jialicheng/unlearn_samsum_t5-base_random_label_2_42

samsum_42

This model is a fine-tuned version of google/t5-v1_1-base on the samsum dataset. It achieves the following results on the evaluation set:

• Loss: 1.4477
• Rouge1: 49.9223
• Rouge2: 25.2665
• Rougel: 41.6791
• Rougelsum: 46.099
• Gen Len: 24.0465
• Test Rougel: 41.6292
• Df Rougel: 42.4048
• Unlearn Overall Rougel: 0.1122
• Unlearn Time: 1987.4286

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: 32
• 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	Overall Rougel	Unlearn Overall Rougel	Time
No log	1.0	461	1.4791	49.1393	24.3432	41.6808	45.2862	24.0257	-0.0160	-0.0160	-1
No log	2.0	922	1.4637	49.2648	24.9894	42.283	45.6087	25.1760	-0.0695	-0.0695	-1
1.8225	3.0	1383	1.4548	49.4165	24.9023	42.3375	45.5909	23.7897	-0.0531	-0.0531	-1
1.8225	4.0	1844	1.4512	49.3989	24.7566	41.9524	45.576	24.0257	0.0196	0.0196	-1
1.7145	5.0	2305	1.4477	49.9223	25.2665	42.4048	46.099	24.0465	0.1122	0.1122	-1

Framework versions

• Transformers 4.39.3
• Pytorch 2.3.0+cu121
• Datasets 2.19.0
• Tokenizers 0.15.2