Hugging Face's logo

dumtjul
/
Ministral-3-3B-Instruct-2512-mlx-mxfp8

MLX
Safetensors
mistral3
8-bit precision
Model card Files
xet
 Community
dumtjul commited on
Commit
6a648ac
·
verified ·
1 Parent(s): a35c99c

Upload 11 files

Browse files
Files changed (9) hide show
  1. .gitattributes +1 -0
  2. config.json +19 -0
  3. model.safetensors +2 -2
  4. model.safetensors.index.json +227 -3
  5. params.json +47 -0
  6. processor_config.json +32 -0
  7. special_tokens_map.json +0 -0
  8. tekken.json +3 -0
  9. tokenizer_config.json +2 -2
.gitattributes CHANGED
@@ -34,3 +34,4 @@ saved_model/**/* 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
  tokenizer.json 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
  tokenizer.json filter=lfs diff=lfs merge=lfs -text
37
+ tekken.json filter=lfs diff=lfs merge=lfs -text
config.json CHANGED
@@ -50,5 +50,24 @@
50
  "vocab_size": 131072
51
  },
52
  "transformers_version": "5.0.0.dev0",
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
53
  "vision_feature_layer": -1
54
  }
 
50
  "vocab_size": 131072
51
  },
52
  "transformers_version": "5.0.0.dev0",
53
+ "vision_config": {
54
+ "attention_dropout": 0.0,
55
+ "head_dim": 64,
56
+ "hidden_act": "silu",
57
+ "hidden_size": 1024,
58
+ "image_size": 1540,
59
+ "initializer_range": 0.02,
60
+ "intermediate_size": 4096,
61
+ "model_type": "pixtral",
62
+ "num_attention_heads": 16,
63
+ "num_channels": 3,
64
+ "num_hidden_layers": 24,
65
+ "patch_size": 14,
66
+ "rope_parameters": {
67
+ "rope_theta": 10000.0,
68
+ "rope_type": "default"
69
+ },
70
+ "rope_theta": 10000.0
71
+ },
72
  "vision_feature_layer": -1
73
  }
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:a2381165997290f9021f7a0aa06b5373a067ac324066cae8c1a6acc709ded96a
3
- size 3536374072
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1618ade97187b782969227e115eefdaa8c54ac190092852cfa6077ff5b23e6c2
3
+ size 4360321853
model.safetensors.index.json CHANGED
@@ -1,7 +1,6 @@
1
  {
2
  "metadata": {
3
- "total_size": 3536320512,
4
- "total_parameters": 3429006336
5
  },
6
  "weight_map": {
7
  "language_model.model.embed_tokens.scales": "model.safetensors",
@@ -422,6 +421,231 @@
422
  "language_model.model.layers.9.self_attn.q_proj.weight": "model.safetensors",
423
  "language_model.model.layers.9.self_attn.v_proj.scales": "model.safetensors",
424
  "language_model.model.layers.9.self_attn.v_proj.weight": "model.safetensors",
425
- "language_model.model.norm.weight": "model.safetensors"
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
426
  }
427
  }
 
1
  {
2
  "metadata": {
3
+ "total_size": 4360235008
 
4
  },
5
  "weight_map": {
6
  "language_model.model.embed_tokens.scales": "model.safetensors",
 
421
  "language_model.model.layers.9.self_attn.q_proj.weight": "model.safetensors",
422
  "language_model.model.layers.9.self_attn.v_proj.scales": "model.safetensors",
423
  "language_model.model.layers.9.self_attn.v_proj.weight": "model.safetensors",
424
+ "language_model.model.norm.weight": "model.safetensors",
425
+ "multi_modal_projector.linear_1.scales": "model.safetensors",
426
+ "multi_modal_projector.linear_1.weight": "model.safetensors",
427
+ "multi_modal_projector.linear_2.scales": "model.safetensors",
428
+ "multi_modal_projector.linear_2.weight": "model.safetensors",
429
+ "multi_modal_projector.norm.weight": "model.safetensors",
430
+ "multi_modal_projector.patch_merger.merging_layer.scales": "model.safetensors",
431
+ "multi_modal_projector.patch_merger.merging_layer.weight": "model.safetensors",
432
