© 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-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 72893
    Prev
    Next
    INDEX
    Explanations

    fiber, stem, acoustic, multiplication

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

    The neuron fires strongly on tokens that are technical or formal type names—especially capitalized identifiers in code or geometry contexts.

    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
     Meksi
    -1.39
    鷽
    -1.28
     incrí
    -1.24
     kalori
    -1.23
     italiens
    -1.23
     klinik
    -1.21
     demokra
    -1.20
     visuel
    -1.20
     kapas
    -1.20
     logiciel
    -1.20
    POSITIVE LOGITS
    It
    1.20
    R
    1.20
     R
    1.19
    there
    1.17
    If
    1.08
    T
    1.08
    To
    1.07
    r
    1.05
    All
    1.04
    When
    1.02
    Activations Density 0.035%

    No Known Activations