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

    Neuronpedia

    INDEX
    Explanations

    understood? Good

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     supply
    0.40
    supply
    0.40
     foss
    0.39
     Mode
    0.38
     Is
    0.38
     mode
    0.38
     membran
    0.37
    and
    0.37
     fossil
    0.37
    是要
    0.37
    POSITIVE LOGITS
     Thought
    0.73
    Thought
    0.71
     thought
    0.70
    thought
    0.65
     pensé
    0.61
     THINK
    0.59
     Exactly
    0.54
    Exactly
    0.53
    と思っていた
    0.53
     pensamiento
    0.52
    Act Density 0.002%

    No Known Activations