Views
No views yet
1merge_method: breadcrumbs_ties
2base_model: meta-llama/Llama-3.2-3B
3tokenizer_source: PJMixers-Dev/LLaMa-3.2-Instruct-JankMix-v0.1-SFT-3B
4dtype: bfloat16
5parameters:
6 normalize: true
7models:
8 - model: meta-llama/Llama-3.2-3B-Instruct
9 parameters:
10 weight: 1
11 density: 0.9
12 gamma: 0.01
13 normalize: true
14 int8_mask: true
15 random_seed: 0
16 temperature: 0.5
17 top_p: 0.65
18 inference: true
19 max_tokens: 999999999
20 stream: true
21 quantization:
22 method: int8
23 value: 100
24 quantization:
25 method: int4
26 value: 100
27 - model: PJMixers-Dev/LLaMa-3.2-Instruct-JankMix-v0.1-SFT-3B
28 parameters:
29 weight: 1
30 density: 0.9
31 gamma: 0.01
32 normalize: true
33 int8_mask: true
34 random_seed: 0
35 temperature: 0.5
36 top_p: 0.65
37 inference: true
38 max_tokens: 999999999
39 stream: true
40 quantization:
41 method: int8
42 value: 100
43 quantization:
44 method: int4
45 value: 100
46 - model: meta-llama/Llama-3.2-3B-Instruct
47 parameters:
48 weight: 1
49 density: 0.9
50 gamma: 0.01
51 normalize: true
52 int8_mask: true
53 random_seed: 0
54 temperature: 0.5
55 top_p: 0.65
56 inference: true
57 max_tokens: 999999999
58 stream: true
59 quantization:
60 method: int8
61 value: 100
62 quantization:
63 method: int4
64 value: 100
65 - model: PJMixers-Dev/LLaMa-3.2-Instruct-JankMix-v0.1-SFT-3B
66 parameters:
67 weight: 1
68 density: 0.9
69 gamma: 0.01
70 normalize: true
71 int8_mask: true
72 random_seed: 0
73 temperature: 0.5
74 top_p: 0.65
75 inference: true
76 max_tokens: 999999999
77 stream: true
78 quantization:
79 method: int8
80 value: 100
81 quantization:
82 method: int4
83 value: 100
84 - model: meta-llama/Llama-3.2-3B-Instruct
85 parameters:
86 weight: 1
87 density: 0.9
88 gamma: 0.01
89 normalize: true
90 int8_mask: true
91 random_seed: 0
92 temperature: 0.5
93 top_p: 0.65
94 inference: true
95 max_tokens: 999999999
96 stream: true
97 quantization:
98 method: int8
99 value: 100
100 quantization:
101 method: int4
102 value: 100
103 - model: PJMixers-Dev/LLaMa-3.2-Instruct-JankMix-v0.1-SFT-3B
104 parameters:
105 weight: 1
106 density: 0.9
107 gamma: 0.01
108 normalize: true
109 int8_mask: true
110 random_seed: 0
111 temperature: 0.5
112 top_p: 0.65
113 inference: true
114 max_tokens: 999999999
115 stream: true
116 quantization:
117 method: int8
118 value: 100
119 quantization:
120 method: int4
121 value: 100
122 - model: meta-llama/Llama-3.2-3B-Instruct
123 parameters:
124 weight: 1
125 density: 0.9
126 gamma: 0.01
127 normalize: true
128 int8_mask: true
129 random_seed: 0
130 temperature: 0.5
131 top_p: 0.65
132 inference: true
133 max_tokens: 999999999
134 stream: true
135 quantization:
136 method: int8
137 value: 100
138 quantization:
139 method: int4
140 value: 100
141 - model: PJMixers-Dev/LLaMa-3.2-Instruct-JankMix-v0.1-SFT-3B
142 parameters:
143 weight: 1
144 density: 0.9
145 gamma: 0.01
146 normalize: true
147 int8_mask: true
148 random_seed: 0
149 temperature: 0.5
150 top_p: 0.65
151 inference: true
152 max_tokens: 999999999
153 stream: true
154 quantization:
155 method: int8
156 value: 100
157 quantization:
158 method: int4
159 value: 100
