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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 1829
    Prev
    Next
    INDEX
    Explanations

    flat arrays

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

    The neuron specifically activates on floating‐point numeric tokens (numbers with decimals).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    extremely large numbers (typically in the range of 1950-2000 or higher).

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408

    references to confrontations or contests, including violent incidents and game results

    oai_token-act-pair · gpt-5Triggered 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
     peč
    0.70
     deswegen
    0.70
     получаем
    0.69
    tokens
    0.68
    TOKEN
    0.68
    ০০
    0.66
     toda
    0.66
    änk
    0.64
    umuz
    0.64
     नल
    0.64
    POSITIVE LOGITS
    It
    0.86
    有用
    0.83
    謝
    0.82
     محبت
    0.80
    Ph
    0.77
     zid
    0.76
     맞는
    0.75
    Those
    0.75
     Racine
    0.71
    Pain
    0.71
    Activations Density 0.001%

    No Known Activations