jialicheng/unlearn_samsum_t5-large_neggrad_4_42

samsum_42

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

• Loss: 87.1946
• Rouge1: 3.236
• Rouge2: 0.0575
• Rougel: 3.2242
• Rougelsum: 3.229
• Gen Len: 2.9648
• Test Rougel: 3.2242
• Df Rougel: 2.9036
• Unlearn Overall Rougel: 0.6603
• Unlearn Time: 2432.0397

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: 1e-05
• train_batch_size: 8
• 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	73	1.6698	42.7849	20.7407	35.3727	39.589	19.8516	0.5366	0.5366	-1
No log	2.0	146	7.5909	13.9304	3.6865	12.4648	13.2181	9.1979	0.4489	0.4489	-1
No log	3.0	219	87.1946	3.236	0.0575	2.9036	3.229	2.9648	0.6603	0.6603	-1
No log	4.0	292	144.7452	0.8156	0.0074	0.9739	0.8157	2.1211	0.4259	0.4259	-1
No log	5.0	365	160.1003	0.6105	0.0	0.6594	0.6071	2.0378	0.4786	0.4786	-1

Framework versions

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