{ "_name_or_path": "/mnt/matylda5/ipoloka/CHIME_CSL/all_data", "activation_dropout": 0.0, "activation_function": "gelu", "additional_layer": false, "additional_self_attention_layer": true, "apply_spec_augment": false, "apply_target_amp_to_n_layers": -1, "architectures": [ "DiCoWForConditionalGeneration" ], "attention_dropout": 0.0, "begin_suppress_tokens": [ 220, 50256 ], "blank_token_id": null, "bos_token_id": 50257, "classifier_proj_size": 256, "ctc_loss_reduction": "mean", "ctc_weight": 0.3, "ctc_zero_infinity": false, "d_model": 1280, "decoder_attention_heads": 20, "decoder_ffn_dim": 5120, "decoder_layerdrop": 0.0, "decoder_layers": 4, "decoder_start_token_id": 50258, "dropout": 0.0, "encoder_attention_heads": 20, "encoder_ffn_dim": 5120, "encoder_layerdrop": 0.0, "encoder_layers": 32, "eos_token_id": 50257, "final_dropout": 0.0, "forced_decoder_ids": null, "init_std": 0.02, "is_encoder_decoder": true, "mask_feature_length": 10, "mask_feature_min_masks": 0, "mask_feature_prob": 0.0, "mask_time_length": 10, "mask_time_min_masks": 2, "mask_time_prob": 0.05, "max_source_positions": 1500, "max_target_positions": 448, "median_filter_width": 7, "model_type": "dicow", "mt_num_speakers": 1, "n_soft_prompts": 16, "num_hidden_layers": 32, "num_mel_bins": 128, "pad_token_id": 50257, "remove_timestamps_from_ctc": true, "scale_embedding": false, "sub_sample": true, "target_amp_bias_only": false, "target_amp_init": "disparagement", "target_amp_is_diagonal": true, "target_amp_use_non_target": true, "target_amp_use_overlap": true, "target_amp_use_silence": true, "target_amp_use_target": true, "torch_dtype": "float32", "transformers_version": "4.39.3", "use_cache": true, "use_target_amplifiers": true, "use_weighted_layer_sum": false, "vocab_size": 51866 }