© 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. 10-GEMMASCOPE-RES-131K
    7. 71186
    Prev
    Next
    INDEX
    Explanations

    dataType, Css, Kodak, odd

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

    The neuron is tuning to programming‐style identifiers—especially CamelCase names and uppercase acronyms.

    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
    -3.30
     But
    -2.88
     sogenannten
    -2.50
     }
    -2.44
     '
    -2.42
     Also
    -2.42
     Hinsicht
    -2.25
    "]
    -2.22
     ...
    -2.20
     For
    -2.17
    POSITIVE LOGITS
    攵
    2.58
    های
    2.36
    雳
    2.20
    𝗦
    2.19
     Siehe
    2.19
    牁
    2.14
     camiset
    2.13
    ’.
    2.11
    隹
    2.11
    up
    2.09
    Activations Density 0.001%

    No Known Activations