flan-t5

This model is a fine-tuned version of google/flan-t5-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.6990
• Rouge1: 67.5685
• Rouge2: 60.4456
• Rougel: 66.1518
• Rougelsum: 66.4684
• Gen Len: 18.9254

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
No log	1.0	135	0.7189	66.6756	59.3564	65.186	65.5624	18.8657
No log	2.0	270	0.7073	67.6204	60.4625	66.1983	66.5708	18.9254
No log	3.0	405	0.7045	67.4291	60.2768	65.9877	66.3244	18.9254
0.6373	4.0	540	0.6990	67.5685	60.4456	66.1518	66.4684	18.9254
0.6373	5.0	675	0.7041	67.5685	60.4456	66.1518	66.4684	18.9254

Framework versions

• Transformers 4.26.1
• Pytorch 1.13.1+cu116
• Datasets 2.10.1
• Tokenizers 0.13.2