flan-t5-xl-clickbait-spoilering

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

• Loss: 0.1774
• Rouge1: 0.4805
• Rouge2: 0.3398
• Rougel: 0.4696
• Bleu: 0.3462
• Gen Len: 24.2312

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: 0.0003
• train_batch_size: 1
• eval_batch_size: 1
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 6

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Bleu	Gen Len
0.1299	1.0	3200	0.1556	0.3671	0.2366	0.3594	0.1378	41.9787
0.0869	2.0	6400	0.1367	0.4271	0.2860	0.4177	0.2620	23.4587
0.0594	3.0	9600	0.1442	0.4436	0.3015	0.4321	0.2795	24.6212
0.0196	4.0	12800	0.1555	0.4378	0.3007	0.4239	0.2425	32.0875
0.0186	5.0	16000	0.1655	0.4565	0.3164	0.4441	0.2904	26.15
0.0032	6.0	19200	0.1774	0.4805	0.3398	0.4696	0.3462	24.2312

Framework versions

• Transformers 4.36.2
• Pytorch 2.1.1+cu121
• Datasets 2.16.1
• Tokenizers 0.15.0