learning_rate: 1e-05
train_batch_size: 16
eval_batch_size: 16
seed: 42
distributed_type: multi-GPU
num_devices: 4
total_train_batch_size: 64
total_eval_batch_size: 64
optimizer: Use OptimizerNames.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_steps: 400
num_epochs: 5

Training Loss	Epoch	Step	Validation Loss	F1	Threshold	F1 At 05
2.7993	0.2825	200	2.7548	0.5499	0.52	0.4928
2.7447	0.5650	400	2.7005	0.5808	0.55	0.5523
2.5547	0.8475	600	2.5330	0.6192	0.55	0.6131
2.4451	1.1299	800	2.4878	0.6397	0.4300	0.6321
2.3049	1.4124	1000	2.5490	0.6525	0.52	0.6487
2.2329	1.6949	1200	2.5617	0.6493	0.4200	0.6336
2.1561	1.9774	1400	2.4758	0.6638	0.47	0.6507
1.8485	2.2599	1600	2.8087	0.6523	0.4300	0.6459
1.7005	2.5424	1800	3.0457	0.6475	0.5	0.6475
1.8078	2.8249	2000	2.8884	0.6492	0.4400	0.6464
1.4174	3.1073	2200	3.1910	0.6542	0.28	0.6438
1.3280	3.3898	2400	3.4488	0.6493	0.27	0.6340
1.3699	3.6723	2600	3.3882	0.6516	0.49	0.6511
1.3491	3.9548	2800	3.2727	0.6521	0.36	0.6453
1.0638	4.2373	3000	3.8112	0.6558	0.51	0.6552
1.0648	4.5198	3200	3.7631	0.6558	0.35	0.6486
1.0734	4.8023	3400	3.7629	0.6549	0.38	0.6481

DrinkIcedT/roberta-large_MBTI_P_MBTI_agg_balanced_25

roberta-large_MBTI_P_MBTI_agg_balanced_25