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

    Neuronpedia

    INDEX
    Explanations

    instances of the word "left" in relation to events or situations

    oai_token-act-pair · gpt-3.5-turbo

    instances where subjects are abandoned or left in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    andise
    -0.77
    Temperature
    -0.71
    dayName
    -0.68
    riott
    -0.67
    insula
    -0.65
     CoC
    -0.65
    %]
    -0.64
    wcsstore
    -0.63
    alez
    -0.62
    displayText
    -0.62
    POSITIVE LOGITS
    overs
    1.10
    wing
    1.03
     undone
    1.02
     untreated
    0.95
    ward
    0.88
     behind
    0.83
    wich
    0.75
     handed
    0.75
    fing
    0.74
     unprotected
    0.73
    Act Density 0.024%

    No Known Activations