{ "architecture_type": "uni-encoder", "architectures": [ "GLiClassModel" ], "class_token_index": 128001, "class_token_pooling": "first", "contrastive_loss_coef": 0.0, "cross_encoder_config": null, "dropout": 0.1, "dtype": "float32", "embed_class_token": true, "encoder_config": { "_attn_implementation_autoset": false, "_name_or_path": "microsoft/deberta-v3-base", "attention_probs_dropout_prob": 0.1, "hidden_act": "gelu", "hidden_dropout_prob": 0.1, "hidden_size": 768, "initializer_range": 0.02, "intermediate_size": 3072, "layer_norm_eps": 1e-07, "legacy": true, "max_position_embeddings": 512, "max_relative_positions": -1, "model_type": "deberta-v2", "norm_rel_ebd": "layer_norm", "num_attention_heads": 12, "num_hidden_layers": 12, "pooler_dropout": 0, "pooler_hidden_act": "gelu", "pooler_hidden_size": 768, "pos_att_type": [ "p2c", "c2p" ], "position_biased_input": false, "position_buckets": 256, "relative_attention": true, "share_att_key": true, "type_vocab_size": 0, "vocab_size": 128003 }, "encoder_layer_id": -1, "encoder_model_name": "microsoft/deberta-v3-base", "example_token_index": 128005, "extract_text_features": false, "focal_loss_alpha": -1, "focal_loss_gamma": -1, "focal_loss_reduction": "none", "hidden_size": 768, "ignore_index": -100, "initializer_range": 0.03, "label_model_config": null, "label_model_name": null, "layer_wise": false, "logit_scale_init_value": 2.6592, "max_labels_alloc": "dynamic", "max_num_classes": 25, "model_type": "GLiClass", "normalize_features": false, "pooling_strategy": "first", "problem_type": "multi_label_classification", "projector_hidden_act": "gelu", "prompt_first": true, "scorer_attn_dropout": 0.1, "scorer_mlp_hidden_size": 1024, "scorer_num_heads": 16, "scorer_type": "mlp", "squeeze_layers": false, "text_token_index": 128002, "transformers_version": "4.57.6", "use_lstm": false, "use_segment_embeddings": false, "vocab_size": 128003 }