| { |
| "architecture_type": "uni-encoder", |
| "architectures": [ |
| "GLiClassModel" |
| ], |
| "bos_token_id": 1, |
| "class_token_index": 250102, |
| "class_token_pooling": "first", |
| "contrastive_loss_coef": 0.1, |
| "dropout": 0.1, |
| "dtype": "float32", |
| "embed_class_token": true, |
| "encoder_config": { |
| "_attn_implementation_autoset": false, |
| "_name_or_path": "microsoft/mdeberta-v3-base", |
| "add_cross_attention": false, |
| "attention_probs_dropout_prob": 0.1, |
| "bos_token_id": null, |
| "cross_attention_hidden_size": null, |
| "decoder_start_token_id": null, |
| "eos_token_id": null, |
| "finetuning_task": null, |
| "hidden_act": "gelu", |
| "hidden_dropout_prob": 0.1, |
| "hidden_size": 768, |
| "initializer_range": 0.02, |
| "intermediate_size": 3072, |
| "is_decoder": false, |
| "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, |
| "output_hidden_states": true, |
| "pad_token_id": 0, |
| "pooler_dropout": 0, |
| "pooler_hidden_act": "gelu", |
| "pooler_hidden_size": 768, |
| "pos_att_type": [ |
| "p2c", |
| "c2p" |
| ], |
| "position_biased_input": false, |
| "position_buckets": 256, |
| "prefix": null, |
| "pruned_heads": {}, |
| "relative_attention": true, |
| "sep_token_id": null, |
| "share_att_key": true, |
| "task_specific_params": null, |
| "tf_legacy_loss": false, |
| "tie_encoder_decoder": false, |
| "tie_word_embeddings": true, |
| "tokenizer_class": null, |
| "torchscript": false, |
| "type_vocab_size": 0, |
| "use_bfloat16": false, |
| "vocab_size": 250104 |
| }, |
| "encoder_layer_id": -1, |
| "encoder_model_name": "microsoft/mdeberta-v3-base", |
| "eos_token_id": 2, |
| "example_token_index": 250106, |
| "extract_text_features": false, |
| "focal_loss_alpha": 0.7, |
| "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": true, |
| "logit_scale_init_value": 2.6592, |
| "max_labels_alloc": "dynamic", |
| "max_num_classes": 25, |
| "model_type": "GLiClass", |
| "normalize_features": false, |
| "pad_token_id": 0, |
| "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": "simple", |
| "shuffle_labels": true, |
| "squeeze_layers": false, |
| "text_token_index": 250103, |
| "transformers_version": "5.1.0", |
| "use_cache": false, |
| "use_lstm": false, |
| "use_segment_embeddings": false, |
| "vocab_size": 250104 |
| } |
|
|
