{ "_name_or_path": "/mnt/task_wrapper/user_output/artifacts/gemma2b_ferret", "architectures": [ "FerretGemmaForCausalLM" ], "auto_map": { "AutoConfig": "modeling.FerretGemmaConfig", "AutoModel": "modeling.FerretGemmaModel", "AutoModelForCausalLM": "modeling.FerretGemmaForCausalLM" }, "attention_bias": false, "attention_dropout": 0.0, "bos_token_id": 2, "eos_token_id": 1, "freeze_mm_mlp_adapter": false, "head_dim": 256, "hidden_act": "gelu_pytorch_tanh", "hidden_activation": "gelu_pytorch_tanh", "hidden_size": 2048, "im_region_fea_token": 256000, "image_aspect_ratio": "anyres", "image_grid_pinpoints": [ [ 336, 672 ], [ 672, 336 ], [ 672, 672 ], [ 1008, 336 ], [ 336, 1008 ] ], "initializer_range": 0.02, "intermediate_size": 16384, "max_position_embeddings": 8192, "mm_hidden_size": 1024, "mm_patch_merge_type": "spatial", "mm_projector_lr": null, "mm_projector_type": "mlp2x_gelu", "mm_use_im_patch_token": false, "mm_use_im_start_end": false, "mm_vision_select_feature": "patch", "mm_vision_select_layer": -2, "mm_vision_tower": "/mnt/task_runtime/work_dirs/openai-clip-vit-large-patch14-336", "mm_vision_tower_lr": null, "model_type": "ferret_gemma", "num_attention_heads": 8, "num_hidden_layers": 18, "num_key_value_heads": 1, "pad_token_id": 0, "region_geo_sampler": true, "rms_norm_eps": 1e-06, "rope_theta": 10000.0, "sampler_pooler_mode": "max", "tokenizer_model_max_length": 4096, "tokenizer_padding_side": "right", "torch_dtype": "bfloat16", "transformers_version": "4.39.0", "tune_mm_mlp_adapter": false, "unfreeze_mm_vision_tower": false, "use_cache": true, "use_mm_proj": true, "vocab_size": 256001 }