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

    Neuronpedia

    INDEX
    Explanations

    words and phrases that convey strength or intensity

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ized
    -0.19
    aled
    -0.16
    ë¡ľ
    -0.16
    ION
    -0.16
    bian
    -0.15
    ollapsed
    -0.15
    led
    -0.15
    jur
    -0.15
    Å¡k
    -0.15
    ohn
    -0.15
    POSITIVE LOGITS
    holds
    0.30
     mẽ
    0.27
    -strong
    0.24
    (er
    0.23
    bow
    0.22
    ,strong
    0.21
    /we
    0.21
    çĥĪ
    0.20
     strong
    0.19
    sville
    0.19
    Act Density 0.036%

    No Known Activations