| { | |
| "_name_or_path": "emu2", | |
| "architectures": [ | |
| "EmuForCausalLM" | |
| ], | |
| "attention_bias": false, | |
| "attention_dropout": 0.0, | |
| "auto_map": { | |
| "AutoConfig": "configuration_emu.EmuConfig", | |
| "AutoModelForCausalLM": "modeling_emu.EmuForCausalLM" | |
| }, | |
| "bos_token_id": 1, | |
| "d_model": 1792, | |
| "eos_token_id": 2, | |
| "hidden_act": "silu", | |
| "hidden_size": 6656, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 17920, | |
| "llama_config_path": "/share/project/quansun/models/llm_models/llama/hf/llama-30b", | |
| "max_position_embeddings": 2048, | |
| "model_version": "base", | |
| "num_attention_heads": 52, | |
| "num_hidden_layers": 60, | |
| "num_key_value_heads": 52, | |
| "pad_token_id": 32000, | |
| "pretraining_tp": 1, | |
| "rms_norm_eps": 1e-06, | |
| "rope_scaling": null, | |
| "rope_theta": 10000.0, | |
| "tie_word_embeddings": false, | |
| "torch_dtype": "float32", | |
| "transformers_version": "4.31.0", | |
| "use_cache": true, | |
| "vision_config": { | |
| "drop_path_rate": 0, | |
| "eva_model_name": "eva-clip-E-14-plus", | |
| "head_width": 112, | |
| "image_size": 448, | |
| "intermediate_size": 15360, | |
| "layer_norm_eps": 1e-06, | |
| "layers": 64, | |
| "mlp_ratio": 8.571428571428571, | |
| "n_query": 64, | |
| "patch_size": 14, | |
| "postnorm": true, | |
| "qkv_bias": true, | |
| "v_query": 64, | |
| "width": 1792, | |
| "xattn": true | |
| }, | |
| "vocab_size": 32272 | |
| } | |