+ "vision_tower.vision_model.ln_pre.weight": "model.safetensors",
433
+ "vision_tower.vision_model.patch_conv.weight": "model.safetensors",
434
+ "vision_tower.vision_model.transformer.layers.0.attention.k_proj.weight": "model.safetensors",
435
+ "vision_tower.vision_model.transformer.layers.0.attention.o_proj.weight": "model.safetensors",
436
+ "vision_tower.vision_model.transformer.layers.0.attention.q_proj.weight": "model.safetensors",
437
+ "vision_tower.vision_model.transformer.layers.0.attention.v_proj.weight": "model.safetensors",
438
+ "vision_tower.vision_model.transformer.layers.0.attention_norm.weight": "model.safetensors",
439
+ "vision_tower.vision_model.transformer.layers.0.feed_forward.down_proj.weight": "model.safetensors",
440
+ "vision_tower.vision_model.transformer.layers.0.feed_forward.gate_proj.weight": "model.safetensors",
441
+ "vision_tower.vision_model.transformer.layers.0.feed_forward.up_proj.weight": "model.safetensors",
442
+ "vision_tower.vision_model.transformer.layers.0.ffn_norm.weight": "model.safetensors",
443
+ "vision_tower.vision_model.transformer.layers.1.attention.k_proj.weight": "model.safetensors",
444
+ "vision_tower.vision_model.transformer.layers.1.attention.o_proj.weight": "model.safetensors",
445
+ "vision_tower.vision_model.transformer.layers.1.attention.q_proj.weight": "model.safetensors",
446
+ "vision_tower.vision_model.transformer.layers.1.attention.v_proj.weight": "model.safetensors",
447
+ "vision_tower.vision_model.transformer.layers.1.attention_norm.weight": "model.safetensors",
448
+ "vision_tower.vision_model.transformer.layers.1.feed_forward.down_proj.weight": "model.safetensors",
449
+ "vision_tower.vision_model.transformer.layers.1.feed_forward.gate_proj.weight": "model.safetensors",
450
+ "vision_tower.vision_model.transformer.layers.1.feed_forward.up_proj.weight": "model.safetensors",
451
+ "vision_tower.vision_model.transformer.layers.1.ffn_norm.weight": "model.safetensors",
452
+ "vision_tower.vision_model.transformer.layers.10.attention.k_proj.weight": "model.safetensors",
453
+ "vision_tower.vision_model.transformer.layers.10.attention.o_proj.weight": "model.safetensors",
454
+ "vision_tower.vision_model.transformer.layers.10.attention.q_proj.weight": "model.safetensors",
455
+ "vision_tower.vision_model.transformer.layers.10.attention.v_proj.weight": "model.safetensors",
456
+ "vision_tower.vision_model.transformer.layers.10.attention_norm.weight": "model.safetensors",
457
+ "vision_tower.vision_model.transformer.layers.10.feed_forward.down_proj.weight": "model.safetensors",
458
+ "vision_tower.vision_model.transformer.layers.10.feed_forward.gate_proj.weight": "model.safetensors",
459
+ "vision_tower.vision_model.transformer.layers.10.feed_forward.up_proj.weight": "model.safetensors",
460
+ "vision_tower.vision_model.transformer.layers.10.ffn_norm.weight": "model.safetensors",
461
+ "vision_tower.vision_model.transformer.layers.11.attention.k_proj.weight": "model.safetensors",
462
+ "vision_tower.vision_model.transformer.layers.11.attention.o_proj.weight": "model.safetensors",
463
+ "vision_tower.vision_model.transformer.layers.11.attention.q_proj.weight": "model.safetensors",
464
+ "vision_tower.vision_model.transformer.layers.11.attention.v_proj.weight": "model.safetensors",
465
+ "vision_tower.vision_model.transformer.layers.11.attention_norm.weight": "model.safetensors",
466
+ "vision_tower.vision_model.transformer.layers.11.feed_forward.down_proj.weight": "model.safetensors",
467
+ "vision_tower.vision_model.transformer.layers.11.feed_forward.gate_proj.weight": "model.safetensors",
