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

    Neuronpedia

    INDEX
    Explanations

    phrasing or contextual information

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     tuples
    0.48
     lan
    0.45
    Lan
    0.41
     Lan
    0.40
    eu
    0.39
    tuples
    0.39
    fois
    0.39
     ίδ
    0.39
     sometimes
    0.39
    lan
    0.38
    POSITIVE LOGITS
     lượt
    0.50
    ữa
    0.44
    TERN
    0.40
    بيك
    0.39
    ⓞ
    0.38
     phrasing
    0.37
     prioritizing
    0.37
    ⱨ
    0.36
    द्री
    0.36
     الإلكتر
    0.36
    Act Density 0.000%

    No Known Activations