tagged small multiples

Los Angeles compared to other cities
contributed by Andy on May 29, 2023 in small multiples, things that go up, bloomberg

Adam, a 9-yr old optimizer, is the go-to for training LLMs (eg, GPT-3, OPT, LLAMA). Introducing Sophia, a new optimizer that is 2x faster than Adam on LLMs. Just a few more lines of code could cut your costs from $2M to $1M (if scaling laws hold). https://t.co/GrMY600lLO 🧵⬇️
contributed by Andy on May 24, 2023 in curves, small multiples

Known reinfections had higher Ct values (indicating lower viral loads) with each variant than ones where no record of a previous symptomatic infection existed, here shown in the unvaccinated. 6/9
— @sbfnk
contributed by Andy on Apr 27, 2022 in covid, small multiples

Vaccines are the single most important tool in our fight against COVID However, I’ve worried our approach using a narrow single protein mRNA Vax may limit our ability to keep up w the virus This new data suggests it might I’ll explain 1/
contributed by Andy on Apr 15, 2022 in covid, small multiples, logs logs logs

Big study of ~70k Covid cases showing Omicron's far lower severity of outcomes vs Delta Symptomatic hospitalisation -53% Median hospital stay length -70% ICU admission -74% Mortality -91% Major effects among both vaccinated and unvaccinated patients. https://t.co/xxxlYanRVL
contributed by Andy on Jan 11, 2022 in covid, small multiples, confidence intervals

New from today's @UKHSA report is the finding that Omicron has faster vaccine effectiveness waning vs symptomatic infection than Delta (N for Moderna small) https://t.co/viVjY37rNQ which is a replication of what has been seen in Israel @BarakRaveh
contributed by Andy on Dec 23, 2021 in covid, small multiples

In neutralization assays, @BalazsLab observes only a 4-fold or 6-fold reduction in titer going from wild-type to Omicron after booster relative to 43-fold or 122-fold after initial mRNA vaccine series (https://t.co/EuHC1fuHWP). 4/14
— @trvrb
contributed by Andy on Dec 20, 2021 in covid, small multiples

Under existing control measures, we project a wave of infections resulting from Omicron that exceeds the January 2021 levels, for all four scenarios. In the optimistic scenario, hospital admissions are projected to remain below Jan 2021 levels.
contributed by Andy on Dec 11, 2021 in small multiples, covid, simulation

'Extreme poverty: how far have we come, how far do we still have to go?' The new post by @MaxCRoser is out https://t.co/hTtqRvq155 This is one the charts from this article.
contributed by Andy on Nov 15, 2021 in small multiples, phase space
end of the notebook.