LLM
-
In this final part of my guide to simulating yourself from your text messages, we take your fine-tuned model and run Text Generation.
-
In part 4 of my guide to fine-tuning an LLM on your text message history, we train using QLoRA and experiment with hyper-parameters.
-
In part 3 of my guide to fine-tuning an LLM on your text messages, we curate the dataset, encrypt and upload to Hugging Face.
-
In part 2 of my guide to fine-tuning an LLM on your text messages, we use our CSV downloads to read, organize and investigate our data.
-
I’m far from the first to think of this. Several people — perhaps inspired by creepy Black Mirror episodes — have tried to fine-tune an LLM on their SMS or WhatsApp history in an effort to create a simulation of themselves. Generally results have disappointed; conversations tend to be mundane (it seems we don’t share…