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

    Neuronpedia

    INDEX
    Explanations

    come followed by directional or connective words

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    按
    0.85
    തിനാ
    0.82
     permanecer
    0.82
     按
    0.81
    directional
    0.79
    ನಿರ್
    0.79
    niches
    0.79
    umbu
    0.78
    BlockFit
    0.76
     actinides
    0.76
    POSITIVE LOGITS
     forth
    1.33
     alive
    1.29
     up
    1.19
     back
    1.15
     closer
    1.13
     ashore
    1.12
     crashing
    1.12
     roaring
    1.08
     undone
    1.08
     together
    1.07
    Act Density 0.074%

    No Known Activations