Image-to-Text
Transformers
Safetensors
Japanese
llava-jp
text-generation
vision
image-captioning
VQA
Inference Endpoints
toshi456 commited on
Commit
ad0c144
1 Parent(s): 5739b78

Upload 8 files

Browse files
config.json ADDED
@@ -0,0 +1,58 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "./output_llava/checkpoints/pretrain-conv-llava-jp-1.3b-stage2-1280",
3
+ "activation_function": "gelu",
4
+ "architectures": [
5
+ "LlavaGpt2ForCausalLM"
6
+ ],
7
+ "attn_pdrop": 0.1,
8
+ "bos_token_id": 7,
9
+ "drop_path_rates": [
10
+ 0.085,
11
+ 0.088,
12
+ 0.091,
13
+ 0.094,
14
+ 0.097,
15
+ 0.1
16
+ ],
17
+ "embd_pdrop": 0.1,
18
+ "eos_token_id": 7,
19
+ "gradient_checkpointing": false,
20
+ "image_aspect_ratio": "square",
21
+ "initializer_range": 0.02,
22
+ "layer_norm_epsilon": 1e-05,
23
+ "mm_dense_connector_type": null,
24
+ "mm_hidden_size": 3072,
25
+ "mm_projector_type": "mlp2x_gelu",
26
+ "mm_vision_resolution": 1280,
27
+ "mm_vision_select_feature": "patch",
28
+ "mm_vision_select_layer": -2,
29
+ "mm_vision_tower": "convnext_large",
30
+ "model_type": "llava-jp",
31
+ "n_ctx": 1024,
32
+ "n_embd": 2048,
33
+ "n_head": 16,
34
+ "n_inner": 8192,
35
+ "n_layer": 24,
36
+ "n_positions": 2048,
37
+ "reorder_and_upcast_attn": false,
38
+ "resid_pdrop": 0.1,
39
+ "scale_attn_by_inverse_layer_idx": false,
40
+ "scale_attn_weights": true,
41
+ "scales": null,
42
+ "summary_activation": null,
43
+ "summary_first_dropout": 0.1,
44
+ "summary_proj_to_labels": true,
45
+ "summary_type": "cls_index",
46
+ "summary_use_proj": true,
47
+ "tokenizer_model_max_length": 1532,
48
+ "tokenizer_padding_side": "right",
49
+ "torch_dtype": "float32",
50
+ "transformers_version": "4.41.2",
51
+ "tune_mm_mlp_adapter": false,
52
+ "use_cache": true,
53
+ "use_mm_proj": true,
54
+ "vision_add_five_stage": 6,
55
+ "vision_encoder_type": "ConvNeXt",
56
+ "vision_five_stage_width": 3072,
57
+ "vocab_size": 50688
58
+ }
generation_config.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 7,
4
+ "eos_token_id": 7,
5
+ "transformers_version": "4.41.2"
6
+ }
model-00001-of-00002.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4f9d7f34ad88e7b37d1dd493817decda61720bfadb07d8ed48772649879fb576
3
+ size 4997886200
model-00002-of-00002.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a325101c74550f438c4e325c968c49d2e71202cc81bcc78607ff3f518d8e9877
3
+ size 2064141360
model.safetensors.index.json ADDED
@@ -0,0 +1,702 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 7061942656
4
+ },
5
+ "weight_map": {
6
+ "lm_head.weight": "model-00002-of-00002.safetensors",
7
+ "model.mm_projector.0.bias": "model-00002-of-00002.safetensors",
8
+ "model.mm_projector.0.weight": "model-00002-of-00002.safetensors",
9
+ "model.mm_projector.2.bias": "model-00002-of-00002.safetensors",
10
+ "model.mm_projector.2.weight": "model-00002-of-00002.safetensors",
11
+ "model.vision_tower.vision_tower.stages.0.blocks.0.conv_dw.bias": "model-00002-of-00002.safetensors",
12
+ "model.vision_tower.vision_tower.stages.0.blocks.0.conv_dw.weight": "model-00002-of-00002.safetensors",
13
+ "model.vision_tower.vision_tower.stages.0.blocks.0.mlp.fc1.bias": "model-00002-of-00002.safetensors",
14
+ "model.vision_tower.vision_tower.stages.0.blocks.0.mlp.fc1.weight": "model-00002-of-00002.safetensors",
15
+ "model.vision_tower.vision_tower.stages.0.blocks.0.mlp.fc2.bias": "model-00002-of-00002.safetensors",
16
+ "model.vision_tower.vision_tower.stages.0.blocks.0.mlp.fc2.weight": "model-00002-of-00002.safetensors",
17
+ "model.vision_tower.vision_tower.stages.0.blocks.0.norm.bias": "model-00002-of-00002.safetensors",
18
+ "model.vision_tower.vision_tower.stages.0.blocks.0.norm.weight": "model-00002-of-00002.safetensors",
19
+ "model.vision_tower.vision_tower.stages.0.blocks.0.weight": "model-00002-of-00002.safetensors",
20
+ "model.vision_tower.vision_tower.stages.0.blocks.1.conv_dw.bias": "model-00002-of-00002.safetensors",
21
+ "model.vision_tower.vision_tower.stages.0.blocks.1.conv_dw.weight": "model-00002-of-00002.safetensors",
22
+ "model.vision_tower.vision_tower.stages.0.blocks.1.mlp.fc1.bias": "model-00002-of-00002.safetensors",
23
+ "model.vision_tower.vision_tower.stages.0.blocks.1.mlp.fc1.weight": "model-00002-of-00002.safetensors",
24
+ "model.vision_tower.vision_tower.stages.0.blocks.1.mlp.fc2.bias": "model-00002-of-00002.safetensors",
25
+ "model.vision_tower.vision_tower.stages.0.blocks.1.mlp.fc2.weight": "model-00002-of-00002.safetensors",
26
+ "model.vision_tower.vision_tower.stages.0.blocks.1.norm.bias": "model-00002-of-00002.safetensors",
27
+ "model.vision_tower.vision_tower.stages.0.blocks.1.norm.weight": "model-00002-of-00002.safetensors",
28
+ "model.vision_tower.vision_tower.stages.0.blocks.1.weight": "model-00002-of-00002.safetensors",
29
+ "model.vision_tower.vision_tower.stages.0.blocks.2.conv_dw.bias": "model-00002-of-00002.safetensors",
30
+ "model.vision_tower.vision_tower.stages.0.blocks.2.conv_dw.weight": "model-00002-of-00002.safetensors",
31
+ "model.vision_tower.vision_tower.stages.0.blocks.2.mlp.fc1.bias": "model-00002-of-00002.safetensors",
32
+ "model.vision_tower.vision_tower.stages.0.blocks.2.mlp.fc1.weight": "model-00002-of-00002.safetensors",
33
+ "model.vision_tower.vision_tower.stages.0.blocks.2.mlp.fc2.bias": "model-00002-of-00002.safetensors",
34
+ "model.vision_tower.vision_tower.stages.0.blocks.2.mlp.fc2.weight": "model-00002-of-00002.safetensors",
35
+ "model.vision_tower.vision_tower.stages.0.blocks.2.norm.bias": "model-00002-of-00002.safetensors",
36
+ "model.vision_tower.vision_tower.stages.0.blocks.2.norm.weight": "model-00002-of-00002.safetensors",
37
+ "model.vision_tower.vision_tower.stages.0.blocks.2.weight": "model-00002-of-00002.safetensors",
38
+ "model.vision_tower.vision_tower.stages.1.blocks.0.conv_dw.bias": "model-00002-of-00002.safetensors",
39
+ "model.vision_tower.vision_tower.stages.1.blocks.0.conv_dw.weight": "model-00002-of-00002.safetensors",
40
+ "model.vision_tower.vision_tower.stages.1.blocks.0.mlp.fc1.bias": "model-00002-of-00002.safetensors",
41
+ "model.vision_tower.vision_tower.stages.1.blocks.0.mlp.fc1.weight": "model-00002-of-00002.safetensors",
42
+ "model.vision_tower.vision_tower.stages.1.blocks.0.mlp.fc2.bias": "model-00002-of-00002.safetensors",
43
+ "model.vision_tower.vision_tower.stages.1.blocks.0.mlp.fc2.weight": "model-00002-of-00002.safetensors",
44
+ "model.vision_tower.vision_tower.stages.1.blocks.0.norm.bias": "model-00002-of-00002.safetensors",
45
+ "model.vision_tower.vision_tower.stages.1.blocks.0.norm.weight": "model-00002-of-00002.safetensors",
46
+ "model.vision_tower.vision_tower.stages.1.blocks.0.weight": "model-00002-of-00002.safetensors",
47
+ "model.vision_tower.vision_tower.stages.1.blocks.1.conv_dw.bias": "model-00002-of-00002.safetensors",
48
+ "model.vision_tower.vision_tower.stages.1.blocks.1.conv_dw.weight": "model-00002-of-00002.safetensors",
49
+ "model.vision_tower.vision_tower.stages.1.blocks.1.mlp.fc1.bias": "model-00002-of-00002.safetensors",
50
+ "model.vision_tower.vision_tower.stages.1.blocks.1.mlp.fc1.weight": "model-00002-of-00002.safetensors",
51
+ "model.vision_tower.vision_tower.stages.1.blocks.1.mlp.fc2.bias": "model-00002-of-00002.safetensors",
52
+ "model.vision_tower.vision_tower.stages.1.blocks.1.mlp.fc2.weight": "model-00002-of-00002.safetensors",
53
+ "model.vision_tower.vision_tower.stages.1.blocks.1.norm.bias": "model-00002-of-00002.safetensors",
54
+ "model.vision_tower.vision_tower.stages.1.blocks.1.norm.weight": "model-00002-of-00002.safetensors",
55
+ "model.vision_tower.vision_tower.stages.1.blocks.1.weight": "model-00002-of-00002.safetensors",
56
+ "model.vision_tower.vision_tower.stages.1.blocks.2.conv_dw.bias": "model-00002-of-00002.safetensors",
57
+ "model.vision_tower.vision_tower.stages.1.blocks.2.conv_dw.weight": "model-00002-of-00002.safetensors",
58
+ "model.vision_tower.vision_tower.stages.1.blocks.2.mlp.fc1.bias": "model-00002-of-00002.safetensors",
59
+ "model.vision_tower.vision_tower.stages.1.blocks.2.mlp.fc1.weight": "model-00002-of-00002.safetensors",
60
+ "model.vision_tower.vision_tower.stages.1.blocks.2.mlp.fc2.bias": "model-00002-of-00002.safetensors",
61
+ "model.vision_tower.vision_tower.stages.1.blocks.2.mlp.fc2.weight": "model-00002-of-00002.safetensors",
62
+ "model.vision_tower.vision_tower.stages.1.blocks.2.norm.bias": "model-00002-of-00002.safetensors",
63
+ "model.vision_tower.vision_tower.stages.1.blocks.2.norm.weight": "model-00002-of-00002.safetensors",
64
+ "model.vision_tower.vision_tower.stages.1.blocks.2.weight": "model-00002-of-00002.safetensors",
65
+ "model.vision_tower.vision_tower.stages.1.downsample.0.bias": "model-00002-of-00002.safetensors",
66
+ "model.vision_tower.vision_tower.stages.1.downsample.0.weight": "model-00002-of-00002.safetensors",
