learning_rate: 2e-05
train_batch_size: 2
eval_batch_size: 1
seed: 42
distributed_type: multi-GPU
gradient_accumulation_steps: 10
total_train_batch_size: 20
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

Training Loss	Epoch	Step	Validation Loss	Mse	Mae	R2	Spearman Correlation
2.3898	1.0	591	0.1214	0.1214	0.1854	-0.1100	nan
1.1133	2.0	1182	0.1133	0.1133	0.1831	-0.0364	nan
1.0841	3.0	1773	0.1271	0.1271	0.2454	-0.1619	nan
1.0847	4.0	2364	0.1099	0.1099	0.1501	-0.0054	nan
1.0609	5.0	2955	0.1116	0.1116	0.1376	-0.0202	nan
1.0033	6.0	3546	0.1205	0.1205	0.1996	-0.1023	-0.0018
0.8877	7.0	4137	0.1248	0.1248	0.1907	-0.1411	-0.0071
0.7012	8.0	4728	0.1227	0.1227	0.1719	-0.1218	-0.0221
0.5185	9.0	5319	0.1290	0.1290	0.1722	-0.1798	-0.0338
0.3866	10.0	5910	0.1358	0.1358	0.1777	-0.2418	-0.0140

athirorg/USS-reward-model-utterance-baseline

USS-reward-model-utterance-baseline