learning_rate: 2e-05
train_batch_size: 16
eval_batch_size: 32
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: linear
lr_scheduler_warmup_steps: 0.1
num_epochs: 10

Training Loss	Epoch	Step	Validation Loss	F1 Macro	F1 Weighted	Accuracy
0.9317	1.0	222	0.9921	0.3071	0.4899	0.5921
0.8962	2.0	444	0.9361	0.3459	0.5165	0.6118
0.7736	3.0	666	0.9622	0.3877	0.5312	0.5829
0.4920	4.0	888	1.2710	0.4271	0.5484	0.5724
0.1894	5.0	1110	1.8321	0.4198	0.5107	0.4974
0.1013	6.0	1332	2.0221	0.4363	0.5451	0.5513
0.0947	7.0	1554	2.1899	0.4209	0.5437	0.5632
0.0207	8.0	1776	2.3020	0.4270	0.5451	0.5605
0.0400	9.0	1998	2.3126	0.4222	0.5436	0.5605
0.0102	10.0	2220	2.3227	0.4165	0.5401	0.5592

akseljoonas/f2llm-sentiment-ablation-A

f2llm-sentiment-ablation-A