Edit model card

bb2bxglm_paraphrase_retrain

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

  • Loss: 0.0001

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: 5e-05
  • train_batch_size: 16
  • eval_batch_size: 16
  • seed: 42
  • gradient_accumulation_steps: 32
  • total_train_batch_size: 512
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 50
  • training_steps: 200
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss
8.59 0.0203 5 9.1062
8.1018 0.0406 10 8.0029
6.5083 0.0610 15 5.7627
4.3652 0.0813 20 3.0824
1.7875 0.1016 25 0.4579
0.2029 0.1219 30 0.0179
0.0116 0.1422 35 0.0019
0.0015 0.1625 40 0.0002
0.0005 0.1829 45 0.0001
0.0005 0.2032 50 0.0001
0.0003 0.2235 55 0.0001

Framework versions

  • Transformers 4.41.2
  • Pytorch 2.3.0+cu121
  • Datasets 2.19.2
  • Tokenizers 0.19.1
Downloads last month
3
Safetensors
Model size
776M params
Tensor type
F32
·
Inference API
This model can be loaded on Inference API (serverless).