add config.json
Browse files- config.json +349 -0
config.json
ADDED
|
@@ -0,0 +1,349 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"Glm5NextForConditionalGeneration"
|
| 4 |
+
],
|
| 5 |
+
"text_config": {
|
| 6 |
+
"attention_bias": false,
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"dtype": "bfloat16",
|
| 9 |
+
"eos_token_id": [
|
| 10 |
+
154820,
|
| 11 |
+
154827,
|
| 12 |
+
154829
|
| 13 |
+
],
|
| 14 |
+
"first_k_dense_replace": 3,
|
| 15 |
+
"hc_eps": 1e-06,
|
| 16 |
+
"hc_mult": 4,
|
| 17 |
+
"hc_sinkhorn_iters": 20,
|
| 18 |
+
"head_dim": 0,
|
| 19 |
+
"hidden_act": "silu",
|
| 20 |
+
"hidden_size": 4096,
|
| 21 |
+
"index_head_dim": 128,
|
| 22 |
+
"index_kpool": 4,
|
| 23 |
+
"index_kpool_always_select_tail": true,
|
| 24 |
+
"index_kpool_compress": true,
|
| 25 |
+
"index_n_heads": 32,
|
| 26 |
+
"index_topk": 2048,
|
| 27 |
+
"index_share_for_mtp_iteration": true,
|
| 28 |
+
"indexer_rope_interleave": true,
|
| 29 |
+
"indexer_types": [
|
| 30 |
+
"full",
|
| 31 |
+
"full",
|
| 32 |
+
"full",
|
| 33 |
+
"full",
|
| 34 |
+
"full",
|
| 35 |
+
"full",
|
| 36 |
+
"full",
|
| 37 |
+
"full",
|
| 38 |
+
"full",
|
| 39 |
+
"full",
|
| 40 |
+
"full",
|
| 41 |
+
"full",
|
| 42 |
+
"full",
|
| 43 |
+
"full",
|
| 44 |
+
"full",
|
| 45 |
+
"full",
|
| 46 |
+
"full",
|
| 47 |
+
"full",
|
| 48 |
+
"full",
|
| 49 |
+
"full",
|
| 50 |
+
"full",
|
| 51 |
+
"full",
|
| 52 |
+
"full",
|
| 53 |
+
"full",
|
| 54 |
+
"full",
|
| 55 |
+
"full",
|
| 56 |
+
"full",
|
| 57 |
+
"full",
|
| 58 |
+
"full",
|
| 59 |
+
"full",
|
| 60 |
+
"full",
|
| 61 |
+
"full",
|
| 62 |
+
"full",
|
| 63 |
+
"full",
|
| 64 |
+
"full",
|
| 65 |
+
"full",
|
| 66 |
+
"full",
|
| 67 |
+
"full",
|
| 68 |
+
"full",
|
| 69 |
+
"full",
|
| 70 |
+
"full",
|
| 71 |
+
"full",
|
| 72 |
+
"full",
|
| 73 |
+
"full",
|
| 74 |
+
"full"
|
| 75 |
+
],
|
| 76 |
+
"initializer_range": 0.02,
|
| 77 |
+
"intermediate_size": 12288,
|
| 78 |
+
"kv_lora_rank": 512,
|
| 79 |
+
"layer_types": [
|
| 80 |
+
"linear_attention",
|
| 81 |
+
"linear_attention",
|
| 82 |
+
"linear_attention",
|
| 83 |
+
"deepseek_sparse_attention",
|
| 84 |
+
"linear_attention",
|
| 85 |
+
"linear_attention",
|
| 86 |
+
"linear_attention",
|
| 87 |
+
"deepseek_sparse_attention",
|
| 88 |
+
"linear_attention",
|
| 89 |
+
"linear_attention",
|
| 90 |
+
"linear_attention",
|
| 91 |
+
"deepseek_sparse_attention",
|
| 92 |
+
"linear_attention",
|
| 93 |
+
