© 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. 13867
    Prev
    Next
    INDEX
    Explanations

    Hebrew and months

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

    The neuron flags individual characters or tokens from non‐ASCII scripts or with diacritical marks (e.g. Hebrew letters or letters like “ŭ”).

    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
     what
    -1.61
     before
    -1.25
     both
    -1.20
     severely
    -1.14
     and
    -1.13
     continuously
    -1.11
     should
    -1.09
    حول
    -1.08
     in
    -1.07
     ironically
    -1.06
    POSITIVE LOGITS
    ↵↵
    1.37
    ܆
    1.17
    筇
    1.14
    为啥
    1.12
    之际
    1.10
    op
    1.09
    焌
    1.09
    蛲
    1.09
     ornate
    1.05
     focuses
    1.04
    Activations Density 0.000%

    No Known Activations