 62.68656716
68 63.23529412
69 63.76811594
70 62.85714286
71 61.97183099
72 61.11111111
73 61.64383562
74 60.81081081
75 61.33333333
76 60.52631579
77 59.74025974
78 60.25641026
79 59.49367089
80 60.00000000
81 60.49382716
82 60.97560976
83 60.24096386
84 60.71428571
85 61.17647059
86 61.62790698
87 60.91954023
88 61.36363636
89 61.79775281
90 62.22222222
91 62.63736264
92 63.04347826
93 62.36559140
94 61.70212766
95 62.10526316
96 62.50000000
97 61.85567010
98 62.24489796
99 62.62626263
100 63.00000000
101 63.36633663
102 63.72549020
103 63.10679612
104 62.50000000
105 62.85714286
106 62.26415094
107 62.61682243
108 62.03703704
109 61.46788991
110 61.81818182
111 62.16216216
112 62.50000000
113 62.83185841
114 63.15789474
115 63.47826087
116 62.93103448
117 63.24786325
118 63.55932203
119 63.02521008
120 63.33333333
121 62.80991736
122 63.11475410
123 63.41463415
124 62.90322581
125 63.20000000
126 63.49206349
127 63.77952756
128 64.06250000
129 63.56589147
130 63.84615385
131 64.12213740
132 64.39393939
133 64.66165414
134 64.17910448
135 63.70370370
136 63.97058824
137 64.23357664
138 64.49275362
139 64.74820144
140 65.00000000
141 64.53900709
142 64.78873239
143 65.03496503
144 65.27777778
145 65.51724138
146 65.06849315
147 64.62585034
148 64.86486486
149 65.10067114
150 65.33333333
151 65.56291391
152 65.78947368
153 66.01307190
154 65.58441558
155 65.80645161
156 66.02564103
157 66.24203822
158 65.82278481
159 66.03773585
160 66.25000000
161 66.45962733
162 66.66666667
163 66.87116564
164 66.46341463
165 66.06060606
166 66.26506024
167 65.86826347
168 66.07142857
169 65.68047337
170 65.88235294
171 66.08187135
172 66.27906977
173 66.47398844
174 66.66666667
175 66.28571429
176 66.47727273
177 66.66666667
178 66.29213483
179 66.48044693
180 66.66666667
181 66.85082873
182 67.03296703
183 66.66666667
184 66.84782609
185 67.02702703
186 67.20430108
187 67.37967914
188 67.55319149
189 67.19576720
190 66.84210526
191 67.01570681
192 67.18750000
193 66.83937824
194 67.01030928
195 67.17948718
196 66.83673469
197 67.00507614
198 67.17171717
199 66.83417085
200 67.00000000
201 67.16417910
202 67.32673267
203 67.48768473
204 67.64705882
205 67.31707317
206 66.99029126
207 67.14975845
208 67.30769231
209 67.46411483
210 67.61904762
211 67.77251185
212 67.45283019
213 67.60563380
214 67.75700935
215 67.90697674
216 67.59259259
217 67.74193548
218 67.43119266
219 67.57990868
220 67.72727273
221 67.42081448
222 67.56756757
223 67.26457399
224 66.96428571
225 66.66666667
226 66.37168142
227 66.51982379
228 66.66666667
229 66.81222707
230 66.95652174
231 66.66666667
232 66.37931034
233 66.52360515
234 66.66666667
235 66.38297872
236 66.52542373
237 66.66666667
238 66.80672269
239 66.94560669
240 67.08333333
241 67.21991701
242 67.35537190
243 67.48971193
244 67.21311475
245 67.34693878
246 67.47967480
247 67.61133603
248 67.74193548
249 67.87148594
250 68.00000000
251 68.12749004
252 68.25396825
253 68.37944664
254 68.11023622
255 68.23529412
256 67.96875000
257 67.70428016
258 67.82945736
259 67.56756757
260 67.69230769
261 67.81609195
262 67.93893130
263 68.06083650
264 68.18181818
265 67.92452830
266 67.66917293
267 67.79026217
268 67.91044776
269 68.02973978
270 68.14814815
271 67.89667897
272 67.64705882
273 67.39926740
274 67.15328467
275 67.27272727
276 67.39130435
277 67.50902527
278 67.26618705
279 67.38351254
280 67.50000000
281 67.61565836
282 67.73049645
283 67.49116608
284 67.60563380
285 67.71929825
286 67.83216783
287 67.94425087
288 68.05555556
289 67.82006920
290 67.93103448
291 68.04123711
292 68.15068493
293 68.25938567
294 68.02721088
295 68.13559322
296 68.24324324
297 68.35016835
