I will fine tune llm with custom data using lora or qlora

Sri Lanka

I speak Sinhala, English, Spanish, French

87 orders completed

Software Engineer

I'm Ramitha Iddamalgoda. I'm a software engineer with years of experience. I'm wiling to designs any kind of websites, applications such as business, company, e-commerce, booking and many more. And I ...

Level 2

Has met high performance criteria and has a proven track record for meeting client expectations.

About this Gig

I will fine-tune your Large Language Models (LLMs) using custom datasets to match your specific domain and requirements.


What I Offer:

  • Custom LLM fine-tuning
  • Dataset creation, cleaning and preprocessing
  • LoRA / QLoRA fine-tuning
  • Supervised Fine-Tuning (SFT)
  • Hyperparameter optimization
  • Model evaluation and performance comparison


Supported Models:

Llama, Qwen, Mistral, Gemma, and other Hugging Face models.


Tools & Technologies:

Python, PyTorch, Hugging Face, Transformers, PEFT, TRL and Datasets.


My Workflow:

  1. Dataset preparation and validation
  2. Pretrained model selection
  3. Fine-tuning and training
  4. Model evaluation
  5. Delivery of trained weights/adapters and source code


No Dataset? No Problem! I can create, clean and structure custom training datasets for your use case.


Every package includes actual model fine-tuning and dataset preparation.


Message me to discuss your requirements!

Programming Language:

Python

•

Javascript

•

Java

•

C++

•

Tensorflow

AI Model Frameworks & Tools:

Hugging Face Transformers

Data Type:

Text

•

Images

•

Multimodal

AI Engine:

GPT

•

Gemini

•

DALL-E

•

DeepSeek

•

Midjourney

•

TensorFlow

•

Llama

My Portfolio