flan-t5-base-finetuned-en-to-no-test

This model is a fine-tuned version of google/flan-t5-base on the opus_books dataset. It achieves the following results on the evaluation set:

• Loss: nan
• Bleu: 0.8116
• Gen Len: 18.1543

Model description

Broken model. Check train results.

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: 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: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu	Gen Len
0.0	1.0	788	nan	0.8116	18.1543
0.0	2.0	1576	nan	0.8116	18.1543
0.0	3.0	2364	nan	0.8116	18.1543
0.0	4.0	3152	nan	0.8116	18.1543
0.0	5.0	3940	nan	0.8116	18.1543
0.0	6.0	4728	nan	0.8116	18.1543
0.0	7.0	5516	nan	0.8116	18.1543
0.0	8.0	6304	nan	0.8116	18.1543
0.0	9.0	7092	nan	0.8116	18.1543
0.0	10.0	7880	nan	0.8116	18.1543

Framework versions

• Transformers 4.15.0
• Pytorch 1.12.1+cu113
• Datasets 2.3.2
• Tokenizers 0.10.3