Upload ./config.json with huggingface_hub
Browse files- config.json +214 -0
config.json
ADDED
|
@@ -0,0 +1,214 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_classification_head": false,
|
| 3 |
+
"anyres_pooling_size": 2,
|
| 4 |
+
"anyres_vit_max_image_size": null,
|
| 5 |
+
"anyres_vit_two_views": false,
|
| 6 |
+
"architectures": [
|
| 7 |
+
"HunYuanMoEV1ForCausalLM"
|
| 8 |
+
],
|
| 9 |
+
"attention_bias": false,
|
| 10 |
+
"attention_dropout": 0.1,
|
| 11 |
+
"attention_head_dim": 128,
|
| 12 |
+
"auto_map": {
|
| 13 |
+
"AutoConfig": "configuration_hunyuan.HunYuanConfig",
|
| 14 |
+
"AutoModel": "hunyuan.HunYuanModel",
|
| 15 |
+
"AutoModelForCausalLM": "hunyuan.HunYuanMoEV1ForCausalLM"
|
| 16 |
+
},
|
| 17 |
+
"bos_token_id": 1,
|
| 18 |
+
"cla_share_factor": 2,
|
| 19 |
+
"class_num": 0,
|
| 20 |
+
"dense_list": [
|
| 21 |
+
4096,
|
| 22 |
+
0
|
| 23 |
+
],
|
| 24 |
+
"eod_token_id": 127967,
|
| 25 |
+
"eos_token_id": 127960,
|
| 26 |
+
"group_limited_greedy": false,
|
| 27 |
+
"hidden_act": "silu",
|
| 28 |
+
"hidden_size": 4096,
|
| 29 |
+
"im_end_id": 6,
|
| 30 |
+
"im_newline_id": 12,
|
| 31 |
+
"im_start_id": 5,
|
| 32 |
+
"image_token_id": 9,
|
| 33 |
+
"initializer_range": 0.02,
|
| 34 |
+
"intermediate_size": 3072,
|
| 35 |
+
"kv_lora_rank": null,
|
| 36 |
+
"mask_init_id": 13,
|
| 37 |
+
"max_position_embeddings": 32768,
|
| 38 |
+
"mlp_bias": false,
|
| 39 |
+
"model_type": "hunyuan",
|
| 40 |
+
"moe_drop_tokens": false,
|
| 41 |
+
"moe_intermediate_size": [
|
| 42 |
+
3072,
|
| 43 |
+
3072,
|
| 44 |
+
3072,
|
| 45 |
+
3072,
|
| 46 |
+
3072,
|
| 47 |
+
3072,
|
| 48 |
+
3072,
|
| 49 |
+
3072,
|
| 50 |
+
3072,
|
| 51 |
+
3072,
|
| 52 |
+
3072,
|
| 53 |
+
3072,
|
| 54 |
+
3072,
|
| 55 |
+
3072,
|
| 56 |
+
3072,
|
| 57 |
+
3072,
|
| 58 |
+
3072,
|
| 59 |
+
3072,
|
| 60 |
+
3072,
|
| 61 |
+
3072,
|
| 62 |
+
3072,
|
| 63 |
+
3072,
|
| 64 |
+
3072,
|
| 65 |
+
3072,
|
| 66 |
+
3072,
|
| 67 |
+
3072,
|
| 68 |
+
3072,
|
| 69 |
+
3072,
|
| 70 |
+
3072,
|
| 71 |
+
3072,
|
| 72 |
+
3072,
|
| 73 |
+
3072
|
| 74 |
+
],
|
| 75 |
+
"moe_layer_num_skipped": 0,
|
| 76 |
+
"moe_random_routing_dropped_token": false,
|
| 77 |
+
"moe_topk": [
|
| 78 |
+
8,
|
| 79 |
+
8,
|
| 80 |
+
8,
|
| 81 |
+
8,
|
| 82 |
+
8,
|
| 83 |
+
8,
|
| 84 |
+
8,
|
| 85 |
+
8,
|
| 86 |
+
8,
|
| 87 |
+
8,
|
| 88 |
+
8,
|
| 89 |
+
8,
|
| 90 |
+
8,
|
| 91 |
+
8,
|
| 92 |
+
8,
|
| 93 |
+
8,
|
| 94 |
+
8,
|
| 95 |
+
8,
|
| 96 |
+
8,
|
| 97 |
+
8,
|
| 98 |
+
8,
|
| 99 |
+
8,
|
| 100 |
+
8,
|
| 101 |
+
8,
|
| 102 |
+
8,
|
| 103 |
+
8,
|
| 104 |
+
8,
|
| 105 |
+
8,
|
| 106 |
+
8,
|
| 107 |
+
8,
|
| 108 |
+
8,
|
| 109 |
+
8
|
| 110 |
+
],
|
| 111 |
+
"n_group": null,
|
| 112 |
+
"norm_topk_prob": true,
|
| 113 |
+
"norm_type": "rms",
|
| 114 |
+
