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

    Neuronpedia

    INDEX
    Explanations

    conjunctions and modifiers that indicate conditional relationships or qualifications

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    but had, but produced, but not

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ſelf
    -0.64
    GraphicsUnit
    -0.63
     itſelf
    -0.56
     Reſ
    -0.53
     houſe
    -0.50
     ſol
    -0.50
     Perſ
    -0.50
     myſelf
    -0.49
     ſtate
    -0.49
     lipat
    -0.49
    POSITIVE LOGITS
     Wort
    0.40
     who
    0.39
     Battalion
    0.39
    つて
    0.38
    klima
    0.38
     MyBatis
    0.38
     Brunner
    0.37
    messer
    0.36
     Nikolaus
    0.36
     fallen
    0.36
    Act Density 0.257%

    No Known Activations