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

    Neuronpedia

    INDEX
    Explanations

    shell command redirection

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    a
    0.76
    া
    0.76
    েও
    0.72
    ه
    0.71
    م
    0.71
    ে
    0.68
    し
    0.68
    i
    0.67
    m
    0.67
    м
    0.67
    POSITIVE LOGITS
    ية
    0.63
     förm
    0.58
    0.55
     früher
    0.54
     quam
    0.53
    ،
    0.51
     vegetables
    0.49
     než
    0.49
    $)
    0.49
     penultimate
    0.48
    Act Density 0.175%

    No Known Activations