Hugging Face's logo

beamaia
/
wandb-test-zephyr-qa

PEFT
Safetensors
Model card Files Community
wandb-test-zephyr-qa / README.md
beamaia's picture
beamaia
Model save
32e84bc verified
preview code
|
raw
history blame
1.55 kB
metadata
license: mit
base_model: HuggingFaceH4/zephyr-7b-beta
tags:
  - trl
  - sft
  - generated_from_trainer
datasets:
  - generator
model-index:
  - name: wandb-test-zephyr-qa
    results: []

wandb-test-zephyr-qa

This model is a fine-tuned version of HuggingFaceH4/zephyr-7b-beta on the generator dataset. It achieves the following results on the evaluation set:

  • Loss: 0.6266

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: 0.0002
  • train_batch_size: 2
  • eval_batch_size: 2
  • seed: 42
  • gradient_accumulation_steps: 8
  • total_train_batch_size: 16
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: cosine
  • lr_scheduler_warmup_ratio: 0.03
  • training_steps: 15
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss
No log 0.92 10 0.6407
No log 1.38 15 0.6266

Framework versions

  • Transformers 4.35.2
  • Pytorch 2.1.0+cu121
  • Datasets 2.13.0
  • Tokenizers 0.14.1