yzhuang commited on
Commit
5c533a7
·
verified ·
1 Parent(s): 14d4768

Training in progress, step 200

Browse files
config.json ADDED
@@ -0,0 +1,861 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "yzhuang/llava-jamba-1.5-Mini-linear-anyres-patch",
3
+ "architectures": [
4
+ "LlavaJambaForCausalLM"
5
+ ],
6
+ "attention_dropout": 0.0,
7
+ "attn_layer_offset": 4,
8
+ "attn_layer_period": 8,
9
+ "bos_token_id": 1,
10
+ "eos_token_id": [
11
+ 2,
12
+ 518
13
+ ],
14
+ "expert_layer_offset": 1,
15
+ "expert_layer_period": 2,
16
+ "freeze_mm_mlp_adapter": false,
17
+ "hidden_act": "silu",
18
+ "hidden_size": 4096,
19
+ "image_aspect_ratio": "anyres",
20
+ "image_grid_pinpoints": [
21
+ [
22
+ 336,
23
+ 336
24
+ ],
25
+ [
26
+ 336,
27
+ 672
28
+ ],
29
+ [
30
+ 336,
31
+ 1008
32
+ ],
33
+ [
34
+ 336,
35
+ 1344
36
+ ],
37
+ [
38
+ 336,
39
+ 1680
40
+ ],
41
+ [
42
+ 336,
43
+ 2016
44
+ ],
45
+ [
46
+ 336,
47
+ 2352
48
+ ],
49
+ [
50
+ 336,
51
+ 2688
52
+ ],
53
+ [
54
+ 336,
55
+ 3024
56
+ ],
57
+ [
58
+ 336,
59
+ 3360
60
+ ],
61
+ [
62
+ 336,
63
+ 3696
64
+ ],
65
+ [
66
+ 336,
67
+ 4032
68
+ ],
69
+ [
70
+ 336,
71
+ 4368
72
+ ],
73
+ [
74
+ 336,
75
+ 4704
76
+ ],
77
+ [
78
+ 336,
79
+ 5040
80
+ ],
81
+ [
82
+ 336,
83
+ 5376
84
+ ],
85
+ [
86
+ 336,
87
+ 5712
88
+ ],
89
+ [
90
+ 336,
91
+ 6048
92
+ ],
93
+ [
94
+ 336,
95
+ 6384
96
+ ],
97
+ [
98
+ 336,
99
+ 6720
100
+ ],
101
+ [
102
+ 336,
103
+ 7056
104
+ ],
105
+ [
106
+ 336,
107
+ 7392
108
+ ],
109
+ [
110
+ 336,
111
+ 7728
112
+ ],
113
+ [
114
+ 336,
115
+ 8064
116
+ ],
117
+ [
118
+ 336,
119
+ 8400
120
+ ],
121
+ [
122
+ 336,
123
+ 8736
124
+ ],
125
+ [
126
+ 336,
127
+ 9072
128
+ ],
129
+ [
130
+ 336,
131
+ 9408
132
+ ],
133
+ [
134
+ 336,
135
+ 9744
136
+ ],
137
+ [
138
+ 336,
139
+ 10080
140
+ ],
141
+ [
142
+ 336,
143
+ 10416
144
+ ],
145
+ [
146
+ 336,
147
+ 10752
148
+ ],
149
+ [
150
+ 336,
151
+ 11088
152
+ ],
153
+ [
154
+ 336,
155
+ 11424
156
+ ],
157
+ [
158
+ 336,
159
+ 11760
160
+ ],
161
+ [
162
+ 336,
163
+ 12096
164
+ ],
165
+ [
166
+ 336,
167
+ 12432
168
+ ],
169
+ [
170
+ 336,
171
+ 12768
172
+ ],
173
+ [
174
+ 336,
175
+ 13104
176
+ ],
177
+ [
178
+ 336,
179
+ 13440
180
+ ],
181
+ [
182
+ 336,
183
+ 13776
184
+ ],
185
+ [
186
+ 336,
187
+ 14112
188
+ ],
189
+ [
190
+ 336,
191
+ 14448
192
+ ],
193
+ [
194
+ 336,
195
+ 14784
196
+ ],
197
+ [
198
+ 336,
199
+ 15120
200
+ ],
201
+ [
202
+ 336,
203
+ 15456
204
+ ],
205
+ [
206
+ 336,
207
+ 15792
208
+ ],
209
+ [
210
+ 336,
211
+ 16128
212
+ ],
213
+ [
214
+ 672,
215
+ 336
216
+ ],
217
+ [
218
+ 672,
219
+ 672
220
+ ],
221
+ [
222
+ 672,
223
+ 1008
224
+ ],
225
+ [
226
+ 672,
227
+ 1344
228
+ ],
229
+ [
230
+ 672,
231
+ 1680
232
+ ],
233
+ [
234
+ 672,
235
+ 2016
236
+ ],
237
+ [
238
+ 672,
239
+ 2352
240
+ ],
241
+ [
242
+ 672,
243
+ 2688
244
+ ],
245
+ [
246
+ 672,
247
+ 3024
248
+ ],
249
+ [
250
+ 672,
251
+ 3360
252
+ ],
253
+ [
254
+ 672,
255
+ 3696
256
+ ],
257
+ [
258
+ 672,
259
+ 4032
260
+ ],
261
+ [
262
+ 672,
263
+ 4368
264
+ ],
265
+ [
266
+ 672,
267
+ 4704
268
+ ],
269
+ [
270
+ 672,
271
+ 5040
272
+ ],
273
+ [
274
+ 672,
275
+ 5376
276
+ ],
277
+ [
278
+ 672,
279
+ 5712
280
+ ],
281
+ [
282
+ 672,
283
+ 6048
284
+ ],
285
+ [
286
+ 672,
287
+ 6384
288
+ ],
289
+ [
290
+ 672,
291
+ 6720
292
+ ],
293
+ [
294
+ 672,
295
+ 7056
296
+ ],
297
+ [
298
+ 672,
299
+ 7392
300
+ ],
301
+ [
302
+ 672,
303
+ 7728
304
+ ],
305
+ [
306
+ 672,
307
+ 8064
308
+ ],
309
+ [
310
+ 1008,
311
+ 336
312
+ ],
313
+ [
314
+ 1008,
315
+ 672
316
+ ],
317
+ [
318
+ 1008,
319
+ 1008
320
+ ],
321
+ [
322
+ 1008,
323
+ 1344
324
+ ],
325
+ [
326
