Hugging Face's logo

unsloth
/
gemma-3-4b-it-unsloth-bnb-4bit

Image-Text-to-Text
Transformers
Safetensors
English
gemma3
unsloth
gemma
google
conversational
text-generation-inference
4-bit precision
bitsandbytes
Model card Files
xet
Community
1
danielhanchen commited on
Commit
a69632d
·
verified ·
1 Parent(s): be907f4

Add files using upload-large-folder tool

Browse files
Files changed (2) hide show
  1. config.json +61 -61
  2. model.safetensors +3 -0
config.json CHANGED
@@ -25,67 +25,67 @@
25
  "multi_modal_projector",
26
  "merger",
27
  "modality_projection",
28
- "language_model.model.layers.4.self_attn",
29
- "language_model.model.layers.6.mlp",
30
- "language_model.model.layers.1.self_attn",
31
- "language_model.model.layers.3.self_attn",
32
- "language_model.model.layers.7.mlp",
33
- "language_model.model.layers.0.self_attn",
34
- "language_model.model.layers.6.self_attn",
35
- "language_model.model.layers.5.self_attn",
36
- "language_model.model.layers.8.mlp",
37
- "language_model.model.layers.8.self_attn",
38
- "language_model.model.layers.10.mlp",
39
- "language_model.model.layers.9.self_attn",
40
- "language_model.model.layers.12.self_attn",
41
- "language_model.model.layers.11.self_attn",
42
- "language_model.model.layers.10.self_attn",
43
- "language_model.model.layers.9.mlp",
44
- "language_model.model.layers.7.self_attn",
45
- "language_model.model.layers.11.mlp",
46
- "language_model.model.layers.13.self_attn",
47
- "language_model.model.layers.12.mlp",
48
- "language_model.model.layers.13.mlp",
49
- "language_model.model.layers.14.mlp",
50
- "language_model.model.layers.22.mlp",
51
- "language_model.model.layers.20.mlp",
52
- "language_model.model.layers.21.mlp",
53
- "language_model.model.layers.23.mlp",
54
- "language_model.model.layers.19.mlp",
55
- "language_model.model.layers.16.mlp",
56
- "language_model.model.layers.18.mlp",
57
- "language_model.model.layers.17.mlp",
58
- "language_model.model.layers.15.mlp",
59
- "language_model.model.layers.24.mlp",
60
- "language_model.model.layers.14.self_attn",
61
- "language_model.model.layers.25.mlp",
62
- "language_model.model.layers.15.self_attn",
63
- "language_model.model.layers.17.self_attn",
64
- "language_model.model.layers.26.mlp",
65
- "language_model.model.layers.27.mlp",
66
- "language_model.model.layers.22.self_attn",
67
- "language_model.model.layers.28.mlp",
68
- "language_model.model.layers.29.mlp",
69
- "language_model.model.layers.16.self_attn",
70
- "language_model.model.layers.30.mlp",
71
- "language_model.model.layers.19.self_attn",
72
- "language_model.model.layers.32.mlp",
73
- "language_model.model.layers.23.self_attn",
74
- "language_model.model.layers.31.mlp",
75
- "language_model.model.layers.18.self_attn",
76
- "language_model.model.layers.31.self_attn",
77
- "language_model.model.layers.20.self_attn",
78
- "language_model.model.layers.26.self_attn",
79
- "language_model.model.layers.21.self_attn",
80
- "language_model.model.layers.29.self_attn",
81
- "language_model.model.layers.32.self_attn",
82
- "language_model.model.layers.30.self_attn",
83
- "language_model.model.layers.25.self_attn",
84
- "language_model.model.layers.28.self_attn",
85
- "language_model.model.layers.24.self_attn",
86
- "language_model.model.layers.27.self_attn",
87
- "language_model.model.layers.33.self_attn",
88
- "language_model.model.layers.33.mlp"
89
  ],
90
  "llm_int8_threshold": 6.0,
91
  "load_in_4bit": true,
 
