© 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. 11740
    Prev
    Next
    INDEX
    Explanations

    **Betting Actions:** * **Fold:**

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

    The neuron responds strongly to numeric tokens—especially large multi-digit numbers or decimal values.

    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
    ट्टर
    0.84
    мана
    0.80
    承诺
    0.79
    Comunic
    0.77
    出
    0.77
    备注
    0.75
    とはいえ
    0.75
    수는
    0.74
    斥
    0.73
    怀孕
    0.73
    POSITIVE LOGITS
    эле
    0.79
    ры
    0.76
    rophe
    0.71
     zde
    0.70
     alue
    0.70
     таблицы
    0.70
     Киев
    0.69
    )}{\
    0.68
     границы
    0.66
    ә
    0.66
    Activations Density 0.000%

    No Known Activations