learning_rate: 3e-05
train_batch_size: 16
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 32
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
num_epochs: 10
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss	Pearson	Spearman	Combined Score
10.9633	1.0	180	1.0169	0.7515	0.7523	0.7519
1.2912	2.0	360	0.6755	0.8444	0.8472	0.8458
0.8139	3.0	540	0.5401	0.8759	0.8765	0.8762
0.6555	4.0	720	0.5099	0.8813	0.8830	0.8821
0.5563	5.0	900	0.4942	0.8865	0.8880	0.8872
0.4952	6.0	1080	0.4800	0.8910	0.8920	0.8915
0.4389	7.0	1260	0.4648	0.8930	0.8940	0.8935
0.3996	8.0	1440	0.4630	0.8928	0.8940	0.8934
0.3746	9.0	1620	0.4543	0.8938	0.8950	0.8944
0.3316	10.0	1800	0.4513	0.8944	0.8954	0.8949

jhj1769/Llama3B_LoRA_stsb

Llama3B_LoRA_stsb