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

    Neuronpedia

    INDEX
    Explanations

    phrases related to taking action or fighting against something

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to fighting back against challenges or adversity

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    address
    -0.80
    chell
    -0.80
    imates
    -0.69
     Helpful
    -0.68
     Reserved
    -0.68
    çīĪ
    -0.66
    ascript
    -0.66
    fee
    -0.66
    20439
    -0.64
    arlane
    -0.64
    POSITIVE LOGITS
     sqor
    0.89
     attrition
    0.86
    enegger
    0.85
     uphill
    0.83
     vain
    0.77
    ardless
    0.75
     urge
    0.73
     raged
    0.73
     valiant
    0.72
     battles
    0.72
    Act Density 0.239%

    No Known Activations