Tag: #llm

2 articles found

Local LLM Output: Why My Em-Dash Preprocessing Was Flawed

I nearly added an em-dash remover to my LLM library. Learn how testing local models revealed unexpected output patterns and reshaped text preprocessing.

#python#ai#llm

Jun 21, 2026

Backend

Debunking LLM Output Assumptions: Em-Dashes & Local Model Quirks

I almost added an em-dash remover to my LLM library, llmclean, based on a faulty assumption. Testing five local models revealed surprising output patterns.

#python#ai#llm

Jun 21, 2026