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

    Neuronpedia

    INDEX
    Explanations

    docker-compose version syntax

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     HI
    0.42
     Valley
    0.41
    ļ
    0.41
     baut
    0.40
     boiled
    0.40
     UL
    0.39
     Mu
    0.38
    }}{
    0.38
    दीप
    0.38
    servername
    0.38
    POSITIVE LOGITS
     felis
    0.39
    ParseException
    0.39
    语法
    0.38
    엄
    0.38
    ObjectModel
    0.37
    Compose
    0.37
    Levi
    0.36
    Jason
    0.36
    🔢
    0.36
    Ober
    0.35
    Act Density 0.005%

    No Known Activations