468
+ "vision_tower.vision_model.transformer.layers.11.feed_forward.up_proj.weight": "model.safetensors",
469
+ "vision_tower.vision_model.transformer.layers.11.ffn_norm.weight": "model.safetensors",
470
+ "vision_tower.vision_model.transformer.layers.12.attention.k_proj.weight": "model.safetensors",
471
+ "vision_tower.vision_model.transformer.layers.12.attention.o_proj.weight": "model.safetensors",
472
+ "vision_tower.vision_model.transformer.layers.12.attention.q_proj.weight": "model.safetensors",
473
+ "vision_tower.vision_model.transformer.layers.12.attention.v_proj.weight": "model.safetensors",
474
+ "vision_tower.vision_model.transformer.layers.12.attention_norm.weight": "model.safetensors",
475
+ "vision_tower.vision_model.transformer.layers.12.feed_forward.down_proj.weight": "model.safetensors",
476
+ "vision_tower.vision_model.transformer.layers.12.feed_forward.gate_proj.weight": "model.safetensors",
477
+ "vision_tower.vision_model.transformer.layers.12.feed_forward.up_proj.weight": "model.safetensors",
478
+ "vision_tower.vision_model.transformer.layers.12.ffn_norm.weight": "model.safetensors",
479
+ "vision_tower.vision_model.transformer.layers.13.attention.k_proj.weight": "model.safetensors",
480
+ "vision_tower.vision_model.transformer.layers.13.attention.o_proj.weight": "model.safetensors",
481
+ "vision_tower.vision_model.transformer.layers.13.attention.q_proj.weight": "model.safetensors",
482
+ "vision_tower.vision_model.transformer.layers.13.attention.v_proj.weight": "model.safetensors",
483
+ "vision_tower.vision_model.transformer.layers.13.attention_norm.weight": "model.safetensors",
484
+ "vision_tower.vision_model.transformer.layers.13.feed_forward.down_proj.weight": "model.safetensors",
485
+ "vision_tower.vision_model.transformer.layers.13.feed_forward.gate_proj.weight": "model.safetensors",
486
+ "vision_tower.vision_model.transformer.layers.13.feed_forward.up_proj.weight": "model.safetensors",
487
+ "vision_tower.vision_model.transformer.layers.13.ffn_norm.weight": "model.safetensors",
488
+ "vision_tower.vision_model.transformer.layers.14.attention.k_proj.weight": "model.safetensors",
489
+ "vision_tower.vision_model.transformer.layers.14.attention.o_proj.weight": "model.safetensors",
490
+ "vision_tower.vision_model.transformer.layers.14.attention.q_proj.weight": "model.safetensors",
491
+ "vision_tower.vision_model.transformer.layers.14.attention.v_proj.weight": "model.safetensors",
492
+ "vision_tower.vision_model.transformer.layers.14.attention_norm.weight": "model.safetensors",
493
+ "vision_tower.vision_model.transformer.layers.14.feed_forward.down_proj.weight": "model.safetensors",
494
+ "vision_tower.vision_model.transformer.layers.14.feed_forward.gate_proj.weight": "model.safetensors",
495
+ "vision_tower.vision_model.transformer.layers.14.feed_forward.up_proj.weight": "model.safetensors",
496
+ "vision_tower.vision_model.transformer.layers.14.ffn_norm.weight": "model.safetensors",
497
+ "vision_tower.vision_model.transformer.layers.15.attention.k_proj.weight": "model.safetensors",
498
+ "vision_tower.vision_model.transformer.layers.15.attention.o_proj.weight": "model.safetensors",
499
+ "vision_tower.vision_model.transformer.layers.15.attention.q_proj.weight": "model.safetensors",
500
+ "vision_tower.vision_model.transformer.layers.15.attention.v_proj.weight": "model.safetensors",
501
+ "vision_tower.vision_model.transformer.layers.15.attention_norm.weight": "model.safetensors",
502
+ "vision_tower.vision_model.transformer.layers.15.feed_forward.down_proj.weight": "model.safetensors",
