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

    Neuronpedia

    INDEX
    Explanations

    HB abbreviation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    es
    1.18
    ة
    1.08
    ed
    1.02
    ુ
    1.02
    人
    1.01
    am
    0.98
    er
    0.98
    en
    0.98
    ität
    0.97
    at
    0.96
    POSITIVE LOGITS
    ANT
    0.99
    お待ち
    0.98
    お金
    0.97
    ある
    0.96
    د
    0.96
    ARE
    0.95
    ،
    0.94
    BT
    0.94
    BE
    0.93
    ANO
    0.92
    Act Density 0.000%

    No Known Activations