| { | |
| "architectures": [ | |
| "Ernie4_5_MoeForCausalLM" | |
| ], | |
| "bos_token_id": 1, | |
| "eos_token_id": 2, | |
| "hidden_act": "silu", | |
| "hidden_size": 2560, | |
| "intermediate_size": 12288, | |
| "max_position_embeddings": 131072, | |
| "model_type": "ernie4_5_moe", | |
| "num_attention_heads": 20, | |
| "num_key_value_heads": 4, | |
| "num_hidden_layers": 28, | |
| "pad_token_id": 0, | |
| "rms_norm_eps": 1e-05, | |
| "use_cache": false, | |
| "vocab_size": 103424, | |
| "rope_theta": 500000, | |
| "use_rmsnorm": true, | |
| "tie_word_embeddings": true, | |
| "use_bias": false, | |
| "moe_num_experts": 64, | |
| "moe_num_shared_experts": 2, | |
| "moe_layer_start_index": 1, | |
| "moe_intermediate_size": 1536, | |
| "moe_capacity": [64,64,64], | |
| "moe_gate": "topk", | |
| "moe_k": 6, | |
| "moe_layer_interval": 1, | |
| "moe_use_aux_free": true, | |
| "dtype": "bfloat16", | |
| "num_nextn_predict_layers": 1, | |
| "multi_token_pred_lambda": 0.3 | |
| } | |