© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 72273
    Prev
    Next
    INDEX
    Explanations

    programming language keywords and syntax

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

    The neuron fires on numeric literal tokens (e.g. numbers used as constants or default parameter values) in code.

    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
     it
    -1.67
     where
    -1.58
    }$
    -1.45
     verá
    -1.34
    </b>
    -1.33
     possibility
    -1.30
    year
    -1.30
    isseurs
    -1.29
    两年
    -1.29
     даже
    -1.28
    POSITIVE LOGITS
    驁
    1.80
    Ꭴ
    1.60
    "$_
    1.59
     speelt
    1.52
     gericht
    1.49
    臾
    1.44
     Основ
    1.41
     arkite
    1.40
     prachtig
    1.39
    “-
    1.38
    Activations Density 0.011%

    No Known Activations