|
--- |
|
library_name: peft |
|
base_model: Qwen/Qwen1.5-0.5B-Chat |
|
language: |
|
- en |
|
pipeline_tag: text-generation |
|
tags: |
|
- chat |
|
widget: |
|
- text: "What is the sum of the first 10 positive integers?" |
|
--- |
|
|
|
# Qwen1.5-0.5B-Chat with EPFL DPO fine-tuning |
|
|
|
Qwen1.5-0.5B-Chat DPO fine-tuned on open-ended and multiple choice questions from different EPFL courses and the Orca Math dataset that consists of ~200K grade school math word problems. |
|
|
|
## Model Details |
|
|
|
### Model Description |
|
|
|
The model was developed during the course Modern Natural Language Processing (CS-552). |
|
Its aim is to fine-tune the base model (Qwen/Qwen1.5-0.5B-Chat) to accurately |
|
answer open-ended and multiple-choice questions from various EPFL courses and Orca Math dataset. |
|
|
|
- **Developed by:** Emma Lise Boehly, Ahmed Aziz Ben Haj Hmida and Jan Kokla |
|
- **Finetuned from model:** Qwen/Qwen1.5-0.5B-Chat |
|
|
|
## Training Details |
|
|
|
### Training Data |
|
|
|
HuggingFace dataset : microsoft/orca-math-word-problems-200k |
|
The EPFL dataset is not publicly available. |
|
|
|
### Training Procedure |
|
|
|
#### Training Hyperparameters |
|
|
|
- **Training regime:** cDPO with bf16 mixed precision, $\beta=0.2$, $lr=3 \times 10^{-6}$, and $label_smoothing=0.2$ |
|
|
|
- PEFT 0.10.0 |
