gpt2-text-simplification_1e4_adafactor_newsela

This model is a fine-tuned version of gpt2 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.3465

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: 0.0001
• train_batch_size: 64
• eval_batch_size: 64
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss
0.7662	1.0	1605	0.8757
0.6538	2.0	3210	0.9019
0.5663	3.0	4815	0.9554
0.4961	4.0	6420	0.9990
0.4299	5.0	8025	1.0271
0.3853	6.0	9630	1.0547
0.3482	7.0	11235	1.1090
0.3152	8.0	12840	1.1387
0.2903	9.0	14445	1.1853
0.2655	10.0	16050	1.2088
0.2477	11.0	17655	1.2168
0.232	12.0	19260	1.2426
0.2192	13.0	20865	1.2522
0.2078	14.0	22470	1.2855
0.198	15.0	24075	1.3048
0.19	16.0	25680	1.3117
0.1834	17.0	27285	1.3262
0.1777	18.0	28890	1.3360
0.1733	19.0	30495	1.3440
0.1702	20.0	32100	1.3465

Framework versions

• Transformers 4.30.2
• Pytorch 2.0.1+cu118
• Datasets 2.13.1
• Tokenizers 0.13.3