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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating successful outcomes or achievements

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "successfully"

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Ancients
    -0.68
     Warcraft
    -0.68
     Rober
    -0.66
     newsletters
    -0.65
     Relief
    -0.64
     Origin
    -0.63
    Orig
    -0.61
    Eth
    -0.61
     Flames
    -0.61
     anonymity
    -0.59
    POSITIVE LOGITS
     succeed
    0.89
     reproduce
    0.88
     successfully
    0.85
     destro
    0.83
     navig
    0.82
     achieved
    0.81
     competed
    0.78
     mastered
    0.77
     exting
    0.76
    TAIN
    0.76
    Act Density 0.005%

    No Known Activations