298 68.45637584
299 68.56187291
300 68.33333333
301 68.10631229
302 68.21192053
303 68.31683168
304 68.42105263
305 68.19672131
306 67.97385621
307 68.07817590
308 68.18181818
309 68.28478964
310 68.38709677
311 68.48874598
312 68.26923077
313 68.37060703
314 68.15286624
315 67.93650794
316 68.03797468
317 68.13880126
318 67.92452830
319 67.71159875
320 67.50000000
321 67.60124611
322 67.70186335
323 67.80185759
324 67.59259259
325 67.69230769
326 67.79141104
327 67.58409786
328 67.68292683
329 67.78115502
330 67.57575758
331 67.67371601
332 67.77108434
333 67.86786787
334 67.66467066
335 67.46268657
336 67.26190476
337 67.35905045
338 67.15976331
339 67.25663717
340 67.35294118
341 67.44868035
342 67.25146199
343 67.34693878
344 67.44186047
345 67.24637681
346 67.05202312
347 66.85878963
348 66.66666667
349 66.76217765
350 66.85714286
351 66.95156695
352 67.04545455
353 66.85552408
354 66.66666667
355 66.47887324
356 66.57303371
357 66.66666667
358 66.75977654
359 66.57381616
360 66.66666667
361 66.75900277
362 66.85082873
363 66.94214876
364 67.03296703
365 67.12328767
366 66.93989071
367 66.75749319
368 66.84782609
369 66.93766938
370 67.02702703
371 66.84636119
372 66.93548387
373 67.02412869
374 67.11229947
375 66.93333333
376 67.02127660
377 67.10875332
378 67.19576720
379 67.01846966
380 67.10526316
381 67.19160105
382 67.01570681
383 66.84073107
384 66.92708333
385 66.75324675
386 66.83937824
387 66.92506460
388 67.01030928
389 66.83804627
390 66.92307692
391 66.75191816
392 66.83673469
393 66.92111959
394 67.00507614
395 67.08860759
396 66.91919192
397 67.00251889
398 66.83417085
399 66.91729323
400 67.00000000
401 66.83291771
402 66.91542289
403 66.99751861
404 67.07920792
405 67.16049383
406 67.24137931
407 67.32186732
408 67.40196078
409 67.23716381
410 67.31707317
411 67.39659367
412 67.47572816
413 67.55447942
414 67.63285024
415 67.46987952
416 67.30769231
417 67.38609113
418 67.46411483
419 67.30310263
420 67.38095238
421 67.22090261
422 67.06161137
423 67.13947991
424 66.98113208
425 67.05882353
426 67.13615023
427 67.21311475
428 67.05607477
429 67.13286713
430 67.20930233
431 67.28538283
432 67.36111111
433 67.20554273
434 67.28110599
435 67.35632184
436 67.43119266
437 67.50572082
438 67.57990868
439 67.65375854
440 67.50000000
441 67.57369615
442 67.42081448
443 67.49435666
444 67.56756757
445 67.64044944
446 67.71300448
447 67.78523490
448 67.85714286
449 67.92873051
450 67.77777778
451 67.84922395
452 67.92035398
453 67.99116998
454 67.84140969
455 67.69230769
456 67.76315789
457 67.83369803
458 67.90393013
459 67.97385621
460 67.82608696
461 67.89587852
462 67.96536797
463 67.81857451
464 67.67241379
465 67.74193548
466 67.81115880
467 67.88008565
468 67.94871795
469 67.80383795
470 67.87234043
471 67.94055202
472 68.00847458
473 67.86469345
474 67.93248945
475 68.00000000
476 68.06722689
477 68.13417191
478 68.20083682
479 68.26722338
480 68.33333333
481 68.19126819
482 68.25726141
483 68.32298137
484 68.38842975
485 68.45360825
486 68.51851852
487 68.37782341
488 68.44262295
489 68.30265849
490 68.16326531
491 68.22810591
492 68.29268293
493 68.15415822
494 68.21862348
495 68.08080808
496 68.14516129
497 68.20925553
498 68.27309237
499 68.33667335
500 68.40000000
501 68.26347305
502 68.32669323
503 68.38966203
504 68.45238095
505 68.51485149
506 68.37944664
507 68.24457594
508 68.30708661
509 68.17288802
510 68.23529412
511 68.10176125
512 67.96875000
513 68.03118908
514 68.09338521
515 68.15533981
516 68.02325581
517 68.08510638
518 68.14671815
519 68.20809249
520 68.07692308
521 68.13819578
522 68.19923372
523 68.06883365
524 68.12977099
525 68.00000000
