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

    Neuronpedia

    INDEX
    Explanations

    allow enabling or options

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     screens
    0.46
     screen
    0.44
     wars
    0.43
     реги
    0.42
     lun
    0.41
     ambient
    0.39
    wars
    0.38
     Screens
    0.37
    මණ
    0.37
     allowed
    0.37
    POSITIVE LOGITS
    瞽
    0.41
    昊
    0.41
     анг
    0.40
    ënten
    0.39
    Again
    0.39
    maligngroup
    0.39
    外国人
    0.38
    援助
    0.38
     Again
    0.38
     ইংরা
    0.38
    Act Density 0.000%

    No Known Activations