25
  "multi_modal_projector",
26
  "merger",
27
  "modality_projection",
28
+ "language_model.model.layers.1.mlp",
29
+ "language_model.model.layers.4.mlp",
30
+ "language_model.model.layers.2.mlp",
31
+ "language_model.model.layers.5.mlp",
32
+ "vision_tower.vision_model.encoder.layers.22.self_attn",
33
+ "language_model.model.layers.3.mlp",
34
+ "vision_tower.vision_model.encoder.layers.26.self_attn",
35
+ "language_model.model.layers.0.mlp",
36
+ "vision_tower.vision_model.encoder.layers.19.self_attn",
37
+ "vision_tower.vision_model.encoder.layers.25.self_attn",
38
+ "vision_tower.vision_model.encoder.layers.23.self_attn",
39
+ "vision_tower.vision_model.encoder.layers.25.mlp",
40
+ "vision_tower.vision_model.encoder.layers.20.self_attn",
41
+ "vision_tower.vision_model.encoder.layers.24.self_attn",
42
+ "language_model.model.layers.2.self_attn",
43
+ "vision_tower.vision_model.encoder.layers.21.self_attn",
44
+ "vision_tower.vision_model.encoder.layers.18.self_attn",
45
+ "vision_tower.vision_model.encoder.layers.19.mlp",
46
+ "vision_tower.vision_model.encoder.layers.20.mlp",
47
+ "vision_tower.vision_model.encoder.layers.21.mlp",
48
+ "vision_tower.vision_model.encoder.layers.15.mlp",
49
+ "vision_tower.vision_model.encoder.layers.22.mlp",
50
+ "vision_tower.vision_model.encoder.layers.23.mlp",
51
+ "vision_tower.vision_model.encoder.layers.17.mlp",
52
+ "vision_tower.vision_model.encoder.layers.18.mlp",
53
+ "vision_tower.vision_model.encoder.layers.24.mlp",
54
+ "vision_tower.vision_model.encoder.layers.16.self_attn",
55
+ "vision_tower.vision_model.encoder.layers.16.mlp",
56
+ "vision_tower.vision_model.encoder.layers.17.self_attn",
57
+ "vision_tower.vision_model.encoder.layers.13.mlp",
58
+ "vision_tower.vision_model.encoder.layers.14.mlp",
59
+ "vision_tower.vision_model.encoder.layers.15.self_attn",
60
+ "vision_tower.vision_model.encoder.layers.12.mlp",
61
+ "vision_tower.vision_model.encoder.layers.14.self_attn",
62
+ "vision_tower.vision_model.encoder.layers.9.mlp",
63
+ "vision_tower.vision_model.encoder.layers.5.mlp",
64
+ "vision_tower.vision_model.encoder.layers.10.self_attn",
65
+ "vision_tower.vision_model.encoder.layers.7.mlp",
66
+ "vision_tower.vision_model.encoder.layers.9.self_attn",
67
+ "vision_tower.vision_model.encoder.layers.8.mlp",
68
+ "vision_tower.vision_model.encoder.layers.10.mlp",
69
+ "vision_tower.vision_model.encoder.layers.13.self_attn",
70
+ "vision_tower.vision_model.encoder.layers.11.mlp",
71
+ "vision_tower.vision_model.encoder.layers.11.self_attn",
72
+ "vision_tower.vision_model.encoder.layers.12.self_attn",
73
+ "vision_tower.vision_model.encoder.layers.3.mlp",
74
+ "vision_tower.vision_model.encoder.layers.8.self_attn",
75
+ "vision_tower.vision_model.encoder.layers.4.self_attn",
76
+ "vision_tower.vision_model.encoder.layers.6.mlp",
77
+ "vision_tower.vision_model.encoder.layers.4.mlp",
78
+ "vision_tower.vision_model.encoder.layers.2.mlp",
79
+ "vision_tower.vision_model.encoder.layers.6.self_attn",
80
+ "vision_tower.vision_model.encoder.layers.7.self_attn",
81
+ "vision_tower.vision_model.encoder.layers.1.self_attn",
82
+ "vision_tower.vision_model.encoder.layers.1.mlp",
83
+ "vision_tower.vision_model.encoder.layers.3.self_attn",
84
+ "vision_tower.vision_model.encoder.layers.5.self_attn",
85
+ "vision_tower.vision_model.encoder.layers.0.self_attn",
86
+ "vision_tower.vision_model.encoder.layers.0.mlp",
87
+ "vision_tower.vision_model.encoder.layers.2.self_attn",
88
+ "vision_tower.vision_model.encoder.layers.26.mlp"
89
  ],
90
  "llm_int8_threshold": 6.0,
91
  "load_in_4bit": true,
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:50fee60341951d70ff28e0480f908480927a1b11e69738457bfe17da34f548c4
3
+ size 4562294331