© 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. 76800
    Prev
    Next
    INDEX
    Explanations

    long-term, effectively, rich metadata

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

    This neuron fires on tokens that are entirely uppercase or are pure digit sequences—e.g. acronyms, all‐caps technical terms, and numeric identifiers.

    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
    odin
    -0.89
    enterOuterAlt
    -0.85
    Indices
    -0.79
    gerichte
    -0.77
    rdag
    -0.77
     pensées
    -0.77
    милия
    -0.76
    tory
    -0.75
    exitRule
    -0.75
    cellaneous
    -0.75
    POSITIVE LOGITS
     CDF
    1.05
     Chicago
    0.90
    CDF
    0.89
     Illinois
    0.88
     Fermi
    0.79
     Rap
    0.74
     Tev
    0.73
    消費者
    0.71
     D
    0.71
     UA
    0.69
    Activations Density 0.020%

    No Known Activations