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

    Neuronpedia

    INDEX
    Explanations

    quantitative data and statistical analysis descriptions

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

    Values reported as mean ± standard deviation

    np_acts-logits-general · gemini-2.0-flash

    mean ±values expressed as mean ±mean ± standard deviation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    доби
    -0.58
    бер
    -0.51
     etc
    -0.45
    ↵↵
    -0.45
     Full
    -0.45
    全
    -0.45
    ör
    -0.45
     zar
    -0.44
     ma
    -0.44
     service
    -0.44
    POSITIVE LOGITS
     ±
    1.47
    ±
    1.34
     ± 
    1.31
     +/-
    1.20
     (±
    1.16
    ± 
    1.13
    +/-
    0.95
    IntoConstraints
    0.94
     +-
    0.90
    0.89
    Act Density 0.989%

    No Known Activations