© 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. 5869
    Prev
    Next
    INDEX
    Explanations

    you / yourself / I

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

    The neuron strongly activates on second-person, directive or persuasive language—especially “you” plus modals or imperatives (e.g. “you will,” “you should,” “play it,” “use every chance,” etc.).

    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
     tzv
    0.69
     interspersed
    0.68
     grouped
    0.67
     hashMap
    0.67
    殆
    0.66
     Closeup
    0.65
    xyplot
    0.64
     fakat
    0.64
     colloquial
    0.63
     okolo
    0.63
    POSITIVE LOGITS
     https
    0.71
    คุณ
    0.71
    you
    0.64
    https
    0.62
    Trusted
    0.61
     ベ
    0.60
     tendrás
    0.60
     you
    0.59
    ほど
    0.59
     mejores
    0.59
    Activations Density 0.728%

    No Known Activations