"num_attention_heads": 32,
|
| 115 |
+
"num_experts": 64,
|
| 116 |
+
"num_hidden_layers": 32,
|
| 117 |
+
"num_key_value_heads": 8,
|
| 118 |
+
"num_media_embeds": 257,
|
| 119 |
+
"num_shared_expert": [
|
| 120 |
+
1,
|
| 121 |
+
1,
|
| 122 |
+
1,
|
| 123 |
+
1,
|
| 124 |
+
1,
|
| 125 |
+
1,
|
| 126 |
+
1,
|
| 127 |
+
1,
|
| 128 |
+
1,
|
| 129 |
+
1,
|
| 130 |
+
1,
|
| 131 |
+
1,
|
| 132 |
+
1,
|
| 133 |
+
1,
|
| 134 |
+
1,
|
| 135 |
+
1,
|
| 136 |
+
1,
|
| 137 |
+
1,
|
| 138 |
+
1,
|
| 139 |
+
1,
|
| 140 |
+
1,
|
| 141 |
+
1,
|
| 142 |
+
1,
|
| 143 |
+
1,
|
| 144 |
+
1,
|
| 145 |
+
1,
|
| 146 |
+
1,
|
| 147 |
+
1,
|
| 148 |
+
1,
|
| 149 |
+
1,
|
| 150 |
+
1,
|
| 151 |
+
1
|
| 152 |
+
],
|
| 153 |
+
"org_vocab_size": 128167,
|
| 154 |
+
"pad_id": 127961,
|
| 155 |
+
"pad_token_id": 127961,
|
| 156 |
+
"pool_type": "last",
|
| 157 |
+
"position_embedding_xdrope": false,
|
| 158 |
+
"pretraining_tp": 1,
|
| 159 |
+
"q_lora_rank": null,
|
| 160 |
+
"qk_nope_head_dim": null,
|
| 161 |
+
"qk_rope_head_dim": null,
|
| 162 |
+
"quantization_config": {
|
| 163 |
+
"bits": 4,
|
| 164 |
+
"checkpoint_format": "gptq",
|
| 165 |
+
"desc_act": true,
|
| 166 |
+
"group_size": 128,
|
| 167 |
+
"model_file_base_name": "model",
|
| 168 |
+
"quant_method": "gptq",
|
| 169 |
+
"static_groups": true,
|
| 170 |
+
"sym": true,
|
| 171 |
+
"true_sequential": true
|
| 172 |
+
},
|
| 173 |
+
"rms_norm_eps": 1e-05,
|
| 174 |
+
"rope_scaling": {
|
| 175 |
+
"alpha": 1000.0,
|
| 176 |
+
"beta_fast": 32,
|
| 177 |
+
"beta_slow": 1,
|
| 178 |
+
"factor": 1.0,
|
| 179 |
+
"mscale": 1.0,
|
| 180 |
+
"mscale_all_dim": 1.0,
|
| 181 |
+
"type": "dynamic"
|
| 182 |
+
},
|
| 183 |
+
"rope_theta": 10000.0,
|
| 184 |
+
"routed_scaling_factor": 1.0,
|
| 185 |
+
"sep_token_id": 127962,
|
| 186 |
+
"skip_cls_token": false,
|
| 187 |
+
"text_end_id": 8,
|
| 188 |
+
"text_start_id": 7,
|
| 189 |
+
"tie_word_embeddings": true,
|
| 190 |
+
"topk_group": null,
|
| 191 |
+
"torch_dtype": "bfloat16",
|
| 192 |
+
"transformers_version": "4.41.2",
|
| 193 |
+
"use_cache": true,
|
| 194 |
+
"use_cla": false,
|
| 195 |
+
"use_mixed_mlp_moe": true,
|
| 196 |
+
"use_mla": false,
|
| 197 |
+
"use_qk_norm": true,
|
| 198 |
+
"use_rotary_pos_emb": true,
|
| 199 |
+
"v_head_dim": null,
|
| 200 |
+
"video_end_id": 11,
|
| 201 |
+
"video_start_id": 10,
|
| 202 |
+
"vit_add_patchemb_bias": false,
|
| 203 |
+
"vit_input_resolution": 224,
|
| 204 |
+
"vit_mapping_type": "resampler",
|
| 205 |
+
"vit_norm_type": "fused",
|
| 206 |
+
"vit_patch": 1,
|
| 207 |
+
"vit_path": null,
|
| 208 |
+
"vit_remove_prenorm": false,
|
| 209 |
+
"vit_token": 64,
|
| 210 |
+
"vit_type": null,
|
| 211 |
+
"vit_used_rms_norm": false,
|
| 212 |
+
"vocab_size": 128167,
|
| 213 |
+
"xdrope_section": null
|
| 214 |
+
}
|