mt5-small-finetuned-mt5-small-poem4e

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

• Loss: nan

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: 2e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 50
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
No log	1.0	121	nan
No log	2.0	242	nan
No log	3.0	363	nan
No log	4.0	484	nan
0.0	5.0	605	nan
0.0	6.0	726	nan
0.0	7.0	847	nan
0.0	8.0	968	nan
0.0	9.0	1089	nan
0.0	10.0	1210	nan
0.0	11.0	1331	nan
0.0	12.0	1452	nan
0.0	13.0	1573	nan
0.0	14.0	1694	nan
0.0	15.0	1815	nan
0.0	16.0	1936	nan
0.0	17.0	2057	nan
0.0	18.0	2178	nan
0.0	19.0	2299	nan
0.0	20.0	2420	nan
0.0	21.0	2541	nan
0.0	22.0	2662	nan
0.0	23.0	2783	nan
0.0	24.0	2904	nan
0.0	25.0	3025	nan
0.0	26.0	3146	nan
0.0	27.0	3267	nan
0.0	28.0	3388	nan
0.0	29.0	3509	nan
0.0	30.0	3630	nan
0.0	31.0	3751	nan
0.0	32.0	3872	nan
0.0	33.0	3993	nan
0.0	34.0	4114	nan
0.0	35.0	4235	nan
0.0	36.0	4356	nan
0.0	37.0	4477	nan
0.0	38.0	4598	nan
0.0	39.0	4719	nan
0.0	40.0	4840	nan
0.0	41.0	4961	nan
0.0	42.0	5082	nan
0.0	43.0	5203	nan
0.0	44.0	5324	nan
0.0	45.0	5445	nan
0.0	46.0	5566	nan
0.0	47.0	5687	nan
0.0	48.0	5808	nan
0.0	49.0	5929	nan
0.0	50.0	6050	nan

Framework versions

• Transformers 4.41.0
• Pytorch 2.2.1+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1