503
+ "vision_tower.vision_model.transformer.layers.15.feed_forward.gate_proj.weight": "model.safetensors",
504
+ "vision_tower.vision_model.transformer.layers.15.feed_forward.up_proj.weight": "model.safetensors",
505
+ "vision_tower.vision_model.transformer.layers.15.ffn_norm.weight": "model.safetensors",
506
+ "vision_tower.vision_model.transformer.layers.16.attention.k_proj.weight": "model.safetensors",
507
+ "vision_tower.vision_model.transformer.layers.16.attention.o_proj.weight": "model.safetensors",
508
+ "vision_tower.vision_model.transformer.layers.16.attention.q_proj.weight": "model.safetensors",
509
+ "vision_tower.vision_model.transformer.layers.16.attention.v_proj.weight": "model.safetensors",
510
+ "vision_tower.vision_model.transformer.layers.16.attention_norm.weight": "model.safetensors",
511
+ "vision_tower.vision_model.transformer.layers.16.feed_forward.down_proj.weight": "model.safetensors",
512
+ "vision_tower.vision_model.transformer.layers.16.feed_forward.gate_proj.weight": "model.safetensors",
513
+ "vision_tower.vision_model.transformer.layers.16.feed_forward.up_proj.weight": "model.safetensors",
514
+ "vision_tower.vision_model.transformer.layers.16.ffn_norm.weight": "model.safetensors",
515
+ "vision_tower.vision_model.transformer.layers.17.attention.k_proj.weight": "model.safetensors",
516
+ "vision_tower.vision_model.transformer.layers.17.attention.o_proj.weight": "model.safetensors",
517
+ "vision_tower.vision_model.transformer.layers.17.attention.q_proj.weight": "model.safetensors",
518
+ "vision_tower.vision_model.transformer.layers.17.attention.v_proj.weight": "model.safetensors",
519
+ "vision_tower.vision_model.transformer.layers.17.attention_norm.weight": "model.safetensors",
520
+ "vision_tower.vision_model.transformer.layers.17.feed_forward.down_proj.weight": "model.safetensors",
521
+ "vision_tower.vision_model.transformer.layers.17.feed_forward.gate_proj.weight": "model.safetensors",
522
+ "vision_tower.vision_model.transformer.layers.17.feed_forward.up_proj.weight": "model.safetensors",
523
+ "vision_tower.vision_model.transformer.layers.17.ffn_norm.weight": "model.safetensors",
524
+ "vision_tower.vision_model.transformer.layers.18.attention.k_proj.weight": "model.safetensors",
525
+ "vision_tower.vision_model.transformer.layers.18.attention.o_proj.weight": "model.safetensors",
526
+ "vision_tower.vision_model.transformer.layers.18.attention.q_proj.weight": "model.safetensors",
527
+ "vision_tower.vision_model.transformer.layers.18.attention.v_proj.weight": "model.safetensors",
528
+ "vision_tower.vision_model.transformer.layers.18.attention_norm.weight": "model.safetensors",
529
+ "vision_tower.vision_model.transformer.layers.18.feed_forward.down_proj.weight": "model.safetensors",
530
+ "vision_tower.vision_model.transformer.layers.18.feed_forward.gate_proj.weight": "model.safetensors",
531
+ "vision_tower.vision_model.transformer.layers.18.feed_forward.up_proj.weight": "model.safetensors",
532
+ "vision_tower.vision_model.transformer.layers.18.ffn_norm.weight": "model.safetensors",
533
+ "vision_tower.vision_model.transformer.layers.19.attention.k_proj.weight": "model.safetensors",
534
+ "vision_tower.vision_model.transformer.layers.19.attention.o_proj.weight": "model.safetensors",
535
+ "vision_tower.vision_model.transformer.layers.19.attention.q_proj.weight": "model.safetensors",
536
+ "vision_tower.vision_model.transformer.layers.19.attention.v_proj.weight": "model.safetensors",
537
