© 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. 22-GEMMASCOPE-RES-131K
    7. 27698
    Prev
    Next
    INDEX
    Explanations

    product names and technical terms

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

    The neuron activates on multi‐token proper names or technical classification labels (e.g. “Mac OS X,” “NCAA Division I,” “SQL Server,” “Oryza sativa,” “DDR4”).

    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
    あー
    -1.41
    必要がある
    -1.41
    pecabe
    -1.32
    の作成
    -1.29
    Ако
    -1.29
    が出来
    -1.28
    ݯ
    -1.28
     знал
    -1.27
    latimes
    -1.25
    の詳細
    -1.25
    POSITIVE LOGITS
    報酬
    1.33
    ,
    1.32
    が
    1.31
    を
    1.27
    そういった
    1.23
    軋
    1.23
     Grüßen
    1.23
     to
    1.16
    Према
    1.14
    abbaye
    1.11
    Activations Density 0.246%

    No Known Activations