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

    Neuronpedia

    INDEX
    Explanations

    negative expressions and forms of the verb "to do."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    currently
    -0.19
     be
    -0.19
     tends
    -0.18
     Currently
    -0.18
     бÑĥдÑĮ
    -0.17
     currently
    -0.17
    Currently
    -0.17
     presently
    -0.16
    éĢļ常
    -0.16
     normally
    -0.15
    POSITIVE LOGITS
     originally
    0.31
     previously
    0.29
     initially
    0.23
     formerly
    0.23
     recently
    0.22
    dling
    0.22
     didn
    0.22
    actic
    0.22
     earlier
    0.21
     did
    0.21
    Act Density 0.117%

    No Known Activations