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

    Neuronpedia

    INDEX
    Explanations

    fault

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    Warn
    -0.07
    ersed
    -0.07
    OutOf
    -0.07
    KH
    -0.07
     ngh
    -0.06
     configurable
    -0.06
     Chap
    -0.06
     msm
    -0.06
    (my
    -0.06
    	search
    -0.06
    POSITIVE LOGITS
    ению
    0.07
     keep
    0.07
     #####
    0.06
    �
    0.06
     Booth
    0.06
    count
    0.06
    ائه
    0.06
    [train
    0.06
     decryption
    0.06
    !).
    0.06
    Act Density 0.025%

    No Known Activations