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

    Neuronpedia

    INDEX
    Explanations

    words related to communication activities, such as corresponding or communication personnel like correspondents

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "correspond" and its variations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    GPU
    -0.82
    STD
    -0.70
    HUD
    -0.65
    nels
    -0.64
    ipple
    -0.63
    umatic
    -0.63
    ghazi
    -0.63
    STRUCT
    -0.62
    STEM
    -0.62
    starter
    -0.61
    POSITIVE LOGITS
    ingly
    1.04
     correspond
    0.99
     Correspond
    0.98
     newsp
    0.96
    ences
    0.94
    hip
    0.93
     correspondence
    0.89
     correspondent
    0.86
    ially
    0.84
    nces
    0.81
    Act Density 0.022%

    No Known Activations