© 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. 12897
    Prev
    Next
    INDEX
    Explanations

    they followed by a verb

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

    The neuron strongly activates on the third‐person plural pronoun “they” (and immediately adjacent “are”), i.e. occurrences of “they” or “they are.”

    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
    اب
    1.04
    не
    0.99
    ը
    0.98
    десят
    0.94
    ᆫ
    0.94
     えっと
    0.93
     aprendido
    0.92
     الابتدائي
    0.91
    И
    0.91
    ان
    0.89
    POSITIVE LOGITS
    t
    1.17
     selves
    1.04
     bodies
    0.95
    mselves
    0.95
     Lordships
    0.93
    selves
    0.91
     ovens
    0.90
    ள்
    0.89
     jums
    0.89
     prisms
    0.89
    Activations Density 0.439%

    No Known Activations