Upload config.json with huggingface_hub
Browse files- config.json +181 -0
config.json
ADDED
|
@@ -0,0 +1,181 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"Gemma3nForConditionalGeneration"
|
| 4 |
+
],
|
| 5 |
+
"audio_config": {
|
| 6 |
+
"conf_attention_chunk_size": 12,
|
| 7 |
+
"conf_attention_context_left": 13,
|
| 8 |
+
"conf_attention_context_right": 0,
|
| 9 |
+
"conf_attention_logit_cap": 50.0,
|
| 10 |
+
"conf_conv_kernel_size": 5,
|
| 11 |
+
"conf_num_attention_heads": 8,
|
| 12 |
+
"conf_num_hidden_layers": 12,
|
| 13 |
+
"conf_positional_bias_size": 256,
|
| 14 |
+
"conf_reduction_factor": 4,
|
| 15 |
+
"conf_residual_weight": 0.5,
|
| 16 |
+
"gradient_clipping": 10000000000.0,
|
| 17 |
+
"hidden_size": 1536,
|
| 18 |
+
"input_feat_size": 128,
|
| 19 |
+
"model_type": "gemma3n_audio",
|
| 20 |
+
"rms_norm_eps": 1e-06,
|
| 21 |
+
"sscp_conv_channel_size": [
|
| 22 |
+
128,
|
| 23 |
+
32
|
| 24 |
+
],
|
| 25 |
+
"sscp_conv_eps": 0.001,
|
| 26 |
+
"sscp_conv_kernel_size": [
|
| 27 |
+
[
|
| 28 |
+
3,
|
| 29 |
+
3
|
| 30 |
+
],
|
| 31 |
+
[
|
| 32 |
+
3,
|
| 33 |
+
3
|
| 34 |
+
]
|
| 35 |
+
],
|
| 36 |
+
"sscp_conv_stride_size": [
|
| 37 |
+
[
|
| 38 |
+
2,
|
| 39 |
+
2
|
| 40 |
+
],
|
| 41 |
+
[
|
| 42 |
+
2,
|
| 43 |
+
2
|
| 44 |
+
]
|
| 45 |
+
],
|
| 46 |
+
"torch_dtype": "bfloat16",
|
| 47 |
+
"vocab_offset": 262272,
|
| 48 |
+
"vocab_size": 128
|
| 49 |
+
},
|
| 50 |
+
"audio_soft_tokens_per_image": 188,
|
| 51 |
+
"audio_token_id": 262273,
|
| 52 |
+
"boa_token_id": 256000,
|
| 53 |
+
"boi_token_id": 255999,
|
| 54 |
+
"eoa_token_id": 262272,
|
| 55 |
+
"eoi_token_id": 262144,
|
| 56 |
+
"eos_token_id": [
|
| 57 |
+
1,
|
| 58 |
+
106
|
| 59 |
+
],
|
| 60 |
+
"image_token_id": 262145,
|
| 61 |
+
"initializer_range": 0.02,
|
| 62 |
+
"model_type": "gemma3n",
|
| 63 |
+
"quantization": {
|
| 64 |
+
"group_size": 64,
|
| 65 |
+
"bits": 8
|
| 66 |
+
},
|
| 67 |
+
"quantization_config": {
|
| 68 |
+
"group_size": 64,
|
| 69 |
+
"bits": 8
|
| 70 |
+
},
|
| 71 |
+
"text_config": {
|
| 72 |
+
"activation_sparsity_pattern": [
|
| 73 |
+
0.95,
|
| 74 |
+
0.95,
|
| 75 |
+
0.95,
|
| 76 |
+
0.95,
|
| 77 |
+
0.95,
|
| 78 |
+
0.95,
|
| 79 |
+
0.95,
|
| 80 |
+
0.95,
|
| 81 |
+
0.95,
|
| 82 |
+
0.95,
|
| 83 |
+
0.0,
|
| 84 |
+
0.0,
|
| 85 |
+
0.0,
|
| 86 |
+
0.0,
|
| 87 |
+
0.0,
|
| 88 |
+
0.0,
|
| 89 |
+
0.0,
|
| 90 |
+
0.0,
|
| 91 |
+
0.0,
|
| 92 |
+
0.0,
|
| 93 |
+
0.0,
|
| 94 |
+
0.0,
|
| 95 |
+
0.0,
|
| 96 |
+
0.0,
|
| 97 |
+
0.0,
|
| 98 |
+
0.0,
|
| 99 |
+
0.0,
|
| 100 |
+
0.0,
|
| 101 |
+
0.0,
|
| 102 |
+
0.0,
|
| 103 |
+
0.0,
|
| 104 |
+
0.0,
|
| 105 |
