© 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. 34-GEMMASCOPE-RES-131K
    7. 31505
    Prev
    Next
    INDEX
    Explanations

    names starting with De

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

    This neuron detects subword tokens that begin with a capital “D,” i.e. fragments of proper names or titles starting with “D.”

    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
    Whether
    -0.91
    whether
    -0.84
     whether
    -0.84
     WHETHER
    -0.84
     Whether
    -0.80
    cycling
    -0.77
    cheur
    -0.73
    aree
    -0.71
    bilità
    -0.70
     kiện
    -0.70
    POSITIVE LOGITS
     Exclu
    0.81
     SIC
    0.79
     Gundam
    0.76
    ัจ
    0.74
    Таким
    0.73
    EDI
    0.73
    EDY
    0.72
     exclusively
    0.72
    ißt
    0.72
     nowadays
    0.71
    Activations Density 0.037%

    No Known Activations