{ "architectures": [ "Exaone4_5_Model" ], "dtype": "float32", "image_token_id": 3, "model_type": "exaone4_5", "pad_token_id": 0, "text_config": { "attention_dropout": 0.0, "bos_token_id": 1, "eos_token_id": 53, "hidden_act": "silu", "hidden_size": 32, "initializer_range": 0.02, "intermediate_size": 32, "layer_types": [ "sliding_attention", "sliding_attention" ], "max_position_embeddings": 512, "model_type": "exaone4", "num_attention_heads": 4, "num_hidden_layers": 2, "num_key_value_heads": 2, "pad_token_id": 0, "rms_norm_eps": 1e-05, "rope_parameters": { "rope_theta": 10000.0, "rope_type": "default" }, "sliding_window": 4096, "sliding_window_pattern": 4, "tie_word_embeddings": false, "use_cache": true, "vocab_size": 153600 }, "tie_word_embeddings": false, "transformers_version": "5.16.0.dev0", "video_token_id": 4, "vision_config": { "depth": 2, "fullatt_block_indexes": [ 7, 15, 23, 31 ], "hidden_act": "silu", "hidden_size": 32, "in_channels": 3, "initializer_range": 0.02, "intermediate_size": 32, "model_type": "exaone4_5_vision", "num_heads": 4, "num_key_value_heads": 2, "out_hidden_size": 32, "patch_size": 16, "spatial_merge_size": 1, "temporal_patch_size": 2, "window_size": 112 }, "vision_end_token_id": 6, "vision_start_token_id": 5 }