67
+ "model.vision_tower.vision_tower.stages.1.downsample.1.bias": "model-00002-of-00002.safetensors",
68
+ "model.vision_tower.vision_tower.stages.1.downsample.1.weight": "model-00002-of-00002.safetensors",
69
+ "model.vision_tower.vision_tower.stages.2.blocks.0.conv_dw.bias": "model-00002-of-00002.safetensors",
70
+ "model.vision_tower.vision_tower.stages.2.blocks.0.conv_dw.weight": "model-00002-of-00002.safetensors",
71
+ "model.vision_tower.vision_tower.stages.2.blocks.0.mlp.fc1.bias": "model-00002-of-00002.safetensors",
72
+ "model.vision_tower.vision_tower.stages.2.blocks.0.mlp.fc1.weight": "model-00002-of-00002.safetensors",
73
+ "model.vision_tower.vision_tower.stages.2.blocks.0.mlp.fc2.bias": "model-00002-of-00002.safetensors",
74
+ "model.vision_tower.vision_tower.stages.2.blocks.0.mlp.fc2.weight": "model-00002-of-00002.safetensors",
75
+ "model.vision_tower.vision_tower.stages.2.blocks.0.norm.bias": "model-00002-of-00002.safetensors",
76
+ "model.vision_tower.vision_tower.stages.2.blocks.0.norm.weight": "model-00002-of-00002.safetensors",
77
+ "model.vision_tower.vision_tower.stages.2.blocks.0.weight": "model-00002-of-00002.safetensors",
78
+ "model.vision_tower.vision_tower.stages.2.blocks.1.conv_dw.bias": "model-00002-of-00002.safetensors",
79
+ "model.vision_tower.vision_tower.stages.2.blocks.1.conv_dw.weight": "model-00002-of-00002.safetensors",
80
+ "model.vision_tower.vision_tower.stages.2.blocks.1.mlp.fc1.bias": "model-00002-of-00002.safetensors",
81
+ "model.vision_tower.vision_tower.stages.2.blocks.1.mlp.fc1.weight": "model-00002-of-00002.safetensors",
82
+ "model.vision_tower.vision_tower.stages.2.blocks.1.mlp.fc2.bias": "model-00002-of-00002.safetensors",
83
+ "model.vision_tower.vision_tower.stages.2.blocks.1.mlp.fc2.weight": "model-00002-of-00002.safetensors",
84
+ "model.vision_tower.vision_tower.stages.2.blocks.1.norm.bias": "model-00002-of-00002.safetensors",
85
+ "model.vision_tower.vision_tower.stages.2.blocks.1.norm.weight": "model-00002-of-00002.safetensors",
86
+ "model.vision_tower.vision_tower.stages.2.blocks.1.weight": "model-00002-of-00002.safetensors",
87
+ "model.vision_tower.vision_tower.stages.2.blocks.10.conv_dw.bias": "model-00002-of-00002.safetensors",
88
+ "model.vision_tower.vision_tower.stages.2.blocks.10.conv_dw.weight": "model-00002-of-00002.safetensors",
89
+ "model.vision_tower.vision_tower.stages.2.blocks.10.mlp.fc1.bias": "model-00002-of-00002.safetensors",
90
+ "model.vision_tower.vision_tower.stages.2.blocks.10.mlp.fc1.weight": "model-00002-of-00002.safetensors",
91
+ "model.vision_tower.vision_tower.stages.2.blocks.10.mlp.fc2.bias": "model-00002-of-00002.safetensors",
92
+ "model.vision_tower.vision_tower.stages.2.blocks.10.mlp.fc2.weight": "model-00002-of-00002.safetensors",
93
+ "model.vision_tower.vision_tower.stages.2.blocks.10.norm.bias": "model-00002-of-00002.safetensors",
94
+ "model.vision_tower.vision_tower.stages.2.blocks.10.norm.weight": "model-00002-of-00002.safetensors",
95
+ "model.vision_tower.vision_tower.stages.2.blocks.10.weight": "model-00002-of-00002.safetensors",
96
+ "model.vision_tower.vision_tower.stages.2.blocks.11.conv_dw.bias": "model-00002-of-00002.safetensors",
97
+ "model.vision_tower.vision_tower.stages.2.blocks.11.conv_dw.weight": "model-00002-of-00002.safetensors",
98
+ "model.vision_tower.vision_tower.stages.2.blocks.11.mlp.fc1.bias": "model-00002-of-00002.safetensors",
99
+ "model.vision_tower.vision_tower.stages.2.blocks.11.mlp.fc1.weight": "model-00002-of-00002.safetensors",
100
+ "model.vision_tower.vision_tower.stages.2.blocks.11.mlp.fc2.bias": "model-00002-of-00002.safetensors",
101
+ "model.vision_tower.vision_tower.stages.2.blocks.11.mlp.fc2.weight": "model-00002-of-00002.safetensors",
102
+ "model.vision_tower.vision_tower.stages.2.blocks.11.norm.bias": "model-00002-of-00002.safetensors",
103
+ "model.vision_tower.vision_tower.stages.2.blocks.11.norm.weight": "model-00002-of-00002.safetensors",
104
+ "model.vision_tower.vision_tower.stages.2.blocks.11.weight": "model-00002-of-00002.safetensors",
105
+ "model.vision_tower.vision_tower.stages.2.blocks.12.conv_dw.bias": "model-00002-of-00002.safetensors",
106
+ "model.vision_tower.vision_tower.stages.2.blocks.12.conv_dw.weight": "model-00002-of-00002.safetensors",
107
+ "model.vision_tower.vision_tower.stages.2.blocks.12.mlp.fc1.bias": "model-00002-of-00002.safetensors",
108
+ "model.vision_tower.vision_tower.stages.2.blocks.12.mlp.fc1.weight": "model-00002-of-00002.safetensors",
109
+ "model.vision_tower.vision_tower.stages.2.blocks.12.mlp.fc2.bias": "model-00002-of-00002.safetensors",
110
+ "model.vision_tower.vision_tower.stages.2.blocks.12.mlp.fc2.weight": "model-00002-of-00002.safetensors",
111
+ "model.vision_tower.vision_tower.stages.2.blocks.12.norm.bias": "model-00002-of-00002.safetensors",
112
+ "model.vision_tower.vision_tower.stages.2.blocks.12.norm.weight": "model-00002-of-00002.safetensors",
113
+ "model.vision_tower.vision_tower.stages.2.blocks.12.weight": "model-00002-of-00002.safetensors",
114
+ "model.vision_tower.vision_tower.stages.2.blocks.13.conv_dw.bias": "model-00002-of-00002.safetensors",
115
+ "model.vision_tower.vision_tower.stages.2.blocks.13.conv_dw.weight": "model-00002-of-00002.safetensors",
116
+ "model.vision_tower.vision_tower.stages.2.blocks.13.mlp.fc1.bias": "model-00002-of-00002.safetensors",
117
+ "model.vision_tower.vision_tower.stages.2.blocks.13.mlp.fc1.weight": "model-00002-of-00002.safetensors",
118
+ "model.vision_tower.vision_tower.stages.2.blocks.13.mlp.fc2.bias": "model-00002-of-00002.safetensors",
119
+ "model.vision_tower.vision_tower.stages.2.blocks.13.mlp.fc2.weight": "model-00002-of-00002.safetensors",
120
+ "model.vision_tower.vision_tower.stages.2.blocks.13.norm.bias": "model-00002-of-00002.safetensors",
121
+ "model.vision_tower.vision_tower.stages.2.blocks.13.norm.weight": "model-00002-of-00002.safetensors",
122
+ "model.vision_tower.vision_tower.stages.2.blocks.13.weight": "model-00002-of-00002.safetensors",
123
+ "model.vision_tower.vision_tower.stages.2.blocks.14.conv_dw.bias": "model-00002-of-00002.safetensors",
124
+ "model.vision_tower.vision_tower.stages.2.blocks.14.conv_dw.weight": "model-00002-of-00002.safetensors",
125
+ "model.vision_tower.vision_tower.stages.2.blocks.14.mlp.fc1.bias": "model-00002-of-00002.safetensors",
126
+ "model.vision_tower.vision_tower.stages.2.blocks.14.mlp.fc1.weight": "model-00002-of-00002.safetensors",
127
+ "model.vision_tower.vision_tower.stages.2.blocks.14.mlp.fc2.bias": "model-00002-of-00002.safetensors",
128
+ "model.vision_tower.vision_tower.stages.2.blocks.14.mlp.fc2.weight": "model-00002-of-00002.safetensors",
129
+ "model.vision_tower.vision_tower.stages.2.blocks.14.norm.bias": "model-00002-of-00002.safetensors",
130
+ "model.vision_tower.vision_tower.stages.2.blocks.14.norm.weight": "model-00002-of-00002.safetensors",
131
+ "model.vision_tower.vision_tower.stages.2.blocks.14.weight": "model-00002-of-00002.safetensors",
132
+ "model.vision_tower.vision_tower.stages.2.blocks.15.conv_dw.bias": "model-00002-of-00002.safetensors",
133
+ "model.vision_tower.vision_tower.stages.2.blocks.15.conv_dw.weight": "model-00002-of-00002.safetensors",
134
+ "model.vision_tower.vision_tower.stages.2.blocks.15.mlp.fc1.bias": "model-00002-of-00002.safetensors",
135
+ "model.vision_tower.vision_tower.stages.2.blocks.15.mlp.fc1.weight": "model-00002-of-00002.safetensors",
136
+ "model.vision_tower.vision_tower.stages.2.blocks.15.mlp.fc2.bias": "model-00002-of-00002.safetensors",
137
+ "model.vision_tower.vision_tower.stages.2.blocks.15.mlp.fc2.weight": "model-00002-of-00002.safetensors",
138
+ "model.vision_tower.vision_tower.stages.2.blocks.15.norm.bias": "model-00002-of-00002.safetensors",
139
+ "model.vision_tower.vision_tower.stages.2.blocks.15.norm.weight": "model-00002-of-00002.safetensors",
140
+ "model.vision_tower.vision_tower.stages.2.blocks.15.weight": "model-00002-of-00002.safetensors",
141
+ "model.vision_tower.vision_tower.stages.2.blocks.16.conv_dw.bias": "model-00002-of-00002.safetensors",
142
+ "model.vision_tower.vision_tower.stages.2.blocks.16.conv_dw.weight": "model-00002-of-00002.safetensors",
143
+ "model.vision_tower.vision_tower.stages.2.blocks.16.mlp.fc1.bias": "model-00002-of-00002.safetensors",
144
+ "model.vision_tower.vision_tower.stages.2.blocks.16.mlp.fc1.weight": "model-00002-of-00002.safetensors",
145
+ "model.vision_tower.vision_tower.stages.2.blocks.16.mlp.fc2.bias": "model-00002-of-00002.safetensors",
146
+ "model.vision_tower.vision_tower.stages.2.blocks.16.mlp.fc2.weight": "model-00002-of-00002.safetensors",
147
+ "model.vision_tower.vision_tower.stages.2.blocks.16.norm.bias": "model-00002-of-00002.safetensors",
148
+ "model.vision_tower.vision_tower.stages.2.blocks.16.norm.weight": "model-00002-of-00002.safetensors",
