© 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. 31-GEMMASCOPE-2-RES-16K
    7. 49
    Prev
    Next
    INDEX
    Explanations

    to + verb indicating action

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

    The neuron fires on “how to”‐style instructional phrases (especially the word “to” in tutorial or guidance contexts).

    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
     The
    0.27
     Offering
    0.26
     A
    0.26
     เพียง
    0.26
     (*
    0.25
     pemberian
    0.24
     Jacobi
    0.24
     Promotion
    0.24
     At
    0.24
     J
    0.23
    POSITIVE LOGITS
     get
    0.59
     manage
    0.57
     navigate
    0.57
     achieve
    0.57
     survive
    0.55
     cope
    0.55
     avoid
    0.54
    做到
    0.54
     overcome
    0.53
    manage
    0.53
    Activations Density 1.023%

    No Known Activations