+ "vision_tower.vision_model.transformer.layers.19.attention_norm.weight": "model.safetensors",
538
+ "vision_tower.vision_model.transformer.layers.19.feed_forward.down_proj.weight": "model.safetensors",
539
+ "vision_tower.vision_model.transformer.layers.19.feed_forward.gate_proj.weight": "model.safetensors",
540
+ "vision_tower.vision_model.transformer.layers.19.feed_forward.up_proj.weight": "model.safetensors",
541
+ "vision_tower.vision_model.transformer.layers.19.ffn_norm.weight": "model.safetensors",
542
+ "vision_tower.vision_model.transformer.layers.2.attention.k_proj.weight": "model.safetensors",
543
+ "vision_tower.vision_model.transformer.layers.2.attention.o_proj.weight": "model.safetensors",
544
+ "vision_tower.vision_model.transformer.layers.2.attention.q_proj.weight": "model.safetensors",
545
+ "vision_tower.vision_model.transformer.layers.2.attention.v_proj.weight": "model.safetensors",
546
+ "vision_tower.vision_model.transformer.layers.2.attention_norm.weight": "model.safetensors",
547
+ "vision_tower.vision_model.transformer.layers.2.feed_forward.down_proj.weight": "model.safetensors",
548
+ "vision_tower.vision_model.transformer.layers.2.feed_forward.gate_proj.weight": "model.safetensors",
549
+ "vision_tower.vision_model.transformer.layers.2.feed_forward.up_proj.weight": "model.safetensors",
550
+ "vision_tower.vision_model.transformer.layers.2.ffn_norm.weight": "model.safetensors",
551
+ "vision_tower.vision_model.transformer.layers.20.attention.k_proj.weight": "model.safetensors",
552
+ "vision_tower.vision_model.transformer.layers.20.attention.o_proj.weight": "model.safetensors",
553
+ "vision_tower.vision_model.transformer.layers.20.attention.q_proj.weight": "model.safetensors",
554
+ "vision_tower.vision_model.transformer.layers.20.attention.v_proj.weight": "model.safetensors",
555
+ "vision_tower.vision_model.transformer.layers.20.attention_norm.weight": "model.safetensors",
556
+ "vision_tower.vision_model.transformer.layers.20.feed_forward.down_proj.weight": "model.safetensors",
557
+ "vision_tower.vision_model.transformer.layers.20.feed_forward.gate_proj.weight": "model.safetensors",
558
+ "vision_tower.vision_model.transformer.layers.20.feed_forward.up_proj.weight": "model.safetensors",
559
+ "vision_tower.vision_model.transformer.layers.20.ffn_norm.weight": "model.safetensors",
560
+ "vision_tower.vision_model.transformer.layers.21.attention.k_proj.weight": "model.safetensors",
561
+ "vision_tower.vision_model.transformer.layers.21.attention.o_proj.weight": "model.safetensors",
562
+ "vision_tower.vision_model.transformer.layers.21.attention.q_proj.weight": "model.safetensors",
563
+ "vision_tower.vision_model.transformer.layers.21.attention.v_proj.weight": "model.safetensors",
564
+ "vision_tower.vision_model.transformer.layers.21.attention_norm.weight": "model.safetensors",
565
+ "vision_tower.vision_model.transformer.layers.21.feed_forward.down_proj.weight": "model.safetensors",
566
+ "vision_tower.vision_model.transformer.layers.21.feed_forward.gate_proj.weight": "model.safetensors",
567
+ "vision_tower.vision_model.transformer.layers.21.feed_forward.up_proj.weight": "model.safetensors",
568
+ "vision_tower.vision_model.transformer.layers.21.ffn_norm.weight": "model.safetensors",
569
+ "vision_tower.vision_model.transformer.layers.22.attention.k_proj.weight": "model.safetensors",
570
+ "vision_tower.vision_model.transformer.layers.22.attention.o_proj.weight": "model.safetensors",
571
+ "vision_tower.vision_model.transformer.layers.22.attention.q_proj.weight": "model.safetensors",
572
