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

    Neuronpedia

    INDEX
    Explanations

    contrasting with alternatives

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Various
    1.12
    Various
    1.12
    Specific
    1.10
    Additional
    1.05
     Additional
    1.01
     besondere
    1.00
     Different
    0.99
     ADDITIONAL
    0.98
    Different
    0.97
     Specific
    0.94
    POSITIVE LOGITS
     competitors
    1.73
     competing
    1.65
     rival
    1.64
     rivals
    1.61
     traditional
    1.52
     comparable
    1.50
     competitor
    1.47
     other
    1.43
    传统的
    1.43
     counterparts
    1.41
    Act Density 0.212%

    No Known Activations