{ "architectures": [ "InstructCell" ], "base_model_config": { "architectures": [ "T5ForConditionalGeneration" ], "d_ff": 3072, "d_model": 768, "decoder_start_token_id": 0, "ignore_index": null, "model_type": "t5", "n_positions": 512, "num_decoder_layers": 12, "num_heads": 12, "num_layers": 12, "num_signal_tokens": 1, "output_past": true, "special_tokens_dict": { "end_tag": "", "placeholder": "", "signal_tokens": [ "" ], "start_tag": "" }, "special_tokens_index_dict": { "end_tag": 32101, "first_signal_token": 32103, "placeholder": 32102, "start_tag": 32100 }, "task_specific_params": { "summarization": { "early_stopping": true, "length_penalty": 2.0, "max_length": 200, "min_length": 30, "no_repeat_ngram_size": 3, "num_beams": 4, "prefix": "summarize: " }, "translation_en_to_de": { "early_stopping": true, "max_length": 300, "num_beams": 4, "prefix": "translate English to German: " }, "translation_en_to_fr": { "early_stopping": true, "max_length": 300, "num_beams": 4, "prefix": "translate English to French: " }, "translation_en_to_ro": { "early_stopping": true, "max_length": 300, "num_beams": 4, "prefix": "translate English to Romanian: " } }, "vocab_size": 32112 }, "feature_decoder_config": { "adaptive_library": true, "condition_input_dim": 256, "dropout_rate": 0.1, "log_variational": true, "n_hidden": 1024, "n_latent": 256, "n_layers": 4, "use_batch_norm": "none", "use_layer_norm": "both" }, "feature_encoder_config": { "count_dim": 18961, "cross_attention_frequency": 1, "hidden_dropout_prob": 0.1, "is_q_former_encoder": true, "num_blocks": 3, "num_hidden_layers": 4, "num_key_value_tokens": 6, "num_query_tokens": 8 }, "log_variational": true, "model_type": "instructcell", "normalize_total": true, "tokenizer_use_fast": false, "torch_dtype": "float32", "transformers_version": "4.33.3" }