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

    Neuronpedia

    INDEX
    Explanations

    instability

    np_max-act · gemini-2.0-flash

    The neuron flags words expressing instability or unpredictability (e.g. “chaotic,” “instability,” “unstable”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     doctoral
    -0.07
    word
    -0.07
     gor
    -0.07
     pow
    -0.07
    医疗
    -0.06
    -highlight
    -0.06
     parole
    -0.06
     Howe
    -0.06
     прос
    -0.06
    Numeric
    -0.06
    POSITIVE LOGITS
     instability
    0.11
     unstable
    0.10
     destabil
    0.09
    št
    0.08
    ACCESS
    0.08
     insecurity
    0.07
     inst
    0.07
     unreliable
    0.07
    BAL
    0.07
    annon
    0.07
    Act Density 0.007%

    No Known Activations