learning_rate: 5e-05
train_batch_size: 4
eval_batch_size: 1
seed: 42
gradient_accumulation_steps: 8
total_train_batch_size: 32
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: 100

Training Loss	Epoch	Step	Validation Loss	Chrf++	Gen Len
12.4017	1.9	38	1.6815	12.0166	124.5147
12.4017	3.8	76	1.5451	11.8662	132.1176
7.7771	5.7	114	1.5389	12.7344	111.2647
7.7771	7.6	152	1.5876	13.5778	104.5809
7.7771	9.5	190	1.7153	14.4162	105.5809
2.1743	11.4	228	1.8695	14.4757	109.4338
2.1743	13.3	266	1.9990	14.9958	103.125
0.6704	15.2	304	2.1507	14.4827	106.4118
0.6704	17.1	342	2.3182	14.3248	118.8456
0.6704	19.0	380	2.3955	14.8547	115.7794

Tsedeniya/finetuned_glosslm_per

finetuned_glosslm_per