{ "architectures": [ "GemmaModel" ], "attention_bias": false, "attention_dropout": 0.0, "bos_token_id": 2, "eos_token_id": 1, "head_dim": 256, "hidden_act": "gelu_pytorch_tanh", "hidden_activation": "gelu_pytorch_tanh", "hidden_size": 512, "initializer_range": 0.02, "intermediate_size": 24576, "max_position_embeddings": 8192, "model_type": "gemma", "num_attention_heads": 16, "num_hidden_layers": 1, "num_key_value_heads": 16, "pad_token_id": 0, "rms_norm_eps": 1e-06, "rope_theta": 10000.0, "torch_dtype": "bfloat16", "transformers_version": "4.40.0.dev0", "use_cache": true, "vocab_size": 512 }