{ "annotation_size": 8943, "architectures": [ "ProteinBertForPreTraining" ], "attention_key_size": 64, "bos_token_id": 1, "conv_kernel_size": 9, "dtype": "float32", "eos_token_id": 2, "global_hidden_size": 512, "head": null, "hidden_act": "gelu", "hidden_size": 128, "id2label": null, "initializer_range": 0.02, "label2id": null, "layer_norm_eps": 0.001, "lm_head": { "act": null, "bias": true, "dropout": 0.0, "hidden_size": null, "layer_norm_eps": 1e-12, "loss_weight": null, "num_labels": 37, "output_name": null, "transform": null, "transform_act": null }, "mask_token_id": 4, "model_type": "proteinbert", "null_token_id": 5, "num_attention_heads": 4, "num_hidden_layers": 6, "num_labels": 1, "pad_token_id": 0, "tie_word_embeddings": false, "transformers_version": "5.9.0", "unk_token_id": 3, "vocab_size": 37, "wide_conv_dilation_rate": 5 }