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

    Neuronpedia

    INDEX
    Explanations

    iconic mechanical robots and transformers

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     early
    0.88
     Early
    0.84
    early
    0.84
    Early
    0.80
     EARLY
    0.78
    棠
    0.72
    藓
    0.72
     спів
    0.72
     mycel
    0.71
    ادو
    0.71
    POSITIVE LOGITS
     Transformers
    1.54
    Transformers
    1.43
     transformers
    1.43
     Transformer
    1.42
     transformer
    1.31
     robot
    1.29
     TRANSFORM
    1.27
     Autob
    1.26
    Transformer
    1.23
     transforming
    1.23
    Act Density 0.054%

    No Known Activations