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

    Neuronpedia

    INDEX
    Explanations

    detailed and descriptive language

    oai_token-act-pair · gpt-3.5-turbo

    forms of the word "fail."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     ILCS
    -0.68
     rep
    -0.61
     Kaf
    -0.61
     additive
    -0.60
    zone
    -0.60
     inject
    -0.59
     Manz
    -0.58
     MP
    -0.57
     Lite
    -0.57
     Nielsen
    -0.56
    POSITIVE LOGITS
    ailed
    4.69
    ailing
    2.73
    ails
    2.68
    ail
    2.05
    ailable
    1.87
    aked
    1.24
    ailability
    1.19
    anged
    1.13
    ained
    1.12
    umbled
    1.12
    Act Density 0.006%

    No Known Activations