+ 1008,
327
+ 1680
328
+ ],
329
+ [
330
+ 1008,
331
+ 2016
332
+ ],
333
+ [
334
+ 1008,
335
+ 2352
336
+ ],
337
+ [
338
+ 1008,
339
+ 2688
340
+ ],
341
+ [
342
+ 1008,
343
+ 3024
344
+ ],
345
+ [
346
+ 1008,
347
+ 3360
348
+ ],
349
+ [
350
+ 1008,
351
+ 3696
352
+ ],
353
+ [
354
+ 1008,
355
+ 4032
356
+ ],
357
+ [
358
+ 1008,
359
+ 4368
360
+ ],
361
+ [
362
+ 1008,
363
+ 4704
364
+ ],
365
+ [
366
+ 1008,
367
+ 5040
368
+ ],
369
+ [
370
+ 1008,
371
+ 5376
372
+ ],
373
+ [
374
+ 1344,
375
+ 336
376
+ ],
377
+ [
378
+ 1344,
379
+ 672
380
+ ],
381
+ [
382
+ 1344,
383
+ 1008
384
+ ],
385
+ [
386
+ 1344,
387
+ 1344
388
+ ],
389
+ [
390
+ 1344,
391
+ 1680
392
+ ],
393
+ [
394
+ 1344,
395
+ 2016
396
+ ],
397
+ [
398
+ 1344,
399
+ 2352
400
+ ],
401
+ [
402
+ 1344,
403
+ 2688
404
+ ],
405
+ [
406
+ 1344,
407
+ 3024
408
+ ],
409
+ [
410
+ 1344,
411
+ 3360
412
+ ],
413
+ [
414
+ 1344,
415
+ 3696
416
+ ],
417
+ [
418
+ 1344,
419
+ 4032
420
+ ],
421
+ [
422
+ 1680,
423
+ 336
424
+ ],
425
+ [
426
+ 1680,
427
+ 672
428
+ ],
429
+ [
430
+ 1680,
431
+ 1008
432
+ ],
433
+ [
434
+ 1680,
435
+ 1344
436
+ ],
437
+ [
438
+ 1680,
439
+ 1680
440
+ ],
441
+ [
442
+ 1680,
443
+ 2016
444
+ ],
445
+ [
446
+ 1680,
447
+ 2352
448
+ ],
449
+ [
450
+ 1680,
451
+ 2688
452
+ ],
453
+ [
454
+ 1680,
455
+ 3024
456
+ ],
457
+ [
458
+ 2016,
459
+ 336
460
+ ],
461
+ [
462
+ 2016,
463
+ 672
464
+ ],
465
+ [
466
+ 2016,
467
+ 1008
468
+ ],
469
+ [
470
+ 2016,
471
+ 1344
472
+ ],
473
+ [
474
+ 2016,
475
+ 1680
476
+ ],
477
+ [
478
+ 2016,
479
+ 2016
480
+ ],
481
+ [
482
+ 2016,
483
+ 2352
484
+ ],
485
+ [
486
+ 2016,
487
+ 2688
488
+ ],
489
+ [
490
+ 2352,
491
+ 336
492
+ ],
493
+ [
494
+ 2352,
495
+ 672
496
+ ],
497
+ [
498
+ 2352,
499
+ 1008
500
+ ],
501
+ [
502
+ 2352,
503
+ 1344
504
+ ],
505
+ [
506
+ 2352,
507
+ 1680
508
+ ],
509
+ [
510
+ 2352,
511
+ 2016
512
+ ],
513
+ [
514
+ 2352,
515
+ 2352
516
+ ],
517
+ [
518
+ 2688,
519
+ 336
520
+ ],
521
+ [
522
+ 2688,
523
+ 672
524
+ ],
525
+ [
526
+ 2688,
527
+ 1008
528
+ ],
529
+ [
530
+ 2688,
531
+ 1344
532
+ ],
533
+ [
534
+ 2688,
535
+ 1680
536
+ ],
537
+ [
538
+ 2688,
539
+ 2016
540
+ ],
541
+ [
542
+ 3024,
543
+ 336
544
+ ],
545
+ [
546
+ 3024,
547
+ 672
548
+ ],
549
+ [
550
+ 3024,
551
+ 1008
552
+ ],
553
+ [
554
+ 3024,
555
+ 1344
556
+ ],
557
+ [
558
+ 3024,
559
+ 1680
560
+ ],
561
+ [
562
+ 3360,
563
+ 336
564
+ ],
565
+ [
566
+ 3360,
567
+ 672
568
+ ],
569
+ [
570
+ 3360,
571
+ 1008
572
+ ],
573
+ [
574
+ 3360,
575
+ 1344
576
+ ],
577
+ [
578
+ 3696,
579
+ 336
580
+ ],
581
+ [
582
+ 3696,
583
+ 672
584
+ ],
585
+ [
586
+ 3696,
587
+ 1008
588
+ ],
589
+ [
590
+ 3696,
591
+ 1344
592
+ ],
593
+ [
594
+ 4032,
595
+ 336
596
+ ],
597
+ [
598
+ 4032,
599
+ 672
600
+ ],
601
+ [
602
+ 4032,
603
+ 1008
604
+ ],
605
+ [
606
+ 4032,
607
+ 1344
608
+ ],
609
+ [
610
+ 4368,
611
+ 336
612
+ ],
613
+ [
614
+ 4368,
615
+ 672
616
+ ],
617
+ [
618
+ 4368,
619
+ 1008
620
+ ],
621
+ [
622
+ 4704,
623
+ 336
624
+ ],
625
+ [
626
+ 4704,
627
+ 672
628
+ ],
629
+ [
630
+ 4704,
631
+ 1008
632
+ ],
633
+ [
634
+ 5040,
635
+ 336
636
+ ],
637
+ [
638
+ 5040,
639
+ 672
640
+ ],
641
+ [
642
+ 5040,
643
+ 1008
644
+ ],
645
+ [
646
+ 5376,
647
+ 336
648
+ ],
649
+ [
650
+ 5376,
651
+ 672
652
+ ],
653
+ [
654
+ 5376,
655
+ 1008
656
+ ],
657
+ [
658
+ 5712,
659
+ 336
660
+ ],
661
+ [
662
+ 5712,
663
+ 672
664
+ ],
665
+ [
666
+ 6048,
667
+ 336
668
+ ],
669
+ [
670
+ 6048,
671
+ 672
672
+ ],
673
+ [
674
+ 6384,
675
+ 336
676
+ ],
677
+ [
678
+ 6384,
679
+ 672
680
+ ],
681
+ [
682
+ 6720,
683
+ 336
684
+ ],
685
+ [
686
+ 6720,
687
+ 672
688
+ ],
689
+ [
