© 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. 10-GEMMASCOPE-RES-131K
    7. 42382
    Prev
    Next
    INDEX
    Explanations

    code comments and headers

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

    the neuron activates on number literals 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_10/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
     when
    -2.41
     because
    -2.30
     they
    -1.91
     after
    -1.89
     then
    -1.84
    して下さい
    -1.68
     during
    -1.66
     before
    -1.65
    0
    -1.64
    いただける
    -1.60
    POSITIVE LOGITS
     visuel
    1.91
    Ⅳ
    1.80
     intrig
    1.77
     rabat
    1.75
     poème
    1.75
     \&
    1.71
     automne
    1.68
    ствия
    1.68
    ке
    1.66
     télécommande
    1.66
    Activations Density 0.003%

    No Known Activations