"linear_attention",
|
| 94 |
+
"linear_attention",
|
| 95 |
+
"deepseek_sparse_attention",
|
| 96 |
+
"linear_attention",
|
| 97 |
+
"linear_attention",
|
| 98 |
+
"linear_attention",
|
| 99 |
+
"deepseek_sparse_attention",
|
| 100 |
+
"linear_attention",
|
| 101 |
+
"linear_attention",
|
| 102 |
+
"linear_attention",
|
| 103 |
+
"deepseek_sparse_attention",
|
| 104 |
+
"linear_attention",
|
| 105 |
+
"linear_attention",
|
| 106 |
+
"linear_attention",
|
| 107 |
+
"deepseek_sparse_attention",
|
| 108 |
+
"linear_attention",
|
| 109 |
+
"linear_attention",
|
| 110 |
+
"linear_attention",
|
| 111 |
+
"deepseek_sparse_attention",
|
| 112 |
+
"linear_attention",
|
| 113 |
+
"linear_attention",
|
| 114 |
+
"linear_attention",
|
| 115 |
+
"deepseek_sparse_attention",
|
| 116 |
+
"linear_attention",
|
| 117 |
+
"linear_attention",
|
| 118 |
+
"linear_attention",
|
| 119 |
+
"deepseek_sparse_attention",
|
| 120 |
+
"linear_attention",
|
| 121 |
+
"linear_attention",
|
| 122 |
+
"linear_attention",
|
| 123 |
+
"deepseek_sparse_attention",
|
| 124 |
+
"linear_attention"
|
| 125 |
+
],
|
| 126 |
+
"linear_attn_config": {
|
| 127 |
+
"num_heads": 64,
|
| 128 |
+
"gate_lower_bound": -5.0,
|
| 129 |
+
"head_dim": 128,
|
| 130 |
+
"short_conv_kernel_size": 4,
|
| 131 |
+
"kda_layers": [
|
| 132 |
+
0,
|
| 133 |
+
1,
|
| 134 |
+
2,
|
| 135 |
+
4,
|
| 136 |
+
5,
|
| 137 |
+
6,
|
| 138 |
+
8,
|
| 139 |
+
9,
|
| 140 |
+
10,
|
| 141 |
+
12,
|
| 142 |
+
13,
|
| 143 |
+
14,
|
| 144 |
+
16,
|
| 145 |
+
17,
|
| 146 |
+
18,
|
| 147 |
+
20,
|
| 148 |
+
21,
|
| 149 |
+
22,
|
| 150 |
+
24,
|
| 151 |
+
25,
|
| 152 |
+
26,
|
| 153 |
+
28,
|
| 154 |
+
29,
|
| 155 |
+
30,
|
| 156 |
+
32,
|
| 157 |
+
33,
|
| 158 |
+
34,
|
| 159 |
+
36,
|
| 160 |
+
37,
|
| 161 |
+
38,
|
| 162 |
+
40,
|
| 163 |
+
41,
|
| 164 |
+
42,
|
| 165 |
+
44
|
| 166 |
+
],
|
| 167 |
+
"full_attn_layers": [
|
| 168 |
+
3,
|
| 169 |
+
7,
|
| 170 |
+
11,
|
| 171 |
+
15,
|
| 172 |
+
19,
|
| 173 |
+
23,
|
| 174 |
+
27,
|
| 175 |
+
31,
|
| 176 |
+
35,
|
| 177 |
+
39,
|
| 178 |
+
43
|
| 179 |
+
]
|
| 180 |
+
},
|
| 181 |
+
"max_position_embeddings": 1048576,
|
| 182 |
+
"mhc": true,
|
| 183 |
+
"mla_use_nope": true,
|
| 184 |
+
"mlp_layer_types": [
|
| 185 |
+
"dense",
|
| 186 |
+
"dense",
|
| 187 |
+
"dense",
|
| 188 |
+
"sparse",
|
| 189 |
+
"sparse",
|
| 190 |
+
"sparse",
|
| 191 |
+
"sparse",
|
| 192 |
+
"sparse",
|
| 193 |
+
"sparse",
|
| 194 |
+
"sparse",
