LLMGUARD

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: 0.6730
  • Accuracy: 0.7628

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: 32
  • eval_batch_size: 8
  • seed: 42
  • 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_ratio: 0.1
  • num_epochs: 32
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Accuracy
2.2334 1.0 876 1.8027 0.4071
1.6018 2.0 1752 1.1836 0.6644
0.9703 3.0 2628 0.8345 0.7433
0.7557 4.0 3504 0.7281 0.7591
0.7028 5.0 4380 0.6809 0.7717
0.6372 6.0 5256 0.6530 0.7768
0.6074 7.0 6132 0.6411 0.7787
0.5809 8.0 7008 0.6292 0.7785
0.5594 9.0 7884 0.6255 0.7832
0.5452 10.0 8760 0.6334 0.7797
0.5334 11.0 9636 0.6225 0.7761
0.5091 12.0 10512 0.6347 0.7734
0.493 13.0 11388 0.6217 0.7794
0.4883 14.0 12264 0.6259 0.7782
0.4746 15.0 13140 0.6265 0.7725
0.4698 16.0 14016 0.6351 0.7728
0.4531 17.0 14892 0.6401 0.7734
0.4579 18.0 15768 0.6435 0.7731
0.4412 19.0 16644 0.6391 0.7710
0.4377 20.0 17520 0.6432 0.7705
0.4362 21.0 18396 0.6500 0.7681
0.4269 22.0 19272 0.6541 0.7674
0.4227 23.0 20148 0.6555 0.7658
0.4196 24.0 21024 0.6569 0.7678
0.4216 25.0 21900 0.6608 0.7660
0.4107 26.0 22776 0.6651 0.7672
0.4118 27.0 23652 0.6629 0.7645
0.4054 28.0 24528 0.6685 0.7624
0.4112 29.0 25404 0.6705 0.7642
0.3999 30.0 26280 0.6724 0.7625
0.405 31.0 27156 0.6721 0.7628
0.394 32.0 28032 0.6730 0.7628

Framework versions

  • Transformers 4.48.0.dev0
  • Pytorch 2.5.1+cu124
  • Datasets 3.2.0
  • Tokenizers 0.21.0
Downloads last month
19
Safetensors
Model size
67M params
Tensor type
F32
·
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Model tree for devhem/LLMGUARD

Finetuned
(7171)
this model