| { |
| "architecture_type": "uni-encoder", |
| "architectures": [ |
| "GLiClassModel" |
| ], |
| "class_token_index": 50368, |
| "class_token_pooling": "first", |
| "contrastive_loss_coef": 0.0, |
| "dropout": 0.1, |
| "dtype": "float32", |
| "embed_class_token": true, |
| "encoder_config": { |
| "_attn_implementation_autoset": false, |
| "_name_or_path": "answerdotai/ModernBERT-base", |
| "add_cross_attention": false, |
| "architectures": [ |
| "ModernBertForMaskedLM" |
| ], |
| "attention_bias": false, |
| "attention_dropout": 0.0, |
| "bos_token_id": 50281, |
| "classifier_activation": "gelu", |
| "classifier_bias": false, |
| "classifier_dropout": 0.0, |
| "classifier_pooling": "mean", |
| "cls_token_id": 50281, |
| "cross_attention_hidden_size": null, |
| "decoder_bias": true, |
| "decoder_start_token_id": null, |
| "deterministic_flash_attn": false, |
| "dtype": "float32", |
| "embedding_dropout": 0.0, |
| "eos_token_id": 50282, |
| "finetuning_task": null, |
| "global_attn_every_n_layers": 3, |
| "gradient_checkpointing": false, |
| "hidden_activation": "gelu", |
| "hidden_size": 768, |
| "initializer_cutoff_factor": 2.0, |
| "initializer_range": 0.02, |
| "intermediate_size": 1152, |
| "is_decoder": false, |
| "layer_norm_eps": 1e-05, |
| "layer_types": [ |
| "full_attention", |
| "sliding_attention", |
| "sliding_attention", |
| "full_attention", |
| "sliding_attention", |
| "sliding_attention", |
| "full_attention", |
| "sliding_attention", |
| "sliding_attention", |
| "full_attention", |
| "sliding_attention", |
| "sliding_attention", |
| "full_attention", |
| "sliding_attention", |
| "sliding_attention", |
| "full_attention", |
| "sliding_attention", |
| "sliding_attention", |
| "full_attention", |
| "sliding_attention", |
| "sliding_attention", |
| "full_attention" |
| ], |
| "local_attention": 128, |
| "max_position_embeddings": 8192, |
| "mlp_bias": false, |
| "mlp_dropout": 0.0, |
| "model_type": "modernbert", |
| "norm_bias": false, |
| "norm_eps": 1e-05, |
| "num_attention_heads": 12, |
| "num_hidden_layers": 22, |
| "pad_token_id": 50283, |
| "position_embedding_type": "absolute", |
| "prefix": null, |
| "pruned_heads": {}, |
| "repad_logits_with_grad": false, |
| "rope_parameters": { |
| "full_attention": { |
| "rope_theta": 160000.0, |
| "rope_type": "default" |
| }, |
| "sliding_attention": { |
| "rope_theta": 10000.0, |
| "rope_type": "default" |
| } |
| }, |
| "sep_token_id": 50282, |
| "sparse_pred_ignore_index": -100, |
| "sparse_prediction": false, |
| "task_specific_params": null, |
| "tf_legacy_loss": false, |
| "tie_encoder_decoder": false, |
| "tie_word_embeddings": true, |
| "tokenizer_class": null, |
| "torchscript": false, |
| "use_bfloat16": false, |
| "vocab_size": 50370 |
| }, |
| "encoder_layer_id": -1, |
| "encoder_model_name": "answerdotai/ModernBERT-base", |
| "example_token_index": 50372, |
| "extract_text_features": false, |
| "focal_loss_alpha": -1, |
| "focal_loss_gamma": -1, |
| "focal_loss_reduction": null, |
| "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, |
| "pad_token_id": 50283, |
| "pooling_strategy": "first", |
| "problem_type": "single_label_classification", |
| "projector_hidden_act": "gelu", |
| "prompt_first": true, |
| "scorer_attn_dropout": 0.1, |
| "scorer_encoder_num_layers": 2, |
| "scorer_mlp_hidden_size": 1024, |
| "scorer_num_heads": 16, |
| "scorer_type": "simple", |
| "squeeze_layers": false, |
| "text_token_index": 50369, |
| "transformers_version": "5.17.0", |
| "use_lstm": false, |
| "use_segment_embeddings": false, |
| "vocab_size": 50370 |
| } |
|
|