160 - model: meta-llama/Llama-3.2-3B-Instruct
161 parameters:
162 weight: 1
163 density: 0.9
164 gamma: 0.01
165 normalize: true
166 int8_mask: true
167 random_seed: 0
168 temperature: 0.5
169 top_p: 0.65
170 inference: true
171 max_tokens: 999999999
172 stream: true
173 quantization:
174 method: int8
175 value: 100
176 quantization:
177 method: int4
178 value: 100
179 - model: PJMixers-Dev/LLaMa-3.2-Instruct-JankMix-v0.1-SFT-3B
180 parameters:
181 weight: 1
182 density: 0.9
183 gamma: 0.01
184 normalize: true
185 int8_mask: true
186 random_seed: 0
187 temperature: 0.5
188 top_p: 0.65
189 inference: true
190 max_tokens: 999999999
191 stream: true
192 quantization:
193 method: int8
194 value: 100
195 quantization:
196 method: int4
197 value: 100
198 - model: meta-llama/Llama-3.2-3B-Instruct
199 parameters:
200 weight: 1
201 density: 0.9
202 gamma: 0.01
203 normalize: true
204 int8_mask: true
205 random_seed: 0
206 temperature: 0.5
207 top_p: 0.65
208 inference: true
209 max_tokens: 999999999
210 stream: true
211 quantization:
212 method: int8
213 value: 100
214 quantization:
215 method: int4
216 value: 100
217 - model: PJMixers-Dev/LLaMa-3.2-Instruct-JankMix-v0.1-SFT-3B
218 parameters:
219 weight: 1
220 density: 0.9
221 gamma: 0.01
222 normalize: true
223 int8_mask: true
224 random_seed: 0
225 temperature: 0.5
226 top_p: 0.65
227 inference: true
228 max_tokens: 999999999
229 stream: true
230 quantization:
231 method: int8
232 value: 100
233 quantization:
234 method: int4
235 value: 100
236 - model: meta-llama/Llama-3.2-3B-Instruct
237 parameters:
238 weight: 1
239 density: 0.9
240 gamma: 0.01
241 normalize: true
242 int8_mask: true
243 random_seed: 0
244 temperature: 0.5
245 top_p: 0.65
246 inference: true
247 max_tokens: 999999999
248 stream: true
249 quantization:
250 method: int8
251 value: 100
252 quantization:
253 method: int4
254 value: 100
255 - model: PJMixers-Dev/LLaMa-3.2-Instruct-JankMix-v0.1-SFT-3B
256 parameters:
257 weight: 1
258 density: 0.9
259 gamma: 0.01
260 normalize: true
261 int8_mask: true
262 random_seed: 0
263 temperature: 0.5
264 top_p: 0.65
265 inference: true
266 max_tokens: 999999999
267 stream: true
268 quantization:
269 method: int8
270 value: 100
271 quantization:
272 method: int4
273 value: 100
274 - model: meta-llama/Llama-3.2-3B-Instruct
275 parameters:
276 weight: 1
277 density: 0.9
278 gamma: 0.01
279 normalize: true
280 int8_mask: true
281 random_seed: 0
282 temperature: 0.5
283 top_p: 0.65
284 inference: true
285 max_tokens: 999999999
286 stream: true
287 quantization:
288 method: int8
289 value: 100
290 quantization:
291 method: int4
292 value: 100
293 - model: PJMixers-Dev/LLaMa-3.2-Instruct-JankMix-v0.1-SFT-3B
294 parameters:
295 weight: 1
296 density: 0.9
297 gamma: 0.01
298 normalize: true
299 int8_mask: true
300 random_seed: 0
301 temperature: 0.5
302 top_p: 0.65
303 inference: true
304 max_tokens: 999999999
305 stream: true
306 quantization:
307 method: int8
308 value: 100
309 quantization:
310 method: int4
311 value: 100
312 - model: meta-llama/Llama-3.2-3B-Instruct
313 parameters:
314 weight: 1
315 density: 0.9
316 gamma: 0.01
317 normalize: true
318 int8_mask: true
319 random_seed: 0
