© 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. 10-GEMMASCOPE-RES-131K
    7. 85403
    Prev
    Next
    INDEX
    Explanations

    code snippets

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

    The neuron strongly activates on numeric literals (numbers) appearing in 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_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
     like
    -1.92
     gives
    -1.77
    when
    -1.56
     delivers
    -1.55
     alleviate
    -1.54
     these
    -1.52
    ﾋ
    -1.51
     ensures
    -1.49
    瑧
    -1.47
     enhances
    -1.46
    POSITIVE LOGITS
     nowy
    1.88
     amazingly
    1.87
     unbelievably
    1.79
     EVEN
    1.73
    镲
    1.73
     astonishing
    1.72
     VERY
    1.72
     strikingly
    1.71
     vijf
    1.67
     wonderfully
    1.65
    Activations Density 0.065%

    No Known Activations