sujalgawas/aya-json-sft — Tiny Aya Global JSON Schema SFT Adapter
Completion-masked QLoRA SFT adapter (r=32, alpha=64, 5 epochs) trained with AlignTune (Lexsi-Labs/aligntune) on sujalgawas/aya-json-sft-dataset.
- Downloads last month
- 128
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support