learning_rate: 0.0005
train_batch_size: 16
eval_batch_size: 8
seed: 42
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: cosine
num_epochs: 2

Training Loss	Epoch	Step	Validation Loss
3.3196	0.1067	200	3.2841
3.1656	0.2133	400	3.1896
3.1198	0.32	600	3.1382
3.0052	0.4267	800	3.1023
3.0033	0.5333	1000	3.0762
3.0341	0.64	1200	3.0543
2.9969	0.7467	1400	3.0364
2.9728	0.8533	1600	3.0241
2.964	0.96	1800	3.0150
2.9292	1.0667	2000	3.0057
2.9275	1.1733	2200	2.9986
2.9083	1.28	2400	2.9915
2.9059	1.3867	2600	2.9884
2.901	1.4933	2800	2.9842
2.8942	1.6	3000	2.9818
2.8835	1.7067	3200	2.9802
2.89	1.8133	3400	2.9792
2.8887	1.92	3600	2.9789

eigenben/smollm-business-news-generator-lora

smollm-business-news-generator-lora