group: document-parsing-json-structured-data title: “Custom Dataset”
Custom Dataset
This page discusses the process and considerations for fine-tuning large language models (LLMs) on custom datasets. The content covers techniques to enhance model performance in specific domains or tasks by leveraging user-generated data.
Integrating Gemma-4 E2B Model with Custom Data
- Tutorial: Fine-Tune Gemma-4 on Your Own Dataset Locally: Step-by-Step Tutorial
- Author/Channel: Fahd Mirza
- URL: Fine-Tune Gemma-4 on Your Own Dataset Locally: Step-by-Step Tutorial
- Summary: Provides a practical guide to fine-tuning Google’s Gemma-4-E2B LLM locally using the
unslothlibrary, transforming it
Integrating OSS-20B Model with Custom Data
- Tutorial: Fine-Tune OSS-20B for Persona: Fahd Mirza
- Author/Channel: Fahd Mirza
- URL: Fine-Tune OSS-20B for Persona: Fahd Mirza
- Summary: Comprehensive guide to fine-tuning OpenAI’s GPT-OSS-20B open-weight model using custom data to embody a specific persona (Fahd Mirza), leveraging Hugging Face’s TRL library for supervised fine-tuning (SFT). Includes Ubuntu 22.04 LTS system setup and step-by-step implementation.
Backlink: 2026 04 14 Fahd Mirza fine tuning weights of OSS 20B
Source Notes
- 2026-04-14: Fahd Mirza - fine tuning weights of OSS-20B
- 2026-04-07: Fine-Tune Gemma-4 on Your Own Dataset Locally: Step-by-Step
- 2026-04-23: Excel