Upload config.json with huggingface_hub
Browse files- config.json +246 -0
config.json
ADDED
|
@@ -0,0 +1,246 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"Lfm2MoeForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"auto_map": {
|
| 6 |
+
"AutoConfig": "configuration_lfm2_moe.Lfm2MoeConfig",
|
| 7 |
+
"AutoModelForCausalLM": "modeling_lfm2_moe.Lfm2MoeForCausalLM"
|
| 8 |
+
},
|
| 9 |
+
"bos_token_id": 1,
|
| 10 |
+
"conv_L_cache": 3,
|
| 11 |
+
"conv_bias": false,
|
| 12 |
+
"dtype": "bfloat16",
|
| 13 |
+
"eos_token_id": 7,
|
| 14 |
+
"hidden_size": 2048,
|
| 15 |
+
"intermediate_size": 7168,
|
| 16 |
+
"layer_types": [
|
| 17 |
+
"conv",
|
| 18 |
+
"conv",
|
| 19 |
+
"full_attention",
|
| 20 |
+
"conv",
|
| 21 |
+
"conv",
|
| 22 |
+
"conv",
|
| 23 |
+
"full_attention",
|
| 24 |
+
"conv",
|
| 25 |
+
"conv",
|
| 26 |
+
"conv",
|
| 27 |
+
"full_attention",
|
| 28 |
+
"conv",
|
| 29 |
+
"conv",
|
| 30 |
+
"conv",
|
| 31 |
+
"full_attention",
|
| 32 |
+
"conv",
|
| 33 |
+
"conv",
|
| 34 |
+
"conv",
|
| 35 |
+
"full_attention",
|
| 36 |
+
"conv",
|
| 37 |
+
"conv",
|
| 38 |
+
"full_attention",
|
| 39 |
+
"conv",
|
| 40 |
+
"conv"
|
| 41 |
+
],
|
| 42 |
+
"max_position_embeddings": 128000,
|
| 43 |
+
"model_type": "lfm2_moe",
|
| 44 |
+
"moe_intermediate_size": 1792,
|
| 45 |
+
"norm_eps": 1e-05,
|
| 46 |
+
"norm_topk_prob": true,
|
| 47 |
+
"num_attention_heads": 32,
|
| 48 |
+
"num_dense_layers": 2,
|
| 49 |
+
"num_experts": 32,
|
| 50 |
+
"num_experts_per_tok": 4,
|
| 51 |
+
"num_hidden_layers": 24,
|
| 52 |
+
"num_key_value_heads": 8,
|
| 53 |
+
"pad_token_id": 0,
|
| 54 |
+
"quantization": {
|
| 55 |
+
"group_size": 64,
|
| 56 |
+
"bits": 8,
|
| 57 |
+
"mode": "affine",
|
| 58 |
+
"model.layers.2.feed_forward.gate": {
|
| 59 |
+
"group_size": 64,
|
| 60 |
+
"bits": 8
|
| 61 |
+
},
|
| 62 |
+
"model.layers.3.feed_forward.gate": {
|
| 63 |
+
"group_size": 64,
|
| 64 |
+
"bits": 8
|
| 65 |
+
},
|
| 66 |
+
"model.layers.4.feed_forward.gate": {
|
| 67 |
+
"group_size": 64,
|
| 68 |
+
"bits": 8
|
| 69 |
+
},
|
| 70 |
+
"model.layers.5.feed_forward.gate": {
|
| 71 |
+
"group_size": 64,
|
| 72 |
+
"bits": 8
|
| 73 |
+
},
|
| 74 |
+
"model.layers.6.feed_forward.gate": {
|
| 75 |
+
"group_size": 64,
|
| 76 |
+
"bits": 8
|
| 77 |
+
},
|
| 78 |
+
"model.layers.7.feed_forward.gate": {
|
| 79 |
+
"group_size": 64,
|
| 80 |
+
"bits": 8
|
| 81 |
+
},
|
| 82 |
+
