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

    Neuronpedia

    INDEX
    Explanations

    quantitative assessments related to performance and expectations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ril
    -0.17
    arkin
    -0.16
    (delegate
    -0.16
    ukkit
    -0.15
    >Returns
    -0.15
    uchs
    -0.14
    enheim
    -0.14
    mlin
    -0.14
     weakest
    -0.14
     âĨĶ
    -0.14
    POSITIVE LOGITS
     exceed
    0.66
     exceeds
    0.60
     exceeded
    0.59
     beyond
    0.59
     excess
    0.58
     exceeding
    0.57
    è¶ħ
    0.54
    è¶ħè¿ĩ
    0.54
     surpass
    0.52
     Beyond
    0.51
    Act Density 0.187%

    No Known Activations