Hugging Face's logo

Holarissun
/
RM-HH-Gemma_harmless_human_20000_gemma2b

PEFT
Safetensors
trl
reward-trainer
Generated from Trainer
Model card Files Community
RM-HH-Gemma_harmless_human_20000_gemma2b / README.md
Holarissun's picture
Holarissun
End of training
83532c7 verified
preview code
|
raw
history blame contribute delete
2.5 kB
metadata
license: gemma
library_name: peft
tags:
  - trl
  - reward-trainer
  - generated_from_trainer
metrics:
  - accuracy
base_model: google/gemma-2b
model-index:
  - name: RM-HH-Gemma_harmless_human_20000_gemma2b
    results: []

RM-HH-Gemma_harmless_human_20000_gemma2b

This model is a fine-tuned version of google/gemma-2b on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.6196
  • Accuracy: 0.6678

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

Training results

Training Loss Epoch Step Validation Loss Accuracy
0.7635 0.06 250 0.7536 0.4972
0.7325 0.11 500 0.7235 0.5378
0.7285 0.17 750 0.7007 0.5673
0.7183 0.22 1000 0.6823 0.5973
0.6913 0.28 1250 0.6703 0.6113
0.6918 0.33 1500 0.6594 0.6263
0.6673 0.39 1750 0.6509 0.6368
0.6753 0.44 2000 0.6432 0.6398
0.6616 0.5 2250 0.6382 0.6503
0.672 0.56 2500 0.6336 0.6488
0.6457 0.61 2750 0.6308 0.6533
0.6705 0.67 3000 0.6276 0.6563
0.6534 0.72 3250 0.6247 0.6593
0.6176 0.78 3500 0.6228 0.6623
0.6597 0.83 3750 0.6210 0.6643
0.6404 0.89 4000 0.6201 0.6663
0.6199 0.94 4250 0.6196 0.6678

Framework versions

  • PEFT 0.9.0
  • Transformers 4.38.2
  • Pytorch 2.1.2
  • Datasets 2.18.0
  • Tokenizers 0.15.2