jialicheng/unlearn_samsum_t5-base_salun_8_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: 1.4901
• Rouge1: 49.5476
• Rouge2: 24.7493
• Rougel: 41.0919
• Rougelsum: 45.8012
• Gen Len: 24.1732
• Test Rougel: 41.0548
• Df Rougel: 42.0937
• Unlearn Overall Rougel: -0.0194
• Unlearn Time: 4698.0502

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: 64
• 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	460	1.4927	48.7926	24.1267	41.6972	45.063	22.6732	-0.0532	-0.0532	-1
No log	2.0	920	1.4901	49.5476	24.7493	42.0937	45.8012	24.1732	-0.0194	-0.0194	-1
1.9781	3.0	1380	1.4806	49.2612	24.5866	41.9899	45.4343	23.9427	-0.0304	-0.0304	-1
1.9781	4.0	1840	1.4745	49.4623	24.6205	42.434	45.5781	23.8815	-0.2274	-0.2274	-1
1.8807	5.0	2300	1.4719	48.9957	24.3369	42.4089	45.2936	23.7513	-0.3002	-0.3002	-1

Framework versions

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