© 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. 34-GEMMASCOPE-RES-131K
    7. 107283
    Prev
    Next
    INDEX
    Explanations

    author Paul Klimov, view class

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

    This neuron fires strongly on code or markup identifiers—things like tag names (e.g. view, text, template), attribute identifiers (e.g. label, mask, row), and function or method names (e.g. create, generators, POST).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     Unger
    -0.82
     Zal
    -0.74
     двигателя
    -0.73
    Xtra
    -0.71
    vol
    -0.71
     presidential
    -0.69
     دهه
    -0.68
     Green
    -0.68
     scopri
    -0.67
    Titus
    -0.67
    POSITIVE LOGITS
     Genshin
    1.78
    Gens
    1.42
     Traveler
    1.16
     Tey
    1.15
     Dend
    1.14
    モンド
    1.10
     Klee
    1.10
    bedo
    1.09
     Ina
    1.09
     _(
    1.09
    Activations Density 0.003%

    No Known Activations