Hugging Face's logo

lthn
/
lemer

Image-Text-to-Text
GGUF
gemma4
lemma
llama.cpp
ollama
multimodal
vision
audio
on-device
conversational
Eval Results
Model card Files
xet
 Community
lthn commited on
Commit
cdeb430
·
verified ·
1 Parent(s): c495358

Upload config.json with huggingface_hub

Browse files
Files changed (1) hide show
  1. config.json +150 -151
config.json CHANGED
@@ -1,158 +1,157 @@
1
  {
2
- "architectures": [
3
- "Gemma4ForConditionalGeneration"
4
- ],
5
- "audio_config": {
6
- "_name_or_path": "",
7
- "architectures": null,
8
- "attention_chunk_size": 12,
9
- "attention_context_left": 13,
10
- "attention_context_right": 0,
11
- "attention_invalid_logits_value": -1000000000.0,
12
- "attention_logit_cap": 50.0,
13
- "chunk_size_feed_forward": 0,
14
- "conv_kernel_size": 5,
15
- "dtype": "bfloat16",
16
- "gradient_clipping": 10000000000.0,
17
- "hidden_act": "silu",
18
- "hidden_size": 1024,
19
- "id2label": {
20
- "0": "LABEL_0",
21
- "1": "LABEL_1"
22
- },
23
- "initializer_range": 0.02,
24
- "is_encoder_decoder": false,
25
- "label2id": {
26
- "LABEL_0": 0,
27
- "LABEL_1": 1
28
- },
29
- "model_type": "gemma4_audio",
30
- "num_attention_heads": 8,
31
- "num_hidden_layers": 12,
32
- "output_attentions": false,
33
- "output_hidden_states": false,
34
- "output_proj_dims": 1536,
35
- "problem_type": null,
36
- "residual_weight": 0.5,
37
- "return_dict": true,
38
- "rms_norm_eps": 1e-06,
39
- "subsampling_conv_channels": [
40
- 128,
41
- 32
42
- ],
43
- "use_clipped_linears": true
44
- },
45
- "audio_token_id": 258881,
46
- "boa_token_id": 256000,
47
- "boi_token_id": 255999,
48
  "dtype": "bfloat16",
49
- "eoa_token_id": 258883,
50
- "eoa_token_index": 258883,
51
- "eoi_token_id": 258882,
52
- "eos_token_id": [
53
- 1,
54
- 106,
55
- 50
56
- ],
57
- "image_token_id": 258880,
58
- "initializer_range": 0.02,
59
- "model_type": "gemma4",
60
- "quantization": {
61
- "group_size": 64,
62
- "bits": 4,
63
- "mode": "affine"
64
  },
65
- "quantization_config": {
66
- "group_size": 64,
67
- "bits": 4,
68
- "mode": "affine"
 