149
+ "model.vision_tower.vision_tower.stages.2.blocks.16.weight": "model-00002-of-00002.safetensors",
150
+ "model.vision_tower.vision_tower.stages.2.blocks.17.conv_dw.bias": "model-00002-of-00002.safetensors",
151
+ "model.vision_tower.vision_tower.stages.2.blocks.17.conv_dw.weight": "model-00002-of-00002.safetensors",
152
+ "model.vision_tower.vision_tower.stages.2.blocks.17.mlp.fc1.bias": "model-00002-of-00002.safetensors",
153
+ "model.vision_tower.vision_tower.stages.2.blocks.17.mlp.fc1.weight": "model-00002-of-00002.safetensors",
154
+ "model.vision_tower.vision_tower.stages.2.blocks.17.mlp.fc2.bias": "model-00002-of-00002.safetensors",
155
+ "model.vision_tower.vision_tower.stages.2.blocks.17.mlp.fc2.weight": "model-00002-of-00002.safetensors",
156
+ "model.vision_tower.vision_tower.stages.2.blocks.17.norm.bias": "model-00002-of-00002.safetensors",
157
+ "model.vision_tower.vision_tower.stages.2.blocks.17.norm.weight": "model-00002-of-00002.safetensors",
158
+ "model.vision_tower.vision_tower.stages.2.blocks.17.weight": "model-00002-of-00002.safetensors",
159
+ "model.vision_tower.vision_tower.stages.2.blocks.18.conv_dw.bias": "model-00002-of-00002.safetensors",
160
+ "model.vision_tower.vision_tower.stages.2.blocks.18.conv_dw.weight": "model-00002-of-00002.safetensors",
161
+ "model.vision_tower.vision_tower.stages.2.blocks.18.mlp.fc1.bias": "model-00002-of-00002.safetensors",
162
+ "model.vision_tower.vision_tower.stages.2.blocks.18.mlp.fc1.weight": "model-00002-of-00002.safetensors",
163
+ "model.vision_tower.vision_tower.stages.2.blocks.18.mlp.fc2.bias": "model-00002-of-00002.safetensors",
164
+ "model.vision_tower.vision_tower.stages.2.blocks.18.mlp.fc2.weight": "model-00002-of-00002.safetensors",
165
+ "model.vision_tower.vision_tower.stages.2.blocks.18.norm.bias": "model-00002-of-00002.safetensors",
166
+ "model.vision_tower.vision_tower.stages.2.blocks.18.norm.weight": "model-00002-of-00002.safetensors",
167
+ "model.vision_tower.vision_tower.stages.2.blocks.18.weight": "model-00002-of-00002.safetensors",
168
+ "model.vision_tower.vision_tower.stages.2.blocks.19.conv_dw.bias": "model-00002-of-00002.safetensors",
169
+ "model.vision_tower.vision_tower.stages.2.blocks.19.conv_dw.weight": "model-00002-of-00002.safetensors",
170
+ "model.vision_tower.vision_tower.stages.2.blocks.19.mlp.fc1.bias": "model-00002-of-00002.safetensors",
171
+ "model.vision_tower.vision_tower.stages.2.blocks.19.mlp.fc1.weight": "model-00002-of-00002.safetensors",
172
+ "model.vision_tower.vision_tower.stages.2.blocks.19.mlp.fc2.bias": "model-00002-of-00002.safetensors",
173
+ "model.vision_tower.vision_tower.stages.2.blocks.19.mlp.fc2.weight": "model-00002-of-00002.safetensors",
174
+ "model.vision_tower.vision_tower.stages.2.blocks.19.norm.bias": "model-00002-of-00002.safetensors",
175
+ "model.vision_tower.vision_tower.stages.2.blocks.19.norm.weight": "model-00002-of-00002.safetensors",
176
+ "model.vision_tower.vision_tower.stages.2.blocks.19.weight": "model-00002-of-00002.safetensors",
177
+ "model.vision_tower.vision_tower.stages.2.blocks.2.conv_dw.bias": "model-00002-of-00002.safetensors",
178
+ "model.vision_tower.vision_tower.stages.2.blocks.2.conv_dw.weight": "model-00002-of-00002.safetensors",
179
+ "model.vision_tower.vision_tower.stages.2.blocks.2.mlp.fc1.bias": "model-00002-of-00002.safetensors",
180
+ "model.vision_tower.vision_tower.stages.2.blocks.2.mlp.fc1.weight": "model-00002-of-00002.safetensors",
181
+ "model.vision_tower.vision_tower.stages.2.blocks.2.mlp.fc2.bias": "model-00002-of-00002.safetensors",
182
+ "model.vision_tower.vision_tower.stages.2.blocks.2.mlp.fc2.weight": "model-00002-of-00002.safetensors",
183
+ "model.vision_tower.vision_tower.stages.2.blocks.2.norm.bias": "model-00002-of-00002.safetensors",
184
+ "model.vision_tower.vision_tower.stages.2.blocks.2.norm.weight": "model-00002-of-00002.safetensors",
185
+ "model.vision_tower.vision_tower.stages.2.blocks.2.weight": "model-00002-of-00002.safetensors",
186
+ "model.vision_tower.vision_tower.stages.2.blocks.20.conv_dw.bias": "model-00002-of-00002.safetensors",
187
+ "model.vision_tower.vision_tower.stages.2.blocks.20.conv_dw.weight": "model-00002-of-00002.safetensors",
188
+ "model.vision_tower.vision_tower.stages.2.blocks.20.mlp.fc1.bias": "model-00002-of-00002.safetensors",
189
+ "model.vision_tower.vision_tower.stages.2.blocks.20.mlp.fc1.weight": "model-00002-of-00002.safetensors",
190
+ "model.vision_tower.vision_tower.stages.2.blocks.20.mlp.fc2.bias": "model-00002-of-00002.safetensors",
191
+ "model.vision_tower.vision_tower.stages.2.blocks.20.mlp.fc2.weight": "model-00002-of-00002.safetensors",
192
+ "model.vision_tower.vision_tower.stages.2.blocks.20.norm.bias": "model-00002-of-00002.safetensors",
193
+ "model.vision_tower.vision_tower.stages.2.blocks.20.norm.weight": "model-00002-of-00002.safetensors",
194
+ "model.vision_tower.vision_tower.stages.2.blocks.20.weight": "model-00002-of-00002.safetensors",
195
+ "model.vision_tower.vision_tower.stages.2.blocks.21.conv_dw.bias": "model-00002-of-00002.safetensors",
196
+ "model.vision_tower.vision_tower.stages.2.blocks.21.conv_dw.weight": "model-00002-of-00002.safetensors",
197
+ "model.vision_tower.vision_tower.stages.2.blocks.21.mlp.fc1.bias": "model-00002-of-00002.safetensors",
198
+ "model.vision_tower.vision_tower.stages.2.blocks.21.mlp.fc1.weight": "model-00002-of-00002.safetensors",
199
+ "model.vision_tower.vision_tower.stages.2.blocks.21.mlp.fc2.bias": "model-00002-of-00002.safetensors",
200
+ "model.vision_tower.vision_tower.stages.2.blocks.21.mlp.fc2.weight": "model-00002-of-00002.safetensors",
201
+ "model.vision_tower.vision_tower.stages.2.blocks.21.norm.bias": "model-00002-of-00002.safetensors",
202
+ "model.vision_tower.vision_tower.stages.2.blocks.21.norm.weight": "model-00002-of-00002.safetensors",
203
+ "model.vision_tower.vision_tower.stages.2.blocks.21.weight": "model-00002-of-00002.safetensors",
204
+ "model.vision_tower.vision_tower.stages.2.blocks.22.conv_dw.bias": "model-00002-of-00002.safetensors",
205
+ "model.vision_tower.vision_tower.stages.2.blocks.22.conv_dw.weight": "model-00002-of-00002.safetensors",
206
+ "model.vision_tower.vision_tower.stages.2.blocks.22.mlp.fc1.bias": "model-00002-of-00002.safetensors",
207
+ "model.vision_tower.vision_tower.stages.2.blocks.22.mlp.fc1.weight": "model-00002-of-00002.safetensors",
208
+ "model.vision_tower.vision_tower.stages.2.blocks.22.mlp.fc2.bias": "model-00002-of-00002.safetensors",
209
+ "model.vision_tower.vision_tower.stages.2.blocks.22.mlp.fc2.weight": "model-00002-of-00002.safetensors",
210
+ "model.vision_tower.vision_tower.stages.2.blocks.22.norm.bias": "model-00002-of-00002.safetensors",
211
+ "model.vision_tower.vision_tower.stages.2.blocks.22.norm.weight": "model-00002-of-00002.safetensors",
212
+ "model.vision_tower.vision_tower.stages.2.blocks.22.weight": "model-00002-of-00002.safetensors",
213
+ "model.vision_tower.vision_tower.stages.2.blocks.23.conv_dw.bias": "model-00002-of-00002.safetensors",
214
+ "model.vision_tower.vision_tower.stages.2.blocks.23.conv_dw.weight": "model-00002-of-00002.safetensors",
215
+ "model.vision_tower.vision_tower.stages.2.blocks.23.mlp.fc1.bias": "model-00002-of-00002.safetensors",
216
+ "model.vision_tower.vision_tower.stages.2.blocks.23.mlp.fc1.weight": "model-00002-of-00002.safetensors",
217
+ "model.vision_tower.vision_tower.stages.2.blocks.23.mlp.fc2.bias": "model-00002-of-00002.safetensors",
218
+ "model.vision_tower.vision_tower.stages.2.blocks.23.mlp.fc2.weight": "model-00002-of-00002.safetensors",
219
+ "model.vision_tower.vision_tower.stages.2.blocks.23.norm.bias": "model-00002-of-00002.safetensors",
220
+ "model.vision_tower.vision_tower.stages.2.blocks.23.norm.weight": "model-00002-of-00002.safetensors",
221
+ "model.vision_tower.vision_tower.stages.2.blocks.23.weight": "model-00002-of-00002.safetensors",
222
+ "model.vision_tower.vision_tower.stages.2.blocks.24.conv_dw.bias": "model-00002-of-00002.safetensors",
223
+ "model.vision_tower.vision_tower.stages.2.blocks.24.conv_dw.weight": "model-00002-of-00002.safetensors",
224
+ "model.vision_tower.vision_tower.stages.2.blocks.24.mlp.fc1.bias": "model-00002-of-00002.safetensors",
225
+ "model.vision_tower.vision_tower.stages.2.blocks.24.mlp.fc1.weight": "model-00002-of-00002.safetensors",
226
+ "model.vision_tower.vision_tower.stages.2.blocks.24.mlp.fc2.bias": "model-00002-of-00002.safetensors",
227
+ "model.vision_tower.vision_tower.stages.2.blocks.24.mlp.fc2.weight": "model-00002-of-00002.safetensors",
228
+ "model.vision_tower.vision_tower.stages.2.blocks.24.norm.bias": "model-00002-of-00002.safetensors",
229
+ "model.vision_tower.vision_tower.stages.2.blocks.24.norm.weight": "model-00002-of-00002.safetensors",