+
0.0,
|
| 106 |
+
0.0,
|
| 107 |
+
0.0
|
| 108 |
+
],
|
| 109 |
+
"altup_active_idx": 0,
|
| 110 |
+
"altup_coef_clip": 120.0,
|
| 111 |
+
"altup_correct_scale": true,
|
| 112 |
+
"altup_lr_multiplier": 1.0,
|
| 113 |
+
"altup_num_inputs": 4,
|
| 114 |
+
"attention_bias": false,
|
| 115 |
+
"attention_dropout": 0.0,
|
| 116 |
+
"final_logit_softcapping": 30.0,
|
| 117 |
+
"head_dim": 256,
|
| 118 |
+
"hidden_activation": "gelu_pytorch_tanh",
|
| 119 |
+
"hidden_size": 2048,
|
| 120 |
+
"hidden_size_per_layer_input": 256,
|
| 121 |
+
"initializer_range": 0.02,
|
| 122 |
+
"intermediate_size": 16384,
|
| 123 |
+
"laurel_rank": 64,
|
| 124 |
+
"layer_types": [
|
| 125 |
+
"sliding_attention",
|
| 126 |
+
"sliding_attention",
|
| 127 |
+
"sliding_attention",
|
| 128 |
+
"sliding_attention",
|
| 129 |
+
"full_attention",
|
| 130 |
+
"sliding_attention",
|
| 131 |
+
"sliding_attention",
|
| 132 |
+
"sliding_attention",
|
| 133 |
+
"sliding_attention",
|
| 134 |
+
"full_attention",
|
| 135 |
+
"sliding_attention",
|
| 136 |
+
"sliding_attention",
|
| 137 |
+
"sliding_attention",
|
| 138 |
+
"sliding_attention",
|
| 139 |
+
"full_attention",
|
| 140 |
+
"sliding_attention",
|
| 141 |
+
"sliding_attention",
|
| 142 |
+
"sliding_attention",
|
| 143 |
+
"sliding_attention",
|
| 144 |
+
"full_attention",
|
| 145 |
+
"sliding_attention",
|
| 146 |
+
"sliding_attention",
|
| 147 |
+
"sliding_attention",
|
| 148 |
+
"sliding_attention",
|
| 149 |
+
"full_attention",
|
| 150 |
+
"sliding_attention",
|
| 151 |
+
"sliding_attention",
|
| 152 |
+
"sliding_attention",
|
| 153 |
+
"sliding_attention",
|
| 154 |
+
"full_attention",
|
| 155 |
+
"sliding_attention",
|
| 156 |
+
"sliding_attention",
|
| 157 |
+
"sliding_attention",
|
| 158 |
+
"sliding_attention",
|
| 159 |
+
"full_attention"
|
| 160 |
+
],
|
| 161 |
+
"max_position_embeddings": 32768,
|
| 162 |
+
"model_type": "gemma3n_text",
|
| 163 |
+
"num_attention_heads": 8,
|
| 164 |
+
"num_hidden_layers": 35,
|
| 165 |
+
"num_key_value_heads": 2,
|
| 166 |
+
"num_kv_shared_layers": 15,
|
| 167 |
+
"query_pre_attn_scalar": 256,
|
| 168 |
+
"rms_norm_eps": 1e-06,
|
| 169 |
+
"rope_local_base_freq": 10000.0,
|
| 170 |
+
"rope_scaling": null,
|
| 171 |
+
"rope_theta": 1000000.0,
|
| 172 |
+
"sliding_window": 512,
|
| 173 |
+
"torch_dtype": "bfloat16",
|
| 174 |
+
"use_cache": true,
|
| 175 |
+
"vocab_size": 262400,
|
| 176 |
+
"vocab_size_per_layer_input": 262144
|
| 177 |
+
},
|
| 178 |
+
"torch_dtype": "bfloat16",
|
| 179 |
+
"transformers_version": "4.53.0.dev0",
|
| 180 |
+
"vision_soft_tokens_per_image": 256
|
| 181 |
+
}
|