{ "architectures": [ "Transformer" ], "bias": false, "bilinear": true, "d_hidden": 2048, "d_model": 512, "gate": null, "n_ctx": 256, "n_head": 8, "n_layer": 6, "repo": null, "tokenizer": "ts-4096", "torch_dtype": "float32", "transformers_version": "4.44.2" }