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

    Neuronpedia

    INDEX
    Explanations

    Python code syntax

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     pres
    -0.11
     /*
    -0.10
    /*
    -0.10
    é¼
    -0.10
     pairs
    -0.09
     \n
    -0.09
     Array
    -0.09
    *Math
    -0.09
    utzer
    -0.09
     true
    -0.09
    POSITIVE LOGITS
     None
    0.12
    .__
    0.12
    (True
    0.11
    opleft
    0.11
     False
    0.10
    _ctxt
    0.10
     True
    0.10
     \\n
    0.10
    False
    0.10
    None
    0.10
    Act Density 0.079%

    No Known Activations