{ "image_size": 16, "in_channels": 4, "patch_size": 2, "dim": 256, "depth": 8, "heads": 4, "cond_dim": 128, "vocab_size": 8192, "text_dim": 128, "text_layers": 2, "text_heads": 4, "max_len": 16, "subword_buckets": 4096, "architectures": [ "IndepthModel" ], "model_type": "indepth" }