© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 40204
    Prev
    Next
    INDEX
    Explanations

    mathematical operations and derivations

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

    The neuron fires on “result” or “output”–style technical terms in scientific writing (e.g. expression, contributions, coefficients, analysis).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     a
    -1.30
     or
    -1.25
     if
    -1.23
     on
    -1.13
     said
    -1.13
     “
    -1.07
     "
    -1.05
     was
    -1.05
     popularly
    -1.04
     jungen
    -1.03
    POSITIVE LOGITS
    🪛
    1.55
     semuanya
    1.35
     poliester
    1.30
     áudio
    1.23
     malang
    1.23
     nobile
    1.23
     everything
    1.23
    deja
    1.21
     bandung
    1.20
    Ils
    1.20
    Activations Density 0.012%

    No Known Activations