© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 15048
    Prev
    Next
    INDEX
    Explanations

    镜像, 进入, 运行状况

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

    The neuron fires on the special leading whitespace or indent markers (the “ ” token) used to denote bulleted or numbered list items.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     on
    -2.63
     in
    -2.14
     this
    -2.13
     Shortly
    -2.13
     When
    -1.95
     after
    -1.91
     went
    -1.90
     before
    -1.88
     with
    -1.86
     some
    -1.81
    POSITIVE LOGITS
     unſ
    2.23
    ⟞
    2.23
     revamped
    2.13
     nové
    2.09
     hefty
    1.98
     simpel
    1.96
     quirky
    1.88
     direkt
    1.86
    ໊
    1.85
    妡
    1.84
    Activations Density 0.002%

    No Known Activations