© 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. 35065
    Prev
    Next
    INDEX
    Explanations

    question/answer pairs

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

    The neuron strongly activates on isolated numeric or punctuation tokens that mark document structure (e.g. page or section break numbers and standalone punctuation), so it’s detecting those article/layout boundary markers rather than regular words.

    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
    Tên
    -1.13
    ΑΡ
    -0.98
     luks
    -0.96
     Durchmesser
    -0.96
     चुनें
    -0.95
     След
    -0.95
     konsumen
    -0.94
     originali
    -0.94
    まず
    -0.93
    這個
    -0.93
    POSITIVE LOGITS
     overall
    1.13
    他没有
    1.10
     upcoming
    1.09
    他想
    1.08
     Friday
    1.06
     stats
    1.03
     tomorrow
    1.02
     Upcoming
    1.01
     Tuesday
    1.01
     stay
    1.01
    Activations Density 0.006%

    No Known Activations