distilbert_qlora

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

• Loss: 0.6173
• Accuracy: 0.8934

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: 0.0002
• train_batch_size: 1
• eval_batch_size: 1
• seed: 42
• gradient_accumulation_steps: 16
• total_train_batch_size: 16
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: constant
• lr_scheduler_warmup_ratio: 0.03
• training_steps: 1875

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.7159	0.0958	187	0.7154	0.8574
0.6637	0.1916	374	0.6786	0.8730
0.6898	0.2874	561	0.6563	0.8795
0.7074	0.3832	748	0.6493	0.8833
0.6035	0.4790	935	0.6531	0.8857
0.6173	0.5748	1122	0.6369	0.8866
0.6508	0.6706	1309	0.6367	0.8889
0.6849	0.7664	1496	0.6180	0.8921
0.6642	0.8622	1683	0.6231	0.8932
0.6412	0.9580	1870	0.6173	0.8934

Framework versions

• Transformers 4.43.4
• Pytorch 2.3.1+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1