© 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
    2. Gemma-3-12B
    3. 31-GEMMASCOPE-2-RES-16K
    4. 5744
    Prev
    Next
    INDEX
    Explanations

    numbers, proper nouns, and job titles

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

    This neuron activates on author surnames in bibliographic citations (i.e., last names in reference lists).

    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_31_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
    可谓
    0.43
    ﹒
    0.41
    私募基金管理人
    0.40
    是一款
    0.39
     portefeuille
    0.38
    ﹐
    0.37
     foams
    0.35
    充满了
    0.35
    ☽
    0.35
     watercolour
    0.35
    POSITIVE LOGITS
    </h1>
    0.50
     execute
    0.40
     Einfluss
    0.40
    гӀ
    0.38
    Execute
    0.38
    |}
    0.36
    ನಗೆ
    0.36
    '}
    0.35
    '){
    0.35
     checksum
    0.35
    Activations Density 0.015%

    No Known Activations