Along with Baguettotron we release the smallest viable language model to date. Monad, a 56M transformer, trained on the English part of SYNTH with non-random performance on MMLU. Desiging Monad an engineering challenge requiring a custom tiny tokenizer. huggingface.co/PleIAs/Monad

Care to explain why?

I’m getting access denied when trying to follow your link. I’m in the UK so guess it’s related to our recently implemented online safety legislation:

We found that forest mice have larger (~2x!) corticospinal tracts (CSTs; the axons of CSNs) compared to prairie mice! This is due to more CSNs in secondary motor/sensory ctx (M2/S2), but not primary sensorimotor ctx (M1/S1)

This led to the question: do forest mice have better hand dexterity?

Really enjoyed reading this, thanks for taking the time to post it.

Great thread! Thanks.

Or maybe a less negative tl;dr = “These technologies may have potential as helpful tools for science writers, but they are not ready for “prime time” at this point for the SciPak team. “

I'm highly skeptical of such income comparisons. The vast discrepancies between what are private and public goods in the countries gets little consideration. For example, here in Germany, good public transportation means less money needed to buy cars. Also, how much is 6-weeks vacation/yr worth?