Hugging Face's logo

cyberpole
/
Meta-Llama-3.1-8B-Instruct-mergedORPO

PEFT
TensorBoard
Safetensors
trl
orpo
Generated from Trainer
Model card Files Metrics Community
Meta-Llama-3.1-8B-Instruct-mergedORPO / README.md
k-r-l's picture
k-r-l
Training in progress, step 1
bf74f4f verified
preview code
|
raw
history blame
1.21 kB
metadata
base_model: bigscience/bloomz-560m
library_name: peft
license: bigscience-bloom-rail-1.0
tags:
  - trl
  - orpo
  - generated_from_trainer
model-index:
  - name: bloomz-560m-mergedORPO
    results: []

bloomz-560m-mergedORPO

This model is a fine-tuned version of bigscience/bloomz-560m on an unknown dataset.

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: 0.0001
  • train_batch_size: 8
  • eval_batch_size: 1
  • seed: 42
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 32
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • training_steps: 5

Training results

Framework versions

  • PEFT 0.12.0
  • Transformers 4.44.0
  • Pytorch 2.4.0+cu121
  • Datasets 2.21.0
  • Tokenizers 0.19.1