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

    Neuronpedia

    INDEX
    Explanations

    material properties and failures

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    correspond
    0.43
    cons
    0.43
     haught
    0.43
    inor
    0.42
    inci
    0.41
    corre
    0.41
    araoh
    0.41
     naught
    0.41
     محسوس
    0.40
    at
    0.40
    POSITIVE LOGITS
    ುವ
    0.54
    ერი
    0.53
    ه
    0.50
    ิม
    0.49
    스
    0.47
    ୃ
    0.46
    哫
    0.46
    點
    0.46
    ία
    0.46
    Grady
    0.45
    Act Density 0.002%

    No Known Activations