https://imbue.com/research/70b-infrastructure From bare metal to a 70B model: infrastructure set-up and scriptsThe Imbue TeamJune 25, 2024 “…we trained a 70B parameter model from scratch on our own infrastructure that outperformed
What I Read: Logarithms, Heteroskedasticity
https://allendowney.substack.com/p/logarithms-and-heteroskedasticity Logarithms and HeteroskedasticityAllen DowneyMay 26, 2024 “A log transform is neither necessary nor sufficient to fix heteroskedasticity”
What I Read: How fast process CSV file
https://datapythonista.me/blog/how-fast-can-we-process-a-csv-file How fast can we process a CSV fileMarc GarciaThu 22 February 2024 “…we’ll see in this blog post how to process a CSV file as fast as possible.”
What I Read: Smooth Noisy Data
https://towardsdatascience.com/the-perfect-way-to-smooth-your-noisy-data-4f3fe6b44440?gi=a6f62aaf2818 The Perfect Way to Smooth Your Noisy DataAndrew BowellOct 25, 2023 “Insanely fast and reliable smoothing and interpolation with the Whittaker-Eilers method.”