© 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. 6455
    Prev
    Next
    INDEX
    Explanations

    TO: and FROM: fields

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

    This neuron responds to uppercase metadata labels or field names (e.g. SUBJECT, TAG, DECL, DIS) often used as headers in documents.

    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
     Abu
    0.75
     abandoning
    0.72
     West
    0.71
     Pamp
    0.71
     દ્વારા
    0.71
     Loma
    0.71
     Mocha
    0.71
    婄
    0.71
     Pleistocene
    0.69
     από
    0.68
    POSITIVE LOGITS
    ですし
    0.65
    EVEN
    0.63
    req
    0.62
    وكان
    0.62
    ']].
    0.61
    阵容
    0.61
    ranked
    0.59
    searchText
    0.59
    ర్థిక
    0.59
    PlatformToolset
    0.58
    Activations Density 0.002%

    No Known Activations