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
6.4728	1.0	180	1.1006	0.7549	0.7603	0.7576
1.2169	2.0	360	0.6991	0.8482	0.8531	0.8507
0.8935	3.0	540	0.5917	0.8619	0.8665	0.8642
0.7535	4.0	720	0.5511	0.8717	0.8745	0.8731
0.6836	5.0	900	0.5198	0.8788	0.8826	0.8807
0.6199	6.0	1080	0.5202	0.8834	0.8853	0.8844
0.5531	7.0	1260	0.4874	0.8869	0.8881	0.8875
0.5299	8.0	1440	0.5102	0.8873	0.8885	0.8879
0.5007	9.0	1620	0.4844	0.8874	0.8886	0.8880
0.4774	10.0	1800	0.4792	0.8872	0.8886	0.8879

jhj1769/Llama1B_Full_stsb

Llama1B_Full_stsb