© 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. 22-GEMMASCOPE-RES-131K
    7. 71648
    Prev
    Next
    INDEX
    Explanations

    code and control flow

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

    The neuron fires on numeric literals (number constants) in source 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_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
    pilih
    -1.30
     reposer
    -1.23
     horrid
    -1.21
    臜
    -1.20
     énorme
    -1.20
    GERMANY
    -1.20
     både
    -1.19
     soooo
    -1.18
     dekol
    -1.17
     rése
    -1.16
    POSITIVE LOGITS
     if
    2.06
     in
    1.70
     with
    1.69
     when
    1.54
     on
    1.48
     after
    1.46
     it
    1.36
     subsequently
    1.34
     at
    1.29
     then
    1.29
    Activations Density 0.007%

    No Known Activations