© 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. 12704
    Prev
    Next
    INDEX
    Explanations

    union

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

    The neuron fires on numeric tokens (e.g. digits, version numbers, line numbers) appearing in code or documentation.

    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
    Tra
    0.63
    TA
    0.56
    混
    0.55
    SEM
    0.55
    Ta
    0.54
    Mix
    0.54
    Palette
    0.53
    A
    0.52
    Ն
    0.52
    MA
    0.52
    POSITIVE LOGITS
     struktur
    0.68
     muestras
    0.67
     ganar
    0.65
    யர்
    0.64
     hadrons
    0.64
     estudar
    0.63
     detener
    0.63
     explorers
    0.62
     तेलंगाना
    0.62
     ahor
    0.61
    Activations Density 0.002%

    No Known Activations