Hugging Face's logo

soob3123
/
amoral-gemma3-12B-v2-qat

Text Generation
Transformers
Safetensors
English
gemma3
image-text-to-text
text-generation-inference
analytical-tasks
bias-neutralization
uncensored
conversational
Model card Files Community
soob3123 commited on
Commit
b73f3b6
·
verified ·
1 Parent(s): b926184

Update config.json

Browse files
Files changed (1) hide show
  1. config.json +56 -31
config.json CHANGED
@@ -1,37 +1,62 @@
1
  {
2
  "architectures": [
3
- "Gemma3ForCausalLM"
4
  ],
5
- "attention_bias": false,
6
- "attention_dropout": 0.0,
7
- "attn_logit_softcapping": null,
8
- "bos_token_id": 2,
9
- "cache_implementation": "hybrid",
10
- "eos_token_id": 1,
11
- "final_logit_softcapping": null,
12
- "head_dim": 256,
13
- "hidden_activation": "gelu_pytorch_tanh",
14
- "hidden_size": 3840,
15
  "initializer_range": 0.02,
16
- "intermediate_size": 15360,
17
- "max_position_embeddings": 131072,
18
- "model_type": "gemma3_text",
19
- "num_attention_heads": 16,
20
- "num_hidden_layers": 48,
21
- "num_key_value_heads": 8,
22
- "pad_token_id": 0,
23
- "query_pre_attn_scalar": 256,
24
- "rms_norm_eps": 1e-06,
25
- "rope_local_base_freq": 10000,
26
- "rope_scaling": {
27
- "factor": 8.0,
28
- "rope_type": "linear"
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
29
  },
30
- "rope_theta": 1000000,
31
- "sliding_window": 1024,
32
- "sliding_window_pattern": 6,
33
  "torch_dtype": "bfloat16",
34
- "transformers_version": "4.50.0.dev0",
35
- "use_cache": true,
36
- "vocab_size": 262208
37
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  {
2
  "architectures": [
3
+ "Gemma3ForConditionalGeneration"
4
  ],
5
+ "boi_token_index": 255999,
6
+ "eoi_token_index": 256000,
7
+ "eos_token_id": [
8
+ 1,
9
+ 106
10
+ ],
11
+ "image_token_index": 262144,
 
 
 
12
  "initializer_range": 0.02,
13
+ "mm_tokens_per_image": 256,
14
+ "model_type": "gemma3",
15
+ "text_config": {
16
+ "attention_bias": false,
17
+ "attention_dropout": 0.0,
18
+ "attn_logit_softcapping": null,
19
+ "cache_implementation": "hybrid",
20
+ "final_logit_softcapping": null,
21
+ "head_dim": 256,
22
+ "hidden_activation": "gelu_pytorch_tanh",
23
+ "hidden_size": 3840,
24
+ "initializer_range": 0.02,
25
+ "intermediate_size": 15360,
26
+ "max_position_embeddings": 131072,
27
+ "model_type": "gemma3_text",
28
+ "num_attention_heads": 16,
29
+ "num_hidden_layers": 48,
30
+ "num_key_value_heads": 8,
31
+ "query_pre_attn_scalar": 256,
32
+ "rms_norm_eps": 1e-06,
33
+ "rope_local_base_freq": 10000,
34
+ "rope_scaling": {
35
+ "factor": 8.0,
36
+ "rope_type": "linear"
37
+ },
38
+ "rope_theta": 1000000,
39
+ "sliding_window": 1024,
40
+ "sliding_window_pattern": 6,
41
+ "torch_dtype": "bfloat16",
42
+ "use_cache": true,
43
+ "vocab_size": 262208
44
  },
 
 
 
45
  "torch_dtype": "bfloat16",
46
+ "transformers_version": "4.52.0.dev0",
47
+ "vision_config": {
48
+ "attention_dropout": 0.0,
49
+ "hidden_act": "gelu_pytorch_tanh",
50
+ "hidden_size": 1152,
51
+ "image_size": 896,
52
+ "intermediate_size": 4304,
53
+ "layer_norm_eps": 1e-06,
54
+ "model_type": "siglip_vision_model",
55
+ "num_attention_heads": 16,
56
+ "num_channels": 3,
57
+ "num_hidden_layers": 27,
58
+ "patch_size": 14,
59
+ "torch_dtype": "bfloat16",
60
+ "vision_use_head": false
61
+ }
62
+ }