© 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. 95539
    Prev
    Next
    INDEX
    Explanations

    price

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

    The neuron fires on curly‐brace delimiters (e.g. “{”, “}”, “},{”) that mark the structure of code or JSON blocks.

    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
     szeret
    -1.42
    تاح
    -1.38
    cestor
    -1.35
     logotipo
    -1.34
     berken
    -1.34
     decorativas
    -1.31
    sepatu
    -1.30
     berangkat
    -1.30
    cestors
    -1.30
     soyez
    -1.28
    POSITIVE LOGITS
     thậm
    1.31
     dessa
    1.27
    el
    1.21
    op
    1.20
    in
    1.20
    la
    1.18
    了一遍
    1.17
    ara
    1.16
    //!
    1.16
     lamentable
    1.15
    Activations Density 0.002%

    No Known Activations