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

    Neuronpedia

    INDEX
    Explanations

    discussions about unnecessary or redundant actions and concepts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Heraus
    -0.54
    zional
    -0.54
     Wicidata
    -0.53
    eph
    -0.53
    ième
    -0.51
    mavi
    -0.51
    ähr
    -0.50
     Νο
    -0.50
    gman
    -0.50
     Violations
    -0.50
    POSITIVE LOGITS
     unnecessary
    1.15
     pointless
    1.14
     useless
    1.05
     needlessly
    1.04
     inneces
    1.00
     unnecessarily
    1.00
     needless
    1.00
     wasting
    0.99
     inutile
    0.98
     waste
    0.97
    Act Density 0.320%

    No Known Activations