© 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. 5324
    Prev
    Next
    INDEX
    Explanations

    promoting harmful content

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

    The neuron strongly activates on words related to promoting or advertising (e.g. “promote,” “promoting,” “promotion”).

    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
    ленного
    0.60
    </h3>
    0.59
    対応
    0.54
    SIZ
    0.54
    થે
    0.54
    定价
    0.54
    >[</
    0.53
    φέρ
    0.53
    Duty
    0.53
    虏
    0.53
    POSITIVE LOGITS
    promote
    1.46
     Promoting
    1.45
     promote
    1.43
     promoting
    1.41
     promoted
    1.38
     Promote
    1.38
     promotes
    1.38
     promouvoir
    1.35
    promoting
    1.34
     promu
    1.26
    Activations Density 0.433%

    No Known Activations