© 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
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 6829
    Prev
    Next
    INDEX
    Explanations

    final answers boxed

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

    The neuron fires on numeric literals (especially floating-point or count values) presented in inline code or backtick-delimited form.

    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
     cứ
    1.09
     appareil
    1.03
     assez
    0.98
     quels
    0.97
    迌
    0.96
     norme
    0.95
     кожного
    0.94
     bertahan
    0.92
     officiel
    0.91
     enda
    0.91
    POSITIVE LOGITS
    י
    1.03
    am
    1.02
    비
    1.02
    ام
    0.98
    abajo
    0.98
    G
    0.96
    re
    0.94
    ียร์
    0.93
    vice
    0.93
    すみ
    0.92
    Activations Density 0.002%

    No Known Activations