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

    Neuronpedia

    INDEX
    Explanations

    phrases related to directness or straightforwardness

    oai_token-act-pair · gpt-3.5-turbo

    variations of the word "straight" and related concepts of directness or simplicity

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    oret
    -0.67
    è¦ļéĨĴ
    -0.65
    chief
    -0.64
     perman
    -0.63
    otos
    -0.63
     Lauder
    -0.63
     mur
    -0.62
    utters
    -0.62
     privately
    -0.59
    mble
    -0.58
    POSITIVE LOGITS
    forward
    1.32
    ened
    1.24
    away
    1.18
    ening
    1.15
    eners
    1.14
    edge
    1.06
    ener
    0.87
    enstein
    0.85
    aways
    0.84
     forward
    0.81
    Act Density 0.035%

    No Known Activations