© 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. 11411
    Prev
    Next
    INDEX
    Explanations

    Set object assignment

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

    The neuron fires most strongly on tokens that are capitalized proper nouns or other words with initial uppercase letters (e.g. names, IDs, brand or code identifiers).

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

    quotation marks appearing in technical or instructional text, particularly in contexts involving code, product descriptions, or structured data.

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

    key topical content words—often core nouns denoting products, metrics, or salient items within informational or commercial text.

    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
    ación
    0.73
     использованием
    0.73
     உருவாக்கு
    0.71
     ఉ
    0.71
    provided
    0.71
     অভ্যন্তরীণ
    0.71
     ateş
    0.71
    aredPreferences
    0.70
    ướng
    0.70
     temas
    0.70
    POSITIVE LOGITS
     jiwa
    0.95
     appart
    0.88
    BLACK
    0.79
     quell
    0.79
     lyr
    0.78
     lads
    0.78
     menswear
    0.77
     coalgebras
    0.77
    𝘖
    0.77
    وک
    0.77
    Activations Density 0.000%

    No Known Activations