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

    Neuronpedia

    INDEX
    Explanations

    use case

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ```
    0.54
     regards
    0.49
    Си
    0.48
     abr
    0.48
     Regards
    0.47
    𝑇
    0.47
    }",
    0.46
    ча
    0.46
     arada
    0.46
    inafter
    0.46
    POSITIVE LOGITS
    mz
    0.64
    0.64
     registró
    0.62
     quirúrg
    0.62
    ংলার
    0.61
     cGraph
    0.59
    diamine
    0.59
    soph
    0.58
     GÉN
    0.57
    fstream
    0.56
    Act Density 0.037%

    No Known Activations