Stories by tosh

Master Plan, Part Deux (2016)

Language models are Super Mario: Absorbing abilities from homologous models

Google Gemma 1.1 2B and 7B instruct

Do we need Attention? A Mamba Primer [video]

A simple puzzle GPTs will NEVER solve

Optuna – A Hyperparameter Optimization Framework

Namecheap is building Namecheap alternative "Spaceship"

SearXNG is a free internet metasearch engine

LLMEasyTools: OpenAI tools and functions with no fuss

The Best Code is No Code At All (2007)

New York City shakes as 4.8 magnitude earthquake hits US north-east

H2O-Danube-1.8B: Technical Report

Yi-9B

Qwen1.5-32B: Fitting the Capstone of the Qwen1.5 Language Model Series

Quantization of LLMs with Llama.cpp

Processing a stream of chunks of JSON with ijson

Before Google Reader was shut down, they were internally looking for maintainers

Google reveals launch date of Android Find My Device network

Buy Your GPUs