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

    Neuronpedia

    INDEX
    Explanations

    phrases or sentences that start with "What is" or similar expressions

    oai_token-act-pair · gpt-3.5-turbo

    expressions of evaluation or assessment regarding conditions or situations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ãĤ¼ãĤ¦ãĤ¹
    -0.72
    estones
    -0.64
    oided
    -0.63
     palms
    -0.61
    ysis
    -0.60
     coffers
    -0.59
    zbollah
    -0.59
    isha
    -0.58
    adh
    -0.58
    Cooldown
    -0.58
    POSITIVE LOGITS
     what
    1.95
    what
    1.59
     whats
    1.54
     WHAT
    1.35
     whatever
    1.21
     What
    1.18
     how
    1.17
     everything
    1.13
    What
    1.09
    whatever
    1.03
    Act Density 0.566%

    No Known Activations