mT5-lithuanian-simplifier

This model is a fine-tuned version of google/mt5-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0687
• Rouge1: 0.7322
• Rouge2: 0.5833
• Rougel: 0.7261
• Gen Len: 34.6325

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: 0.0001
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 8

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Gen Len
23.3756	0.48	200	16.6054	0.011	0.0006	0.0105	512.0
1.435	0.96	400	0.5767	0.1436	0.0387	0.1307	39.0358
0.2269	1.44	600	0.1658	0.4744	0.317	0.4631	34.6325
0.1709	1.91	800	0.1030	0.5865	0.4231	0.577	34.6325
0.1311	2.39	1000	0.0923	0.6245	0.4626	0.6173	34.6325
0.1196	2.87	1200	0.0851	0.6759	0.5159	0.6681	34.6325
0.1156	3.35	1400	0.0761	0.6927	0.5288	0.6845	34.6325
0.0897	3.83	1600	0.0756	0.698	0.5352	0.6906	34.6325
0.1149	4.31	1800	0.0738	0.7085	0.55	0.7022	34.6325
0.0967	4.78	2000	0.0725	0.7187	0.5632	0.712	34.6325
0.1097	5.26	2200	0.0703	0.7214	0.5645	0.7143	34.6325
0.0852	5.74	2400	0.0708	0.7241	0.5724	0.7176	34.6325
0.0537	6.22	2600	0.0693	0.7285	0.578	0.7217	34.6325
0.0866	6.7	2800	0.0698	0.7277	0.5795	0.7216	34.6325
0.0692	7.18	3000	0.0685	0.7336	0.5849	0.7274	34.6325
0.0786	7.66	3200	0.0687	0.7322	0.5833	0.7261	34.6325

Framework versions

• Transformers 4.36.2
• Pytorch 2.1.1
• Datasets 2.16.1
• Tokenizers 0.15.0