Edit model card

deberta-v3-large-harmfulness

This model is a fine-tuned version of microsoft/deberta-v3-large on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 1.1439
  • Mse: 1.1439

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-05
  • train_batch_size: 8
  • eval_batch_size: 4
  • seed: 42
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 32
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: constant
  • num_epochs: 1.0

Training results

Training Loss Epoch Step Validation Loss Mse
2.1055 0.06 50 1.7626 1.7626
1.6818 0.12 100 1.4659 1.4659
1.4062 0.18 150 1.6323 1.6323
1.4777 0.24 200 1.3776 1.3776
1.4106 0.3 250 1.3044 1.3044
1.2702 0.37 300 1.3792 1.3792
1.4448 0.43 350 1.3048 1.3048
1.3582 0.49 400 1.2485 1.2485
1.2357 0.55 450 1.3709 1.3709
1.1075 0.61 500 1.2538 1.2538
1.2406 0.67 550 1.1996 1.1996
1.2309 0.73 600 1.2671 1.2671
1.194 0.79 650 1.4326 1.4326
1.2135 0.85 700 1.1763 1.1763
1.2196 0.91 750 1.1439 1.1439
1.1983 0.97 800 1.1711 1.1711

Framework versions

  • Transformers 4.30.1
  • Pytorch 2.0.1+cu117
  • Datasets 2.12.0
  • Tokenizers 0.13.3
Downloads last month
415
Safetensors
Model size
435M params
Tensor type
I64
·
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.