|
| 195 |
+
"sparse",
|
| 196 |
+
"sparse",
|
| 197 |
+
"sparse",
|
| 198 |
+
"sparse",
|
| 199 |
+
"sparse",
|
| 200 |
+
"sparse",
|
| 201 |
+
"sparse",
|
| 202 |
+
"sparse",
|
| 203 |
+
"sparse",
|
| 204 |
+
"sparse",
|
| 205 |
+
"sparse",
|
| 206 |
+
"sparse",
|
| 207 |
+
"sparse",
|
| 208 |
+
"sparse",
|
| 209 |
+
"sparse",
|
| 210 |
+
"sparse",
|
| 211 |
+
"sparse",
|
| 212 |
+
"sparse",
|
| 213 |
+
"sparse",
|
| 214 |
+
"sparse",
|
| 215 |
+
"sparse",
|
| 216 |
+
"sparse",
|
| 217 |
+
"sparse",
|
| 218 |
+
"sparse",
|
| 219 |
+
"sparse",
|
| 220 |
+
"sparse",
|
| 221 |
+
"sparse",
|
| 222 |
+
"sparse",
|
| 223 |
+
"sparse",
|
| 224 |
+
"sparse",
|
| 225 |
+
"sparse",
|
| 226 |
+
"sparse",
|
| 227 |
+
"sparse",
|
| 228 |
+
"sparse",
|
| 229 |
+
"sparse"
|
| 230 |
+
],
|
| 231 |
+
"model_type": "glm5_next_text",
|
| 232 |
+
"moe_intermediate_size": 2048,
|
| 233 |
+
"moe_router_dtype": "float32",
|
| 234 |
+
"n_group": 1,
|
| 235 |
+
"n_routed_experts": 288,
|
| 236 |
+
"n_shared_experts": 1,
|
| 237 |
+
"norm_topk_prob": true,
|
| 238 |
+
"num_attention_heads": 64,
|
| 239 |
+
"num_experts_per_tok": 8,
|
| 240 |
+
"num_hidden_layers": 45,
|
| 241 |
+
"num_key_value_heads": 64,
|
| 242 |
+
"num_nextn_predict_layers": 1,
|
| 243 |
+
"output_router_logits": false,
|
| 244 |
+
"pad_token_id": 154820,
|
| 245 |
+
"q_lora_rank": 1536,
|
| 246 |
+
"qk_head_dim": 256,
|
| 247 |
+
"qk_nope_head_dim": 256,
|
| 248 |
+
"qk_rope_head_dim": 0,
|
| 249 |
+
"rms_norm_eps": 1e-05,
|
| 250 |
+
"routed_scaling_factor": 2.5,
|
| 251 |
+
"router_aux_loss_coef": 0.001,
|
| 252 |
+
"scoring_func": "sigmoid",
|
| 253 |
+
"swiglu_limit": 10.0,
|
| 254 |
+
"tie_word_embeddings": false,
|
| 255 |
+
"topk_group": 1,
|
| 256 |
+
"topk_method": "noaux_tc",
|
| 257 |
+
"use_cache": true,
|
| 258 |
+
"v_head_dim": 256,
|
| 259 |
+
"vocab_size": 154880
|
| 260 |
+
},
|
| 261 |
+
"vision_config": {
|
| 262 |
+
"attention_bias": true,
|
| 263 |
+
"attention_dropout": 0.0,
|
| 264 |
+
"depth": 24,
|
| 265 |
+
"hidden_act": "silu",
|
| 266 |
+
"hidden_size": 1024,
|
| 267 |
+
"image_size": 448,
|
| 268 |
+
"in_channels": 3,
|
| 269 |
+
"initializer_range": 0.02,
|
| 270 |
+
"intermediate_size": 4096,
|
| 271 |
+
"model_type": "glm5_next_vision",
|
| 272 |
+
"num_heads": 16,
|
| 273 |
+
"out_hidden_size": 4096,
|
| 274 |
+
"patch_size": 14,
|
| 275 |
+
"projection_intermediate_size": 10240,
|
| 276 |
+
"rms_norm_eps": 1e-05,