690
+ 7056,
691
+ 336
692
+ ],
693
+ [
694
+ 7056,
695
+ 672
696
+ ],
697
+ [
698
+ 7392,
699
+ 336
700
+ ],
701
+ [
702
+ 7392,
703
+ 672
704
+ ],
705
+ [
706
+ 7728,
707
+ 336
708
+ ],
709
+ [
710
+ 7728,
711
+ 672
712
+ ],
713
+ [
714
+ 8064,
715
+ 336
716
+ ],
717
+ [
718
+ 8064,
719
+ 672
720
+ ],
721
+ [
722
+ 8400,
723
+ 336
724
+ ],
725
+ [
726
+ 8736,
727
+ 336
728
+ ],
729
+ [
730
+ 9072,
731
+ 336
732
+ ],
733
+ [
734
+ 9408,
735
+ 336
736
+ ],
737
+ [
738
+ 9744,
739
+ 336
740
+ ],
741
+ [
742
+ 10080,
743
+ 336
744
+ ],
745
+ [
746
+ 10416,
747
+ 336
748
+ ],
749
+ [
750
+ 10752,
751
+ 336
752
+ ],
753
+ [
754
+ 11088,
755
+ 336
756
+ ],
757
+ [
758
+ 11424,
759
+ 336
760
+ ],
761
+ [
762
+ 11760,
763
+ 336
764
+ ],
765
+ [
766
+ 12096,
767
+ 336
768
+ ],
769
+ [
770
+ 12432,
771
+ 336
772
+ ],
773
+ [
774
+ 12768,
775
+ 336
776
+ ],
777
+ [
778
+ 13104,
779
+ 336
780
+ ],
781
+ [
782
+ 13440,
783
+ 336
784
+ ],
785
+ [
786
+ 13776,
787
+ 336
788
+ ],
789
+ [
790
+ 14112,
791
+ 336
792
+ ],
793
+ [
794
+ 14448,
795
+ 336
796
+ ],
797
+ [
798
+ 14784,
799
+ 336
800
+ ],
801
+ [
802
+ 15120,
803
+ 336
804
+ ],
805
+ [
806
+ 15456,
807
+ 336
808
+ ],
809
+ [
810
+ 15792,
811
+ 336
812
+ ],
813
+ [
814
+ 16128,
815
+ 336
816
+ ]
817
+ ],
818
+ "initializer_range": 0.02,
819
+ "intermediate_size": 14336,
820
+ "mamba_conv_bias": true,
821
+ "mamba_d_conv": 4,
822
+ "mamba_d_state": 16,
823
+ "mamba_dt_rank": 256,
824
+ "mamba_expand": 2,
825
+ "mamba_proj_bias": false,
826
+ "max_position_embeddings": 262144,
827
+ "mm_hidden_size": 1024,
828
+ "mm_patch_merge_type": "spatial_unpad",
829
+ "mm_projector_lr": 2e-05,
830
+ "mm_projector_type": "linear",
831
+ "mm_use_im_patch_token": false,
832
+ "mm_use_im_start_end": false,
833
+ "mm_vision_select_feature": "patch",
834
+ "mm_vision_select_layer": -2,
835
+ "mm_vision_tower": "openai/clip-vit-large-patch14-336",
836
+ "mm_vision_tower_lr": 2e-06,
837
+ "model_type": "jamba",
838
+ "num_attention_heads": 32,
839
+ "num_experts": 16,
840
+ "num_experts_per_tok": 2,
841
+ "num_hidden_layers": 32,
842
+ "num_key_value_heads": 8,
843
+ "num_logits_to_keep": 1,
844
+ "output_router_logits": false,
845
+ "pad_token_id": 0,
846
+ "rms_norm_eps": 1e-06,
847
+ "router_aux_loss_coef": 0.001,
848
+ "sliding_window": null,
849
+ "tie_word_embeddings": false,
850
+ "tokenizer_model_max_length": 32896,
851
+ "tokenizer_padding_side": "right",
852
+ "torch_dtype": "bfloat16",
853
+ "transformers_version": "4.45.0.dev0",
854
+ "tune_mm_mlp_adapter": false,
855
+ "unfreeze_mm_vision_tower": true,
856
+ "use_cache": false,
857
+ "use_image_rope": true,
858
+ "use_mamba_kernels": true,
859
+ "use_mm_proj": true,
860
+ "vocab_size": 65536
861
+ }
model-00001-of-00022.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a1cfbb380ad8b4e1076ea9661ece0569f2936afd82212e90ee15c8ff640a5460
3
+ size 4951245440
model-00002-of-00022.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:098cbc2adb3d68287d61cbbccca4da5f5e6469ea404529efed18dbf652fdcd11
3
+ size 4884145352
model-00003-of-00022.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:172907986a7b689fef0083daeef8c71e3781b6007cdf9f1826572f9a9a77fad0
3
+ size 4992294984
model-00004-of-00022.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3a817f7d413816900866634aadf09481c4590577b53785afc1a09b2a036954a0
3
+ size 4958591424
model-00005-of-00022.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f457edc36965fc843896a704208ead577c511d17b92a9f8d3c1293525acc118c
3
+ size 4975501696
model-00006-of-00022.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5ac139ea5c91bcde10265a0187d44b420b82966876743b6d815c7d5b4cb91e22
