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

    Neuronpedia

    INDEX
    Explanations

    memory management

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    in
    0.53
    important
    0.50
    remember
    0.49
    forgot
    0.47
    messages
    0.47
    anía
    0.44
    osts
    0.44
    ortes
    0.43
     forgot
    0.43
     forgotten
    0.43
    POSITIVE LOGITS
     foam
    0.47
     banks
    0.46
    хол
    0.42
     Foam
    0.41
     explains
    0.41
    ورية
    0.41
     lee
    0.41
     teaser
    0.39
     Creek
    0.39
    дит
    0.39
    Act Density 0.022%

    No Known Activations