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

    Neuronpedia

    INDEX
    Explanations

    phrases related to modifications or enhancements

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating accumulation or increase

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    anches
    -0.76
    aine
    -0.75
    SPONSORED
    -0.75
    ILA
    -0.72
    oi
    -0.70
    ģĸ
    -0.70
    ORGE
    -0.69
    ommel
    -0.69
    aida
    -0.69
    ...]
    -0.68
    POSITIVE LOGITS
     versions
    0.95
    ness
    0.89
     version
    0.87
     sized
    0.76
    ly
    0.69
     wording
    0.69
     crit
    0.68
     amounts
    0.68
    nes
    0.67
     approach
    0.67
    Act Density 0.070%

    No Known Activations