Finetuning

👋 Sign in for the ability to sort posts by relevant, latest, or top.

PRANJUL RATHOUR

Sep 6

From adapter to deployment: merging LoRA weights and serving with vLLM or a Space

#finetuning #deployment #vllm #serving

3 min read

PRANJUL RATHOUR

Sep 6

Small language models (1B–3B): what they're good for after fine-tuning

#smallmodels #finetuning #llm #deployment

3 min read

PRANJUL RATHOUR

Sep 6

Fine-tuning on a laptop or a free GPU: a VRAM budget walkthrough

#finetuning #vram #qlora #budget

3 min read

PRANJUL RATHOUR

Sep 6

Ten fine-tuning mistakes I see students make (and made myself)

#finetuning #mistakes #students #guide

3 min read

PRANJUL RATHOUR

Sep 6

Instruction tuning vs domain adaptation: two different fine-tuning goals

#finetuning #instructiontuning #domainadaptation #llm

3 min read

PRANJUL RATHOUR

Sep 6

Validate your dataset before you burn a GPU hour: a checklist

#finetuning #datasets #validation #checklist

3 min read

PRANJUL RATHOUR

Sep 6

Fine-tuning datasets: Alpaca, ShareGPT and chat templates without the confusion

#finetuning #datasets #chattemplates #guide

3 min read

PRANJUL RATHOUR

Sep 6

QLoRA hyperparameters that actually matter (rank, alpha, learning rate, epochs)

#qlora #hyperparameters #finetuning #llm

3 min read

PRANJUL RATHOUR

Sep 6

How to read a loss curve during fine-tuning (and when to stop)

#finetuning #losscurve #training #guide

3 min read

PRANJUL RATHOUR

Sep 6

RAG vs fine-tuning: which one your problem actually needs

#rag #finetuning #architecture #decisionguide

3 min read

PRANJUL RATHOUR

Sep 6

LoRA vs QLoRA vs full fine-tuning: cost, quality and when each makes sense

#finetuning #lora #qlora #llm

3 min read

PRANJUL RATHOUR

Sep 5

Fine-tuning a 1.7B model at 3.2 GB VRAM — building FineTune Studio

#finetuning #qlora #llm #opensource

2 min read

Seth Wheeler

Aug 23

How a Dedup Pass Deleted My Training Curriculum

#llm #measurement #finetuning #security

6 min read

Rickesh T N

Aug 24

A generic fine-tuning playbook, written after doing it wrong several times

#llm #finetuning #machinelearning #mlops

6 min read

Wuic Framework

Aug 31

Teaching a local coding agent from its own mistakes: DPO on a 30B model

#dpo #qlora #finetuning #llm

7 min read

👋 Sign in for the ability to sort posts by relevant, latest, or top.