320 temperature: 0.5
321 top_p: 0.65
322 inference: true
323 max_tokens: 999999999
324 stream: true
325 quantization:
326 method: int8
327 value: 100
328 quantization:
329 method: int4
330 value: 100
331 - model: PJMixers-Dev/LLaMa-3.2-Instruct-JankMix-v0.1-SFT-3B
332 parameters:
333 weight: 1
334 density: 0.9
335 gamma: 0.01
336 normalize: true
337 int8_mask: true
338 random_seed: 0
339 temperature: 0.5
340 top_p: 0.65
341 inference: true
342 max_tokens: 999999999
343 stream: true
344 quantization:
345 method: int8
346 value: 100
347 quantization:
348 method: int4
349 value: 100
350 - model: meta-llama/Llama-3.2-3B-Instruct
351 parameters:
352 weight: 1
353 density: 0.9
354 gamma: 0.01
355 normalize: true
356 int8_mask: true
357 random_seed: 0
358 temperature: 0.5
359 top_p: 0.65
360 inference: true
361 max_tokens: 999999999
362 stream: true
363 quantization:
364 method: int8
365 value: 100
366 quantization:
367 method: int4
368 value: 100
369 - model: PJMixers-Dev/LLaMa-3.2-Instruct-JankMix-v0.1-SFT-3B
370 parameters:
371 weight: 1
372 density: 0.9
373 gamma: 0.01
374 normalize: true
375 int8_mask: true
376 random_seed: 0
377 temperature: 0.5
378 top_p: 0.65
379 inference: true
380 max_tokens: 999999999
381 stream: true
382 quantization:
383 method: int8
384 value: 100
385 quantization:
386 method: int4
387 value: 100
388 - model: meta-llama/Llama-3.2-3B-Instruct
389 parameters:
390 weight: 1
391 density: 0.9
392 gamma: 0.01
393 normalize: true
394 int8_mask: true
395 random_seed: 0
396 temperature: 0.5
397 top_p: 0.65
398 inference: true
399 max_tokens: 999999999
400 stream: true
401 quantization:
402 method: int8
403 value: 100
404 quantization:
405 method: int4
406 value: 100
407 - model: PJMixers-Dev/LLaMa-3.2-Instruct-JankMix-v0.1-SFT-3B
408 parameters:
409 weight: 1
410 density: 0.9
411 gamma: 0.01
412 normalize: true
413 int8_mask: true
414 random_seed: 0
415 temperature: 0.5
416 top_p: 0.65
417 inference: true
418 max_tokens: 999999999
419 stream: true
420 quantization:
421 method: int8
422 value: 100
423 quantization:
424 method: int4
425 value: 100
426 - model: meta-llama/Llama-3.2-3B-Instruct
427 parameters:
428 weight: 1
429 density: 0.9
430 gamma: 0.01
431 normalize: true
432 int8_mask: true
433 random_seed: 0
434 temperature: 0.5
435 top_p: 0.65
436 inference: true
437 max_tokens: 999999999
438 stream: true
439 quantization:
440 method: int8
441 value: 100
442 quantization:
443 method: int4
444 value: 100
445 - model: PJMixers-Dev/LLaMa-3.2-Instruct-JankMix-v0.1-SFT-3B
446 parameters:
447 weight: 1
448 density: 0.9
449 gamma: 0.01
450 normalize: true
451 int8_mask: true
452 random_seed: 0
453 temperature: 0.5
454 top_p: 0.65
455 inference: true
456 max_tokens: 999999999
457 stream: true
458 quantization:
459 method: int8
460 value: 100
461 quantization:
462 method: int4
463 value: 100
464 - model: meta-llama/Llama-3.2-3B-Instruct
465 parameters:
466 weight: 1
467 density: 0.9
468 gamma: 0.01
469 normalize: true
470 int8_mask: true
471 random_seed: 0
472 temperature: 0.5
473 top_p: 0.65
474 inference: true
475 max_tokens: 999999999
476 stream: true
477 quantization:
478 method: int8
