cagataydev/doer-qwen3-1.7b-test

doer — fine-tune of Qwen/Qwen3-1.7B

Trained on cagataydev/doer-training — 468 text turns.

• base: Qwen/Qwen3-1.7B
• method: LoRA r=16 α=32 (merged)
• steps: 50, lr=0.0002
• dataset: 468 train / 53 eval records

Use

pip install doer-cli
DOER_PROVIDER=transformers \
  DOER_MODEL=cagataydev/doer-qwen3-1.7b-test \
  doer "what is doer"

Generated by doer/hf_jobs/train_text_lora.py on HF Jobs.