distilbert-1k

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

• Loss: 1.0986
• Accuracy: 0.3333

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-07
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 25

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.0989	1.0	6000	1.0986	0.3333
1.0993	2.0	12000	1.0986	0.3333
1.0992	3.0	18000	1.0986	0.3333
1.0993	4.0	24000	1.0986	0.3333
1.0985	5.0	30000	1.0986	0.3333
1.0988	6.0	36000	1.0986	0.3333
1.0993	7.0	42000	1.0986	0.3333
1.0983	8.0	48000	1.0986	0.3333
1.0983	9.0	54000	1.0986	0.3333
1.0982	10.0	60000	1.0986	0.3333
1.0986	11.0	66000	1.0986	0.3333
1.0985	12.0	72000	1.0986	0.3333
1.0983	13.0	78000	1.0986	0.3333
1.0987	14.0	84000	1.0986	0.3333
1.0992	15.0	90000	1.0986	0.3333
1.099	16.0	96000	1.0986	0.3333
1.0991	17.0	102000	1.0986	0.3333
1.0982	18.0	108000	1.0986	0.3333
1.0989	19.0	114000	1.0986	0.3333
1.099	20.0	120000	1.0986	0.3333
1.0991	21.0	126000	1.0986	0.3333
1.099	22.0	132000	1.0986	0.3333
1.099	23.0	138000	1.0986	0.3333
1.0984	24.0	144000	1.0986	0.3333
1.0985	25.0	150000	1.0986	0.3333

Framework versions

• Transformers 4.40.0
• Pytorch 2.0.0+cu117
• Datasets 2.19.0
• Tokenizers 0.19.1