distilbert_lda_100_v1

This model is a fine-tuned version of on the gokulsrinivasagan/processed_wikitext-103-raw-v1-ld-100 dataset. It achieves the following results on the evaluation set:

• Loss: 5.9919
• Accuracy: 0.5720

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: 96
• eval_batch_size: 96
• seed: 10
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 10000
• num_epochs: 25

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
10.6393	4.1982	10000	10.5707	0.1522
7.6607	8.3963	20000	7.4539	0.3924
6.5199	12.5945	30000	6.4104	0.5232
6.2188	16.7926	40000	6.1147	0.5560
6.0786	20.9908	50000	6.0022	0.5686

Framework versions

• Transformers 4.46.1
• Pytorch 2.2.0+cu121
• Datasets 3.1.0
• Tokenizers 0.20.1