3
+ size 4884145344
model-00007-of-00022.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c1c25d2886b6bdbbe28cf9f3567bffaf7a6dd9a78aab7a1e039fa5d3e1e91c4c
3
+ size 4884145368
model-00008-of-00022.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2004b400ff97cf34fffb6bd0bf77de81d503b395bcbc73eea5dbe61dac3450a4
3
+ size 4992295040
model-00009-of-00022.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:65a00fb5ee1a482148427bb96c4abc6136cbd9158ed46d920d282c4813d7172a
3
+ size 4932507176
model-00010-of-00022.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ac265198c0bd4749931143fe87dbe50cee821b239c16fd17962ac7a3d8f7c936
3
+ size 4884145400
model-00011-of-00022.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:09445ffaf16ca042c8a7a955da04ceb86e28c7f347e060548f5c51cdbd7feafd
3
+ size 4884145424
model-00012-of-00022.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1c2425467d643017a7f08a70e7bd2d417fae42811530d293819f80a31167ae9d
3
+ size 4884145416
model-00013-of-00022.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bc736be73de8d0d9360b000b4273297581a1f2ccb1d64bc0d23deeee3b35a623
3
+ size 4932507176
model-00014-of-00022.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9fde50f37073dc2c13c4e0bed07fa3440092397805ae07110d61f57ee335d8a5
3
+ size 4992295016
model-00015-of-00022.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:acd039f3480600f082b8c39ee23037976ba148050ffea7d4e4f227596dd28c52
3
+ size 4884145416
model-00016-of-00022.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3cab047cf3b747a8560f5cd602acc7cc62699a28f7126db21c1ab73d6b3293e4
3
+ size 4884145416
model-00017-of-00022.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:498a91f1f50f279d50f0167696ff0cf0c63dde32f54d9d45ee17835ed42b1f4f
3
+ size 4908260720
model-00018-of-00022.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:72690a84c00834a692030f610e348ba1dfc3fc69968eb6a838f187be5f629681
3
+ size 4908391864
model-00019-of-00022.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f4baa0c1c006c8161545116a4ee4195e755ac135e31ff18480568671d81c9ac4
3
+ size 4992295032
model-00020-of-00022.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4103cc99e23798b9225f2ec2d8ff4acffb362caaa739b8b60dc5a49e202732c0
3
+ size 4884145424
model-00021-of-00022.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8f83f405e98fb45fcf4fd8bd042f3f4129ae5b34bb5a2a86ff0b18473e65b52f
3
+ size 4725920848
model-00022-of-00022.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a6355f74567b654092bb340a39f59a80948c2de1f20540d2f15c2b147755ac8c
3
+ size 536872144
model.safetensors.index.json ADDED
The diff for this file is too large to render. See raw diff
 
special_tokens_map.json ADDED
@@ -0,0 +1,48 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<|eom|>",
4
+ "<|bom|>",
5
+ "<|system|>",
6
+ "<|user|>",
7
+ "<|assistant|>",
8
+ "<|tool|>",
9
+ "<documents>",
10
+ "</documents>",
11
+ "<tool_definitions>",
12
+ "</tool_definitions>",
13
+ "<active_output_modes>",
14
+ "</active_output_modes>",
15
+ "<citations>",
16
+ "</citations>",
17
+ "<tool_calls>",
18
+ "</tool_calls>"
19
+ ],
20
+ "bos_token": {
21
+ "content": "<|startoftext|>",
22
+ "lstrip": false,
23
+ "normalized": false,
24
+ "rstrip": false,
25
+ "single_word": false
26
+ },
27
+ "eos_token": {
28
+ "content": "<|endoftext|>",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false
33
+ },
34
+ "pad_token": {
35
+ "content": "<|pad|>",
36
+ "lstrip": false,
37
+ "normalized": false,
38
+ "rstrip": false,
39
+ "single_word": false
40
+ },
41
