© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 7399
    Prev
    Next
    INDEX
    Explanations

    data limits and selections

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

    The neuron fires on tokens used to specify or limit the number of results (e.g. “top,” “take,” “head,” “limit”).

    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
     propiet
    -1.42
     to
    -1.40
     aspira
    -1.39
    یکر
    -1.33
    PSU
    -1.32
     ronde
    -1.31
     reforma
    -1.30
    ’
    -1.30
    OCH
    -1.29
     hispan
    -1.29
    POSITIVE LOGITS
    خوب
    1.62
    '
    1.56
    ممنون
    1.55
     néhány
    1.45
    replyreport
    1.37
    わたしは
    1.35
    Nå
    1.34
    شما
    1.34
     действительно
    1.34
     ''
    1.33
    Activations Density 0.021%

    No Known Activations