© 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. 13437
    Prev
    Next
    INDEX
    Explanations

    profile picture, cover photo, bio, information

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

    The neuron is specialized to detect imperative calls to add something, i.e. occurrences of the phrase “add a” (as in “Add a Vermont business…”).

    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
    区分
    0.64
     Shows
    0.60
     shows
    0.56
     kör
    0.56
     މ
    0.56
     Researchers
    0.55
     Executes
    0.55
     Surgeons
    0.55
     shares
    0.55
     Proceeds
    0.55
    POSITIVE LOGITS
     barren
    0.64
    뻘
    0.64
    naie
    0.61
    SeekBar
    0.60
     дописавши
    0.60
    ед
    0.56
     caldera
    0.55
    AppDesigner
    0.55
    لین
    0.55
    setContent
    0.55
    Activations Density 0.002%

    No Known Activations