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

    Neuronpedia

    INDEX
    Explanations

    in conclusion

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     summary
    0.66
     summaries
    0.57
     summarize
    0.55
    summary
    0.55
    Summary
    0.53
     summarized
    0.51
     Summary
    0.49
     summarizing
    0.49
     few
    0.49
     summarizes
    0.49
    POSITIVE LOGITS
     Ultimately
    0.53
    Ultimately
    0.51
     ultimately
    0.49
    adto
    0.45
     экстре
    0.45
     चरम
    0.44
     таких
    0.43
     Extreme
    0.43
     extreme
    0.43
     เช่น
    0.43
    Act Density 0.005%

    No Known Activations