© 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. 129249
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on locale or language-region codes (e.g. “ak_GH,” “sq_AL,” “de_AT”) formatted as language_abbreviation followed by an underscore and country code.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    lists starting with numbers or letters

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     at
    -1.87
     before
    -1.77
     eight
    -1.61
     five
    -1.59
     until
    -1.59
     when
    -1.58
     during
    -1.56
     seven
    -1.52
     because
    -1.51
     three
    -1.50
    POSITIVE LOGITS
     myriad
    1.63
     wondrous
    1.53
     exceedingly
    1.52
     delightfully
    1.52
     gigantic
    1.52
     momentous
    1.42
     revamped
    1.41
     wonderfully
    1.41
     ludicrous
    1.40
     bizarre
    1.39
    Activations Density 0.004%

    No Known Activations