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

    Neuronpedia

    INDEX
    Explanations

    instances of the word 'withdraw' or its variations

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "withdraw" and its variations indicating the action of pulling back or removing support

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Collider
    -0.76
     Cru
    -0.74
    Cru
    -0.68
     Nose
    -0.66
     Cosmic
    -0.66
    iola
    -0.66
    Core
    -0.65
     Mavericks
    -0.63
     Hole
    -0.62
    vironments
    -0.62
    POSITIVE LOGITS
     withdrawals
    0.88
     withdraw
    0.86
     withdrawal
    0.79
     withdrawing
    0.78
    uary
    0.74
    ruck
    0.74
     gobl
    0.73
    ends
    0.71
    avorite
    0.70
     aback
    0.69
    Act Density 0.013%

    No Known Activations