config.json

{"embed_dim": 1024, "vision_cfg": {"image_size": 224, "layers": 32, "width": 1280, "head_width": 80, "patch_size": 14}, "text_cfg": {"hf_model_name": "xlm-roberta-large", "hf_tokenizer_name": "xlm-roberta-large", "hf_pooler_type": "mean_pooler"}}