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

    Neuronpedia

    INDEX
    Explanations

    parentheses, asterisks, and commas

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     possessions
    0.50
     temporarily
    0.48
     plugged
    0.47
     decision
    0.47
     downed
    0.47
     conditioned
    0.47
     hashed
    0.47
     animals
    0.46
     horizontally
    0.45
     acquisitions
    0.45
    POSITIVE LOGITS
    (
    0.52
    ض
    0.47
    Schools
    0.47
    安
    0.47
    未来
    0.47
     Stati
    0.47
    וי
    0.46
    Tis
    0.45
    Invit
    0.45
    谋
    0.45
    Act Density 0.000%

    No Known Activations