© 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
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 130135
    Prev
    Next
    INDEX
    Explanations

    The neuron responds most strongly to programming identifiers and keywords written with uppercase letters or CamelCase (e.g. class/method names and SQL keywords).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    list items, option calls

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     its
    -1.10
    なんですね
    -0.96
    oplayer
    -0.95
     vêtement
    -0.91
    ремя
    -0.91
    🥩
    -0.91
    ].)
    -0.90
     segundo
    -0.90
     bienvenue
    -0.89
    ฿
    -0.88
    POSITIVE LOGITS
     Lastly
    1.13
    Lastly
    1.09
     lastly
    1.05
     schließlich
    1.03
    最後に
    0.97
    Kesimpulan
    0.81
    至于
    0.79
    ुष
    0.78
    ？（
    0.76
    嗣
    0.76
    Activations Density 0.002%

    No Known Activations