p-tuning-gpt2-large-with-sst2

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

• Loss: 0.3065
• Accuracy: 0.8968

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: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.4485	1.0	4210	0.3556	0.8406
0.2677	2.0	8420	0.3760	0.8693
0.2717	3.0	12630	0.3264	0.8727
0.3687	4.0	16840	0.3410	0.8807
0.3483	5.0	21050	0.3075	0.8865
0.2969	6.0	25260	0.3315	0.8888
0.3577	7.0	29470	0.2875	0.8853
0.4036	8.0	33680	0.3143	0.8899
0.404	9.0	37890	0.2858	0.8911
0.2797	10.0	42100	0.3035	0.8876
0.3328	11.0	46310	0.3168	0.8876
0.2345	12.0	50520	0.3063	0.8933
0.3154	13.0	54730	0.2972	0.8911
0.2937	14.0	58940	0.2994	0.8933
0.2123	15.0	63150	0.2938	0.8899
0.295	16.0	67360	0.3087	0.8945
0.1924	17.0	71570	0.3031	0.8933
0.2415	18.0	75780	0.3067	0.8979
0.1876	19.0	79990	0.3080	0.8979
0.3891	20.0	84200	0.3065	0.8968

Framework versions

• PEFT 0.7.1
• Transformers 4.40.1
• Pytorch 2.5.0
• Datasets 3.0.1
• Tokenizers 0.19.1