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

    Neuronpedia

    INDEX
    Explanations

    numerical values that are specified as being approximate or around a specific number

    oai_token-act-pair · gpt-3.5-turbo

    expressions of approximation or estimation

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    enders
    -0.82
    ieu
    -0.79
    iled
    -0.76
    hered
    -0.74
    oli
    -0.72
    Express
    -0.70
    woods
    -0.69
    olog
    -0.69
    zzle
    -0.68
    actic
    -0.67
    POSITIVE LOGITS
    Ĥİ
    1.03
     approximately
    0.85
    approximately
    0.80
     eleph
    0.77
     Approximately
    0.77
     approximate
    0.76
     (~
    0.75
    atility
    0.73
     800
    0.72
    ©¶æ
    0.72
    Act Density 0.007%

    No Known Activations