69
  },
70
- "text_config": {
71
- "attention_bias": false,
72
- "attention_dropout": 0.0,
73
- "attention_k_eq_v": false,
74
- "bos_token_id": 2,
75
- "dtype": "bfloat16",
76
- "enable_moe_block": false,
77
- "eos_token_id": 1,
78
- "expert_intermediate_size": null,
79
- "final_logit_softcapping": 30.0,
80
- "global_head_dim": 512,
81
- "head_dim": 256,
82
- "hidden_activation": "gelu_pytorch_tanh",
83
- "hidden_size": 1536,
84
- "hidden_size_per_layer_input": 256,
85
- "initializer_range": 0.02,
86
- "intermediate_size": 6144,
87
- "layer_types": [
88
- "sliding_attention",
89
- "sliding_attention",
90
- "sliding_attention",
91
- "sliding_attention",
92
- "full_attention",
93
- "sliding_attention",
94
- "sliding_attention",
95
- "sliding_attention",
96
- "sliding_attention",
97
- "full_attention",
98
- "sliding_attention",
99
- "sliding_attention",
100
- "sliding_attention",
101
- "sliding_attention",
102
- "full_attention",
103
- "sliding_attention",
104
- "sliding_attention",
105
- "sliding_attention",
106
- "sliding_attention",
107
- "full_attention",
108
- "sliding_attention",
109
- "sliding_attention",
110
- "sliding_attention",
111
- "sliding_attention",
112
- "full_attention",
113
- "sliding_attention",
114
- "sliding_attention",
115
- "sliding_attention",
116
- "sliding_attention",
117
- "full_attention",
118
- "sliding_attention",
119
- "sliding_attention",
120
- "sliding_attention",
121
- "sliding_attention",
122
- "full_attention"
123
- ],
124
- "max_position_embeddings": 131072,
125
- "model_type": "gemma4_text",
126
- "num_attention_heads": 8,
127
- "num_experts": null,
128
- "num_global_key_value_heads": null,
129
- "num_hidden_layers": 35,
130
- "num_key_value_heads": 1,
131
- "num_kv_shared_layers": 20,
132
- "pad_token_id": 0,
133
- "rms_norm_eps": 1e-06,
134
- "rope_parameters": {
135
- "full_attention": {
136
- "partial_rotary_factor": 0.25,
137
- "rope_theta": 1000000.0,
138
- "rope_type": "proportional"
139
- },
140
- "sliding_attention": {
141
- "rope_theta": 10000.0,
142
- "rope_type": "default"
143
- }
144
- },
145
- "sliding_window": 512,
146
- "tie_word_embeddings": true,
147
- "top_k_experts": null,
148
- "use_bidirectional_attention": null,
149
- "use_cache": true,
150
- "use_double_wide_mlp": true,
151
- "vocab_size": 262144,
152
- "vocab_size_per_layer_input": 262144
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
153
  },
 
154
  "tie_word_embeddings": true,
155
- "transformers_version": "5.5.0.dev0",
156
- "video_token_id": 258884,
157
- "vision_soft_tokens_per_image": 280
 
 
 
 
 
 
 
 
158
  }
 
1
  {
2
+ "architectures": [
3
+ "Gemma4ForConditionalGeneration"
4
+ ],
5
+ "audio_config": {
6
+ "_name_or_path": "",
7
+ "architectures": null,
8
+ "attention_chunk_size": 12,
9
+ "attention_context_left": 13,
10
+ "attention_context_right": 0,
11
+ "attention_invalid_logits_value": -1000000000.0,
12
+ "attention_logit_cap": 50.0,
13
+ "chunk_size_feed_forward": 0,
14
+ "conv_kernel_size": 5,
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
15
  "dtype": "bfloat16",
16
+ "gradient_clipping": 10000000000.0,
17
+ "hidden_act": "silu",
18
+ "hidden_size": 1024,
19
+ "id2label": {
20
+ "0": "LABEL_0",
21
+ "1": "LABEL_1"
 
 
 
 
 
 
 
 
 