230
+ "model.vision_tower.vision_tower.stages.2.blocks.24.weight": "model-00002-of-00002.safetensors",
231
+ "model.vision_tower.vision_tower.stages.2.blocks.25.conv_dw.bias": "model-00002-of-00002.safetensors",
232
+ "model.vision_tower.vision_tower.stages.2.blocks.25.conv_dw.weight": "model-00002-of-00002.safetensors",
233
+ "model.vision_tower.vision_tower.stages.2.blocks.25.mlp.fc1.bias": "model-00002-of-00002.safetensors",
234
+ "model.vision_tower.vision_tower.stages.2.blocks.25.mlp.fc1.weight": "model-00002-of-00002.safetensors",
235
+ "model.vision_tower.vision_tower.stages.2.blocks.25.mlp.fc2.bias": "model-00002-of-00002.safetensors",
236
+ "model.vision_tower.vision_tower.stages.2.blocks.25.mlp.fc2.weight": "model-00002-of-00002.safetensors",
237
+ "model.vision_tower.vision_tower.stages.2.blocks.25.norm.bias": "model-00002-of-00002.safetensors",
238
+ "model.vision_tower.vision_tower.stages.2.blocks.25.norm.weight": "model-00002-of-00002.safetensors",
239
+ "model.vision_tower.vision_tower.stages.2.blocks.25.weight": "model-00002-of-00002.safetensors",
240
+ "model.vision_tower.vision_tower.stages.2.blocks.26.conv_dw.bias": "model-00002-of-00002.safetensors",
241
+ "model.vision_tower.vision_tower.stages.2.blocks.26.conv_dw.weight": "model-00002-of-00002.safetensors",
242
+ "model.vision_tower.vision_tower.stages.2.blocks.26.mlp.fc1.bias": "model-00002-of-00002.safetensors",
243
+ "model.vision_tower.vision_tower.stages.2.blocks.26.mlp.fc1.weight": "model-00002-of-00002.safetensors",
244
+ "model.vision_tower.vision_tower.stages.2.blocks.26.mlp.fc2.bias": "model-00002-of-00002.safetensors",
245
+ "model.vision_tower.vision_tower.stages.2.blocks.26.mlp.fc2.weight": "model-00002-of-00002.safetensors",
246
+ "model.vision_tower.vision_tower.stages.2.blocks.26.norm.bias": "model-00002-of-00002.safetensors",
247
+ "model.vision_tower.vision_tower.stages.2.blocks.26.norm.weight": "model-00002-of-00002.safetensors",
248
+ "model.vision_tower.vision_tower.stages.2.blocks.26.weight": "model-00002-of-00002.safetensors",
249
+ "model.vision_tower.vision_tower.stages.2.blocks.3.conv_dw.bias": "model-00002-of-00002.safetensors",
250
+ "model.vision_tower.vision_tower.stages.2.blocks.3.conv_dw.weight": "model-00002-of-00002.safetensors",
251
+ "model.vision_tower.vision_tower.stages.2.blocks.3.mlp.fc1.bias": "model-00002-of-00002.safetensors",
252
+ "model.vision_tower.vision_tower.stages.2.blocks.3.mlp.fc1.weight": "model-00002-of-00002.safetensors",
253
+ "model.vision_tower.vision_tower.stages.2.blocks.3.mlp.fc2.bias": "model-00002-of-00002.safetensors",
254
+ "model.vision_tower.vision_tower.stages.2.blocks.3.mlp.fc2.weight": "model-00002-of-00002.safetensors",
255
+ "model.vision_tower.vision_tower.stages.2.blocks.3.norm.bias": "model-00002-of-00002.safetensors",
256
+ "model.vision_tower.vision_tower.stages.2.blocks.3.norm.weight": "model-00002-of-00002.safetensors",
257
+ "model.vision_tower.vision_tower.stages.2.blocks.3.weight": "model-00002-of-00002.safetensors",
258
+ "model.vision_tower.vision_tower.stages.2.blocks.4.conv_dw.bias": "model-00002-of-00002.safetensors",
259
+ "model.vision_tower.vision_tower.stages.2.blocks.4.conv_dw.weight": "model-00002-of-00002.safetensors",
260
+ "model.vision_tower.vision_tower.stages.2.blocks.4.mlp.fc1.bias": "model-00002-of-00002.safetensors",
261
+ "model.vision_tower.vision_tower.stages.2.blocks.4.mlp.fc1.weight": "model-00002-of-00002.safetensors",
262
+ "model.vision_tower.vision_tower.stages.2.blocks.4.mlp.fc2.bias": "model-00002-of-00002.safetensors",
263
+ "model.vision_tower.vision_tower.stages.2.blocks.4.mlp.fc2.weight": "model-00002-of-00002.safetensors",
264
+ "model.vision_tower.vision_tower.stages.2.blocks.4.norm.bias": "model-00002-of-00002.safetensors",
265
+ "model.vision_tower.vision_tower.stages.2.blocks.4.norm.weight": "model-00002-of-00002.safetensors",
266
+ "model.vision_tower.vision_tower.stages.2.blocks.4.weight": "model-00002-of-00002.safetensors",
267
+ "model.vision_tower.vision_tower.stages.2.blocks.5.conv_dw.bias": "model-00002-of-00002.safetensors",
268
+ "model.vision_tower.vision_tower.stages.2.blocks.5.conv_dw.weight": "model-00002-of-00002.safetensors",
269
+ "model.vision_tower.vision_tower.stages.2.blocks.5.mlp.fc1.bias": "model-00002-of-00002.safetensors",
270
+ "model.vision_tower.vision_tower.stages.2.blocks.5.mlp.fc1.weight": "model-00002-of-00002.safetensors",
271
+ "model.vision_tower.vision_tower.stages.2.blocks.5.mlp.fc2.bias": "model-00002-of-00002.safetensors",
272
+ "model.vision_tower.vision_tower.stages.2.blocks.5.mlp.fc2.weight": "model-00002-of-00002.safetensors",
273
+ "model.vision_tower.vision_tower.stages.2.blocks.5.norm.bias": "model-00002-of-00002.safetensors",
274
+ "model.vision_tower.vision_tower.stages.2.blocks.5.norm.weight": "model-00002-of-00002.safetensors",
275
+ "model.vision_tower.vision_tower.stages.2.blocks.5.weight": "model-00002-of-00002.safetensors",
276
+ "model.vision_tower.vision_tower.stages.2.blocks.6.conv_dw.bias": "model-00002-of-00002.safetensors",
277
+ "model.vision_tower.vision_tower.stages.2.blocks.6.conv_dw.weight": "model-00002-of-00002.safetensors",
278
+ "model.vision_tower.vision_tower.stages.2.blocks.6.mlp.fc1.bias": "model-00002-of-00002.safetensors",
279
+ "model.vision_tower.vision_tower.stages.2.blocks.6.mlp.fc1.weight": "model-00002-of-00002.safetensors",
280
+ "model.vision_tower.vision_tower.stages.2.blocks.6.mlp.fc2.bias": "model-00002-of-00002.safetensors",
281
+ "model.vision_tower.vision_tower.stages.2.blocks.6.mlp.fc2.weight": "model-00002-of-00002.safetensors",
282
+ "model.vision_tower.vision_tower.stages.2.blocks.6.norm.bias": "model-00002-of-00002.safetensors",
283
+ "model.vision_tower.vision_tower.stages.2.blocks.6.norm.weight": "model-00002-of-00002.safetensors",
284
+ "model.vision_tower.vision_tower.stages.2.blocks.6.weight": "model-00002-of-00002.safetensors",
285
+ "model.vision_tower.vision_tower.stages.2.blocks.7.conv_dw.bias": "model-00002-of-00002.safetensors",
286
+ "model.vision_tower.vision_tower.stages.2.blocks.7.conv_dw.weight": "model-00002-of-00002.safetensors",
287
+ "model.vision_tower.vision_tower.stages.2.blocks.7.mlp.fc1.bias": "model-00002-of-00002.safetensors",
288
+ "model.vision_tower.vision_tower.stages.2.blocks.7.mlp.fc1.weight": "model-00002-of-00002.safetensors",
289
+ "model.vision_tower.vision_tower.stages.2.blocks.7.mlp.fc2.bias": "model-00002-of-00002.safetensors",
290
+ "model.vision_tower.vision_tower.stages.2.blocks.7.mlp.fc2.weight": "model-00002-of-00002.safetensors",
291
+ "model.vision_tower.vision_tower.stages.2.blocks.7.norm.bias": "model-00002-of-00002.safetensors",
292
+ "model.vision_tower.vision_tower.stages.2.blocks.7.norm.weight": "model-00002-of-00002.safetensors",
293
+ "model.vision_tower.vision_tower.stages.2.blocks.7.weight": "model-00002-of-00002.safetensors",
294
+ "model.vision_tower.vision_tower.stages.2.blocks.8.conv_dw.bias": "model-00002-of-00002.safetensors",
295
+ "model.vision_tower.vision_tower.stages.2.blocks.8.conv_dw.weight": "model-00002-of-00002.safetensors",
296
+ "model.vision_tower.vision_tower.stages.2.blocks.8.mlp.fc1.bias": "model-00002-of-00002.safetensors",
297
+ "model.vision_tower.vision_tower.stages.2.blocks.8.mlp.fc1.weight": "model-00002-of-00002.safetensors",
298
+ "model.vision_tower.vision_tower.stages.2.blocks.8.mlp.fc2.bias": "model-00002-of-00002.safetensors",
299
+ "model.vision_tower.vision_tower.stages.2.blocks.8.mlp.fc2.weight": "model-00002-of-00002.safetensors",
300
+ "model.vision_tower.vision_tower.stages.2.blocks.8.norm.bias": "model-00002-of-00002.safetensors",
301
+ "model.vision_tower.vision_tower.stages.2.blocks.8.norm.weight": "model-00002-of-00002.safetensors",
302
+ "model.vision_tower.vision_tower.stages.2.blocks.8.weight": "model-00002-of-00002.safetensors",
303
+ "model.vision_tower.vision_tower.stages.2.blocks.9.conv_dw.bias": "model-00002-of-00002.safetensors",
304
+ "model.vision_tower.vision_tower.stages.2.blocks.9.conv_dw.weight": "model-00002-of-00002.safetensors",
305
+ "model.vision_tower.vision_tower.stages.2.blocks.9.mlp.fc1.bias": "model-00002-of-00002.safetensors",
306
+ "model.vision_tower.vision_tower.stages.2.blocks.9.mlp.fc1.weight": "model-00002-of-00002.safetensors",
307
+ "model.vision_tower.vision_tower.stages.2.blocks.9.mlp.fc2.bias": "model-00002-of-00002.safetensors",
308
+ "model.vision_tower.vision_tower.stages.2.blocks.9.mlp.fc2.weight": "model-00002-of-00002.safetensors",
309
+ "model.vision_tower.vision_tower.stages.2.blocks.9.norm.bias": "model-00002-of-00002.safetensors",
310
+ "model.vision_tower.vision_tower.stages.2.blocks.9.norm.weight": "model-00002-of-00002.safetensors",
311
+ "model.vision_tower.vision_tower.stages.2.blocks.9.weight": "model-00002-of-00002.safetensors",
