© 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
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 15957
    Prev
    Next
    INDEX
    Explanations

    keywords from programming languages

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

    The neuron fires on numeric literal tokens (numbers) in code or comments.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    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
    嬉しいです
    -1.13
     citazioni
    -1.08
     it
    -1.03
    看著
    -1.00
    hradun
    -0.98
    liputi
    -0.98
     wist
    -0.97
    both
    -0.96
    だったら
    -0.96
    由于
    -0.95
    POSITIVE LOGITS
     at
    1.20
     együtt
    1.08
     an
    1.06
    pixmap
    1.04
     például
    1.03
    ：</
    1.03
     CONCLUSIONS
    0.98
     памя
    0.94
     tía
    0.94
     my
    0.94
    Activations Density 0.008%

    No Known Activations