© 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. 12430
    Prev
    Next
    INDEX
    Explanations

    teaching

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

    The neuron is highly responsive to the lemma “teach” in its various forms (e.g. teach, teaching).

    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
    
    0.65
     administ
    0.63
    manager
    0.63
    şte
    0.60
    confusion
    0.60
    hagen
    0.60
    ויות
    0.59
     Gruy
    0.57
    管理器
    0.57
    riften
    0.57
    POSITIVE LOGITS
     dạy
    0.73
    ६
    0.66
     Петра
    0.65
    ৬
    0.63
     lec
    0.63
    녓
    0.63
     parking
    0.62
     homage
    0.62
     mennesker
    0.62
    PrefabInstance
    0.61
    Activations Density 0.088%

    No Known Activations