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

    Neuronpedia

    INDEX
    Explanations

    phrases related to evaluating or discussing progress and outcomes

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

    Tokens before "last," "previously," or related words

    np_acts-logits-general · gemini-2.0-flash

    previous or earlier occurrences

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    XMLSchema
    -0.58
    وفاته
    -0.58
     pinulongan
    -0.49
     propOrder
    -0.49
    eaways
    -0.49
    forName
    -0.48
    finally
    -0.47
    fromCharCode
    -0.46
    äiv
    -0.46
    ậc
    -0.46
    POSITIVE LOGITS
     previous
    2.02
     earlier
    1.89
     previously
    1.79
    previous
    1.78
     Previous
    1.68
    Previous
    1.66
    earlier
    1.60
     Earlier
    1.60
     Previously
    1.57
    Previously
    1.56
    Act Density 0.885%

    No Known Activations