© 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. 67210
    Prev
    Next
    INDEX
    Explanations

    import statements and code

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

    The neuron primarily detects numeric literals (digits) 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
     moeilijk
    -1.11
    IZON
    -1.02
     stappen
    -0.97
     pijn
    -0.96
     neemt
    -0.94
     triko
    -0.94
    exitRule
    -0.94
    clerosis
    -0.94
    ATIONS
    -0.93
     kyse
    -0.93
    POSITIVE LOGITS
     múl
    0.87
     Krzysz
    0.83
     but
    0.83
    lyse
    0.81
     quí
    0.81
    ,
    0.81
    flügel
    0.80
     alignItems
    0.80
     potem
    0.80
    owanie
    0.80
    Activations Density 0.005%

    No Known Activations