{ "architectures": [ "BitNetForCausalLM" ], "attention_bias": false, "attention_dropout": 0.0, "bos_token_id": 1, "dtype": "float32", "eos_token_id": 2, "hidden_act": "relu2", "hidden_size": 256, "initializer_range": 0.02, "intermediate_size": 512, "max_position_embeddings": 128, "model_type": "bitnet", "num_attention_heads": 4, "num_hidden_layers": 4, "num_key_value_heads": 2, "pad_token_id": 3, "rms_norm_eps": 1e-05, "rope_theta": 10000.0, "tie_word_embeddings": true, "transformers_version": "4.57.6", "use_cache": true, "vocab_size": 2048 }