american biologist

ich weiß es nicht
post-darwinian evangelist

[american biologist @95thoughts]

this misses the most important part!

Inside of us is a process that bridges evolution and learning to create an architecture capable of recursive self-improvement. No life without it.

what is this metalearning process ?

it’s not the genome, or the brain…

[roon @tszzl]

people forget about the Baldwin effect when reasoning about the difference between evolution and learning

when you recapitulate learning that your ancestors did, when you learn to see, walk, and speak in an unbroken line for thousands of generations, it gets easier each time

[american biologist @95thoughts]

@wow_tru @micsolana Filter feeding whales have been around for ~30M years, true chonking only arrived in the Plio-Pleistocene, ~3–4.5Mya with the onset of intense seasonal upwelling (high density krill patches)

So love of the filter-feeding game + oceanic productivity necessary for maximum chonk

[american biologist @95thoughts]

@micsolana Needs to be marine. Q comes down to predation habits + how they scale with size

The largest marine reptiles (icthyosaurs) were macropredators, hunting for big fish. This gets less efficient with size

blue whales’ filter feeding is more efficient with size

qed blue whale = GOAT

[american biologist @95thoughts]

@ModernAmbroseB @micsolana Completeness of the fossil record aside-

the positive scaling of the blue whale’s filter feeding strat vs the negative scaling of aquatic reptiles’ raptorial feeding is sufficient for the GOATing of the blue whale

[american biologist @95thoughts]

@micsolana Needs to be marine. Q comes down to predation habits + how they scale with size

The largest marine reptiles (icthyosaurs) were macropredators, hunting for big fish. This gets less efficient with size

blue whales’ filter feeding is more efficient with size

qed blue whale = GOAT

[american biologist @95thoughts]

@micsolana Oceanic productivity also factors in. the fact that no marine reptiles developed filter feeding is probably downstream of lower / less dense oceanic productivity in their time

So while they reached similar length (~90+ ft) they were unable to cultivate mass like the GOAT

[american biologist @95thoughts]

@micsolana Needs to be marine. Q comes down to predation habits + how they scale with size

The largest marine reptiles (icthyosaurs) were macropredators, hunting for big fish. This gets less efficient with size

blue whales’ filter feeding is more efficient with size

qed blue whale = GOAT

[american biologist @95thoughts]

@lecong https://t.co/UCYn78KQOe

[Anshul Kundaje @anshulkundaje]

@lecong @arjunrajlab I think you forgot to link to the paper

[american biologist @95thoughts]

@cwrichardson @robinhanson If cellular structure is emergent from these, you need to know the underlying attractors dynamics to recapitulate LTM -

which you cant get from a snapshot (or even continuous snapshots) of the structural state

[american biologist @95thoughts]

@cwrichardson @robinhanson Comes down to what can persist long enough: Higher order cellular structures are more stable over time than the molecular features

but molecular features can form attractor states (feedback loops) that persist on longer timescales

Especially true for epigenetic mods

[american biologist @95thoughts]

@almostlikethat @Sauers_ well I can tell you it will talk about brine shrimp as much as you’d like but as soon as you start bringing remotely anything mol bio adjacent along for with the brine shrimp subject matter you’re in danger. Even outside of methylation, which I thought was the tripwire, but - no

[american biologist @95thoughts]

@josiezayner Finally, a reasonable cost to sequence my pet lungfish’s genome!

https://t.co/MOr449WNB0

[american biologist @95thoughts]

@remusrisnov @Plinz @EtiolatedLimb @QiaochuYuan With a few tricks you can even get them to reconstitute the entire organism from the initial embryo, rather than having to wait for the chimeric generation to use one of the iPSC-origin cells in a gamete

(This one is paywalled but here’s the first page)

[american biologist @95thoughts]

@remusrisnov @Plinz @EtiolatedLimb @QiaochuYuan If the skin cell -turned-stem cell is injected into another embryo, these cells can go on to form all potential new cell types of the organism- skin, gut, muscle, even germline- and procreate at the organismal level indefinitely

[american biologist @95thoughts]

@remusrisnov @Plinz @EtiolatedLimb @QiaochuYuan This Nobel winning work from Yamanaka showed a skin cell can be de-differentiated to an embryonic-like state and the re-differentiate into all germ layers

Not natural, but demonstrable

https://t.co/l8XamA6k3f

[american biologist @95thoughts]

@Plinz @QiaochuYuan then milk is the only ethically permissible food under this framework, passing the ontological filter of non-cellularity

fruits are an edge case - ontologically cellular but teleologically destined for consumption

But consuming nuts violates their telos
~ eating children

[american biologist @95thoughts]

@owl_posting wonder if penalizing CoT could have something to do with it

[Tim Hua 🇺🇦 @Tim_Hua_]

Anthropic accidentally trained against the chain of thought in Claude Mythos, Opus 4.6, and Sonnet 4.6

[american biologist @95thoughts]

@AskYatharth @deepfates @DanielleFong Interesting. Not just cast aside but cast to the substrate from which Adam was made?

Is that the implication here, how truth is meant to spring from the earth?

[american biologist @95thoughts]

@deepfates @DanielleFong https://t.co/sfFvIKxf3X

[american biologist @95thoughts]

@deepfates @DanielleFong the group might enjoy The Phenomenon of Man

specifically the retrocausal framing of the “omega point”

[american biologist @95thoughts]

@TSBurkhardt https://t.co/RcxKkUY4oy

[euan ashley @euanashley]

Something about the “mirage reasoning” mode seemed to put the model in a different epistemological framework; one in which it performs significantly better than guessing.

[american biologist @95thoughts]

schizophrenia has never been diagnosed in a person who was born blind

[euan ashley @euanashley]

Something about the “mirage reasoning” mode seemed to put the model in a different epistemological framework; one in which it performs significantly better than guessing.