+ "vision_tower.vision_model.transformer.layers.22.attention.v_proj.weight": "model.safetensors",
573
+ "vision_tower.vision_model.transformer.layers.22.attention_norm.weight": "model.safetensors",
574
+ "vision_tower.vision_model.transformer.layers.22.feed_forward.down_proj.weight": "model.safetensors",
575
+ "vision_tower.vision_model.transformer.layers.22.feed_forward.gate_proj.weight": "model.safetensors",
576
+ "vision_tower.vision_model.transformer.layers.22.feed_forward.up_proj.weight": "model.safetensors",
577
+ "vision_tower.vision_model.transformer.layers.22.ffn_norm.weight": "model.safetensors",
578
+ "vision_tower.vision_model.transformer.layers.23.attention.k_proj.weight": "model.safetensors",
579
+ "vision_tower.vision_model.transformer.layers.23.attention.o_proj.weight": "model.safetensors",
580
+ "vision_tower.vision_model.transformer.layers.23.attention.q_proj.weight": "model.safetensors",
581
+ "vision_tower.vision_model.transformer.layers.23.attention.v_proj.weight": "model.safetensors",
582
+ "vision_tower.vision_model.transformer.layers.23.attention_norm.weight": "model.safetensors",
583
+ "vision_tower.vision_model.transformer.layers.23.feed_forward.down_proj.weight": "model.safetensors",
584
+ "vision_tower.vision_model.transformer.layers.23.feed_forward.gate_proj.weight": "model.safetensors",
585
+ "vision_tower.vision_model.transformer.layers.23.feed_forward.up_proj.weight": "model.safetensors",
586
+ "vision_tower.vision_model.transformer.layers.23.ffn_norm.weight": "model.safetensors",
587
+ "vision_tower.vision_model.transformer.layers.3.attention.k_proj.weight": "model.safetensors",
588
+ "vision_tower.vision_model.transformer.layers.3.attention.o_proj.weight": "model.safetensors",
589
+ "vision_tower.vision_model.transformer.layers.3.attention.q_proj.weight": "model.safetensors",
590
+ "vision_tower.vision_model.transformer.layers.3.attention.v_proj.weight": "model.safetensors",
591
+ "vision_tower.vision_model.transformer.layers.3.attention_norm.weight": "model.safetensors",
592
+ "vision_tower.vision_model.transformer.layers.3.feed_forward.down_proj.weight": "model.safetensors",
593
+ "vision_tower.vision_model.transformer.layers.3.feed_forward.gate_proj.weight": "model.safetensors",
594
+ "vision_tower.vision_model.transformer.layers.3.feed_forward.up_proj.weight": "model.safetensors",
595
+ "vision_tower.vision_model.transformer.layers.3.ffn_norm.weight": "model.safetensors",
596
+ "vision_tower.vision_model.transformer.layers.4.attention.k_proj.weight": "model.safetensors",
597
+ "vision_tower.vision_model.transformer.layers.4.attention.o_proj.weight": "model.safetensors",
598
+ "vision_tower.vision_model.transformer.layers.4.attention.q_proj.weight": "model.safetensors",
599
+ "vision_tower.vision_model.transformer.layers.4.attention.v_proj.weight": "model.safetensors",
600
+ "vision_tower.vision_model.transformer.layers.4.attention_norm.weight": "model.safetensors",
601
+ "vision_tower.vision_model.transformer.layers.4.feed_forward.down_proj.weight": "model.safetensors",
602
+ "vision_tower.vision_model.transformer.layers.4.feed_forward.gate_proj.weight": "model.safetensors",
603
+ "vision_tower.vision_model.transformer.layers.4.feed_forward.up_proj.weight": "model.safetensors",
604
+ "vision_tower.vision_model.transformer.layers.4.ffn_norm.weight": "model.safetensors",
605
+ "vision_tower.vision_model.transformer.layers.5.attention.k_proj.weight": "model.safetensors",
606
+ "vision_tower.vision_model.transformer.layers.5.attention.o_proj.weight": "model.safetensors",
607