+ "unk_token": {
42
+ "content": "<|unk|>",
43
+ "lstrip": false,
44
+ "normalized": false,
45
+ "rstrip": false,
46
+ "single_word": false
47
+ }
48
+ }
tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8b0df4fb43262c452ef37061951a06df4c63ca191d02a60ea08f14428af24376
3
+ size 1124714
tokenizer_config.json ADDED
@@ -0,0 +1,198 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "add_prefix_space": true,
5
+ "added_tokens_decoder": {
6
+ "0": {
7
+ "content": "<|pad|>",
8
+ "lstrip": false,
9
+ "normalized": false,
10
+ "rstrip": false,
11
+ "single_word": false,
12
+ "special": true
13
+ },
14
+ "1": {
15
+ "content": "<|startoftext|>",
16
+ "lstrip": false,
17
+ "normalized": false,
18
+ "rstrip": false,
19
+ "single_word": false,
20
+ "special": true
21
+ },
22
+ "2": {
23
+ "content": "<|endoftext|>",
24
+ "lstrip": false,
25
+ "normalized": false,
26
+ "rstrip": false,
27
+ "single_word": false,
28
+ "special": true
29
+ },
30
+ "3": {
31
+ "content": "<|unk|>",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false,
36
+ "special": true
37
+ },
38
+ "518": {
39
+ "content": "<|eom|>",
40
+ "lstrip": false,
41
+ "normalized": false,
42
+ "rstrip": false,
43
+ "single_word": false,
44
+ "special": true
45
+ },
46
+ "519": {
47
+ "content": "<|bom|>",
48
+ "lstrip": false,
49
+ "normalized": false,
50
+ "rstrip": false,
51
+ "single_word": false,
52
+ "special": true
53
+ },
54
+ "520": {
55
+ "content": "<|system|>",
56
+ "lstrip": false,
57
+ "normalized": false,
58
+ "rstrip": false,
59
+ "single_word": false,
60
+ "special": true
61
+ },
62
+ "521": {
63
+ "content": "<|user|>",
64
+ "lstrip": false,
65
+ "normalized": false,
66
+ "rstrip": false,
67
+ "single_word": false,
68
+ "special": true
69
+ },
70
+ "522": {
71
+ "content": "<|assistant|>",
72
+ "lstrip": false,
73
+ "normalized": false,
74
+ "rstrip": false,
75
+ "single_word": false,
76
+ "special": true
77
+ },
78
+ "523": {
79
+ "content": "<|tool|>",
80
+ "lstrip": false,
81
+ "normalized": false,
82
+ "rstrip": false,
83
+ "single_word": false,
84
+ "special": true
85
+ },
86
+ "524": {
87
+ "content": "<documents>",
88
+ "lstrip": false,
89
+ "normalized": false,
90
+ "rstrip": false,
91
+ "single_word": false,
92
+ "special": true
93
+ },
94
+ "525": {
95
+ "content": "</documents>",
96
+ "lstrip": false,
97
+ "normalized": false,
98
+ "rstrip": false,
99
+ "single_word": false,
100
+ "special": true
101
+ },
102
+ "526": {
103
+ "content": "<tool_definitions>",
104
+ "lstrip": false,
105
+ "normalized": false,
106
+ "rstrip": false,
107
+ "single_word": false,
108
+ "special": true
109
+ },
110
+ "527": {
111
+ "content": "</tool_definitions>",
112
+ "lstrip": false,
113
+ "normalized": false,
114
+ "rstrip": false,
115
+ "single_word": false,
116
+ "special": true
117
+ },
118
+ "528": {
119
+ "content": "<active_output_modes>",
120
+ "lstrip": false,
121
+ "normalized": false,
122
+ "rstrip": false,
123
+ "single_word": false,
124
+ "special": true
125
+ },
126
+ "529": {
127
+ "content": "</active_output_modes>",
128
+ "lstrip": false,
129
+ "normalized": false,
130
+ "rstrip": false,
131
+ "single_word": false,
132
+ "special": true
133
+ },
134
+ "530": {
135
+ "content": "<citations>",
136
+ "lstrip": false,
137
+ "normalized": false,
138
+ "rstrip": false,
139
+ "single_word": false,
140
+ "special": true
141
+ },
142
+ "531": {
143
+ "content": "</citations>",
144
+ "lstrip": false,
145
+ "normalized": false,
146
+ "rstrip": false,
147
+ "single_word": false,
148
+ "special": true
149
+ },
150
+ "532": {
151
+ "content": "<tool_calls>",
152
