Writing logs to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/distilbert-base-uncased-glue:rte-2020-06-29-14:58/log.txt. Loading nlp dataset glue, subset rte, split train. Loading nlp dataset glue, subset rte, split validation. Loaded dataset. Found: 2 labels: ([0, 1]) Loading transformers AutoModelForSequenceClassification: distilbert-base-uncased Tokenizing training data. (len: 2490) Tokenizing eval data (len: 277) Loaded data and tokenized in 10.953861474990845s Training model across 1 GPUs ***** Running training ***** Num examples = 2490 Batch size = 16 Max sequence length = 128 Num steps = 775 Num epochs = 5 Learning rate = 2e-05 Eval accuracy: 59.205776173285194% Best acc found. Saved model to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/distilbert-base-uncased-glue:rte-2020-06-29-14:58/. Eval accuracy: 60.28880866425993% Best acc found. Saved model to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/distilbert-base-uncased-glue:rte-2020-06-29-14:58/. Eval accuracy: 62.8158844765343% Best acc found. Saved model to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/distilbert-base-uncased-glue:rte-2020-06-29-14:58/. Eval accuracy: 65.34296028880865% Best acc found. Saved model to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/distilbert-base-uncased-glue:rte-2020-06-29-14:58/. Eval accuracy: 65.70397111913357% Best acc found. Saved model to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/distilbert-base-uncased-glue:rte-2020-06-29-14:58/. Saved tokenizer to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/distilbert-base-uncased-glue:rte-2020-06-29-14:58/. Wrote README to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/distilbert-base-uncased-glue:rte-2020-06-29-14:58/README.md. Wrote training args to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/distilbert-base-uncased-glue:rte-2020-06-29-14:58/train_args.json.