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

    Neuronpedia

    INDEX
    Explanations

    references to eyewitness testimonies

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

    witness testimony

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     AssemblyCompany
    -0.38
    DockStyle
    -0.36
     defStyle
    -0.35
     št
    -0.35
    autoconfigure
    -0.35
     Plenum
    -0.35
    Bucket
    -0.34
     entstanden
    -0.33
    acceptance
    -0.33
     بدان
    -0.32
    POSITIVE LOGITS
     witness
    0.81
     witnesses
    0.79
     eyewitness
    0.71
     WITNESS
    0.71
     Witnesses
    0.70
    witness
    0.69
    Witness
    0.68
     Witness
    0.67
    Witnesses
    0.67
     kasarigan
    0.66
    Act Density 0.384%

    No Known Activations