© 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
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 13591
    Prev
    Next
    INDEX
    Explanations

    person/human/player

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

    This neuron detects the leading words and key role titles that introduce and label formal clauses or sections (e.g. “A,” “The,” “To,” “Student,” “Teacher”) in policy-style text.

    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
    做什么
    0.53
    wilderness
    0.50
    物語
    0.48
    さは
    0.48
    arse
    0.48
    algebras
    0.48
    вЂ
    0.48
    🍒
    0.47
    닼
    0.47
     ব্যায়াম
    0.46
    POSITIVE LOGITS
    ผู้
    1.40
     jugador
    1.35
     henkil
    1.34
     PERSON
    1.32
     jogador
    1.31
     person
    1.27
    person
    1.26
     người
    1.26
     Spieler
    1.25
     ผู้
    1.23
    Activations Density 0.478%

    No Known Activations