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

    Neuronpedia

    INDEX
    Explanations

    consequence/result phrases

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     hatta
    0.61
     downright
    0.60
     even
    0.57
     thereby
    0.57
    even
    0.55
    therefore
    0.53
     bahkan
    0.52
     dadurch
    0.52
     thậm
    0.51
     something
    0.51
    POSITIVE LOGITS
     också
    0.56
     indire
    0.43
     aid
    0.42
     også
    0.41
     myös
    0.40
    也会
    0.39
     όμως
    0.39
     也
    0.39
     얻
    0.39
     외부
    0.37
    Act Density 0.007%

    No Known Activations