Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    contractions with missing letters, often indicating informal speech

    oai_token-act-pair · gpt-3.5-turbo

    special characters and punctuation, specifically various forms of apostrophes and grave accents

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Pastebin
    -0.70
     guiName
    -0.61
    èĢħ
    -0.61
     telesc
    -0.60
    odium
    -0.60
     Nir
    -0.60
     Crush
    -0.60
     guiActiveUn
    -0.58
     elim
    -0.58
     polyg
    -0.57
    POSITIVE LOGITS
    s
    0.82
    then
    0.80
    ves
    0.76
    issance
    0.75
    ses
    0.71
    meric
    0.71
    spect
    0.70
    ÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤ
    0.70
    only
    0.70
    sticks
    0.68
    Act Density 0.178%

    No Known Activations