"model.layers.8.feed_forward.gate": {
|
| 83 |
+
"group_size": 64,
|
| 84 |
+
"bits": 8
|
| 85 |
+
},
|
| 86 |
+
"model.layers.9.feed_forward.gate": {
|
| 87 |
+
"group_size": 64,
|
| 88 |
+
"bits": 8
|
| 89 |
+
},
|
| 90 |
+
"model.layers.10.feed_forward.gate": {
|
| 91 |
+
"group_size": 64,
|
| 92 |
+
"bits": 8
|
| 93 |
+
},
|
| 94 |
+
"model.layers.11.feed_forward.gate": {
|
| 95 |
+
"group_size": 64,
|
| 96 |
+
"bits": 8
|
| 97 |
+
},
|
| 98 |
+
"model.layers.12.feed_forward.gate": {
|
| 99 |
+
"group_size": 64,
|
| 100 |
+
"bits": 8
|
| 101 |
+
},
|
| 102 |
+
"model.layers.13.feed_forward.gate": {
|
| 103 |
+
"group_size": 64,
|
| 104 |
+
"bits": 8
|
| 105 |
+
},
|
| 106 |
+
"model.layers.14.feed_forward.gate": {
|
| 107 |
+
"group_size": 64,
|
| 108 |
+
"bits": 8
|
| 109 |
+
},
|
| 110 |
+
"model.layers.15.feed_forward.gate": {
|
| 111 |
+
"group_size": 64,
|
| 112 |
+
"bits": 8
|
| 113 |
+
},
|
| 114 |
+
"model.layers.16.feed_forward.gate": {
|
| 115 |
+
"group_size": 64,
|
| 116 |
+
"bits": 8
|
| 117 |
+
},
|
| 118 |
+
"model.layers.17.feed_forward.gate": {
|
| 119 |
+
"group_size": 64,
|
| 120 |
+
"bits": 8
|
| 121 |
+
},
|
| 122 |
+
"model.layers.18.feed_forward.gate": {
|
| 123 |
+
"group_size": 64,
|
| 124 |
+
"bits": 8
|
| 125 |
+
},
|
| 126 |
+
"model.layers.19.feed_forward.gate": {
|
| 127 |
+
"group_size": 64,
|
| 128 |
+
"bits": 8
|
| 129 |
+
},
|
| 130 |
+
"model.layers.20.feed_forward.gate": {
|
| 131 |
+
"group_size": 64,
|
| 132 |
+
"bits": 8
|
| 133 |
+
},
|
| 134 |
+
"model.layers.21.feed_forward.gate": {
|
| 135 |
+
"group_size": 64,
|
| 136 |
+
"bits": 8
|
| 137 |
+
},
|
| 138 |
+
"model.layers.22.feed_forward.gate": {
|
| 139 |
+
"group_size": 64,
|
| 140 |
+
"bits": 8
|
| 141 |
+
},
|
| 142 |
+
"model.layers.23.feed_forward.gate": {
|
| 143 |
+
"group_size": 64,
|
| 144 |
+
"bits": 8
|
| 145 |
+
}
|
| 146 |
+
},
|
| 147 |
+
"quantization_config": {
|
| 148 |
+
"group_size": 64,
|
| 149 |
+
"bits": 8,
|
| 150 |
+
"mode": "affine",
|
| 151 |
+
"model.layers.2.feed_forward.gate": {
|
| 152 |
+
"group_size": 64,
|
| 153 |
+
"bits": 8
|
| 154 |
+
},
|
| 155 |
+
"model.layers.3.feed_forward.gate": {
|
| 156 |
+
"group_size": 64,
|
| 157 |
+
"bits": 8
|
| 158 |
+
},
|
| 159 |
+
"model.layers.4.feed_forward.gate": {
|
| 160 |
+
"group_size": 64,
|
| 161 |
+
"bits": 8
|
| 162 |
+
},
|
| 163 |
+
"model.layers.5.feed_forward.gate": {
|
| 164 |
+
"group_size": 64,
