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

    Neuronpedia

    INDEX
    Explanations

    how to check and inspect

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Metadata
    0.46
     năng
    0.44
    失敗
    0.43
    Payload
    0.42
    metadata
    0.41
     mantienen
    0.41
     HUOBI
    0.41
     energie
    0.40
     başar
    0.40
    AndPassword
    0.40
    POSITIVE LOGITS
     inspect
    0.66
     inspection
    0.65
     Inspect
    0.63
     Inspection
    0.62
     inspecting
    0.54
     INSPE
    0.54
     examination
    0.53
    inspection
    0.51
     inspected
    0.50
    Inspection
    0.50
    Act Density 0.017%

    No Known Activations