Hugging Face's logo

zoebchhatriwala
/
Zesyde-2B-Multimodal-Router

Feature Extraction
PyTorch
Safetensors
ONNX
gemma4
multimodal
decision-router
agent-routing
gemma
multilingual
Model card Files
xet
 Community
zoebchhatriwala commited on
Commit
8e7be00
·
verified ·
1 Parent(s): f0aaacc

Upload config.json with huggingface_hub

Browse files
Files changed (1) hide show
  1. config.json +176 -0
config.json ADDED
@@ -0,0 +1,176 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Gemma4ForConditionalGeneration"
4
+ ],
5
+ "audio_config": {
6
+ "attention_chunk_size": 12,
7
+ "attention_context_left": 13,
8
+ "attention_context_right": 0,
9
+ "attention_invalid_logits_value": -1000000000.0,
10
+ "attention_logit_cap": 50.0,
11
+ "conv_kernel_size": 5,
12
+ "dtype": "bfloat16",
13
+ "gradient_clipping": 10000000000.0,
14
+ "hidden_act": "silu",
15
+ "hidden_size": 1024,
16
+ "initializer_range": 0.02,
17
+ "model_type": "gemma4_audio",
18
+ "num_attention_heads": 8,
19
+ "num_hidden_layers": 12,
20
+ "output_proj_dims": 1536,
21
+ "residual_weight": 0.5,
22
+ "rms_norm_eps": 1e-06,
23
+ "subsampling_conv_channels": [
24
+ 128,
25
+ 32
26
+ ],
27
+ "use_clipped_linears": true
28
+ },
29
+ "audio_token_id": 258881,
30
+ "boa_token_id": 256000,
31
+ "boi_token_id": 255999,
32
+ "dtype": "bfloat16",
33
+ "eoa_token_id": 258883,
34
+ "eoa_token_index": 258883,
35
+ "eoi_token_id": 258882,
36
+ "image_token_id": 258880,
37
+ "initializer_range": 0.02,
38
+ "model_type": "gemma4",
39
+ "text_config": {
40
+ "attention_bias": false,
41
+ "attention_dropout": 0.0,
42
+ "attention_k_eq_v": false,
43
+ "bos_token_id": 2,
44
+ "dtype": "bfloat16",
45
+ "enable_moe_block": false,
46
+ "eos_token_id": 1,
47
+ "expert_intermediate_size": null,
48
+ "final_logit_softcapping": 30.0,
49
+ "head_dim": 256,
50
+ "hidden_activation": "gelu_pytorch_tanh",
51
+ "hidden_size": 1536,
52
+ "hidden_size_per_layer_input": 256,
53
+ "initializer_range": 0.02,
54
+ "intermediate_size": 6144,
55
+ "layer_types": [
56
+ "sliding_attention",
57
+ "sliding_attention",
58
+ "sliding_attention",
59
+ "sliding_attention",
60
+ "full_attention",
61
+ "sliding_attention",
62
+ "sliding_attention",
63
+ "sliding_attention",
64
+ "sliding_attention",
65
+ "full_attention",
66
+ "sliding_attention",
67
+ "sliding_attention",
68
+ "sliding_attention",
69
+ "sliding_attention",
70
+ "full_attention",
71
+ "sliding_attention",
72
+ "sliding_attention",
73
+ "sliding_attention",
74
+ "sliding_attention",
75
+ "full_attention",
76
+ "sliding_attention",
77
+ "sliding_attention",
78
+ "sliding_attention",
79
+ "sliding_attention",
80
+ "full_attention",
81
+ "sliding_attention",
82
+ "sliding_attention",
83
+ "sliding_attention",
84
+ "sliding_attention",
85
+ "full_attention",
86
+ "sliding_attention",
87
+ "sliding_attention",
88
+ "sliding_attention",
89
+ "sliding_attention",
90
+ "full_attention"
91
+ ],
92
+ "max_position_embeddings": 131072,
93
+ "model_type": "gemma4_text",
94
+ "moe_intermediate_size": null,
95
+ "num_attention_heads": 8,
96
+ "num_experts": null,
97
+ "num_hidden_layers": 35,
98
+ "num_key_value_heads": 1,
99
+ "num_kv_shared_layers": 20,
100
+ "pad_token_id": 0,
101
+ "per_layer_config": {
102
+ "04": {
103
+ "head_dim": 512
104
+ },
105
+ "09": {
106
+ "head_dim": 512
107
+ },
108
+ "14": {
109
+ "head_dim": 512
110
+ },
111
+ "19": {
112
+ "head_dim": 512
113
+ },
114
+ "24": {
115
+ "head_dim": 512
116
+ },
117
+ "29": {
118
+ "head_dim": 512
119
+ },
120
+ "34": {
121
+ "head_dim": 512
122
+ }
123
+ },
124
+ "rms_norm_eps": 1e-06,
125
+ "rope_parameters": {
126
+ "full_attention": {
127
+ "partial_rotary_factor": 0.25,
128
+ "rope_theta": 1000000.0,
129
+ "rope_type": "proportional"
130
+ },
131
+ "sliding_attention": {
132
+ "rope_theta": 10000.0,
133
+ "rope_type": "default"
134
+ }
135
+ },
136
+ "sliding_window": 512,
137
+ "tie_word_embeddings": true,
138
+ "top_k_experts": null,
139
+ "use_bidirectional_attention": null,
140
+ "use_cache": false,
141
+ "use_double_wide_mlp": true,
142
+ "vocab_size": 262144,
143
+ "vocab_size_per_layer_input": 262144
144
+ },
145
+ "tie_word_embeddings": true,
146
+ "transformers_version": "5.20.2",
147
+ "video_token_id": 258884,
148
+ "vision_config": {
149
+ "attention_bias": false,
150
+ "attention_dropout": 0.0,
151
+ "default_output_length": 280,
152
+ "dtype": "bfloat16",
153
+ "global_head_dim": 64,
154
+ "head_dim": 64,
155
+ "hidden_activation": "gelu_pytorch_tanh",
156
+ "hidden_size": 768,
157
+ "initializer_range": 0.02,
158
+ "intermediate_size": 3072,
159
+ "max_position_embeddings": 131072,
160
+ "model_type": "gemma4_vision",
161
+ "num_attention_heads": 12,
162
+ "num_hidden_layers": 16,
163
+ "num_key_value_heads": 12,
164
+ "patch_size": 16,
165
+ "pooling_kernel_size": 3,
166
+ "position_embedding_size": 10240,
167
+ "rms_norm_eps": 1e-06,
168
+ "rope_parameters": {
169
+ "rope_theta": 100.0,
170
+ "rope_type": "axial"
171
+ },
172
+ "standardize": false,
173
+ "use_clipped_linears": true
174
+ },
175
+ "vision_soft_tokens_per_image": 280
176
+ }