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

    Neuronpedia

    INDEX
    Explanations

    phrases related to targeting or pursuing something or someone

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to pursuing or targeting individuals or entities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    Released
    -0.81
    uten
    -0.78
    ascript
    -0.74
    undo
    -0.74
    Chart
    -0.74
    ynski
    -0.73
    kept
    -0.69
    abet
    -0.69
    Split
    -0.68
    chell
    -0.68
    POSITIVE LOGITS
     whistleblowers
    0.84
     somebody
    0.76
     timber
    0.75
     blockers
    0.70
     puberty
    0.67
     payday
    0.67
     someone
    0.67
     fossil
    0.65
     pedestrians
    0.65
     loopholes
    0.65
    Act Density 0.028%

    No Known Activations