© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 69690
    Prev
    Next
    INDEX
    Explanations

    serialized version

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

    The neuron activates on occurrences of the term “serialize” (and its variants like “serializing” or “serialized”) in code or text.

    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
    僦
    -3.16
    全都
    -2.95
     "
    -2.94
    ⢊
    -2.73
    й
    -2.70
    饪
    -2.63
    廰
    -2.58
     podem
    -2.55
     uſe
    -2.50
     ´
    -2.45
    POSITIVE LOGITS
    —”
    2.89
    ıları
    2.38
    𖡼
    2.27
    沇
    2.19
     décor
    2.19
     Crédito
    2.17
    ଽ
    2.17
    ники
    2.17
     土産
    2.16
    篳
    2.16
    Activations Density 0.008%

    No Known Activations