base_model, tags
| base_model | tags | |||
|---|---|---|---|---|
| Qwen/Qwen3-1.7B |
|
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. It is still being updated and trained.
The model is designed to generate various riddles out of a list of topics that are targeted for various ages. Upon generating the riddle, the model will also generate an answer but not state it. The user can input an answer, and the model will respond with Correct/Incorrect.
Limitations
Limited amount of riddle topics. Trained only on short answers for the riddle. Mostly short riddles will be generated. Model is strict on the answers, and can't evaluate the user's answer respectably.