© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 10998
    Prev
    Next
    INDEX
    Explanations

    PSU, surface pressure, preparation, or Krishna

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron fires on numeric tokens—especially multi­digit sequences or decimal numbers (e.g. IDs, floats, version numbers) embedded in the text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_24_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     catheters
    0.81
    I
    0.79
     cervix
    0.75
    er
    0.75
    ER
    0.75
    tees
    0.73
     Shiva
    0.72
     rhin
    0.71
    U
    0.70
    هو
    0.69
    POSITIVE LOGITS
     жүктөп
    1.01
    recomend
    0.98
     recomend
    0.90
    륭
    0.84
     recomendaciones
    0.83
     congrat
    0.81
    verläss
    0.79
     übernimmt
    0.78
    ímos
    0.78
    astico
    0.76
    Activations Density 0.002%

    No Known Activations