{ "architectures": [ "Aimv2VisionModel" ], "attention_dropout": 0.1, "dropout": 0.1, "dtype": "float32", "hidden_act": "silu", "hidden_size": 32, "image_size": 30, "initializer_range": 0.02, "intermediate_size": 37, "is_native": false, "mlp_bias": false, "model_type": "aimv2_vision_model", "num_attention_heads": 4, "num_channels": 3, "num_hidden_layers": 2, "patch_size": 2, "projection_dim": 32, "qkv_bias": false, "rms_norm_eps": 1e-05, "transformers_version": "5.15.0.dev0", "use_head": true }