© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 12930
    Prev
    Next
    INDEX
    Explanations

    start of a code block

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

    The neuron activates on uppercase method or function names that mark the start or end of operations (e.g. “Begin”, “Draw”, “End”, “beginTransmission”, “endTransmission”).

    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
     our
    -1.72
     poniendo
    -1.30
    正常的
    -1.27
    gatsby
    -1.25
     I
    -1.25
    ينا
    -1.23
     millón
    -1.22
    ,’”
    -1.22
     Sat
    -1.19
    reactstrap
    -1.19
    POSITIVE LOGITS
    }$.
    1.31
    signable
    1.30
    您可以
    1.27
     ของ
    1.22
     meilleures
    1.20
    ,
    1.20
    這
    1.14
     Generale
    1.13
     Ogni
    1.13
     menyen
    1.13
    Activations Density 0.011%

    No Known Activations