Edit model card

RM-HH-AllMix_helpful_gpt3_20000_gemma2b_shuffleFalse_extractchosenFalse

This model is a fine-tuned version of google/gemma-2b on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 0.0839
  • Accuracy: 0.9876

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: 1.41e-05
  • train_batch_size: 4
  • eval_batch_size: 8
  • seed: 42
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 16
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 2.0

Training results

Training Loss Epoch Step Validation Loss Accuracy
0.7074 0.17 250 0.3710 0.8750
0.6147 0.33 500 0.1958 0.9673
0.5749 0.5 750 0.1424 0.9763
0.5776 0.67 1000 0.1249 0.9827
0.5601 0.84 1250 0.1087 0.9868
0.5549 1.0 1500 0.0982 0.9887
0.5465 1.17 1750 0.0941 0.9876
0.5494 1.34 2000 0.0887 0.9872
0.54 1.51 2250 0.0858 0.9895
0.5375 1.67 2500 0.0848 0.9891
0.5266 1.84 2750 0.0839 0.9876

Framework versions

  • PEFT 0.10.0
  • Transformers 4.38.2
  • Pytorch 2.1.2+cu121
  • Datasets 2.18.0
  • Tokenizers 0.15.2
Downloads last month
0
Inference API
Unable to determine this model’s pipeline type. Check the docs .

Model tree for Holarissun/RM-HH-AllMix_helpful_gpt3_20000_gemma2b_shuffleFalse_extractchosenFalse

Base model

google/gemma-2b
Adapter
(23294)
this model