flan-t5-base-samsum

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

• Loss: 1.3717
• Rouge1: 47.4138
• Rouge2: 23.7051
• Rougel: 40.0689
• Rougelsum: 43.6335
• Gen Len: 17.0745

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
1.4543	1.0	1842	1.3866	46.7678	22.9561	39.1377	43.068	17.5531
1.3399	2.0	3684	1.3731	47.2896	24.0391	40.0258	43.7449	17.3297
1.2788	3.0	5526	1.3717	47.4138	23.7051	40.0689	43.6335	17.0745
1.2239	4.0	7368	1.3752	47.6634	24.1918	40.1481	44.0084	17.4676
1.1997	5.0	9210	1.3755	47.3995	23.7499	39.8569	43.6053	17.3932

Framework versions

• Transformers 4.39.3
• Pytorch 2.2.1+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2