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

    Neuronpedia

    INDEX
    Explanations

    left with remaining

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     erste
    0.46
     første
    0.43
     allere
    0.43
    İlk
    0.41
     eerste
    0.41
     أول
    0.40
     afgelopen
    0.40
     první
    0.39
     rück
    0.38
    နေ့
    0.38
    POSITIVE LOGITS
     remaining
    1.16
    remaining
    1.05
    剩下
    0.98
    Remaining
    0.97
    剩下的
    0.96
     Remaining
    0.95
     অবশিষ্ট
    0.94
    剩
    0.83
     остав
    0.82
     остались
    0.81
    Act Density 0.038%

    No Known Activations