526 67.87072243
527 67.74193548
528 67.61363636
529 67.48582231
530 67.54716981
531 67.60828625
532 67.66917293
533 67.72983114
534 67.60299625
535 67.47663551
536 67.53731343
537 67.59776536
538 67.47211896
539 67.53246753
540 67.40740741
541 67.46765250
542 67.52767528
543 67.58747698
544 67.64705882
545 67.70642202
546 67.58241758
547 67.45886654
548 67.51824818
549 67.57741348
550 67.63636364
551 67.51361162
552 67.39130435
553 67.45027125
554 67.50902527
555 67.56756757
556 67.44604317
557 67.50448833
558 67.38351254
559 67.44186047
560 67.50000000
561 67.55793226
562 67.61565836
563 67.67317940
564 67.55319149
565 67.43362832
566 67.49116608
567 67.54850088
568 67.60563380
569 67.48681898
570 67.36842105
571 67.42556918
572 67.30769231
573 67.36474695
574 67.42160279
575 67.47826087
576 67.53472222
577 67.59098787
578 67.64705882
579 67.70293610
580 67.58620690
581 67.64199656
582 67.69759450
583 67.75300172
584 67.80821918
585 67.86324786
586 67.74744027
587 67.80238501
588 67.85714286
589 67.91171477
590 67.96610169
591 68.02030457
592 68.07432432
593 68.12816189
594 68.18181818
595 68.06722689
596 67.95302013
597 67.83919598
598 67.72575251
599 67.77963272
600 67.83333333
601 67.88685524
602 67.77408638
603 67.66169154
604 67.71523179
605 67.76859504
606 67.82178218
607 67.71004942
608 67.76315789
609 67.65188834
610 67.70491803
611 67.59410802
612 67.64705882
613 67.53670473
614 67.58957655
615 67.64227642
616 67.53246753
617 67.58508914
618 67.47572816
619 67.36672052
620 67.41935484
621 67.47181965
622 67.36334405
623 67.41573034
624 67.46794872
625 67.36000000
626 67.25239617
627 67.30462520
628 67.35668790
629 67.24960254
630 67.30158730
631 67.35340729
632 67.24683544
633 67.14060032
634 67.19242902
635 67.24409449
636 67.29559748
637 67.18995290
638 67.24137931
639 67.29264476
640 67.34375000
641 67.39469579
642 67.28971963
643 67.18506998
644 67.23602484
645 67.13178295
646 67.18266254
647 67.23338485
648 67.12962963
649 67.02619414
650 67.07692308
651 67.12749616
652 67.17791411
653 67.07503828
654 66.97247706
655 66.87022901
656 66.76829268
657 66.66666667
658 66.56534954
659 66.46433991
660 66.51515152
661 66.41452345
662 66.46525680
663 66.51583710
664 66.56626506
665 66.61654135
666 66.66666667
667 66.71664168
668 66.61676647
669 66.51718984
670 66.56716418
671 66.46795827
672 66.51785714
673 66.41901932
674 66.32047478
675 66.37037037
676 66.27218935
677 66.32200886
678 66.37168142
679 66.27393225
680 66.17647059
681 66.22613803
682 66.12903226
683 66.03221083
684 66.08187135
685 66.13138686
686 66.03498542
687 65.93886463
688 65.98837209
689 65.89259797
690 65.94202899
691 65.84659913
692 65.89595376
693 65.94516595
694 65.99423631
695 65.89928058
696 65.80459770
697 65.71018651
698 65.61604585
699 65.52217454
700 65.57142857
701 65.47788873
702 65.52706553
703 65.57610242
704 65.62500000
705 65.67375887
706 65.72237960
707 65.77086280
708 65.81920904
709 65.86741890
710 65.77464789
711 65.82278481
712 65.87078652
713 65.91865358
714 65.82633053
715 65.73426573
716 65.78212291
717 65.82984658
718 65.73816156
719 65.64673157
720 65.69444444
721 65.74202497
722 65.78947368
723 65.83679115
724 65.74585635
725 65.79310345
726 65.70247934
727 65.61210454
728 65.65934066
729 65.70644719
730 65.75342466
731 65.80027360
732 65.84699454
733 65.89358799
734 65.94005450
735 65.98639456
736 66.03260870
737 65.94301221
738 65.98915989
739 66.03518268
740 66.08108108
741 66.12685560
742 66.17250674
743 66.21803499
744 66.12903226
745 66.17449664
746 66.21983914
747 66.13119143
748 66.04278075
749 65.95460614
750 66.00000000
Final result: 66.0000 +/- 1.7309
Random chance: 25.0083 +/- 1.5824