{ "cache_dir": null, "connector_type": "mlp2x_gelu", "hidden_size": 1280, "ignore_index": -100, "image_aspect_ratio": "square", "image_token_index": -200, "llm_model_name_or_path": "apple/OpenELM-270M-Instruct", "model_type": "tinyllava", "num_queries": 128, "num_resampler_layers": 3, "pad_token": "", "pad_token_id": 0, "resampler_hidden_size": 768, "text_config": { "_name_or_path": "apple/OpenELM-270M-Instruct", "activation_fn_name": "swish", "architectures": [ "OpenELMForCausalLM" ], "auto_map": { "AutoConfig": "apple/OpenELM-270M-Instruct--configuration_openelm.OpenELMConfig", "AutoModelForCausalLM": "apple/OpenELM-270M-Instruct--modeling_openelm.OpenELMForCausalLM" }, "ffn_dim_divisor": 256, "ffn_multipliers": [ 0.5, 0.73, 0.97, 1.2, 1.43, 1.67, 1.9, 2.13, 2.37, 2.6, 2.83, 3.07, 3.3, 3.53, 3.77, 4.0 ], "ffn_with_glu": true, "head_dim": 64, "max_context_length": 2048, "model_dim": 1280, "model_type": "openelm", "normalization_layer_name": "rms_norm", "normalize_qk_projections": true, "num_gqa_groups": 4, "num_kv_heads": [ 3, 3, 3, 3, 3, 4, 4, 4, 4, 4, 4, 4, 5, 5, 5, 5 ], "num_query_heads": [ 12, 12, 12, 12, 12, 16, 16, 16, 16, 16, 16, 16, 20, 20, 20, 20 ], "num_transformer_layers": 16, "qkv_multipliers": [ 0.5, 1.0 ], "rope_freq_constant": 10000, "rope_max_length": 4096, "share_input_output_layers": true, "tie_word_embeddings": true, "torch_dtype": "float16" }, "tokenizer_model_max_length": 2048, "tokenizer_name_or_path": "meta-llama/Llama-2-7b-hf", "tokenizer_padding_side": "right", "tokenizer_use_fast": false, "transformers_version": "4.40.1", "tune_type_connector": "full", "tune_type_llm": "lora", "tune_type_vision_tower": "frozen", "tune_vision_tower_from_layer": 0, "use_cache": true, "vision_config": { "_name_or_path": "apple/aimv2-large-patch14-224-distilled", "architectures": [ "AIMv2Model" ], "auto_map": { "AutoConfig": "apple/aimv2-large-patch14-224-distilled--configuration_aimv2.AIMv2Config", "AutoModel": "apple/aimv2-large-patch14-224-distilled--modeling_aimv2.AIMv2Model", "FlaxAutoModel": "apple/aimv2-large-patch14-224-distilled--modeling_flax_aimv2.FlaxAIMv2Model" }, "image_size": 224, "intermediate_size": 2816, "model_name_or_path": "apple/aimv2-large-patch14-224-distilled", "model_name_or_path2": "", "model_type": "aimv2", "num_attention_heads": 8, "projection_dropout": 0.0, "qkv_bias": false, "rms_norm_eps": 1e-05, "torch_dtype": "float32", "use_bias": false }, "vision_feature_layer": -2, "vision_feature_select_strategy": "patch", "vision_hidden_size": 1024, "vision_model_name_or_path": "apple/aimv2-large-patch14-224-distilled", "vision_model_name_or_path2": "", "vocab_size": 32000 }