© 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. 10197
    Prev
    Next
    INDEX
    Explanations

    punctuation and quotes

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

    The neuron strongly fires on numeric tokens and punctuation typical of scholarly citations (e.g. volume–page numbers, years, DOIs), effectively spotting reference or bibliography entries.

    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
     અને
    0.73
    <0x80>
    0.73
    *
    0.72
     ਅਤੇ
    0.71
    ский
    0.70
     здійс
    0.70
    ية
    0.68
     மற்றும்
    0.68
    amek
    0.68
    }}>
    0.67
    POSITIVE LOGITS
     Walgreens
    0.93
     VoIP
    0.92
     Aloe
    0.89
     Corps
    0.89
     Sega
    0.89
     kyll
    0.88
    kans
    0.87
    čas
    0.87
     Chevy
    0.86
     Cuerpo
    0.86
    Activations Density 0.001%

    No Known Activations