+ "vision_tower.vision_model.transformer.layers.5.attention.q_proj.weight": "model.safetensors",
608
+ "vision_tower.vision_model.transformer.layers.5.attention.v_proj.weight": "model.safetensors",
609
+ "vision_tower.vision_model.transformer.layers.5.attention_norm.weight": "model.safetensors",
610
+ "vision_tower.vision_model.transformer.layers.5.feed_forward.down_proj.weight": "model.safetensors",
611
+ "vision_tower.vision_model.transformer.layers.5.feed_forward.gate_proj.weight": "model.safetensors",
612
+ "vision_tower.vision_model.transformer.layers.5.feed_forward.up_proj.weight": "model.safetensors",
613
+ "vision_tower.vision_model.transformer.layers.5.ffn_norm.weight": "model.safetensors",
614
+ "vision_tower.vision_model.transformer.layers.6.attention.k_proj.weight": "model.safetensors",
615
+ "vision_tower.vision_model.transformer.layers.6.attention.o_proj.weight": "model.safetensors",
616
+ "vision_tower.vision_model.transformer.layers.6.attention.q_proj.weight": "model.safetensors",
617
+ "vision_tower.vision_model.transformer.layers.6.attention.v_proj.weight": "model.safetensors",
618
+ "vision_tower.vision_model.transformer.layers.6.attention_norm.weight": "model.safetensors",
619
+ "vision_tower.vision_model.transformer.layers.6.feed_forward.down_proj.weight": "model.safetensors",
620
+ "vision_tower.vision_model.transformer.layers.6.feed_forward.gate_proj.weight": "model.safetensors",
621
+ "vision_tower.vision_model.transformer.layers.6.feed_forward.up_proj.weight": "model.safetensors",
622
+ "vision_tower.vision_model.transformer.layers.6.ffn_norm.weight": "model.safetensors",
623
+ "vision_tower.vision_model.transformer.layers.7.attention.k_proj.weight": "model.safetensors",
624
+ "vision_tower.vision_model.transformer.layers.7.attention.o_proj.weight": "model.safetensors",
625
+ "vision_tower.vision_model.transformer.layers.7.attention.q_proj.weight": "model.safetensors",
626
+ "vision_tower.vision_model.transformer.layers.7.attention.v_proj.weight": "model.safetensors",
627
+ "vision_tower.vision_model.transformer.layers.7.attention_norm.weight": "model.safetensors",
628
+ "vision_tower.vision_model.transformer.layers.7.feed_forward.down_proj.weight": "model.safetensors",
629
+ "vision_tower.vision_model.transformer.layers.7.feed_forward.gate_proj.weight": "model.safetensors",
630
+ "vision_tower.vision_model.transformer.layers.7.feed_forward.up_proj.weight": "model.safetensors",
631
+ "vision_tower.vision_model.transformer.layers.7.ffn_norm.weight": "model.safetensors",
632
+ "vision_tower.vision_model.transformer.layers.8.attention.k_proj.weight": "model.safetensors",
633
+ "vision_tower.vision_model.transformer.layers.8.attention.o_proj.weight": "model.safetensors",
634
+ "vision_tower.vision_model.transformer.layers.8.attention.q_proj.weight": "model.safetensors",
635
+ "vision_tower.vision_model.transformer.layers.8.attention.v_proj.weight": "model.safetensors",
636
+ "vision_tower.vision_model.transformer.layers.8.attention_norm.weight": "model.safetensors",
637
+ "vision_tower.vision_model.transformer.layers.8.feed_forward.down_proj.weight": "model.safetensors",
638
+ "vision_tower.vision_model.transformer.layers.8.feed_forward.gate_proj.weight": "model.safetensors",
639
+ "vision_tower.vision_model.transformer.layers.8.feed_forward.up_proj.weight": "model.safetensors",
640
+ "vision_tower.vision_model.transformer.layers.8.ffn_norm.weight": "model.safetensors",
641
+ "vision_tower.vision_model.transformer.layers.9.attention.k_proj.weight": "model.safetensors",
642
