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

    Neuronpedia

    INDEX
    Explanations

    break rule, already present, alternatives

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     hes
    0.42
     JJ
    0.41
     Cre
    0.41
     Jay
    0.40
     Leaf
    0.40
     ving
    0.39
     hepat
    0.37
    sear
    0.37
    retryWrites
    0.37
     tief
    0.37
    POSITIVE LOGITS
     Claim
    0.39
    筁
    0.38
    ENAME
    0.38
    んですよ
    0.37
    Initialization
    0.37
    Disp
    0.36
    Claim
    0.36
    殓
    0.35
    Gly
    0.35
    RefCount
    0.35
    Act Density 0.001%

    No Known Activations