learning_rate: 3e-05
train_batch_size: 8
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 32
optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
num_epochs: 10
mixed_precision_training: Native AMP
label_smoothing_factor: 0.1

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
2.5858	1.0	313	0.5200	0.9184	0.9185
1.8127	2.0	626	0.5246	0.9179	0.9178
1.5712	3.0	939	0.5342	0.9188	0.9186
1.4496	4.0	1252	0.5358	0.9201	0.9200
1.4135	5.0	1565	0.5396	0.9214	0.9214
1.4042	6.0	1878	0.5420	0.92	0.9200
1.4004	7.0	2191	0.5407	0.9196	0.9195
1.3981	8.0	2504	0.5414	0.9207	0.9205
1.3969	9.0	2817	0.5413	0.92	0.9199
1.3957	10.0	3130	0.5419	0.9195	0.9193

MariaFGI/Desired_run

desired_run