22
  },
23
+ "initializer_range": 0.02,
24
+ "is_encoder_decoder": false,
25
+ "label2id": {
26
+ "LABEL_0": 0,
27
+ "LABEL_1": 1
28
  },
29
+ "model_type": "gemma4_audio",
30
+ "num_attention_heads": 8,
31
+ "num_hidden_layers": 12,
32
+ "output_attentions": false,
33
+ "output_hidden_states": false,
34
+ "output_proj_dims": 1536,
35
+ "problem_type": null,
36
+ "residual_weight": 0.5,
37
+ "return_dict": true,
38
+ "rms_norm_eps": 1e-06,
39
+ "subsampling_conv_channels": [
40
+ 128,
41
+ 32
42
+ ],
43
+ "use_clipped_linears": true
44
+ },
45
+ "audio_token_id": 258881,
46
+ "boa_token_id": 256000,
47
+ "boi_token_id": 255999,
48
+ "dtype": "bfloat16",
49
+ "eoa_token_id": 258883,
50
+ "eoa_token_index": 258883,
51
+ "eoi_token_id": 258882,
52
+ "eos_token_id": [
53
+ 1,
54
+ 106
55
+ ],
56
+ "image_token_id": 258880,
57
+ "initializer_range": 0.02,
58
+ "model_type": "gemma4",
59
+ "quantization": {
60
+ "group_size": 64,
61
+ "bits": 4,
62
+ "mode": "affine"
63
+ },
64
+ "quantization_config": {
65
+ "group_size": 64,
66
+ "bits": 4,
67
+ "mode": "affine"
68
+ },
69
+ "text_config": {
70
+ "attention_bias": false,
71
+ "attention_dropout": 0.0,
72
+ "attention_k_eq_v": false,
73
+ "bos_token_id": 2,
74
+ "dtype": "bfloat16",
75
+ "enable_moe_block": false,
76
+ "eos_token_id": 1,
77
+ "expert_intermediate_size": null,
78
+ "final_logit_softcapping": 30.0,
79
+ "global_head_dim": 512,
80
+ "head_dim": 256,
81
+ "hidden_activation": "gelu_pytorch_tanh",
82
+ "hidden_size": 1536,
83
+ "hidden_size_per_layer_input": 256,
84
+ "initializer_range": 0.02,
85
+ "intermediate_size": 6144,
86
+ "layer_types": [
87
+ "sliding_attention",
88
+ "sliding_attention",
89
+ "sliding_attention",
90
+ "sliding_attention",
91
+ "full_attention",
92
+ "sliding_attention",
93
+ "sliding_attention",
94
+ "sliding_attention",
95
+ "sliding_attention",
96
+ "full_attention",
97
+ "sliding_attention",
98
+ "sliding_attention",
99
+ "sliding_attention",
100
+ "sliding_attention",
101
+ "full_attention",
102
+ "sliding_attention",
103
+ "sliding_attention",
104
+ "sliding_attention",
105
+ "sliding_attention",
106
+ "full_attention",
107
+ "sliding_attention",
108
+ "sliding_attention",
109
+ "sliding_attention",
110
+ "sliding_attention",
111
+ "full_attention",
112
+ "sliding_attention",
113
+ "sliding_attention",
114
+ "sliding_attention",
115
+ "sliding_attention",
116
+ "full_attention",
117
+ "sliding_attention",
118
+ "sliding_attention",
119
+ "sliding_attention",
120
+ "sliding_attention",
121
+ "full_attention"
122
+ ],
123
+ "max_position_embeddings": 131072,
124
+ "model_type": "gemma4_text",
125
+ "num_attention_heads": 8,
126
+ "num_experts": null,
127
+ "num_global_key_value_heads": null,
128
+ "num_hidden_layers": 35,
129
+ "num_key_value_heads": 1,
130
+ "num_kv_shared_layers": 20,
131
+ "pad_token_id": 0,
132
+ "rms_norm_eps": 1e-06,
133
+ "rope_parameters": {
134
+ "full_attention": {
135
+ "partial_rotary_factor": 0.25,
136
+ "rope_theta": 1000000.0,
137
+ "rope_type": "proportional"
138
+ },
139
+ "sliding_attention": {
140
+ "rope_theta": 10000.0,
141
+ "rope_type": "default"
142
+ }
143
  },
144
+ "sliding_window": 512,
145
  "tie_word_embeddings": true,
146
+ "top_k_experts": null,
147
+ "use_bidirectional_attention": null,
148
+ "use_cache": true,
149
+ "use_double_wide_mlp": true,
150
+ "vocab_size": 262144,
151
+ "vocab_size_per_layer_input": 262144
152
+ },
153
+ "tie_word_embeddings": true,
154
+ "transformers_version": "5.5.0.dev0",
155
+ "video_token_id": 258884,
156
+ "vision_soft_tokens_per_image": 280
157
  }