dkat0's picture
Update README.md
ab0c8ca verified
metadata
license: other
base_model: Qwen/Qwen2-7B
tags:
  - llama-factory
  - full
  - generated_from_trainer
metrics:
  - accuracy
model-index:
  - name: sft2
    results: []

qwen-2-7b-instruct-self-align

This model is a fine-tuned version of Qwen/Qwen2-7B on the qwen2-self-align-instruct dataset.

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

  • learning_rate: 0.0001
  • train_batch_size: 1
  • eval_batch_size: 1
  • seed: 42
  • distributed_type: multi-GPU
  • num_devices: 4
  • gradient_accumulation_steps: 2
  • total_train_batch_size: 8
  • total_eval_batch_size: 4
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: cosine
  • lr_scheduler_warmup_ratio: 0.1
  • num_epochs: 3.0

Framework versions

  • Transformers 4.42.3
  • Pytorch 2.3.1+cu121
  • Datasets 2.20.0
  • Tokenizers 0.19.1