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

    Neuronpedia

    INDEX
    Explanations

    malfunctions and fixes

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     textile
    0.42
     textiles
    0.41
    texto
    0.40
    政策
    0.39
     padx
    0.39
     uninformed
    0.39
     erset
    0.39
     TEXT
    0.38
     textual
    0.38
     Danger
    0.38
    POSITIVE LOGITS
     glitch
    0.79
     glitches
    0.75
     malfunctions
    0.73
     malfunctioning
    0.72
     malfunction
    0.69
     problèmes
    0.68
     reliability
    0.67
    Reli
    0.65
     problémy
    0.65
     problems
    0.64
    Act Density 0.265%

    No Known Activations