Upload Gemma3ForConditionalGeneration
Browse files- config.json +6 -6
- model.safetensors +2 -2
config.json
CHANGED
@@ -6,7 +6,7 @@
|
|
6 |
"eoi_token_index": 256000,
|
7 |
"image_token_index": 262144,
|
8 |
"initializer_range": 0.02,
|
9 |
-
"mm_tokens_per_image":
|
10 |
"model_type": "gemma3",
|
11 |
"text_config": {
|
12 |
"attention_bias": false,
|
@@ -16,13 +16,13 @@
|
|
16 |
"final_logit_softcapping": null,
|
17 |
"head_dim": 256,
|
18 |
"hidden_activation": "gelu_pytorch_tanh",
|
19 |
-
"hidden_size":
|
20 |
"initializer_range": 0.02,
|
21 |
"intermediate_size": 512,
|
22 |
"max_position_embeddings": 131072,
|
23 |
"model_type": "gemma3_text",
|
24 |
"num_attention_heads": 4,
|
25 |
-
"num_hidden_layers":
|
26 |
"num_key_value_heads": 2,
|
27 |
"query_pre_attn_scalar": 256,
|
28 |
"rms_norm_eps": 1e-06,
|
@@ -39,14 +39,14 @@
|
|
39 |
"vision_config": {
|
40 |
"attention_dropout": 0.1,
|
41 |
"hidden_act": "gelu_pytorch_tanh",
|
42 |
-
"hidden_size":
|
43 |
-
"image_size":
|
44 |
"intermediate_size": 512,
|
45 |
"layer_norm_eps": 1e-06,
|
46 |
"model_type": "siglip_vision_model",
|
47 |
"num_attention_heads": 4,
|
48 |
"num_channels": 3,
|
49 |
"num_hidden_layers": 4,
|
50 |
-
"patch_size":
|
51 |
}
|
52 |
}
|
|
|
6 |
"eoi_token_index": 256000,
|
7 |
"image_token_index": 262144,
|
8 |
"initializer_range": 0.02,
|
9 |
+
"mm_tokens_per_image": 64,
|
10 |
"model_type": "gemma3",
|
11 |
"text_config": {
|
12 |
"attention_bias": false,
|
|
|
16 |
"final_logit_softcapping": null,
|
17 |
"head_dim": 256,
|
18 |
"hidden_activation": "gelu_pytorch_tanh",
|
19 |
+
"hidden_size": 128,
|
20 |
"initializer_range": 0.02,
|
21 |
"intermediate_size": 512,
|
22 |
"max_position_embeddings": 131072,
|
23 |
"model_type": "gemma3_text",
|
24 |
"num_attention_heads": 4,
|
25 |
+
"num_hidden_layers": 8,
|
26 |
"num_key_value_heads": 2,
|
27 |
"query_pre_attn_scalar": 256,
|
28 |
"rms_norm_eps": 1e-06,
|
|
|
39 |
"vision_config": {
|
40 |
"attention_dropout": 0.1,
|
41 |
"hidden_act": "gelu_pytorch_tanh",
|
42 |
+
"hidden_size": 128,
|
43 |
+
"image_size": 32,
|
44 |
"intermediate_size": 512,
|
45 |
"layer_norm_eps": 1e-06,
|
46 |
"model_type": "siglip_vision_model",
|
47 |
"num_attention_heads": 4,
|
48 |
"num_channels": 3,
|
49 |
"num_hidden_layers": 4,
|
50 |
+
"patch_size": 2
|
51 |
}
|
52 |
}
|
model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:53c279eae6c9dab986b2b03645a0cc9d6ce0bd54af0552d913e805a2bfc12946
|
3 |
+
size 157351984
|