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

    Neuronpedia

    INDEX
    Explanations

    seem troubled or simple, keep asking

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     (“
    1.14
    （
    1.05
    ：（
    1.03
     hilarious
    1.01
    awesome
    0.99
    amazing
    0.99
     lovingly
    0.96
    Awesome
    0.95
     （
    0.95
     hugely
    0.95
    POSITIVE LOGITS
    ..."
    1.43
    …"
    1.42
    …”
    1.39
    ...”
    1.31
    ."
    1.27
    ...")
    1.23
    .""
    1.21
    .."
    1.19
    ,"
    1.17
    ...
    1.16
    Act Density 0.148%

    No Known Activations