{ "architectures": [ "DiscreteDiffusionModel" ], "argmax_decoding": true, "attention_strategy": "full", "audio_encoder_name": "UsefulSensors/moonshine-streaming-medium", "auto_map": { "AutoConfig": "configuration_dlm.DiscreteDiffusionConfig", "AutoModel": "modeling_dlm.DiscreteDiffusionModel", "AutoModelForMaskedLM": "modeling_dlm.DiscreteDiffusionModel" }, "backbone_config": { "_name_or_path": "FacebookAI/xlm-roberta-base", "add_cross_attention": false, "architectures": [ "XLMRobertaForMaskedLM" ], "attention_probs_dropout_prob": 0.1, "bos_token_id": 0, "chunk_size_feed_forward": 0, "classifier_dropout": null, "dtype": null, "eos_token_id": 2, "hidden_act": "gelu", "hidden_dropout_prob": 0.1, "hidden_size": 768, "id2label": { "0": "LABEL_0", "1": "LABEL_1" }, "initializer_range": 0.02, "intermediate_size": 3072, "is_decoder": false, "is_encoder_decoder": false, "label2id": { "LABEL_0": 0, "LABEL_1": 1 }, "layer_norm_eps": 1e-05, "max_position_embeddings": 514, "model_type": "xlm-roberta", "num_attention_heads": 12, "num_hidden_layers": 12, "output_attentions": false, "output_hidden_states": false, "output_past": true, "pad_token_id": 1, "position_embedding_type": "absolute", "problem_type": null, "return_dict": true, "tie_word_embeddings": true, "transformers_version": "5.11.0", "type_vocab_size": 1, "use_cache": true, "vocab_size": 250005 }, "bos_token_id": 0, "dataset_type": "speech_translation_multitask", "diffusion_type": "absorbing", "dtype": "float32", "eos_token_id": 2, "hidden_size": 768, "intermediate_size": 3072, "lora": false, "lora_alpha": 16, "lora_bias": "none", "lora_dropout": 0, "lora_rank": 16, "lora_target_modules": [ "query", "value" ], "mask_token_id": 250001, "max_position_embeddings": 514, "model_type": "discrete_diffusion", "num_attention_heads": 12, "num_diffusion_timesteps": 50, "pad_token_id": 1, "tie_word_embeddings": true, "transformers_version": "5.11.0", "vocab_pad_to_multiple": 1 }