{ "architectures": [ "CIDDiffusionForMaskedLM" ], "attention_bias": false, "attention_dropout": 0.0, "base_model_name_or_path": "openbmb/MiniCPM5-2B-Base", "bos_token_id": 0, "cid_diffusion_base": true, "diffusion_attention": "bidirectional", "diffusion_objective": "masked-diffusion", "dtype": "bfloat16", "eos_token_id": 1, "head_dim": 128, "hidden_act": "silu", "hidden_size": 2048, "initializer_range": 0.02, "intermediate_size": 6144, "mask_token_id": 130560, "max_position_embeddings": 524288, "mlp_bias": false, "model_type": "llama", "num_attention_heads": 16, "num_hidden_layers": 42, "num_key_value_heads": 2, "pad_token_id": 1, "pretraining_tp": 1, "rms_norm_eps": 1e-06, "rope_scaling": null, "rope_theta": 5000000, "tie_word_embeddings": false, "transformers_version": "4.57.6", "use_cache": false, "vocab_size": 130561, "auto_map": { "AutoModelForCausalLM": "modeling_cid_diffusion.CIDDiffusionForMaskedLM" } }