Writing logs to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/distilbert-base-cased-snli-2020-06-24-14:03/log.txt.
Loading nlp dataset snli, split train.
Loading nlp dataset snli, split validation.
Filtering samples with labels outside of [0, 1, 2].
Filtered 550152 train samples to 549367 points.
Filtered 10000 dev samples to 9842 points.
Loaded dataset. Found: 3 labels: ([0, 1, 2])
Loading transformers AutoModelForSequenceClassification: distilbert-base-cased
Tokenizing training data. (len: 549367)
Tokenizing eval data (len: 9842)
Loaded data and tokenized in 102.62434101104736s
Training model across 4 GPUs
***** Running training *****
	Num examples = 549367
	Batch size = 256
	Max sequence length = 128
	Num steps = 6435
	Num epochs = 3
	Learning rate = 2e-05
Eval accuracy: 86.09022556390977%
Best acc found. Saved model to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/distilbert-base-cased-snli-2020-06-24-14:03/.
Eval accuracy: 86.97419223735014%
Best acc found. Saved model to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/distilbert-base-cased-snli-2020-06-24-14:03/.
Eval accuracy: 87.68542979069295%
Best acc found. Saved model to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/distilbert-base-cased-snli-2020-06-24-14:03/.
Saved tokenizer <textattack.models.tokenizers.auto_tokenizer.AutoTokenizer object at 0x7fed9c3a8fa0> to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/distilbert-base-cased-snli-2020-06-24-14:03/.
Wrote README to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/distilbert-base-cased-snli-2020-06-24-14:03/README.md.
Wrote training args to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/distilbert-base-cased-snli-2020-06-24-14:03/train_args.json.