Training on code improves LLM performance on non-coding tasks
Adding code to your training data makes your LLM better at non-coding tasks too
aimodels-fyi
🧠Training on code improves LLM performance on non-coding tasks
Adding code to your training data makes your LLM better at non-coding tasks too
LLMs can speak in JPEG
By studying “secret” messages (JPEGs), LLMs can eventually learn to write them.
Different now
The week the internet actually changed forever
A change to make diffusion models 80%Â faster
New discovery: leaner, meaner diffusion
The death of creativity
How to improve your semantic search with hypothetical document embeddings
How to use a simple LLM call to dramatically improve the quality of your semantic search results