Qwen3-1.7B Backward LoRA Assignment 3
This is the backward model for Assignment 3.
- Base model: Qwen/Qwen3-1.7B
- Method: LoRA finetuning
- Training data: openassistant-guanaco
- Objective: generate a plausible user instruction from an assistant response
This model is used in Task 2 for self-augmentation.
