© 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. 129076
    Prev
    Next
    INDEX
    Explanations

    The neuron detects numeric literals (number constants) in code.

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

    variable assignment and access

    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
    åter
    -0.98
     salah
    -0.90
     aplik
    -0.87
    rodz
    -0.87
    płat
    -0.85
    dekor
    -0.85
     alebo
    -0.85
    vaig
    -0.85
     komfor
    -0.85
    dö
    -0.85
    POSITIVE LOGITS
     zároveň
    0.91
     jednocześnie
    0.90
     همچنین
    0.88
     also
    0.88
     ergänzt
    0.85
    by
    0.80
     anschließend
    0.79
     vermittelt
    0.78
    Sa
    0.78
    Bre
    0.77
    Activations Density 0.003%

    No Known Activations