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

    Neuronpedia

    INDEX
    Explanations

    abrasive sensory comparisons

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     faint
    0.43
     mysterious
    0.42
    不思議
    0.41
    晕
    0.40
     Senate
    0.40
     breath
    0.39
     perfectly
    0.39
     აღმასრულებელი
    0.39
     ubiquitous
    0.38
     Analyses
    0.38
    POSITIVE LOGITS
     razor
    0.64
     sandpaper
    0.61
     abrasive
    0.56
     nails
    0.53
    nails
    0.52
     lemons
    0.52
    nail
    0.51
     knives
    0.50
     grinder
    0.50
     nail
    0.50
    Act Density 0.038%

    No Known Activations