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

    Neuronpedia

    INDEX
    Explanations

    infinitive verbs that indicate actions or tasks

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

    Comes before instructions

    np_acts-logits-general · gemini-2.0-flash

    how to instructions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ViewFeatures
    -0.68
    aarrggbb
    -0.67
    SequentialGroup
    -0.65
    OGND
    -0.64
    readFileSync
    -0.64
    ForTesting
    -0.61
    лтемелер
    -0.60
    Hauptartikel
    -0.60
    出版年
    -0.59
     studier
    -0.59
    POSITIVE LOGITS
     best
    0.99
     safely
    0.79
     avoid
    0.78
    best
    0.75
     better
    0.75
     mejores
    0.70
     mejor
    0.68
     properly
    0.68
     melhor
    0.67
     maximize
    0.65
    Act Density 0.147%

    No Known Activations