© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 93982
    Prev
    Next
    INDEX
    Explanations

    motorcycles or racing

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

    The neuron activates on numeric tokens and alphanumeric model identifiers—essentially spotting bike model numbers, years, and other digit-based references.

    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
    alım
    -0.83
    msford
    -0.82
     Bảo
    -0.79
    uário
    -0.77
    Cano
    -0.77
    Selector
    -0.74
    ký
    -0.73
     αὐ
    -0.71
     Tư
    -0.71
    𝟯
    -0.71
    POSITIVE LOGITS
    JR
    0.87
     Thousand
    0.82
     thou
    0.81
     Mille
    0.76
     race
    0.71
     Track
    0.70
    renn
    0.70
     Pani
    0.69
    千
    0.69
     entertain
    0.68
    Activations Density 0.006%

    No Known Activations