jialicheng/unlearn_samsum_t5-large_neggrad_6_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: 290.3438
• Rouge1: 0.2901
• Rouge2: 0.0
• Rougel: 0.294
• Rougelsum: 0.2889
• Gen Len: 2.0
• Test Rougel: 0.294
• Df Rougel: 0.2241
• Unlearn Overall Rougel: 0.5350
• Unlearn Time: 1103.6610

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	110	2.4633	19.9456	10.1732	18.2153	18.7859	7.7370	0.4185	0.4185	-1
No log	2.0	220	165.9415	0.3318	0.0	0.2978	0.3277	2.0052	0.5184	0.5184	-1
No log	3.0	330	290.3438	0.2901	0.0	0.2241	0.2889	2.0	0.5350	0.5350	-1
No log	4.0	440	343.0455	0.2678	0.0	0.2125	0.2684	2.0	0.5295	0.5295	-1
No log	5.0	550	352.4384	0.2678	0.0	0.2125	0.2684	2.0	0.5295	0.5295	-1

Framework versions

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