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

    Neuronpedia

    INDEX
    Explanations

    multilingual text features

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Params
    0.45
     granola
    0.42
     metab
    0.41
     pouch
    0.41
     checksum
    0.41
     thermos
    0.41
     jogging
    0.40
     negligible
    0.40
     gripper
    0.40
     turnkey
    0.39
    POSITIVE LOGITS
    ikke
    0.45
     Sett
    0.44
    对着
    0.41
    ér
    0.41
    탤
    0.41
     талант
    0.40
    ے۔
    0.39
        
    0.39
     роках
    0.39
    ény
    0.39
    Act Density 0.000%

    No Known Activations