# Supervised Fine-Tuning (SFT): Pre-trained Base Model'i Instruct'a Dönüştürme — Llama-3-Instruct Anatomisi

> Source: https://sukruyusufkaya.com/learn/llm-muhendisligi/supervised-finetuning-sft-base-to-instruct
> Updated: 2026-08-14T19:08:33.735Z
> Category: LLM Mühendisliği
> Module: Modül 14: Fine-tuning — SFT, LoRA, QLoRA
**TLDR:** Supervised Fine-Tuning (SFT) anatomi: pre-trained base model → instruction-following model. Instruction dataset (Alpaca, OASST, Dolly), chat template uygulaması, loss masking (sadece response üzerinde loss), hyperparameter farkları (lr 1/10 of pre-train), Llama-3-Instruct production recipe, Türkçe için pratik fine-tune.