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

    Neuronpedia

    INDEX
    Explanations

    instances of the verb "keep" in various forms

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    /animate
    -0.15
    ernes
    -0.15
    ekim
    -0.15
    raci
    -0.14
    annels
    -0.14
    ennes
    -0.14
    .kernel
    -0.14
    ediator
    -0.14
    å·ŀå¸Ĥ
    -0.14
    ÑĢим
    -0.14
    POSITIVE LOGITS
    werk
    0.15
     multim
    0.15
    -utils
    0.15
    aged
    0.14
    oo
    0.14
    [
    0.14
     forward
    0.13
    lei
    0.13
    tele
    0.13
     format
    0.13
    Act Density 0.034%

    No Known Activations