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

    Neuronpedia

    INDEX
    Explanations

    Temperature conversion

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    Hong
    -0.07
    _ma
    -0.07
    钱包
    -0.07
    勃勃
    -0.06
     Hong
    -0.06
     mies
    -0.06
    🌆
    -0.06
    	TArray
    -0.06
     Al
    -0.06
     AAA
    -0.06
    POSITIVE LOGITS
    关键是
    0.07
    调剂
    0.07
    атур
    0.07
    DIM
    0.07
    expected
    0.07
    declare
    0.06
    𝕂
    0.06
    qrst
    0.06
    _aw
    0.06
     Quant
    0.06
    Act Density 0.017%

    No Known Activations