learning_rate: 5e-05
train_batch_size: 64
eval_batch_size: 64
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 256
optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.1
num_epochs: 20

Training Loss	Epoch	Step	Validation Loss	Accuracy
2.6034	0.9873	39	2.0544	0.4520
1.4429	2.0	79	0.7736	0.7849
0.8307	2.9873	118	0.5456	0.8413
0.6814	4.0	158	0.4881	0.8516
0.6199	4.9873	197	0.4614	0.8528
0.5578	6.0	237	0.4419	0.8615
0.5198	6.9873	276	0.4485	0.8603
0.4811	8.0	316	0.4355	0.8659
0.4568	8.9873	355	0.4182	0.8651
0.4268	10.0	395	0.4094	0.8702
0.4281	10.9873	434	0.4158	0.8706
0.4143	12.0	474	0.4078	0.8734
0.4009	12.9873	513	0.4066	0.8714
0.3642	14.0	553	0.4131	0.8683
0.3659	14.9873	592	0.4047	0.8726
0.3487	16.0	632	0.4054	0.8710
0.35	16.9873	671	0.4107	0.8722
0.3291	18.0	711	0.4099	0.8698
0.338	18.9873	750	0.4063	0.8718
0.3419	19.7468	780	0.4066	0.8702

Dev176/21BAI1229

21BAI1229