+ "vision_tower.vision_model.transformer.layers.9.attention.o_proj.weight": "model.safetensors",
643
+ "vision_tower.vision_model.transformer.layers.9.attention.q_proj.weight": "model.safetensors",
644
+ "vision_tower.vision_model.transformer.layers.9.attention.v_proj.weight": "model.safetensors",
645
+ "vision_tower.vision_model.transformer.layers.9.attention_norm.weight": "model.safetensors",
646
+ "vision_tower.vision_model.transformer.layers.9.feed_forward.down_proj.weight": "model.safetensors",
647
+ "vision_tower.vision_model.transformer.layers.9.feed_forward.gate_proj.weight": "model.safetensors",
648
+ "vision_tower.vision_model.transformer.layers.9.feed_forward.up_proj.weight": "model.safetensors",
649
+ "vision_tower.vision_model.transformer.layers.9.ffn_norm.weight": "model.safetensors"
650
  }
651
  }
params.json ADDED
@@ -0,0 +1,47 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dim": 3072,
3
+ "n_layers": 26,
4
+ "head_dim": 128,
5
+ "hidden_dim": 9216,
6
+ "n_heads": 32,
7
+ "n_kv_heads": 8,
8
+ "rope_theta": 1000000.0,
9
+ "norm_eps": 1e-05,
10
+ "vocab_size": 131072,
11
+ "tied_embeddings": true,
12
+ "max_position_embeddings": 262144,
13
+ "llama_4_scaling": {
14
+ "original_max_position_embeddings": 16384,
15
+ "beta": 0.1
16
+ },
17
+ "q_lora_rank": null,
18
+ "qk_rope_head_dim": null,
19
+ "qk_nope_head_dim": null,
20
+ "kv_lora_rank": null,
21
+ "v_head_dim": null,
22
+ "yarn": {
23
+ "original_max_position_embeddings": 16384,
24
+ "factor": 16,
25
+ "apply_scale": false,
26
+ "beta": 32,
27
+ "alpha": 1
28
+ },
29
+ "vision_encoder": {
30
+ "image_token_id": 10,
31
+ "intermediate_size": 4096,
32
+ "num_hidden_layers": 24,
33
+ "num_attention_heads": 16,
34
+ "mm_projector_id": "patch_merge",
35
+ "spatial_merge_size": 2,
36
+ "hidden_size": 1024,
37
+ "num_channels": 3,
38
+ "image_size": 1540,
39
+ "max_image_size": 1540,
40
+ "patch_size": 14,
41
+ "rope_theta": 10000.0,
42
+ "add_pre_mm_projector_layer_norm": true,
43
+ "adapter_bias": false,
44
+ "image_break_token_id": 12,
45
+ "image_end_token_id": 13
46
+ }
47
+ }
processor_config.json ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "image_break_token": "[IMG_BREAK]",
3
+ "image_end_token": "[IMG_END]",
4
+ "image_processor": {
5
+ "data_format": "channels_first",
6
+ "do_convert_rgb": true,
7
+ "do_normalize": true,
8
+ "do_rescale": true,
9
+ "do_resize": true,
10
+ "image_mean": [
11
+ 0.48145466,
12
+ 0.4578275,
13
+ 0.40821073
14
+ ],
15
+ "image_processor_type": "PixtralImageProcessorFast",
16
+ "image_std": [
17
+ 0.26862954,
18
+ 0.26130258,
19
+ 0.27577711
20
+ ],
21
+ "patch_size": 14,
22
+ "resample": 3,
23
+ "rescale_factor": 0.00392156862745098,
24
+ "size": {
25
+ "longest_edge": 1540
26
+ }
27
+ },
28
+ "image_token": "[IMG]",
29
+ "patch_size": 14,
30
+ "processor_class": "PixtralProcessor",
31
+ "spatial_merge_size": 2
32
+ }
special_tokens_map.json ADDED
The diff for this file is too large to render. See raw diff
 
tekken.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e29d19ea32eb7e26e6c0572d57cb7f9eca0f4420e0e0fe6ae1cf3be94da1c0d6
3
+ size 16753777
tokenizer_config.json CHANGED
@@ -10,7 +10,7 @@
10
  "pad_token": "<pad>",
11
  "processor_class": "PixtralProcessor",
12
  "tokenizer_class": "TokenizersBackend",
13
- "tool_parser_type": "mistral",
14
  "unk_token": "<unk>",
15
- "use_default_system_prompt": false
 
16
  }
 
10
  "pad_token": "<pad>",
11
  "processor_class": "PixtralProcessor",
12
  "tokenizer_class": "TokenizersBackend",
 
13
  "unk_token": "<unk>",
14
+ "use_default_system_prompt": false,
15
+ "use_fast": true
16
  }