learning_rate: 6e-05
train_batch_size: 32
eval_batch_size: 8
seed: 42
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: cosine
num_epochs: 5

Training Loss	Epoch	Step	Validation Loss	Accuracy
7.8963	0.2096	100	2.9564	0.7577
1.6375	0.4193	200	1.0919	0.8806
0.8239	0.6289	300	0.6403	0.9335
0.5738	0.8386	400	0.5019	0.9448
0.3915	1.0482	500	0.4918	0.9452
0.1938	1.2579	600	0.4370	0.9548
0.2045	1.4675	700	0.4937	0.9435
0.1874	1.6771	800	0.4477	0.9568
0.1804	1.8868	900	0.4118	0.9581
0.1237	2.0964	1000	0.3573	0.9616
0.076	2.3061	1100	0.3772	0.9574
0.0834	2.5157	1200	0.3337	0.9652
0.0713	2.7254	1300	0.3032	0.9658
0.0514	2.9350	1400	0.3009	0.9661
0.0448	3.1447	1500	0.2892	0.9661
0.0425	3.3543	1600	0.2864	0.9671
0.0341	3.5639	1700	0.2859	0.9642
0.0389	3.7736	1800	0.2763	0.9677
0.0409	3.9832	1900	0.2682	0.9668
0.0266	4.1929	2000	0.2624	0.9674
0.0265	4.4025	2100	0.2610	0.9684
0.0267	4.6122	2200	0.2592	0.9684
0.027	4.8218	2300	0.2589	0.9681

WilliamHH/Assignment4_Distilled_ModernBERT

Assignment4_Distilled_ModernBERT