|
| 277 |
+
"spatial_merge_size": 2,
|
| 278 |
+
"swiglu_limit": 10.0,
|
| 279 |
+
"temporal_patch_size": 2
|
| 280 |
+
},
|
| 281 |
+
"image_token_id": 154854,
|
| 282 |
+
"video_token_id": 154855,
|
| 283 |
+
"image_start_token_id": 154830,
|
| 284 |
+
"image_end_token_id": 154831,
|
| 285 |
+
"video_start_token_id": 154832,
|
| 286 |
+
"video_end_token_id": 154833,
|
| 287 |
+
"tie_word_embeddings": false,
|
| 288 |
+
"model_type": "glm5_next",
|
| 289 |
+
"transformers_version": "5.16.0",
|
| 290 |
+
"quantization_config": {
|
| 291 |
+
"config_groups": {
|
| 292 |
+
"group_0": {
|
| 293 |
+
"targets": [
|
| 294 |
+
"Linear"
|
| 295 |
+
],
|
| 296 |
+
"weights": {
|
| 297 |
+
"num_bits": 4,
|
| 298 |
+
"type": "float",
|
| 299 |
+
"group_size": 16,
|
| 300 |
+
"dynamic": false,
|
| 301 |
+
"symmetric": true
|
| 302 |
+
},
|
| 303 |
+
"input_activations": null,
|
| 304 |
+
"output_activations": null
|
| 305 |
+
}
|
| 306 |
+
},
|
| 307 |
+
"quant_algo": "NVFP4",
|
| 308 |
+
"kv_cache_scheme": null,
|
| 309 |
+
"producer": {
|
| 310 |
+
"name": "modelopt",
|
| 311 |
+
"version": "0.45.0"
|
| 312 |
+
},
|
| 313 |
+
"quant_method": "modelopt",
|
| 314 |
+
"ignore": [
|
| 315 |
+
"lm_head",
|
| 316 |
+
"model.language_model.embed_tokens",
|
| 317 |
+
"*.self_attn.in_proj_qkvbfg_a",
|
| 318 |
+
"*.self_attn.q_proj",
|
| 319 |
+
"*.self_attn.k_proj",
|
| 320 |
+
"*.self_attn.v_proj",
|
| 321 |
+
"*.self_attn.b_proj",
|
| 322 |
+
"*.self_attn.f_a_proj",
|
| 323 |
+
"*.self_attn.g_a_proj",
|
| 324 |
+
"*.self_attn.f_b_proj",
|
| 325 |
+
"*.self_attn.g_b_proj",
|
| 326 |
+
"*.self_attn.fused_qkv_a_proj",
|
| 327 |
+
"*.self_attn.q_a_proj",
|
| 328 |
+
"*.self_attn.kv_a_proj_with_mqa",
|
| 329 |
+
"*.self_attn.q_b_proj",
|
| 330 |
+
"*.self_attn.kv_b_proj",
|
| 331 |
+
"*.self_attn.indexer.wk_weights_proj",
|
| 332 |
+
"*.self_attn.indexer.wk",
|
| 333 |
+
"*.self_attn.indexer.weights_proj",
|
| 334 |
+
"*.self_attn.indexer.wq_b",
|
| 335 |
+
"*.self_attn.o_proj",
|
| 336 |
+
"*.mlp.gate",
|
| 337 |
+
"*.mlp.gate_up_proj",
|
| 338 |
+
"*.mlp.down_proj",
|
| 339 |
+
"*.mlp.shared_experts.gate_up_proj",
|
| 340 |
+
"*.mlp.shared_experts.down_proj",
|
| 341 |
+
"*.mlp.shared_experts.gate_proj",
|
| 342 |
+
"*.mlp.shared_experts.up_proj",
|
| 343 |
+
"*.mlp.gate_proj",
|
| 344 |
+
"*.mlp.up_proj",
|
| 345 |
+
"*.eh_proj",
|
| 346 |
+
"model.visual.*"
|
| 347 |
+
]
|
| 348 |
+
}
|
| 349 |
+
}
|