© 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. 24-GEMMASCOPE-2-RES-16K
    7. 1307
    Prev
    Next
    INDEX
    Explanations

    selection choosing options

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

    The main thing this neuron does is detect promotional discount and coupon language (e.g., mentions of promo codes, coupons, percentages off, savings, and deals).

    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_24_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
     a
    0.72
    K
    0.70
     potem
    0.68
     m
    0.67
    S
    0.64
     f
    0.64
     k
    0.63
    U
    0.63
    C
    0.61
    5
    0.60
    POSITIVE LOGITS
     மற்றும்
    0.70
     ਅਤੇ
    0.68
     एवं
    0.67
     অথবা
    0.67
     ወይም
    0.66
    angebot
    0.66
     یا
    0.65
     mishaps
    0.64
     અથવા
    0.64
     بابەت
    0.64
    Activations Density 0.107%

    No Known Activations