312
+ "model.vision_tower.vision_tower.stages.2.downsample.0.bias": "model-00002-of-00002.safetensors",
313
+ "model.vision_tower.vision_tower.stages.2.downsample.0.weight": "model-00002-of-00002.safetensors",
314
+ "model.vision_tower.vision_tower.stages.2.downsample.1.bias": "model-00002-of-00002.safetensors",
315
+ "model.vision_tower.vision_tower.stages.2.downsample.1.weight": "model-00002-of-00002.safetensors",
316
+ "model.vision_tower.vision_tower.stages.3.blocks.0.conv_dw.bias": "model-00002-of-00002.safetensors",
317
+ "model.vision_tower.vision_tower.stages.3.blocks.0.conv_dw.weight": "model-00002-of-00002.safetensors",
318
+ "model.vision_tower.vision_tower.stages.3.blocks.0.mlp.fc1.bias": "model-00002-of-00002.safetensors",
319
+ "model.vision_tower.vision_tower.stages.3.blocks.0.mlp.fc1.weight": "model-00002-of-00002.safetensors",
320
+ "model.vision_tower.vision_tower.stages.3.blocks.0.mlp.fc2.bias": "model-00002-of-00002.safetensors",
321
+ "model.vision_tower.vision_tower.stages.3.blocks.0.mlp.fc2.weight": "model-00002-of-00002.safetensors",
322
+ "model.vision_tower.vision_tower.stages.3.blocks.0.norm.bias": "model-00002-of-00002.safetensors",
323
+ "model.vision_tower.vision_tower.stages.3.blocks.0.norm.weight": "model-00002-of-00002.safetensors",
324
+ "model.vision_tower.vision_tower.stages.3.blocks.0.weight": "model-00002-of-00002.safetensors",
325
+ "model.vision_tower.vision_tower.stages.3.blocks.1.conv_dw.bias": "model-00002-of-00002.safetensors",
326
+ "model.vision_tower.vision_tower.stages.3.blocks.1.conv_dw.weight": "model-00002-of-00002.safetensors",
327
+ "model.vision_tower.vision_tower.stages.3.blocks.1.mlp.fc1.bias": "model-00002-of-00002.safetensors",
328
+ "model.vision_tower.vision_tower.stages.3.blocks.1.mlp.fc1.weight": "model-00002-of-00002.safetensors",
329
+ "model.vision_tower.vision_tower.stages.3.blocks.1.mlp.fc2.bias": "model-00002-of-00002.safetensors",
330
+ "model.vision_tower.vision_tower.stages.3.blocks.1.mlp.fc2.weight": "model-00002-of-00002.safetensors",
331
+ "model.vision_tower.vision_tower.stages.3.blocks.1.norm.bias": "model-00002-of-00002.safetensors",
332
+ "model.vision_tower.vision_tower.stages.3.blocks.1.norm.weight": "model-00002-of-00002.safetensors",
333
+ "model.vision_tower.vision_tower.stages.3.blocks.1.weight": "model-00002-of-00002.safetensors",
334
+ "model.vision_tower.vision_tower.stages.3.blocks.2.conv_dw.bias": "model-00002-of-00002.safetensors",
335
+ "model.vision_tower.vision_tower.stages.3.blocks.2.conv_dw.weight": "model-00002-of-00002.safetensors",
336
+ "model.vision_tower.vision_tower.stages.3.blocks.2.mlp.fc1.bias": "model-00002-of-00002.safetensors",
337
+ "model.vision_tower.vision_tower.stages.3.blocks.2.mlp.fc1.weight": "model-00002-of-00002.safetensors",
338
+ "model.vision_tower.vision_tower.stages.3.blocks.2.mlp.fc2.bias": "model-00002-of-00002.safetensors",
339
+ "model.vision_tower.vision_tower.stages.3.blocks.2.mlp.fc2.weight": "model-00002-of-00002.safetensors",
340
+ "model.vision_tower.vision_tower.stages.3.blocks.2.norm.bias": "model-00002-of-00002.safetensors",
341
+ "model.vision_tower.vision_tower.stages.3.blocks.2.norm.weight": "model-00002-of-00002.safetensors",
342
+ "model.vision_tower.vision_tower.stages.3.blocks.2.weight": "model-00002-of-00002.safetensors",
343
+ "model.vision_tower.vision_tower.stages.3.downsample.0.bias": "model-00002-of-00002.safetensors",
344
+ "model.vision_tower.vision_tower.stages.3.downsample.0.weight": "model-00002-of-00002.safetensors",
345
+ "model.vision_tower.vision_tower.stages.3.downsample.1.bias": "model-00002-of-00002.safetensors",
346
+ "model.vision_tower.vision_tower.stages.3.downsample.1.weight": "model-00002-of-00002.safetensors",
347
+ "model.vision_tower.vision_tower.stages.4.blocks.0.conv_dw.bias": "model-00002-of-00002.safetensors",
348
+ "model.vision_tower.vision_tower.stages.4.blocks.0.conv_dw.weight": "model-00002-of-00002.safetensors",
349
+ "model.vision_tower.vision_tower.stages.4.blocks.0.mlp.fc1.bias": "model-00002-of-00002.safetensors",
350
+ "model.vision_tower.vision_tower.stages.4.blocks.0.mlp.fc1.weight": "model-00002-of-00002.safetensors",
351
+ "model.vision_tower.vision_tower.stages.4.blocks.0.mlp.fc2.bias": "model-00002-of-00002.safetensors",
352
+ "model.vision_tower.vision_tower.stages.4.blocks.0.mlp.fc2.weight": "model-00002-of-00002.safetensors",
353
+ "model.vision_tower.vision_tower.stages.4.blocks.0.norm.bias": "model-00002-of-00002.safetensors",
354
+ "model.vision_tower.vision_tower.stages.4.blocks.0.norm.weight": "model-00002-of-00002.safetensors",
355
+ "model.vision_tower.vision_tower.stages.4.blocks.0.weight": "model-00002-of-00002.safetensors",
356
+ "model.vision_tower.vision_tower.stages.4.blocks.1.conv_dw.bias": "model-00002-of-00002.safetensors",
357
+ "model.vision_tower.vision_tower.stages.4.blocks.1.conv_dw.weight": "model-00002-of-00002.safetensors",
358
+ "model.vision_tower.vision_tower.stages.4.blocks.1.mlp.fc1.bias": "model-00002-of-00002.safetensors",
359
+ "model.vision_tower.vision_tower.stages.4.blocks.1.mlp.fc1.weight": "model-00002-of-00002.safetensors",
360
+ "model.vision_tower.vision_tower.stages.4.blocks.1.mlp.fc2.bias": "model-00002-of-00002.safetensors",
361
+ "model.vision_tower.vision_tower.stages.4.blocks.1.mlp.fc2.weight": "model-00002-of-00002.safetensors",
362
+ "model.vision_tower.vision_tower.stages.4.blocks.1.norm.bias": "model-00002-of-00002.safetensors",
363
+ "model.vision_tower.vision_tower.stages.4.blocks.1.norm.weight": "model-00002-of-00002.safetensors",
364
+ "model.vision_tower.vision_tower.stages.4.blocks.1.weight": "model-00002-of-00002.safetensors",
365
+ "model.vision_tower.vision_tower.stages.4.blocks.2.conv_dw.bias": "model-00002-of-00002.safetensors",
366
+ "model.vision_tower.vision_tower.stages.4.blocks.2.conv_dw.weight": "model-00002-of-00002.safetensors",
367
+ "model.vision_tower.vision_tower.stages.4.blocks.2.mlp.fc1.bias": "model-00002-of-00002.safetensors",
368
+ "model.vision_tower.vision_tower.stages.4.blocks.2.mlp.fc1.weight": "model-00002-of-00002.safetensors",
369
+ "model.vision_tower.vision_tower.stages.4.blocks.2.mlp.fc2.bias": "model-00002-of-00002.safetensors",
370
+ "model.vision_tower.vision_tower.stages.4.blocks.2.mlp.fc2.weight": "model-00002-of-00002.safetensors",
371
+ "model.vision_tower.vision_tower.stages.4.blocks.2.norm.bias": "model-00002-of-00002.safetensors",
372
+ "model.vision_tower.vision_tower.stages.4.blocks.2.norm.weight": "model-00002-of-00002.safetensors",
373
+ "model.vision_tower.vision_tower.stages.4.blocks.2.weight": "model-00002-of-00002.safetensors",
374
+ "model.vision_tower.vision_tower.stages.4.blocks.3.conv_dw.bias": "model-00002-of-00002.safetensors",
375
+ "model.vision_tower.vision_tower.stages.4.blocks.3.conv_dw.weight": "model-00002-of-00002.safetensors",
376
+ "model.vision_tower.vision_tower.stages.4.blocks.3.mlp.fc1.bias": "model-00002-of-00002.safetensors",
377
+ "model.vision_tower.vision_tower.stages.4.blocks.3.mlp.fc1.weight": "model-00002-of-00002.safetensors",
378
+ "model.vision_tower.vision_tower.stages.4.blocks.3.mlp.fc2.bias": "model-00002-of-00002.safetensors",
379
+ "model.vision_tower.vision_tower.stages.4.blocks.3.mlp.fc2.weight": "model-00002-of-00002.safetensors",
380
+ "model.vision_tower.vision_tower.stages.4.blocks.3.norm.bias": "model-00002-of-00002.safetensors",
381
+ "model.vision_tower.vision_tower.stages.4.blocks.3.norm.weight": "model-00002-of-00002.safetensors",
382
+ "model.vision_tower.vision_tower.stages.4.blocks.3.weight": "model-00002-of-00002.safetensors",
383
+ "model.vision_tower.vision_tower.stages.4.blocks.4.conv_dw.bias": "model-00002-of-00002.safetensors",
384
+ "model.vision_tower.vision_tower.stages.4.blocks.4.conv_dw.weight": "model-00002-of-00002.safetensors",
385
+ "model.vision_tower.vision_tower.stages.4.blocks.4.mlp.fc1.bias": "model-00002-of-00002.safetensors",
386
+ "model.vision_tower.vision_tower.stages.4.blocks.4.mlp.fc1.weight": "model-00002-of-00002.safetensors",
387
+ "model.vision_tower.vision_tower.stages.4.blocks.4.mlp.fc2.bias": "model-00002-of-00002.safetensors",
388
+ "model.vision_tower.vision_tower.stages.4.blocks.4.mlp.fc2.weight": "model-00002-of-00002.safetensors",
389
+ "model.vision_tower.vision_tower.stages.4.blocks.4.norm.bias": "model-00002-of-00002.safetensors",
390
+ "model.vision_tower.vision_tower.stages.4.blocks.4.norm.weight": "model-00002-of-00002.safetensors",
391
+ "model.vision_tower.vision_tower.stages.4.blocks.4.weight": "model-00002-of-00002.safetensors",
392
+ "model.vision_tower.vision_tower.stages.4.blocks.5.conv_dw.bias": "model-00002-of-00002.safetensors",
393
+ "model.vision_tower.vision_tower.stages.4.blocks.5.conv_dw.weight": "model-00002-of-00002.safetensors",
