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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating difficulty or impossibility

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating a sense of impossibility or difficulty

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    illon
    -0.85
    Ö¼
    -0.79
    ngth
    -0.74
    artments
    -0.73
    cmp
    -0.70
    achus
    -0.69
     è£ıè¦ļéĨĴ
    -0.69
     Springs
    -0.68
    sofar
    -0.68
    baugh
    -0.66
    POSITIVE LOGITS
     impossible
    1.30
     nonexistent
    1.15
     extinct
    1.10
     unstoppable
    1.06
     indistinguishable
    1.05
     unheard
    1.05
     invisible
    1.01
     invincible
    0.99
     identical
    0.99
     synonymous
    0.95
    Act Density 0.080%

    No Known Activations