© 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
    2. Gemma-3-12B
    3. 31-GEMMASCOPE-2-RES-16K
    4. 2178
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on domain-specific proper names and technical service or product terms (e.g. “Web Development,” “Plugin Development,” “Server Management,” “Magento e-Commerce,” “Logistic issues,” “Key players,” etc.).

    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
    uron
    0.65
    arno
    0.63
    ḡ
    0.62
    úl
    0.61
    undra
    0.61
    ándolo
    0.60
    মিটার
    0.59
    ორი
    0.59
    गोलिक
    0.59
    ěn
    0.59
    POSITIVE LOGITS
    [`
    0.61
     samt
    0.61
    Serialized
    0.61
     Leistungen
    0.55
     attir
    0.55
     +/-
    0.54
    いただき
    0.53
    [{{
    0.53
     readonly
    0.52
     serta
    0.52
    Activations Density 0.076%

    No Known Activations