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

    Neuronpedia

    INDEX
    Explanations

    adding late or during

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    orphan
    0.39
    epsilon
    0.39
    ǔ
    0.38
    かけた
    0.38
     বিবি
    0.38
     muque
    0.37
     Cuz
    0.37
     துணை
    0.36
     rq
    0.36
    سین
    0.36
    POSITIVE LOGITS
     midway
    0.80
     halfway
    0.79
     Midway
    0.60
     With
    0.50
    With
    0.46
     WITH
    0.45
     with
    0.43
     approaching
    0.43
     midpoint
    0.42
     mid
    0.42
    Act Density 0.024%

    No Known Activations