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

    Neuronpedia

    INDEX
    Explanations

    reading ability

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    os
    0.74
    elie
    0.72
     starboard
    0.68
     raining
    0.66
    Ar
    0.64
     blank
    0.64
    crazy
    0.64
     ideologies
    0.63
     বাড়ির
    0.63
    alal
    0.62
    POSITIVE LOGITS
    atable
    0.89
    thedocs
    0.87
     aloud
    0.86
    iness
    0.84
    mittance
    0.82
    ixir
    0.81
    write
    0.80
    ahrung
    0.79
    ଼
    0.78
    mitted
    0.78
    Act Density 0.085%

    No Known Activations