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

    Neuronpedia

    INDEX
    Explanations

    encoding and data types

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    nike
    1.88
    из
    1.67
    statt
    1.66
    drag
    1.60
    Richt
    1.60
     NKG
    1.59
    জনকে
    1.57
     unsurpassed
    1.57
    𝑥
    1.56
    cited
    1.53
    POSITIVE LOGITS
    ()=>{
    1.99
    ب
    1.90
     ја
    1.90
     querer
    1.86
    𝕥
    1.78
    𝘤
    1.77
     collett
    1.76
    การ
    1.76
    𝘶
    1.75
    𝘭
    1.73
    Act Density 0.120%

    No Known Activations