{ "architectures": [ "MiniCPM3ForCausalLM" ], "attention_bias": false, "attention_dropout": 0.0, "bos_token_id": 1, "dim_model_base": 256, "dtype": "float32", "eos_token_id": 73440, "head_dim": 64, "hidden_act": "gelu", "hidden_size": 32, "initializer_range": 0.02, "intermediate_size": 32, "is_decoder": true, "kv_lora_rank": 32, "max_position_embeddings": 512, "mlp_bias": false, "model_type": "minicpm3", "num_attention_heads": 2, "num_hidden_layers": 2, "num_key_value_heads": 2, "pad_token_id": 0, "pretraining_tp": 1, "q_lora_rank": 16, "qk_head_dim": 128, "qk_nope_head_dim": 64, "qk_rope_head_dim": 64, "rms_norm_eps": 1e-05, "rope_parameters": { "rope_theta": 10000.0, "rope_type": "default" }, "scale_depth": 1.4, "scale_emb": 12, "tie_word_embeddings": false, "transformers_version": "5.16.0.dev0", "use_cache": true, "v_head_dim": 64, "vocab_size": 73448 }