learning_rate: 0.0002
train_batch_size: 2
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 8
total_train_batch_size: 16
optimizer: Use 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: 2
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss
1.6047	0.1016	100	1.5466
1.3265	0.2032	200	1.1507
1.0355	0.3048	300	0.8742
0.8164	0.4063	400	0.6808
0.6442	0.5079	500	0.5399
0.5446	0.6095	600	0.4273
0.4727	0.7111	700	0.3645
0.4179	0.8127	800	0.3171
0.3762	0.9143	900	0.2857
0.3514	1.0152	1000	0.2714
0.3191	1.1168	1100	0.2552
0.301	1.2184	1200	0.2464
0.2894	1.32	1300	0.2395
0.29	1.4216	1400	0.2353
0.2724	1.5232	1500	0.2300
0.2628	1.6248	1600	0.2283
0.2629	1.7263	1700	0.2251
0.2612	1.8279	1800	0.2228
0.2452	1.9295	1900	0.2209

vishnu2003/phi2-finetuned-textbook-21

phi2-finetuned-textbook-21