394
+ "model.vision_tower.vision_tower.stages.4.blocks.5.mlp.fc1.bias": "model-00002-of-00002.safetensors",
395
+ "model.vision_tower.vision_tower.stages.4.blocks.5.mlp.fc1.weight": "model-00002-of-00002.safetensors",
396
+ "model.vision_tower.vision_tower.stages.4.blocks.5.mlp.fc2.bias": "model-00002-of-00002.safetensors",
397
+ "model.vision_tower.vision_tower.stages.4.blocks.5.mlp.fc2.weight": "model-00002-of-00002.safetensors",
398
+ "model.vision_tower.vision_tower.stages.4.blocks.5.norm.bias": "model-00002-of-00002.safetensors",
399
+ "model.vision_tower.vision_tower.stages.4.blocks.5.norm.weight": "model-00002-of-00002.safetensors",
400
+ "model.vision_tower.vision_tower.stages.4.blocks.5.weight": "model-00002-of-00002.safetensors",
401
+ "model.vision_tower.vision_tower.stages.4.downsample.0.bias": "model-00002-of-00002.safetensors",
402
+ "model.vision_tower.vision_tower.stages.4.downsample.0.weight": "model-00002-of-00002.safetensors",
403
+ "model.vision_tower.vision_tower.stages.4.downsample.1.bias": "model-00002-of-00002.safetensors",
404
+ "model.vision_tower.vision_tower.stages.4.downsample.1.weight": "model-00002-of-00002.safetensors",
405
+ "model.vision_tower.vision_tower.stem.0.bias": "model-00002-of-00002.safetensors",
406
+ "model.vision_tower.vision_tower.stem.0.weight": "model-00002-of-00002.safetensors",
407
+ "model.vision_tower.vision_tower.stem.1.bias": "model-00002-of-00002.safetensors",
408
+ "model.vision_tower.vision_tower.stem.1.weight": "model-00002-of-00002.safetensors",
409
+ "transformer.h.0.attn.c_attn.bias": "model-00001-of-00002.safetensors",
410
+ "transformer.h.0.attn.c_attn.weight": "model-00001-of-00002.safetensors",
411
+ "transformer.h.0.attn.c_proj.bias": "model-00001-of-00002.safetensors",
412
+ "transformer.h.0.attn.c_proj.weight": "model-00001-of-00002.safetensors",
413
+ "transformer.h.0.ln_1.bias": "model-00001-of-00002.safetensors",
414
+ "transformer.h.0.ln_1.weight": "model-00001-of-00002.safetensors",
415
+ "transformer.h.0.ln_2.bias": "model-00001-of-00002.safetensors",
416
+ "transformer.h.0.ln_2.weight": "model-00001-of-00002.safetensors",
417
+ "transformer.h.0.mlp.c_fc.bias": "model-00001-of-00002.safetensors",
418
+ "transformer.h.0.mlp.c_fc.weight": "model-00001-of-00002.safetensors",
419
+ "transformer.h.0.mlp.c_proj.bias": "model-00001-of-00002.safetensors",
420
+ "transformer.h.0.mlp.c_proj.weight": "model-00001-of-00002.safetensors",
421
+ "transformer.h.1.attn.c_attn.bias": "model-00001-of-00002.safetensors",
422
+ "transformer.h.1.attn.c_attn.weight": "model-00001-of-00002.safetensors",
423
+ "transformer.h.1.attn.c_proj.bias": "model-00001-of-00002.safetensors",
424
+ "transformer.h.1.attn.c_proj.weight": "model-00001-of-00002.safetensors",
425
+ "transformer.h.1.ln_1.bias": "model-00001-of-00002.safetensors",
426
+ "transformer.h.1.ln_1.weight": "model-00001-of-00002.safetensors",
427
+ "transformer.h.1.ln_2.bias": "model-00001-of-00002.safetensors",
428
+ "transformer.h.1.ln_2.weight": "model-00001-of-00002.safetensors",
429
+ "transformer.h.1.mlp.c_fc.bias": "model-00001-of-00002.safetensors",
430
+ "transformer.h.1.mlp.c_fc.weight": "model-00001-of-00002.safetensors",
431
+ "transformer.h.1.mlp.c_proj.bias": "model-00001-of-00002.safetensors",
432
+ "transformer.h.1.mlp.c_proj.weight": "model-00001-of-00002.safetensors",
433
+ "transformer.h.10.attn.c_attn.bias": "model-00001-of-00002.safetensors",
434
+ "transformer.h.10.attn.c_attn.weight": "model-00001-of-00002.safetensors",
435
+ "transformer.h.10.attn.c_proj.bias": "model-00001-of-00002.safetensors",
436
+ "transformer.h.10.attn.c_proj.weight": "model-00001-of-00002.safetensors",
437
+ "transformer.h.10.ln_1.bias": "model-00001-of-00002.safetensors",
438
+ "transformer.h.10.ln_1.weight": "model-00001-of-00002.safetensors",
439
+ "transformer.h.10.ln_2.bias": "model-00001-of-00002.safetensors",
440
+ "transformer.h.10.ln_2.weight": "model-00001-of-00002.safetensors",
441
+ "transformer.h.10.mlp.c_fc.bias": "model-00001-of-00002.safetensors",
442
+ "transformer.h.10.mlp.c_fc.weight": "model-00001-of-00002.safetensors",
443
+ "transformer.h.10.mlp.c_proj.bias": "model-00001-of-00002.safetensors",
444
+ "transformer.h.10.mlp.c_proj.weight": "model-00001-of-00002.safetensors",
445
+ "transformer.h.11.attn.c_attn.bias": "model-00001-of-00002.safetensors",
446
+ "transformer.h.11.attn.c_attn.weight": "model-00001-of-00002.safetensors",
447
+ "transformer.h.11.attn.c_proj.bias": "model-00001-of-00002.safetensors",
448
+ "transformer.h.11.attn.c_proj.weight": "model-00001-of-00002.safetensors",
449
+ "transformer.h.11.ln_1.bias": "model-00001-of-00002.safetensors",
450
+ "transformer.h.11.ln_1.weight": "model-00001-of-00002.safetensors",
451
+ "transformer.h.11.ln_2.bias": "model-00001-of-00002.safetensors",
452
+ "transformer.h.11.ln_2.weight": "model-00001-of-00002.safetensors",
453
+ "transformer.h.11.mlp.c_fc.bias": "model-00001-of-00002.safetensors",
454
+ "transformer.h.11.mlp.c_fc.weight": "model-00001-of-00002.safetensors",
455
+ "transformer.h.11.mlp.c_proj.bias": "model-00001-of-00002.safetensors",
456
+ "transformer.h.11.mlp.c_proj.weight": "model-00001-of-00002.safetensors",
457
+ "transformer.h.12.attn.c_attn.bias": "model-00001-of-00002.safetensors",
458
+ "transformer.h.12.attn.c_attn.weight": "model-00001-of-00002.safetensors",
459
+ "transformer.h.12.attn.c_proj.bias": "model-00001-of-00002.safetensors",
460
+ "transformer.h.12.attn.c_proj.weight": "model-00001-of-00002.safetensors",
461
+ "transformer.h.12.ln_1.bias": "model-00001-of-00002.safetensors",
462
+ "transformer.h.12.ln_1.weight": "model-00001-of-00002.safetensors",
463
+ "transformer.h.12.ln_2.bias": "model-00001-of-00002.safetensors",
464
+ "transformer.h.12.ln_2.weight": "model-00001-of-00002.safetensors",
465
+ "transformer.h.12.mlp.c_fc.bias": "model-00001-of-00002.safetensors",
466
+ "transformer.h.12.mlp.c_fc.weight": "model-00001-of-00002.safetensors",
467
+ "transformer.h.12.mlp.c_proj.bias": "model-00001-of-00002.safetensors",
468
+ "transformer.h.12.mlp.c_proj.weight": "model-00001-of-00002.safetensors",
469
+ "transformer.h.13.attn.c_attn.bias": "model-00001-of-00002.safetensors",
470
+ "transformer.h.13.attn.c_attn.weight": "model-00001-of-00002.safetensors",
471
+ "transformer.h.13.attn.c_proj.bias": "model-00001-of-00002.safetensors",
472
+ "transformer.h.13.attn.c_proj.weight": "model-00001-of-00002.safetensors",
473
+ "transformer.h.13.ln_1.bias": "model-00001-of-00002.safetensors",
474
+ "transformer.h.13.ln_1.weight": "model-00001-of-00002.safetensors",
475
+ "transformer.h.13.ln_2.bias": "model-00001-of-00002.safetensors",
476
+ "transformer.h.13.ln_2.weight": "model-00001-of-00002.safetensors",
477
+ "transformer.h.13.mlp.c_fc.bias": "model-00001-of-00002.safetensors",
478
+ "transformer.h.13.mlp.c_fc.weight": "model-00001-of-00002.safetensors",
479
+ "transformer.h.13.mlp.c_proj.bias": "model-00001-of-00002.safetensors",
480
+ "transformer.h.13.mlp.c_proj.weight": "model-00001-of-00002.safetensors",
481
+ "transformer.h.14.attn.c_attn.bias": "model-00001-of-00002.safetensors",
482
+ "transformer.h.14.attn.c_attn.weight": "model-00001-of-00002.safetensors",
483
+ "transformer.h.14.attn.c_proj.bias": "model-00001-of-00002.safetensors",
484
+ "transformer.h.14.attn.c_proj.weight": "model-00001-of-00002.safetensors",
485
+ "transformer.h.14.ln_1.bias": "model-00001-of-00002.safetensors",
486
+ "transformer.h.14.ln_1.weight": "model-00001-of-00002.safetensors",
487
+ "transformer.h.14.ln_2.bias": "model-00001-of-00002.safetensors",
488
+ "transformer.h.14.ln_2.weight": "model-00001-of-00002.safetensors",
489
+ "transformer.h.14.mlp.c_fc.bias": "model-00001-of-00002.safetensors",
490
+ "transformer.h.14.mlp.c_fc.weight": "model-00001-of-00002.safetensors",
491
+ "transformer.h.14.mlp.c_proj.bias": "model-00001-of-00002.safetensors",
492
+ "transformer.h.14.mlp.c_proj.weight": "model-00001-of-00002.safetensors",
493
+ "transformer.h.15.attn.c_attn.bias": "model-00001-of-00002.safetensors",
494
+ "transformer.h.15.attn.c_attn.weight": "model-00001-of-00002.safetensors",
495
+ "transformer.h.15.attn.c_proj.bias": "model-00001-of-00002.safetensors",
496
+ "transformer.h.15.attn.c_proj.weight": "model-00001-of-00002.safetensors",
497
+ "transformer.h.15.ln_1.bias": "model-00001-of-00002.safetensors",
498
+ "transformer.h.15.ln_1.weight": "model-00001-of-00002.safetensors",
499
+ "transformer.h.15.ln_2.bias": "model-00001-of-00002.safetensors",
500
+ "transformer.h.15.ln_2.weight": "model-00001-of-00002.safetensors",
501
+ "transformer.h.15.mlp.c_fc.bias": "model-00001-of-00002.safetensors",
502
+ "transformer.h.15.mlp.c_fc.weight": "model-00001-of-00002.safetensors",
503
+ "transformer.h.15.mlp.c_proj.bias": "model-00001-of-00002.safetensors",