|
| 165 |
+
"bits": 8
|
| 166 |
+
},
|
| 167 |
+
"model.layers.6.feed_forward.gate": {
|
| 168 |
+
"group_size": 64,
|
| 169 |
+
"bits": 8
|
| 170 |
+
},
|
| 171 |
+
"model.layers.7.feed_forward.gate": {
|
| 172 |
+
"group_size": 64,
|
| 173 |
+
"bits": 8
|
| 174 |
+
},
|
| 175 |
+
"model.layers.8.feed_forward.gate": {
|
| 176 |
+
"group_size": 64,
|
| 177 |
+
"bits": 8
|
| 178 |
+
},
|
| 179 |
+
"model.layers.9.feed_forward.gate": {
|
| 180 |
+
"group_size": 64,
|
| 181 |
+
"bits": 8
|
| 182 |
+
},
|
| 183 |
+
"model.layers.10.feed_forward.gate": {
|
| 184 |
+
"group_size": 64,
|
| 185 |
+
"bits": 8
|
| 186 |
+
},
|
| 187 |
+
"model.layers.11.feed_forward.gate": {
|
| 188 |
+
"group_size": 64,
|
| 189 |
+
"bits": 8
|
| 190 |
+
},
|
| 191 |
+
"model.layers.12.feed_forward.gate": {
|
| 192 |
+
"group_size": 64,
|
| 193 |
+
"bits": 8
|
| 194 |
+
},
|
| 195 |
+
"model.layers.13.feed_forward.gate": {
|
| 196 |
+
"group_size": 64,
|
| 197 |
+
"bits": 8
|
| 198 |
+
},
|
| 199 |
+
"model.layers.14.feed_forward.gate": {
|
| 200 |
+
"group_size": 64,
|
| 201 |
+
"bits": 8
|
| 202 |
+
},
|
| 203 |
+
"model.layers.15.feed_forward.gate": {
|
| 204 |
+
"group_size": 64,
|
| 205 |
+
"bits": 8
|
| 206 |
+
},
|
| 207 |
+
"model.layers.16.feed_forward.gate": {
|
| 208 |
+
"group_size": 64,
|
| 209 |
+
"bits": 8
|
| 210 |
+
},
|
| 211 |
+
"model.layers.17.feed_forward.gate": {
|
| 212 |
+
"group_size": 64,
|
| 213 |
+
"bits": 8
|
| 214 |
+
},
|
| 215 |
+
"model.layers.18.feed_forward.gate": {
|
| 216 |
+
"group_size": 64,
|
| 217 |
+
"bits": 8
|
| 218 |
+
},
|
| 219 |
+
"model.layers.19.feed_forward.gate": {
|
| 220 |
+
"group_size": 64,
|
| 221 |
+
"bits": 8
|
| 222 |
+
},
|
| 223 |
+
"model.layers.20.feed_forward.gate": {
|
| 224 |
+
"group_size": 64,
|
| 225 |
+
"bits": 8
|
| 226 |
+
},
|
| 227 |
+
"model.layers.21.feed_forward.gate": {
|
| 228 |
+
"group_size": 64,
|
| 229 |
+
"bits": 8
|
| 230 |
+
},
|
| 231 |
+
"model.layers.22.feed_forward.gate": {
|
| 232 |
+
"group_size": 64,
|
| 233 |
+
"bits": 8
|
| 234 |
+
},
|
| 235 |
+
"model.layers.23.feed_forward.gate": {
|
| 236 |
+
"group_size": 64,
|
| 237 |
+
"bits": 8
|
| 238 |
+
}
|
| 239 |
+
},
|
| 240 |
+
"rope_theta": 1000000.0,
|
| 241 |
+
"routed_scaling_factor": 1.0,
|
| 242 |
+
"transformers_version": "4.56.1",
|
| 243 |
+
"use_cache": true,
|
| 244 |
+
"use_expert_bias": true,
|
| 245 |
+
"vocab_size": 65536
|
| 246 |
+
}
|