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

    Neuronpedia

    INDEX
    Explanations

    verbs related to clearing a path or making preparations for an event or process

    oai_token-act-pair · gpt-3.5-turbo

    phrases that indicate paving the way for future events or actions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    iosyncr
    -0.67
    kov
    -0.66
     mathemat
    -0.64
    IFIED
    -0.62
    ividual
    -0.62
     livest
    -0.62
     pecul
    -0.59
    hemat
    -0.59
     cov
    -0.59
    querque
    -0.58
    POSITIVE LOGITS
    ward
    1.00
    fare
    0.90
    WARD
    0.90
     toward
    0.76
    steps
    0.73
    forward
    0.73
     towards
    0.70
     ahead
    0.70
     Path
    0.70
    finding
    0.70
    Act Density 0.022%

    No Known Activations