479 value: 100
480 quantization:
481 method: int4
482 value: 100
483 - model: PJMixers-Dev/LLaMa-3.2-Instruct-JankMix-v0.1-SFT-3B
484 parameters:
485 weight: 1
486 density: 0.9
487 gamma: 0.01
488 normalize: true
489 int8_mask: true
490 random_seed: 0
491 temperature: 0.5
492 top_p: 0.65
493 inference: true
494 max_tokens: 999999999
495 stream: true
496 quantization:
497 method: int8
498 value: 100
499 quantization:
500 method: int4
501 value: 100
502 - model: meta-llama/Llama-3.2-3B-Instruct
503 parameters:
504 weight: 1
505 density: 0.9
506 gamma: 0.01
507 normalize: true
508 int8_mask: true
509 random_seed: 0
510 temperature: 0.5
511 top_p: 0.65
512 inference: true
513 max_tokens: 999999999
514 stream: true
515 quantization:
516 method: int8
517 value: 100
518 quantization:
519 method: int4
520 value: 100
521 - model: PJMixers-Dev/LLaMa-3.2-Instruct-JankMix-v0.1-SFT-3B
522 parameters:
523 weight: 1
524 density: 0.9
525 gamma: 0.01
526 normalize: true
527 int8_mask: true
528 random_seed: 0
529 temperature: 0.5
530 top_p: 0.65
531 inference: true
532 max_tokens: 999999999
533 stream: true
534 quantization:
535 method: int8
536 value: 100
537 quantization:
538 method: int4
539 value: 100
540 - model: meta-llama/Llama-3.2-3B-Instruct
541 parameters:
542 weight: 1
543 density: 0.9
544 gamma: 0.01
545 normalize: true
546 int8_mask: true
547 random_seed: 0
548 temperature: 0.5
549 top_p: 0.65
550 inference: true
551 max_tokens: 999999999
552 stream: true
553 quantization:
554 method: int8
555 value: 100
556 quantization:
557 method: int4
558 value: 100
559 - model: PJMixers-Dev/LLaMa-3.2-Instruct-JankMix-v0.1-SFT-3B
560 parameters:
561 weight: 1
562 density: 0.9
563 gamma: 0.01
564 normalize: true
565 int8_mask: true
566 random_seed: 0
567 temperature: 0.5
568 top_p: 0.65
569 inference: true
570 max_tokens: 999999999
571 stream: true
572 quantization:
573 method: int8
574 value: 100
575 quantization:
576 method: int4
577 value: 100
578 - model: meta-llama/Llama-3.2-3B-Instruct
579 parameters:
580 weight: 1
581 density: 0.9
582 gamma: 0.01
583 normalize: true
584 int8_mask: true
585 random_seed: 0
586 temperature: 0.5
587 top_p: 0.65
588 inference: true
589 max_tokens: 999999999
590 stream: true
591 quantization:
592 method: int8
593 value: 100
594 quantization:
595 method: int4
596 value: 100
597 - model: PJMixers-Dev/LLaMa-3.2-Instruct-JankMix-v0.1-SFT-3B
598 parameters:
599 weight: 1
600 density: 0.9
601 gamma: 0.01
602 normalize: true
603 int8_mask: true
604 random_seed: 0
605 temperature: 0.5
606 top_p: 0.65
607 inference: true
608 max_tokens: 999999999
609 stream: true
610 quantization:
611 method: int8
612 value: 100
613 quantization:
614 method: int4
615 value: 100
616 - model: meta-llama/Llama-3.2-3B-Instruct
617 parameters:
618 weight: 1
619 density: 0.9
620 gamma: 0.01
621 normalize: true
622 int8_mask: true
623 random_seed: 0
624 temperature: 0.5
625 top_p: 0.65
626 inference: true
627 max_tokens: 999999999
628 stream: true
629 quantization:
630 method: int8
631 value: 100
632 quantization:
633 method: int4
634 value: 100
635 - model: PJMixers-Dev/LLaMa-3.2-Instruct-JankMix-v0.1-SFT-3B
636 parameters:
