{ "architectures": [ "MonoidForCausalLM" ], "attention_bias": false, "auto_map": { "AutoConfig": "MonoidForCausalLM.MonoidConfig", "AutoModelForCausalLM": "MonoidForCausalLM.MonoidForCausalLM" }, "bos_token_id": 128000, "dtype": "bfloat16", "eos_token_id": 128009, "head_dim": 64, "hidden_act": "silu", "hidden_size": 2048, "initializer_range": 0.041666666666666664, "intermediate_size": 8192, "max_position_embeddings": 131072, "mlp_bias": false, "model_type": "monoid", "num_attention_heads": 32, "num_hidden_layers": 16, "pad_token_id": 128009, "rms_norm_eps": 1e-05, "transformers_version": "4.57.6", "vocab_size": 128256 }