504
+ "transformer.h.15.mlp.c_proj.weight": "model-00001-of-00002.safetensors",
505
+ "transformer.h.16.attn.c_attn.bias": "model-00001-of-00002.safetensors",
506
+ "transformer.h.16.attn.c_attn.weight": "model-00001-of-00002.safetensors",
507
+ "transformer.h.16.attn.c_proj.bias": "model-00001-of-00002.safetensors",
508
+ "transformer.h.16.attn.c_proj.weight": "model-00001-of-00002.safetensors",
509
+ "transformer.h.16.ln_1.bias": "model-00001-of-00002.safetensors",
510
+ "transformer.h.16.ln_1.weight": "model-00001-of-00002.safetensors",
511
+ "transformer.h.16.ln_2.bias": "model-00001-of-00002.safetensors",
512
+ "transformer.h.16.ln_2.weight": "model-00001-of-00002.safetensors",
513
+ "transformer.h.16.mlp.c_fc.bias": "model-00001-of-00002.safetensors",
514
+ "transformer.h.16.mlp.c_fc.weight": "model-00001-of-00002.safetensors",
515
+ "transformer.h.16.mlp.c_proj.bias": "model-00001-of-00002.safetensors",
516
+ "transformer.h.16.mlp.c_proj.weight": "model-00001-of-00002.safetensors",
517
+ "transformer.h.17.attn.c_attn.bias": "model-00001-of-00002.safetensors",
518
+ "transformer.h.17.attn.c_attn.weight": "model-00001-of-00002.safetensors",
519
+ "transformer.h.17.attn.c_proj.bias": "model-00001-of-00002.safetensors",
520
+ "transformer.h.17.attn.c_proj.weight": "model-00001-of-00002.safetensors",
521
+ "transformer.h.17.ln_1.bias": "model-00001-of-00002.safetensors",
522
+ "transformer.h.17.ln_1.weight": "model-00001-of-00002.safetensors",
523
+ "transformer.h.17.ln_2.bias": "model-00001-of-00002.safetensors",
524
+ "transformer.h.17.ln_2.weight": "model-00001-of-00002.safetensors",
525
+ "transformer.h.17.mlp.c_fc.bias": "model-00001-of-00002.safetensors",
526
+ "transformer.h.17.mlp.c_fc.weight": "model-00001-of-00002.safetensors",
527
+ "transformer.h.17.mlp.c_proj.bias": "model-00001-of-00002.safetensors",
528
+ "transformer.h.17.mlp.c_proj.weight": "model-00001-of-00002.safetensors",
529
+ "transformer.h.18.attn.c_attn.bias": "model-00001-of-00002.safetensors",
530
+ "transformer.h.18.attn.c_attn.weight": "model-00001-of-00002.safetensors",
531
+ "transformer.h.18.attn.c_proj.bias": "model-00001-of-00002.safetensors",
532
+ "transformer.h.18.attn.c_proj.weight": "model-00001-of-00002.safetensors",
533
+ "transformer.h.18.ln_1.bias": "model-00001-of-00002.safetensors",
534
+ "transformer.h.18.ln_1.weight": "model-00001-of-00002.safetensors",
535
+ "transformer.h.18.ln_2.bias": "model-00001-of-00002.safetensors",
536
+ "transformer.h.18.ln_2.weight": "model-00001-of-00002.safetensors",
537
+ "transformer.h.18.mlp.c_fc.bias": "model-00001-of-00002.safetensors",
538
+ "transformer.h.18.mlp.c_fc.weight": "model-00001-of-00002.safetensors",
539
+ "transformer.h.18.mlp.c_proj.bias": "model-00001-of-00002.safetensors",
540
+ "transformer.h.18.mlp.c_proj.weight": "model-00001-of-00002.safetensors",
541
+ "transformer.h.19.attn.c_attn.bias": "model-00001-of-00002.safetensors",
542
+ "transformer.h.19.attn.c_attn.weight": "model-00001-of-00002.safetensors",
543
+ "transformer.h.19.attn.c_proj.bias": "model-00001-of-00002.safetensors",
544
+ "transformer.h.19.attn.c_proj.weight": "model-00001-of-00002.safetensors",
545
+ "transformer.h.19.ln_1.bias": "model-00001-of-00002.safetensors",
546
+ "transformer.h.19.ln_1.weight": "model-00001-of-00002.safetensors",
547
+ "transformer.h.19.ln_2.bias": "model-00001-of-00002.safetensors",
548
+ "transformer.h.19.ln_2.weight": "model-00001-of-00002.safetensors",
549
+ "transformer.h.19.mlp.c_fc.bias": "model-00001-of-00002.safetensors",
550
+ "transformer.h.19.mlp.c_fc.weight": "model-00001-of-00002.safetensors",
551
+ "transformer.h.19.mlp.c_proj.bias": "model-00001-of-00002.safetensors",
552
+ "transformer.h.19.mlp.c_proj.weight": "model-00001-of-00002.safetensors",
553
+ "transformer.h.2.attn.c_attn.bias": "model-00001-of-00002.safetensors",
554
+ "transformer.h.2.attn.c_attn.weight": "model-00001-of-00002.safetensors",
555
+ "transformer.h.2.attn.c_proj.bias": "model-00001-of-00002.safetensors",
556
+ "transformer.h.2.attn.c_proj.weight": "model-00001-of-00002.safetensors",
557
+ "transformer.h.2.ln_1.bias": "model-00001-of-00002.safetensors",
558
+ "transformer.h.2.ln_1.weight": "model-00001-of-00002.safetensors",
559
+ "transformer.h.2.ln_2.bias": "model-00001-of-00002.safetensors",
560
+ "transformer.h.2.ln_2.weight": "model-00001-of-00002.safetensors",
561
+ "transformer.h.2.mlp.c_fc.bias": "model-00001-of-00002.safetensors",
562
+ "transformer.h.2.mlp.c_fc.weight": "model-00001-of-00002.safetensors",
563
+ "transformer.h.2.mlp.c_proj.bias": "model-00001-of-00002.safetensors",
564
+ "transformer.h.2.mlp.c_proj.weight": "model-00001-of-00002.safetensors",
565
+ "transformer.h.20.attn.c_attn.bias": "model-00001-of-00002.safetensors",
566
+ "transformer.h.20.attn.c_attn.weight": "model-00001-of-00002.safetensors",
567
+ "transformer.h.20.attn.c_proj.bias": "model-00001-of-00002.safetensors",
568
+ "transformer.h.20.attn.c_proj.weight": "model-00001-of-00002.safetensors",
569
+ "transformer.h.20.ln_1.bias": "model-00001-of-00002.safetensors",
570
+ "transformer.h.20.ln_1.weight": "model-00001-of-00002.safetensors",
571
+ "transformer.h.20.ln_2.bias": "model-00001-of-00002.safetensors",
572
+ "transformer.h.20.ln_2.weight": "model-00001-of-00002.safetensors",
573
+ "transformer.h.20.mlp.c_fc.bias": "model-00001-of-00002.safetensors",
574
+ "transformer.h.20.mlp.c_fc.weight": "model-00001-of-00002.safetensors",
575
+ "transformer.h.20.mlp.c_proj.bias": "model-00001-of-00002.safetensors",
576
+ "transformer.h.20.mlp.c_proj.weight": "model-00001-of-00002.safetensors",
577
+ "transformer.h.21.attn.c_attn.bias": "model-00001-of-00002.safetensors",
578
+ "transformer.h.21.attn.c_attn.weight": "model-00001-of-00002.safetensors",
579
+ "transformer.h.21.attn.c_proj.bias": "model-00001-of-00002.safetensors",
580
+ "transformer.h.21.attn.c_proj.weight": "model-00001-of-00002.safetensors",
581
+ "transformer.h.21.ln_1.bias": "model-00001-of-00002.safetensors",
582
+ "transformer.h.21.ln_1.weight": "model-00001-of-00002.safetensors",
583
+ "transformer.h.21.ln_2.bias": "model-00001-of-00002.safetensors",
584
+ "transformer.h.21.ln_2.weight": "model-00001-of-00002.safetensors",
585
+ "transformer.h.21.mlp.c_fc.bias": "model-00001-of-00002.safetensors",
586
+ "transformer.h.21.mlp.c_fc.weight": "model-00001-of-00002.safetensors",
587
+ "transformer.h.21.mlp.c_proj.bias": "model-00001-of-00002.safetensors",
588
+ "transformer.h.21.mlp.c_proj.weight": "model-00001-of-00002.safetensors",
589
+ "transformer.h.22.attn.c_attn.bias": "model-00001-of-00002.safetensors",
590
+ "transformer.h.22.attn.c_attn.weight": "model-00001-of-00002.safetensors",
591
+ "transformer.h.22.attn.c_proj.bias": "model-00001-of-00002.safetensors",
592
+ "transformer.h.22.attn.c_proj.weight": "model-00001-of-00002.safetensors",
593
+ "transformer.h.22.ln_1.bias": "model-00001-of-00002.safetensors",
594
+ "transformer.h.22.ln_1.weight": "model-00001-of-00002.safetensors",
595
+ "transformer.h.22.ln_2.bias": "model-00001-of-00002.safetensors",
596
+ "transformer.h.22.ln_2.weight": "model-00001-of-00002.safetensors",
597
+ "transformer.h.22.mlp.c_fc.bias": "model-00001-of-00002.safetensors",
598
+ "transformer.h.22.mlp.c_fc.weight": "model-00001-of-00002.safetensors",
599
+ "transformer.h.22.mlp.c_proj.bias": "model-00002-of-00002.safetensors",
600
+ "transformer.h.22.mlp.c_proj.weight": "model-00002-of-00002.safetensors",
601
+ "transformer.h.23.attn.c_attn.bias": "model-00002-of-00002.safetensors",
602
+ "transformer.h.23.attn.c_attn.weight": "model-00002-of-00002.safetensors",
603
+ "transformer.h.23.attn.c_proj.bias": "model-00002-of-00002.safetensors",
604
+ "transformer.h.23.attn.c_proj.weight": "model-00002-of-00002.safetensors",
605
+ "transformer.h.23.ln_1.bias": "model-00002-of-00002.safetensors",
606
+ "transformer.h.23.ln_1.weight": "model-00002-of-00002.safetensors",
607
+ "transformer.h.23.ln_2.bias": "model-00002-of-00002.safetensors",
608
+ "transformer.h.23.ln_2.weight": "model-00002-of-00002.safetensors",
609
+ "transformer.h.23.mlp.c_fc.bias": "model-00002-of-00002.safetensors",
610
+ "transformer.h.23.mlp.c_fc.weight": "model-00002-of-00002.safetensors",
611
+ "transformer.h.23.mlp.c_proj.bias": "model-00002-of-00002.safetensors",
612
+ "transformer.h.23.mlp.c_proj.weight": "model-00002-of-00002.safetensors",
613
+ "transformer.h.3.attn.c_attn.bias": "model-00001-of-00002.safetensors",
614
+ "transformer.h.3.attn.c_attn.weight": "model-00001-of-00002.safetensors",
615
+ "transformer.h.3.attn.c_proj.bias": "model-00001-of-00002.safetensors",
616
+ "transformer.h.3.attn.c_proj.weight": "model-00001-of-00002.safetensors",
617
+ "transformer.h.3.ln_1.bias": "model-00001-of-00002.safetensors",
618
