28 lines
664 B
Markdown
28 lines
664 B
Markdown
|
---
|
||
|
|
base_model: Qwen/Qwen3-1.7B
|
||
|
|
tags:
|
||
|
|
- peft
|
||
|
|
- lora
|
||
|
|
- text-generation
|
||
|
|
---
|
||
|
|
|
||
|
|
# Qwen3-1.7B-riddles
|
||
|
|
|
||
|
|
## Model Description
|
||
|
|
Fine-tuned from `Qwen/Qwen3-1.7B` using QLoRA (4-bit) with supervised fine-tuning.
|
||
|
|
|
||
|
|
## Training Details
|
||
|
|
- Dataset: `jujuloaiza/riddletraining`
|
||
|
|
- LoRA rank: 16, alpha: 32
|
||
|
|
- Epochs: 3, Learning rate: 0.0002
|
||
|
|
|
||
|
|
## Intended Use
|
||
|
|
|
||
|
|
This model is a test model used for the CS-394/594 class at DigiPen.
|
||
|
|
|
||
|
|
The model is designed to generate various riddles. It has been trained to produce riddles, with an answer in mind. Then to evaluate the based off the ridle.
|
||
|
|
|
||
|
|
## Limitations
|
||
|
|
|
||
|
|
Limited amount of riddle topics. Trained only on short answers for the riddle.
|