Param

building @VersineApp (YC F26) | previously cs @uwaterloo, vector search @databricks, deep research @glean, ads ml @quora

[Param @xparamt]

cs job market is so back

[Param @xparamt]

@uxduck @UXDuckHQ @ycombinator this is sick!

[Param @xparamt]

@mcgeedan42 love graphide, makes our CI actually usable

[Param @xparamt]

@DarrylLuo HOOOOLY

[Param @xparamt]

and RSI in 1965

[Param @xparamt]

imagine predicting the singularity before moore’s law was even a thing

[Param @xparamt]

@Atrioc when is @spacexai being renamed back

[Param @xparamt]

@MitcheIl would love to be in it!

[Param @xparamt]

RSI is here

[Rikiya Takehi @rikiyatakehi]

We asked GPT-6 Astra to edit Qwen3.5-0.8B weights to improve math.
It directly changed only 0.016% of weights, fixing 0.8B's reasoning loops & improving math up to 6x and even instruction following 3x, without gradients or training.

We call this Weight-Space Programming: frontier LLMs can now natively read/write neural network weights! 🧵
Blog: https://t.co/bOWCcFhCVq

[Param @xparamt]

@pavitrabhalla congrats!!

[Param @xparamt]

@AarushYadav47 wow this is so cool

[Param @xparamt]

@aravmathur23 (I have it too)

[Param @xparamt]

@aravmathur23 they detect your brainrot level

[Param @xparamt]

i finally get what they do now

[benahorowitz.eth @bhorowitz]

https://t.co/kiJ0J1dIJ6

[Param @xparamt]

@MTSlive https://t.co/C8PwYCOklb

[Param @xparamt]

this was way too easy to do

[Param @xparamt]

@iamtianyigu congrats!

[Param @xparamt]

@j_olschwang congrats!

[Param @xparamt]

@PavanKatta1998 @hopper_research congrats!

[Param @xparamt]

@ad1tyaranjan @ycombinator @graypassinc FIRE

[Param @xparamt]

@aravmathur23 looks like they need graypass

[Param @xparamt]

@h3nryXYZ LM Studio

[Param @xparamt]

this was way too easy to do

[lyra @_lyraaaa_]

if you have a qwen3.5 4b around, please try steering neurons L14 N8949 +1 and L13 N7956 -1
now you have evil qwen

recipe: dose × median(‖resid‖ leaving layer L, tokens 1:) / ‖W_down[:, n]‖` at the down_proj at every input