© 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. 22-GEMMASCOPE-RES-131K
    4. 70512
    Prev
    Next
    INDEX
    Explanations

    using method/tool

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

    The neuron primarily responds to three-digit numeric tokens.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    ’
    -2.31
    や
    -2.28
     was
    -2.23
    を
    -1.98
    vereignty
    -1.73
    では
    -1.71
    參考資料
    -1.70
    をはじめ
    -1.69
    на
    -1.68
     produisons
    -1.64
    POSITIVE LOGITS
    ごめん
    2.39
     the
    2.34
    すみません
    2.31
    ごめんなさい
    2.28
    ウチ
    2.08
    ええ
    2.00
    えっ
    1.97
    喏
    1.95
    からお
    1.91
    ====
    1.91
    Activations Density 0.081%

    No Known Activations