+ "transformer.h.3.ln_1.weight": "model-00001-of-00002.safetensors",
619
+ "transformer.h.3.ln_2.bias": "model-00001-of-00002.safetensors",
620
+ "transformer.h.3.ln_2.weight": "model-00001-of-00002.safetensors",
621
+ "transformer.h.3.mlp.c_fc.bias": "model-00001-of-00002.safetensors",
622
+ "transformer.h.3.mlp.c_fc.weight": "model-00001-of-00002.safetensors",
623
+ "transformer.h.3.mlp.c_proj.bias": "model-00001-of-00002.safetensors",
624
+ "transformer.h.3.mlp.c_proj.weight": "model-00001-of-00002.safetensors",
625
+ "transformer.h.4.attn.c_attn.bias": "model-00001-of-00002.safetensors",
626
+ "transformer.h.4.attn.c_attn.weight": "model-00001-of-00002.safetensors",
627
+ "transformer.h.4.attn.c_proj.bias": "model-00001-of-00002.safetensors",
628
+ "transformer.h.4.attn.c_proj.weight": "model-00001-of-00002.safetensors",
629
+ "transformer.h.4.ln_1.bias": "model-00001-of-00002.safetensors",
630
+ "transformer.h.4.ln_1.weight": "model-00001-of-00002.safetensors",
631
+ "transformer.h.4.ln_2.bias": "model-00001-of-00002.safetensors",
632
+ "transformer.h.4.ln_2.weight": "model-00001-of-00002.safetensors",
633
+ "transformer.h.4.mlp.c_fc.bias": "model-00001-of-00002.safetensors",
634
+ "transformer.h.4.mlp.c_fc.weight": "model-00001-of-00002.safetensors",
635
+ "transformer.h.4.mlp.c_proj.bias": "model-00001-of-00002.safetensors",
636
+ "transformer.h.4.mlp.c_proj.weight": "model-00001-of-00002.safetensors",
637
+ "transformer.h.5.attn.c_attn.bias": "model-00001-of-00002.safetensors",
638
+ "transformer.h.5.attn.c_attn.weight": "model-00001-of-00002.safetensors",
639
+ "transformer.h.5.attn.c_proj.bias": "model-00001-of-00002.safetensors",
640
+ "transformer.h.5.attn.c_proj.weight": "model-00001-of-00002.safetensors",
641
+ "transformer.h.5.ln_1.bias": "model-00001-of-00002.safetensors",
642
+ "transformer.h.5.ln_1.weight": "model-00001-of-00002.safetensors",
643
+ "transformer.h.5.ln_2.bias": "model-00001-of-00002.safetensors",
644
+ "transformer.h.5.ln_2.weight": "model-00001-of-00002.safetensors",
645
+ "transformer.h.5.mlp.c_fc.bias": "model-00001-of-00002.safetensors",
646
+ "transformer.h.5.mlp.c_fc.weight": "model-00001-of-00002.safetensors",
647
+ "transformer.h.5.mlp.c_proj.bias": "model-00001-of-00002.safetensors",
648
+ "transformer.h.5.mlp.c_proj.weight": "model-00001-of-00002.safetensors",
649
+ "transformer.h.6.attn.c_attn.bias": "model-00001-of-00002.safetensors",
650
+ "transformer.h.6.attn.c_attn.weight": "model-00001-of-00002.safetensors",
651
+ "transformer.h.6.attn.c_proj.bias": "model-00001-of-00002.safetensors",
652
+ "transformer.h.6.attn.c_proj.weight": "model-00001-of-00002.safetensors",
653
+ "transformer.h.6.ln_1.bias": "model-00001-of-00002.safetensors",
654
+ "transformer.h.6.ln_1.weight": "model-00001-of-00002.safetensors",
655
+ "transformer.h.6.ln_2.bias": "model-00001-of-00002.safetensors",
656
+ "transformer.h.6.ln_2.weight": "model-00001-of-00002.safetensors",
657
+ "transformer.h.6.mlp.c_fc.bias": "model-00001-of-00002.safetensors",
658
+ "transformer.h.6.mlp.c_fc.weight": "model-00001-of-00002.safetensors",
659
+ "transformer.h.6.mlp.c_proj.bias": "model-00001-of-00002.safetensors",
660
+ "transformer.h.6.mlp.c_proj.weight": "model-00001-of-00002.safetensors",
661
+ "transformer.h.7.attn.c_attn.bias": "model-00001-of-00002.safetensors",
662
+ "transformer.h.7.attn.c_attn.weight": "model-00001-of-00002.safetensors",
663
+ "transformer.h.7.attn.c_proj.bias": "model-00001-of-00002.safetensors",
664
+ "transformer.h.7.attn.c_proj.weight": "model-00001-of-00002.safetensors",
665
+ "transformer.h.7.ln_1.bias": "model-00001-of-00002.safetensors",
666
+ "transformer.h.7.ln_1.weight": "model-00001-of-00002.safetensors",
667
+ "transformer.h.7.ln_2.bias": "model-00001-of-00002.safetensors",
668
+ "transformer.h.7.ln_2.weight": "model-00001-of-00002.safetensors",
669
+ "transformer.h.7.mlp.c_fc.bias": "model-00001-of-00002.safetensors",
670
+ "transformer.h.7.mlp.c_fc.weight": "model-00001-of-00002.safetensors",
671
+ "transformer.h.7.mlp.c_proj.bias": "model-00001-of-00002.safetensors",
672
+ "transformer.h.7.mlp.c_proj.weight": "model-00001-of-00002.safetensors",
673
+ "transformer.h.8.attn.c_attn.bias": "model-00001-of-00002.safetensors",
674
+ "transformer.h.8.attn.c_attn.weight": "model-00001-of-00002.safetensors",
675
+ "transformer.h.8.attn.c_proj.bias": "model-00001-of-00002.safetensors",
676
+ "transformer.h.8.attn.c_proj.weight": "model-00001-of-00002.safetensors",
677
+ "transformer.h.8.ln_1.bias": "model-00001-of-00002.safetensors",
678
+ "transformer.h.8.ln_1.weight": "model-00001-of-00002.safetensors",
679
+ "transformer.h.8.ln_2.bias": "model-00001-of-00002.safetensors",
680
+ "transformer.h.8.ln_2.weight": "model-00001-of-00002.safetensors",
681
+ "transformer.h.8.mlp.c_fc.bias": "model-00001-of-00002.safetensors",
682
+ "transformer.h.8.mlp.c_fc.weight": "model-00001-of-00002.safetensors",
683
+ "transformer.h.8.mlp.c_proj.bias": "model-00001-of-00002.safetensors",
684
+ "transformer.h.8.mlp.c_proj.weight": "model-00001-of-00002.safetensors",
685
+ "transformer.h.9.attn.c_attn.bias": "model-00001-of-00002.safetensors",
686
+ "transformer.h.9.attn.c_attn.weight": "model-00001-of-00002.safetensors",
687
+ "transformer.h.9.attn.c_proj.bias": "model-00001-of-00002.safetensors",
688
+ "transformer.h.9.attn.c_proj.weight": "model-00001-of-00002.safetensors",
689
+ "transformer.h.9.ln_1.bias": "model-00001-of-00002.safetensors",
690
+ "transformer.h.9.ln_1.weight": "model-00001-of-00002.safetensors",
691
+ "transformer.h.9.ln_2.bias": "model-00001-of-00002.safetensors",
692
+ "transformer.h.9.ln_2.weight": "model-00001-of-00002.safetensors",
693
+ "transformer.h.9.mlp.c_fc.bias": "model-00001-of-00002.safetensors",
694
+ "transformer.h.9.mlp.c_fc.weight": "model-00001-of-00002.safetensors",
695
+ "transformer.h.9.mlp.c_proj.bias": "model-00001-of-00002.safetensors",
696
+ "transformer.h.9.mlp.c_proj.weight": "model-00001-of-00002.safetensors",
697
+ "transformer.ln_f.bias": "model-00002-of-00002.safetensors",
698
+ "transformer.ln_f.weight": "model-00002-of-00002.safetensors",
699
+ "transformer.wpe.weight": "model-00001-of-00002.safetensors",
700
+ "transformer.wte.weight": "model-00001-of-00002.safetensors"
701
+ }
702
+ }
special_tokens_map.json ADDED
@@ -0,0 +1,48 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "</s|LLM-jp>"
4
+ ],
5
+ "bos_token": {
6
+ "content": "<s|LLM-jp>",
7
+ "lstrip": false,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false
11
+ },
12
+ "cls_token": {
13
+ "content": "<CLS|LLM-jp>",
14
+ "lstrip": false,
15
+ "normalized": false,
16
+ "rstrip": false,
17
+ "single_word": false
18
+ },
19
+ "eos_token": {
20
+ "content": "<EOD|LLM-jp>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false
25
+ },
26
+ "mask_token": {
27
+ "content": "<mask|LLM-jp>",
28
+ "lstrip": false,
29
+ "normalized": false,
30
+ "rstrip": false,
31
+ "single_word": false
32
+ },
33
+ "pad_token": "<unk|LLM-jp>",
34
+ "sep_token": {
35
+ "content": "<SEP|LLM-jp>",
36
+ "lstrip": false,
37
+ "normalized": false,
38
+ "rstrip": false,
39
+ "single_word": false
40
+ },
41
+ "unk_token": {
42
+ "content": "<unk|LLM-jp>",
43
+ "lstrip": false,
44
+ "normalized": false,
45
+ "rstrip": false,
46
+ "single_word": false
47
+ }
48
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,85 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "<unk|LLM-jp>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "<s|LLM-jp>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "</s|LLM-jp>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "<mask|LLM-jp>",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "4": {
36
+ "content": "<pad|LLM-jp>",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ },
43
+ "5": {
44
+ "content": "<CLS|LLM-jp>",
45
+ "lstrip": false,
46
+ "normalized": false,
47
+ "rstrip": false,
48
+ "single_word": false,
49
+ "special": true
50
+ },
51
+ "6": {
52
+ "content": "<SEP|LLM-jp>",
53
+ "lstrip": false,
54
+ "normalized": false,
55
+ "rstrip": false,
56
+ "single_word": false,
57
+ "special": true
58
+ },
59
+ "7": {
60
+ "content": "<EOD|LLM-jp>",
61
+ "lstrip": false,
62
+ "normalized": false,
63
+ "rstrip": false,
64
+ "single_word": false,
65
+ "special": true
66
+ }
67
+ },
68
+ "additional_special_tokens": [
69
+ "</s|LLM-jp>"
70
+ ],
71
+ "bos_token": "<s|LLM-jp>",
72
+ "clean_up_tokenization_spaces": false,
73
+ "cls_token": "<CLS|LLM-jp>",
74
+ "eod_token": "<EOD|LLM-jp>",
75
+ "eos_token": "<EOD|LLM-jp>",
76
+ "extra_ids": 0,
77
+ "mask_token": "<mask|LLM-jp>",
78
+ "model_max_length": 1532,
79
+ "pad_token": "<unk|LLM-jp>",
80
+ "padding_side": "right",
81
+ "sep_token": "<SEP|LLM-jp>",
82
+ "sp_model_kwargs": {},
83
+ "tokenizer_class": "PreTrainedTokenizerFast",
84
+ "unk_token": "<unk|LLM-jp>"
85
+ }