dnd

This model is a fine-tuned version of neal49/distilbert-sst2-runglue on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.4907
• Accuracy: 0.8246

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: 2e-06
• 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: linear
• num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	15	0.6813	0.5789
No log	2.0	30	0.6725	0.5789
No log	3.0	45	0.6588	0.6140
No log	4.0	60	0.6536	0.6140
No log	5.0	75	0.6524	0.6140
No log	6.0	90	0.6426	0.6140
No log	7.0	105	0.6333	0.6316
No log	8.0	120	0.6148	0.6491
No log	9.0	135	0.6081	0.6491
No log	10.0	150	0.5724	0.7018
No log	11.0	165	0.5984	0.6842
No log	12.0	180	0.5328	0.7368
No log	13.0	195	0.5419	0.7719
No log	14.0	210	0.5271	0.7719
No log	15.0	225	0.5188	0.7719
No log	16.0	240	0.5283	0.7719
No log	17.0	255	0.5012	0.7719
No log	18.0	270	0.4863	0.7895
No log	19.0	285	0.5329	0.7895
No log	20.0	300	0.4861	0.8070
No log	21.0	315	0.5065	0.8246
No log	22.0	330	0.4864	0.8070
No log	23.0	345	0.5060	0.8246
No log	24.0	360	0.4752	0.8246
No log	25.0	375	0.4983	0.8246
No log	26.0	390	0.4925	0.8246
No log	27.0	405	0.4774	0.8246
No log	28.0	420	0.4804	0.8246
No log	29.0	435	0.4927	0.8246
No log	30.0	450	0.4907	0.8246

Framework versions

• Transformers 4.32.1
• Pytorch 2.3.0+cu121
• Datasets 2.20.0
• Tokenizers 0.13.2