mt5-base-ft-rf-02

This model is a fine-tuned version of google/mt5-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.4229

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

Training results

Training Loss	Epoch	Step	Validation Loss
43.082	0.24	50	37.1069
34.6827	0.49	100	28.8296
21.0188	0.73	150	19.9344
18.3905	0.98	200	12.0120
14.342	1.22	250	9.2877
6.2116	1.46	300	6.1602
6.5474	1.71	350	4.6816
1.9222	1.95	400	2.6431
2.0579	2.2	450	1.2741
1.1028	2.44	500	0.9638
1.3341	2.68	550	0.8896
0.6531	2.93	600	0.8461
0.9805	3.17	650	0.7652
0.7167	3.41	700	0.7544
1.0224	3.66	750	0.7493
0.5367	3.9	800	0.7188
0.9352	4.15	850	0.6844
0.4927	4.39	900	0.6595
0.7141	4.63	950	0.6458
0.5773	4.88	1000	0.5911
0.4791	5.12	1050	0.5691
0.498	5.37	1100	0.5572
0.4306	5.61	1150	0.5315
0.334	5.85	1200	0.5123
0.3783	6.1	1250	0.4970
0.7719	6.34	1300	0.4774
0.3732	6.59	1350	0.4591
0.6203	6.83	1400	0.4482
0.4669	7.07	1450	0.4434
0.5568	7.32	1500	0.4307
0.6352	7.56	1550	0.4257
1.4137	7.8	1600	0.4229

Framework versions

• Transformers 4.31.0
• Pytorch 2.0.1+cu118
• Datasets 2.14.4
• Tokenizers 0.13.3