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

    Neuronpedia

    INDEX
    Explanations

    references to organizations, locations, and scheduled events

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

    locations and organizations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     siyang
    -0.31
     rather
    -0.31
    ⓧ
    -0.31
     sometimes
    -0.27
    Fatalf
    -0.26
     ceea
    -0.26
     Calvo
    -0.26
     persoons
    -0.25
     prze
    -0.25
     complicada
    -0.24
    POSITIVE LOGITS
    parsedMessage
    0.77
     beſ
    0.71
     témoig
    0.71
     beſte
    0.69
    <unused74>
    0.68
    <unused14>
    0.68
    <unused41>
    0.68
    <unused8>
    0.68
    [@BOS@]
    0.68
    <unused3>
    0.68
    Act Density 0.133%

    No Known Activations