Finetuned Glosslm Ain

finetuned_glosslm_ain This model is a fine-tuned version of lecslab/glosslm on an unknown dataset. It achieves the following results on the evaluation set: - Loss: 0.3371 - Chrf++: 74.6308 - Gen Len: 36.5459 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - 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 results | Training Loss | Epoch | Step | Validation Loss | Chrf++ | Gen Len | |:-------------:|:-------:|:----:|:---------------:|:-------:|:-------:| | 3.0812 | 1.9941 | 420 | 0.4395 | 53.4455 | 33.5321 | | 1.2481 | 3.9845 | 840 | 0.2439 | 68.2833 | 36.6101 | | 0.5980 | 5.9750 | 1260 | 0.2333 | 74.0715 | 37.4358 | | 0.3184 | 7.9655 | 1680 | 0.2656 | 74.7345 | 37.445 | | 0.2239 | 9.9560 | 2100 | 0.2351 | 77.5073 | 37.6422 | | 0.1788 | 11.9465 | 2520 | 0.3110 | 74.8554 | 37.0 | | 0.1477 | 13.9370 | 2940 | 0.2845 | 77.1494 | 37.3853 | | 0.1343 | 15.9275 | 3360 | 0.2478 | 78.0687 | 37.3119 | | 0.1266 | 17.9180 | 3780 | 0.2438 | 78.2725 | 39.3716 | | 0.1301 | 19.9084 | 4200 | 0.2342 | 77.3823 | 37.2248 | | 0.1299 | 21.8989 | 4620 | 0.2580 | 79.0047 | 37.367 | | 0.1243 | 23.8894 | 5040 | 0.3013 | 75.2974 | 37.5734 | | 0.1282 | 25.8799 | 5460 | 0.5167 | 74.9527 | 37.2982 | | 0.1245 | 27.8704 | 5880 | 0.3371 | 74.6308 | 36.5459 | ### Framework versions - Transformers 5.13.0 - Pytorch 2.7.1+cu126 - Datasets 5.0.0 - Tokenizers 0.22.2