637 weight: 1
638 density: 0.9
639 gamma: 0.01
640 normalize: true
641 int8_mask: true
642 random_seed: 0
643 temperature: 0.5
644 top_p: 0.65
645 inference: true
646 max_tokens: 999999999
647 stream: true
648 quantization:
649 method: int8
650 value: 100
651 quantization:
652 method: int4
653 value: 100
654 - model: meta-llama/Llama-3.2-3B-Instruct
655 parameters:
656 weight: 1
657 density: 0.9
658 gamma: 0.01
659 normalize: true
660 int8_mask: true
661 random_seed: 0
662 temperature: 0.5
663 top_p: 0.65
664 inference: true
665 max_tokens: 999999999
666 stream: true
667 quantization:
668 method: int8
669 value: 100
670 quantization:
671 method: int4
672 value: 100
673 - model: PJMixers-Dev/LLaMa-3.2-Instruct-JankMix-v0.1-SFT-3B
674 parameters:
675 weight: 1
676 density: 0.9
677 gamma: 0.01
678 normalize: true
679 int8_mask: true
680 random_seed: 0
681 temperature: 0.5
682 top_p: 0.65
683 inference: true
684 max_tokens: 999999999
685 stream: true
686 quantization:
687 method: int8
688 value: 100
689 quantization:
690 method: int4
691 value: 100
692 - model: meta-llama/Llama-3.2-3B-Instruct
693 parameters:
694 weight: 1
695 density: 0.9
696 gamma: 0.01
697 normalize: true
698 int8_mask: true
699 random_seed: 0
700 temperature: 0.5
701 top_p: 0.65
702 inference: true
703 max_tokens: 999999999
704 stream: true
705 quantization:
706 method: int8
707 value: 100
708 quantization:
709 method: int4
710 value: 100
711 - model: PJMixers-Dev/LLaMa-3.2-Instruct-JankMix-v0.1-SFT-3B
712 parameters:
713 weight: 1
714 density: 0.9
715 gamma: 0.01
716 normalize: true
717 int8_mask: true
718 random_seed: 0
719 temperature: 0.5
720 top_p: 0.65
721 inference: true
722 max_tokens: 999999999
723 stream: true
724 quantization:
725 method: int8
726 value: 100
727 quantization:
728 method: int4
729 value: 100
730 - model: meta-llama/Llama-3.2-3B-Instruct
731 parameters:
732 weight: 1
733 density: 0.9
734 gamma: 0.01
735 normalize: true
736 int8_mask: true
737 random_seed: 0
738 temperature: 0.5
739 top_p: 0.65
740 inference: true
741 max_tokens: 999999999
742 stream: true
743 quantization:
744 method: int8
745 value: 100
746 quantization:
747 method: int4
748 value: 100
749 - model: PJMixers-Dev/LLaMa-3.2-Instruct-JankMix-v0.1-SFT-3B
750 parameters:
751 weight: 1
752 density: 0.9
753 gamma: 0.01
754 normalize: true
755 int8_mask: true
756 random_seed: 0
757 temperature: 0.5
758 top_p: 0.65
759 inference: true
760 max_tokens: 999999999
761 stream: true
762 quantization:
763 method: int8
764 value: 100
765 quantization:
766 method: int4
767 value: 100
768 - model: meta-llama/Llama-3.2-3B-Instruct
769 parameters:
770 weight: 1
771 density: 0.9
772 gamma: 0.01
773 normalize: true
774 int8_mask: true
775 random_seed: 0
776 temperature: 0.5
777 top_p: 0.65
778 inference: true
779 max_tokens: 999999999
780 stream: true
781 quantization:
782 method: int8
783 value: 100
784 quantization:
785 method: int4
786 value: 100
787 - model: PJMixers-Dev/LLaMa-3.2-Instruct-JankMix-v0.1-SFT-3B
788 parameters:
789 weight: 1
790 density: 0.9
791 gamma: 0.01
792 normalize: true
793 int8_mask: true
794 random_seed: 0
795 temperature: 0.5
796 top_p: 0.65