+ "lstrip": false,
153
+ "normalized": false,
154
+ "rstrip": false,
155
+ "single_word": false,
156
+ "special": true
157
+ },
158
+ "533": {
159
+ "content": "</tool_calls>",
160
+ "lstrip": false,
161
+ "normalized": false,
162
+ "rstrip": false,
163
+ "single_word": false,
164
+ "special": true
165
+ }
166
+ },
167
+ "additional_special_tokens": [
168
+ "<|eom|>",
169
+ "<|bom|>",
170
+ "<|system|>",
171
+ "<|user|>",
172
+ "<|assistant|>",
173
+ "<|tool|>",
174
+ "<documents>",
175
+ "</documents>",
176
+ "<tool_definitions>",
177
+ "</tool_definitions>",
178
+ "<active_output_modes>",
179
+ "</active_output_modes>",
180
+ "<citations>",
181
+ "</citations>",
182
+ "<tool_calls>",
183
+ "</tool_calls>"
184
+ ],
185
+ "bos_token": "<|startoftext|>",
186
+ "chat_template": "{# Variables #}\n{% set ns = namespace(message_count=0, is_last_checked_defined=False) %}\n{##}\n{% set bom_str = bom_str or \"<|bom|>\" %}\n{% set eom_str = eom_str or \"<|eom|>\" %}\n{% set default_system_message = \"\" %}\n{##}\n{% set documents_prefix = \"<documents>\" %}\n{% set documents_suffix = \"</documents>\" %}\n{% set tool_definitions_prefix = \"<tool_definitions>\" %}\n{% set tool_definitions_suffix = \"</tool_definitions>\" %}\n{% set active_modes_prefix = \"<active_output_modes>\" %}\n{% set active_modes_suffix = \"</active_output_modes>\" %}\n{##}\n{% set tool_calls_prefix = \"<tool_calls>\" %}\n{% set tool_calls_suffix = \"</tool_calls>\" %}\n{% set citations_prefix = \"<citations>\" %}\n{% set citations_suffix = \"</citations>\" %}\n{##}\n{% if add_generation_prompt is not defined %}\n {% set add_generation_prompt = True %}\n{% endif %}\n{% set role_to_predict = role_to_predict or \"assistant\" %}\n{% if messages|length > 0 and messages[0].role == \"system\" %}\n {% set system_message = messages[0].content %}\n {% set loop_messages = messages[1:] %}\n{% else %}\n {% set system_message = default_system_message %}\n {% set loop_messages = messages %}\n{% endif %}\n{##}\n{##}\n{# Macros #}\n{% macro handle_tool_definitions(tools) %}\n {{- tool_definitions_prefix -}}\n {{- \"\\n# Tools\" -}}\n {{- \"\\n\\n## Functions\" -}}\n {% for tool in tools %}\n {% set _ = is_param_set(tool, field=\"type\") %}\n {% set is_tool_type_set = ns.is_last_checked_defined %}\n {% if is_tool_type_set %}\n {% if tool.type == \"function\" %}\n {% set tool = tool.function %}\n {% else %}\n {{ raise_exception(\"Currently, the only supported tool type is `function`\") }}\n {% endif %}\n {% endif %}\n {{- \"\\n\\n\" + (tool|tojson(indent=2)) -}}\n {% endfor %}\n {{- \"\\n\" + tool_definitions_suffix -}}\n{% endmacro %}\n{##}\n{% macro handle_first_system_message(system_message, tools) %}\n {{- bom_str + handle_role(\"system\") -}}\n {% set _ = is_param_set(system_message) %}\n {% set is_system_message_set = ns.is_last_checked_defined %}\n {% if is_system_message_set %}\n {{- system_message -}}\n {% endif %}\n {% set _ = is_param_set(tools, is_list=True) %}\n {% set is_tools_set = ns.is_last_checked_defined %}\n {% if is_tools_set %}\n {% if system_message %}\n {{- \"\\n\\n\" -}}\n {% endif %}\n {{- handle_tool_definitions(tools) -}}\n {% endif %}\n {% set ns.message_count = ns.message_count + 1 %}\n{% endmacro %}\n{##}\n{% macro handle_tool_calls(tool_calls) %}\n {{- tool_calls_prefix + \"[\\n\" -}}\n {% for tool_call in tool_calls %}\n {% set _ = is_param_set(tool_call, field=\"function\") %}\n {% set is_tool_call_function_set = ns.is_last_checked_defined %}\n {% if is_tool_call_function_set %}\n {%- set tool_call = tool_call.function %}\n {%- endif %}\n {% set arguments = tool_call.arguments %}\n {% if arguments is not string %}\n {%- set arguments = arguments|tojson -%}\n {%- endif %}\n {{ \"{\\\"name\\\": \\\"\" + tool_call.name + \"\\\", \\\"arguments\\\": \" + arguments + \"}\" -}}\n {% if not loop.last %}\n {{- \",\" }}\n {% endif %}\n {% endfor %}\n {{- \"\\n]\" + tool_calls_suffix -}}\n{% endmacro %}\n{##}\n{% macro handle_documents(documents) %}\n {{- documents_prefix -}}\n {{- \"\\n# Documents\" -}}\n {{- \"\\n\\nYou can use the following documents for reference:\" -}}\n {% for doc in documents %}\n {{- \"\\n\\n## Document ID: \" + loop.index0|string -}}\n {% set _ = is_param_set(doc, field=\"title\") %}\n {% set is_doc_title_set = ns.is_last_checked_defined %}\n {% if is_doc_title_set %}\n {{- \"\\nTitle: \" + doc.title -}}\n {% endif %}\n {% for key, value in doc.items() %}\n {% if key not in [\"title\", \"text\"] %}\n {{- \"\\n\" + key|title + \": \" + value|string -}}\n {% endif %}\n {% endfor %}\n {{- \"\\nText: \" + doc.text -}}\n {% endfor %}\n {{- \"\\n\" + documents_suffix -}}\n{% endmacro %}\n{##}\n{% macro handle_knobs(knobs) %}\n {{- active_modes_prefix -}}\n {{- \"\\n# Active Modes\" -}}\n {{ \"\\n\\nThe following modes configure the format or style of your responses. You should adhere to all currently\" -}}\n {{ \" active modes simultaneously.\" -}}\n {% if knobs.citation_mode == \"fast\" %}\n {{- \"\\n\\n## Citation Mode\" -}}\n {{- \"\\n\\nProvide a list of references only for the documents you base your response on. Format your response\" -}}\n {{ \" with the original answer followed by a citation section. Use this template:\" -}}\n {{ \" `{answer}\" + citations_prefix + \"DOCUMENT_IDS\" + citations_suffix + \"`, where DOCUMENT_IDS are the relevant document numbers\" -}}\n {{ \" (e.g. [2, 5, 9]), or [] if the answer cannot be supported by the provided documents.\" -}}\n {% endif %}\n {% if knobs.response_format == \"json_object\" %}\n {{- \"\\n\\n## JSON Mode\" -}}\n {{ \"\\n\\nProvide your response in JSON format. Adhere strictly to any schema given by the user.\" -}}\n {{ \" If an appropriate JSON format exists, use it without modification.\" -}}\n {% endif %}\n {{- \"\\n\" + active_modes_suffix -}}\n{% endmacro %}\n{##}\n{% macro get_last_user_index(messages) %}\n {% set ns.last_user_index = 0 %}\n {% for message in messages %}\n {% if message.role == 'user' %}\n {% set ns.last_user_index = loop.index0 %}\n {% endif %}\n {% endfor %}\n {{- ns.last_user_index -}}\n{% endmacro %}\n{##}\n{% macro handle_last_system_message(documents, knobs, use_documents, use_knobs) %}\n {{- bom_str + handle_role(\"system\") -}}\n {% set macros_to_call = [] %}\n {% set params_for_macros = [] %}\n {% if use_documents %}\n {% set macros_to_call = macros_to_call + [handle_documents] %}\n {% set params_for_macros = params_for_macros + [[documents]] %}\n {% endif %}\n {% if use_knobs %}\n {% set macros_to_call = macros_to_call + [handle_knobs] %}\n {% set params_for_macros = params_for_macros + [[knobs]] %}\n {% endif %}\n {% for i in range(macros_to_call|length) %}\n {% if i > 0 %}\n {{- \"\\n\\n\" -}}\n {% endif %}\n {{- macros_to_call[i](*params_for_macros[i]) -}}\n {% endfor %}\n {% set ns.message_count = ns.message_count + 1 %}\n{% endmacro %}\n{##}\n{% macro handle_role(role, add_space=True) %}\n {{- \"<|\" + role + \"|>\" -}}\n {% if add_space %}\n {{- \" \" -}}\n {% endif %}\n{% endmacro %}\n{##}\n{% macro is_param_set(param, field=none, is_list=False) %}\n {% if field is not none %}\n {% if field in param %}\n {% set param = param[field] %}\n {% else %}\n {% set param = none %}\n {% endif %}\n {% endif %}\n {% set is_defined = param is defined and param is not none %}\n {% if is_list %}\n {% set ns.is_last_checked_defined = is_defined and param|length > 0 %}\n {% else %}\n {% set ns.is_last_checked_defined = is_defined %}\n {% endif %}\n{% endmacro %}\n{##}\n{##}\n{# Template #}\n{{- \"<|startoftext|>\" -}}\n{% set _ = is_param_set(system_message) %}\n{% set is_system_message_set = ns.is_last_checked_defined %}\n{% set _ = is_param_set(tools, is_list=True) %}\n{% set is_tools_set = ns.is_last_checked_defined %}\n{% set has_system_message = (is_system_message_set or is_tools_set) %}\n{% if has_system_message %}\n {{- handle_first_system_message(system_message, tools) -}}\n{% endif %}\n{% set last_user_index = get_last_user_index(loop_messages)|int %}\n{% for message in loop_messages %}\n {% if loop.index0 == last_user_index %}\n {% set _ = is_param_set(documents, is_list=True) %}\n {% set use_documents = ns.is_last_checked_defined %}\n {% set _ = is_param_set(knobs) %}\n {% set use_knobs = ns.is_last_checked_defined and knobs.is_set %}\n {% set add_last_system_message = use_documents or use_knobs %}\n {% if add_last_system_message %}\n {% if ns.message_count > 0 %}\n {{- eom_str -}}\n {% endif %}\n {{- handle_last_system_message(documents, knobs, use_documents, use_knobs) -}}\n {% endif %}\n {% endif %}\n {% set role = message.role %}\n {% set _ = is_param_set(message, field=\"name\") %}\n {% set is_message_name_set = ns.is_last_checked_defined %}\n {% if is_message_name_set %}\n {% set message_prefix = handle_role(role) + \"(\" + message.name + \")\" %}\n {% else %}\n {% set message_prefix = handle_role(role) %}\n {% endif %}\n {% set content = (message.content or \"\") %}\n {% if content is not string %}\n {% set content = content|tojson %}\n {% endif %}\n {% if ns.message_count > 0 %}\n {{- eom_str -}}\n {% endif %}\n {{- bom_str + message_prefix + content -}}\n {% set _ = is_param_set(message, field=\"tool_calls\", is_list=True) %}\n {% set is_tool_calls_set = ns.is_last_checked_defined %}\n {% if role == \"assistant\" and is_tool_calls_set %}\n {{- handle_tool_calls(message.tool_calls) -}}\n {% endif %}\n {% set _ = is_param_set(message, field=\"citations\", is_list=True) %}\n {% set is_citations_set = ns.is_last_checked_defined %}\n {% if role == \"assistant\" and is_citations_set %}\n {{- citations_prefix + message.citations|map(attribute=\"document_id\")|list|string + citations_suffix -}}\n {% endif %}\n {% set ns.message_count = ns.message_count + 1 %}\n{% endfor %}\n{% if add_generation_prompt %}\n {% if ns.message_count > 0 %}\n {{- eom_str -}}\n {% endif %}\n {{- bom_str + handle_role(role_to_predict, add_space=False) -}}\n {% set _ = is_param_set(generation_preamble) %}\n {% set is_generation_preamble_set = ns.is_last_checked_defined %}\n {% if is_generation_preamble_set and generation_preamble.strip() != \"\" %}\n {{- \" \" + generation_preamble -}}\n {% endif %}\n {% set ns.message_count = ns.message_count + 1 %}\n{% else %}\n {% if ns.message_count > 0 %}\n {{- eom_str -}}\n {% endif %}\n{% endif %}\n",
187
+ "clean_up_tokenization_spaces": false,
188
+ "eos_token": "<|endoftext|>",
189
+ "legacy": true,
190
+ "model_max_length": 32896,
191
+ "pad_token": "<|pad|>",
192
+ "padding_side": "right",
193
+ "sp_model_kwargs": {},
194
+ "spaces_between_special_tokens": false,
195
+ "tokenizer_class": "LlamaTokenizer",
196
+ "unk_token": "<|unk|>",
197
+ "use_default_system_prompt": false
198
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:23e00acd99f0c631b6ea896e57d736a08a7e0acc5c2d97c0acb751ca012a50e3
3
+ size 7672