© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 84958
    Prev
    Next
    INDEX
    Explanations

    convert data or text

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

    The neuron is picking out domain-specific technical or content-bearing terms (e.g. “real,” “life,” “game,” “print,” “data,” “import,” “<,” “div,” “toxic,” etc.).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     to
    -1.65
     each
    -1.46
     olmak
    -1.45
     these
    -1.41
     dziś
    -1.36
     this
    -1.34
     it
    -1.30
     persönlichen
    -1.27
     Möglichkeit
    -1.25
     from
    -1.23
    POSITIVE LOGITS
    oció
    1.55
    躂
    1.49
    izacao
    1.42
    而不是
    1.37
     instead
    1.37
     вместо
    1.34
     حتی
    1.31
    tyfik
    1.29
    ଃ
    1.28
     sadece
    1.27
    Activations Density 0.147%

    No Known Activations