797 inference: true
798 max_tokens: 999999999
799 stream: true
800 quantization:
801 method: int8
802 value: 100
803 quantization:
804 method: int4
805 value: 100
806 - model: meta-llama/Llama-3.2-3B-Instruct
807 parameters:
808 weight: 1
809 density: 0.9
810 gamma: 0.01
811 normalize: true
812 int8_mask: true
813 random_seed: 0
814 temperature: 0.5
815 top_p: 0.65
816 inference: true
817 max_tokens: 999999999
818 stream: true
819 quantization:
820 method: int8
821 value: 100
822 quantization:
823 method: int4
824 value: 100
825 - model: PJMixers-Dev/LLaMa-3.2-Instruct-JankMix-v0.1-SFT-3B
826 parameters:
827 weight: 1
828 density: 0.9
829 gamma: 0.01
830 normalize: true
831 int8_mask: true
832 random_seed: 0
833 temperature: 0.5
834 top_p: 0.65
835 inference: true
836 max_tokens: 999999999
837 stream: true
838 quantization:
839 method: int8
840 value: 100
841 quantization:
842 method: int4
843 value: 100
844 - model: meta-llama/Llama-3.2-3B-Instruct
845 parameters:
846 weight: 1
847 density: 0.9
848 gamma: 0.01
849 normalize: true
850 int8_mask: true
851 random_seed: 0
852 temperature: 0.5
853 top_p: 0.65
854 inference: true
855 max_tokens: 999999999
856 stream: true
857 quantization:
858 method: int8
859 value: 100
860 quantization:
861 method: int4
862 value: 100
863 - model: PJMixers-Dev/LLaMa-3.2-Instruct-JankMix-v0.1-SFT-3B
864 parameters:
865 weight: 1
866 density: 0.9
867 gamma: 0.01
868 normalize: true
869 int8_mask: true
870 random_seed: 0
871 temperature: 0.5
872 top_p: 0.65
873 inference: true
874 max_tokens: 999999999
875 stream: true
876 quantization:
877 method: int8
878 value: 100
879 quantization:
880 method: int4
881 value: 100
882 - model: meta-llama/Llama-3.2-3B-Instruct
883 parameters:
884 weight: 1
885 density: 0.9
886 gamma: 0.01
887 normalize: true
888 int8_mask: true
889 random_seed: 0
890 temperature: 0.5
891 top_p: 0.65
892 inference: true
893 max_tokens: 999999999
894 stream: true
895 quantization:
896 method: int8
897 value: 100
898 quantization:
899 method: int4
900 value: 100
901 - model: PJMixers-Dev/LLaMa-3.2-Instruct-JankMix-v0.1-SFT-3B
902 parameters:
903 weight: 1
904 density: 0.9
905 gamma: 0.01
906 normalize: true
907 int8_mask: true
908 random_seed: 0
909 temperature: 0.5
910 top_p: 0.65
911 inference: true
912 max_tokens: 999999999
913 stream: true
914 quantization:
915 method: int8
916 value: 100
917 quantization:
918 method: int4
919 value: 100
920 - model: meta-llama/Llama-3.2-3B-Instruct
921 parameters:
922 weight: 1
923 density: 0.9
924 gamma: 0.01
925 normalize: true
926 int8_mask: true
927 random_seed: 0
928 temperature: 0.5
929 top_p: 0.65
930 inference: true
931 max_tokens: 999999999
932 stream: true
933 quantization:
934 method: int8
935 value: 100
936 quantization:
937 method: int4
938 value: 100
939 - model: PJMixers-Dev/LLaMa-3.2-Instruct-JankMix-v0.1-SFT-3B
940 parameters:
941 weight: 1
942 density: 0.9
943 gamma: 0.01
944 normalize: true
945 int8_mask: true
946 random_seed: 0
947 temperature: 0.5
948 top_p: 0.65
949 inference: true
950 max_tokens: 999999999
951 stream: true
952 quantization:
953 method: int8
954 value: 100
955 quantization:
956 method: int4
957 value: 100
958 - model: meta-llama/Llama-3.2-3B-Instruct
959 parameters:
960 weight: 1
961 density: 0.9
962 gamma: 0.01
963 normalize: true
964 int8_mask: true
965 random_seed: 0
966 temperature: 0.5
967 top_p: 0.65
968 inference: true
969 max_tokens: 999999999
970 stream: true
971 quantization:
972 method: int8
973 value: 100
974 quantization:
975 method: int4
976 value: 100
977 - model: PJMixers-Dev/LLaMa-3.2-Instruct-JankMix-v0.1-SFT-3B
978 parameters:
979 weight: 1
980 density: 0.9
981 gamma: 0.01
982 normalize: true
983 int8_mask: true
984 random_seed: 0
985 temperature: 0.5
986 top_p: 0.65
987 inference: true
988 max_tokens: 999999999
989 stream: true
990 quantization:
991 method: int8
992 value: 100
993 quantization:
994 method: int4
995 value: 100
996 - model: meta-llama/Llama-3.2-3B-Instruct
997 parameters:
998 weight: 1
999 density: 0.9
1000 gamma: 0.01
1001 normalize: true
1002 int8_mask: true
1003 random_seed: 0
1004 temperature: 0.5
1005 top_p: 0.65
1006 inference: true
1007 max_tokens: 999999999
1008 stream: true
1009 quantization:
1010 method: int8
1011 value: 100
1012 quantization:
1013 method: int4
1014 value: 100
1015 - model: PJMixers-Dev/LLaMa-3.2-Instruct-JankMix-v0.1-SFT-3B
1016 parameters:
1017 weight: 1
1018 density: 0.9
1019 gamma: 0.01
1020 normalize: true
1021 int8_mask: true
1022 random_seed: 0
1023 temperature: 0.5
1024 top_p: 0.65
1025 inference: true
1026 max_tokens: 999999999
1027 stream: true
1028 quantization:
1029 method: int8
1030 value: 100
1031 quantization:
1032 method: int4
1033 value: 100
1034 - model: meta-llama/Llama-3.2-3B-Instruct
1035 parameters:
1036 weight: 1
1037 density: 0.9
1038 gamma: 0.01
1039 normalize: true
1040 int8_mask: true
1041 random_seed: 0
1042 temperature: 0.5
1043 top_p: 0.65
1044 inference: true
1045 max_tokens: 999999999
1046 stream: true
1047 quantization:
1048 method: int8
1049 value: 100
1050 quantization:
1051 method: int4
1052 value: 100
1053 - model: PJMixers-Dev/LLaMa-3.2-Instruct-JankMix-v0.1-SFT-3B
1054 parameters:
1055 weight: 1
1056 density: 0.9
1057 gamma: 0.01
1058 normalize: true
1059 int8_mask: true
1060 random_seed: 0
1061 temperature: 0.5
1062 top_p: 0.65
1063 inference: true
1064 max_tokens: 999999999
1065 stream: true
1066 quantization:
1067 method: int8
1068 value: 100
1069 quantization:
1070 method: int4
1071 value: 100
1072 - model: meta-llama/Llama-3.2-3B-Instruct
1073 parameters:
1074 weight: 1
1075 density: 0.9
1076 gamma: 0.01
1077 normalize: true
1078 int8_mask: true
1079 random_seed: 0
1080 temperature: 0.5
1081 top_p: 0.65
1082 inference: true
1083 max_tokens: 999999999
1084 stream: true
1085 quantization:
1086 method: int8
1087 value: 100
1088 quantization:
1089 method: int4
1090 value: 100
1091 - model: PJMixers-Dev/LLaMa-3.2-Instruct-JankMix-v0.1-SFT-3B
1092 parameters:
1093 weight: 1
1094 density: 0.9
1095 gamma: 0.01
1096 normalize: true
1097 int8_mask: true
1098 random_seed: 0
1099 temperature: 0.5
1100 top_p: 0.65
1101 inference: true
1102 max_tokens: 999999999
1103 stream: true
1104 quantization:
1105 method: int8
1106 value: 100
1107 quantization:
1108 method: int4
1109 value: 100
1110