© 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
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 92944
    Prev
    Next
    INDEX
    Explanations

    mathematical definitions or descriptions

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

    The neuron strongly activates on structural or metadata markers—e.g. colons, commas, domain or tag keywords (like “type,” “com,” “unicode_”), and other punctuation used in headers or code-like metadata.

    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
    tuu
    -1.07
     verdo
    -0.97
    Writ
    -0.97
    jsh
    -0.94
     permett
    -0.94
     permettra
    -0.94
     最后
    -0.91
    gamis
    -0.91
    ducati
    -0.91
    pedro
    -0.90
    POSITIVE LOGITS
    Якщо
    1.27
    れません
    1.02
     Integrals
    0.99
    Напомним
    0.98
    Основ
    0.96
     then
    0.94
    اگر
    0.94
    Уже
    0.93
    렐
    0.93
    nymi
    0.91
    Activations Density 0.000%

    No Known Activations