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

    Neuronpedia

    INDEX
    Explanations

    verb/adj followed by preposition/adverb

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    geq
    0.79
    don
    0.78
    ku
    0.77
    nie
    0.77
    arten
    0.76
    cache
    0.74
    crime
    0.73
    OK
    0.72
    sampling
    0.72
    correlation
    0.72
    POSITIVE LOGITS
    에
    0.77
     kepada
    0.74
     terhadap
    0.72
    ция
    0.71
    t
    0.71
    ہ
    0.69
     heralded
    0.66
     timestamp
    0.66
     terminator
    0.66
     signific
    0.65
    Act Density 0.937%

    No Known Activations