aieng-lab/t5-3b_comment-type-java

1{2  "_name_or_path": "t5-3b",3  "architectures": [4    "T5ForSequenceClassification"5  ],6  "classifier_dropout": 0.0,7  "d_ff": 16384,8  "d_kv": 128,9  "d_model": 1024,10  "decoder_start_token_id": 0,11  "dense_act_fn": "relu",12  "dropout_rate": 0.1,13  "eos_token_id": 1,14  "feed_forward_proj": "relu",15  "id2label": {16    "0": "LABEL_0",17    "1": "LABEL_1",18    "2": "LABEL_2",19    "3": "LABEL_3",20    "4": "LABEL_4",21    "5": "LABEL_5",22    "6": "LABEL_6"23  },24  "initializer_factor": 1.0,25  "is_encoder_decoder": true,26  "is_gated_act": false,27  "label2id": {28    "LABEL_0": 0,29    "LABEL_1": 1,30    "LABEL_2": 2,31    "LABEL_3": 3,32    "LABEL_4": 4,33    "LABEL_5": 5,34    "LABEL_6": 635  },36  "layer_norm_epsilon": 1e-06,37  "model_type": "t5",38  "n_positions": 512,39  "num_decoder_layers": 24,40  "num_heads": 32,41  "num_layers": 24,42  "output_past": true,43  "pad_token_id": 0,44  "problem_type": "multi_label_classification",45  "relative_attention_max_distance": 128,46  "relative_attention_num_buckets": 32,47  "task_specific_params": {48    "summarization": {49      "early_stopping": true,50      "length_penalty": 2.0,51      "max_length": 200,52      "min_length": 30,53      "no_repeat_ngram_size": 3,54      "num_beams": 4,55      "prefix": "summarize: "56    },57    "translation_en_to_de": {58      "early_stopping": true,59      "max_length": 300,60      "num_beams": 4,61      "prefix": "translate English to German: "62    },63    "translation_en_to_fr": {64      "early_stopping": true,65      "max_length": 300,66      "num_beams": 4,67      "prefix": "translate English to French: "68    },69    "translation_en_to_ro": {70      "early_stopping": true,71      "max_length": 300,72      "num_beams": 4,73      "prefix": "translate English to Romanian: "74    }75  },76  "torch_dtype": "bfloat16",77  "transformers_version": "4.48.3",78  "use_cache": true,79  "vocab_size": 3212880}81