© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 108838
    Prev
    Next
    INDEX
    Explanations

    software frameworks

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

    The neuron is detecting long runs of punctuation used as decorative separators (e.g. lines of dashes or pipes) in code 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_34/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
     raw
    -0.81
     formats
    -0.73
     Mel
    -0.73
     πλη
    -0.70
     drivers
    -0.70
     now
    -0.69
     punch
    -0.69
    蒐
    -0.69
     correlation
    -0.68
    IGraph
    -0.68
    POSITIVE LOGITS
     framework
    2.06
     Framework
    1.95
    框架
    1.79
    framework
    1.77
    Framework
    1.66
     FRAMEWORK
    1.63
     frameworks
    1.59
    フレーム
    1.22
     Frameworks
    1.20
    FRAME
    1.19
    Activations Density 0.180%

    No Known Activations