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

    Neuronpedia

    INDEX
    Explanations

    argument structure: consequent, sequence, fallacy

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    0.57
    лимп
    0.56
    浈
    0.53
    嚗
    0.53
    颞
    0.52
    ⦕
    0.51
     způ
    0.50
     notValid
    0.50
     resTmp
    0.50
     смартфо
    0.49
    POSITIVE LOGITS
    S
    0.63
     .
    0.56
    Y
    0.55
    W
    0.55
    ↵
    0.54
     a
    0.52
     data
    0.51
    N
    0.50
    Z
    0.49
    test
    0.49
    Act Density 0.010%

    No Known Activations