mt5-base-lit-simplif

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: 1.1660
• Rouge1: 45.9067
• Rouge2: 33.3486
• Rougel: 43.6217
• Rougelsum: 43.7128
• Gen Len: 17.3389
• Bleu: 11.5761
• Corpus Bleu: 8.3337

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: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len	Bleu	Corpus Bleu
No log	1.0	209	1.7570	22.6673	14.3228	21.6876	21.634	11.9379	4.355	5.0525
No log	2.0	418	1.3636	40.2937	28.0627	38.6255	38.8148	16.9737	10.3859	8.7596
4.4247	3.0	627	1.2516	40.7671	28.2819	38.4458	38.598	18.0692	8.7679	8.1542
4.4247	4.0	836	1.1865	45.1478	32.7329	42.788	42.9494	17.4988	11.2435	8.2933
1.7397	5.0	1045	1.1660	45.9067	33.3486	43.6217	43.7128	17.3389	11.5761	8.3337

Framework versions

• Transformers 4.33.0
• Pytorch 2.1.1
• Datasets 2.14.4
• Tokenizers 0.13.3