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

    Neuronpedia

    INDEX
    Explanations

    references to technical issues or limitations in software and coding contexts

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

    Limitations or constraints

    np_acts-logits-general · gemini-2.0-flash

    cannot, unable, or restricted

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     AssemblyTitle
    -0.60
     Wiktionnaire
    -0.54
    ographed
    -0.54
     FontWeight
    -0.50
     الحره
    -0.49
    letzt
    -0.49
    nessed
    -0.49
     orthon
    -0.48
     وتسجيلات
    -0.47
     Erwart
    -0.47
    POSITIVE LOGITS
    只能
    1.16
     cannot
    1.03
    也只能
    1.00
     unable
    0.97
     inability
    0.96
    cannot
    0.92
     impossibility
    0.90
    没办法
    0.90
     only
    0.89
     impossible
    0.88
    Act Density 0.538%

    No Known Activations