© 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. 84394
    Prev
    Next
    INDEX
    Explanations

    atorfree

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

    The neuron fires on numeric tokens (digits, subscripts, and other number‐bearing fragments) in mathematical or technical text.

    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
    場合は
    -0.99
     капю
    -0.98
     laterale
    -0.97
     posteriore
    -0.96
    ras
    -0.96
    Ḳ
    -0.94
    uniform
    -0.93
     potuto
    -0.90
    变得
    -0.90
    rometry
    -0.90
    POSITIVE LOGITS
     hábit
    1.18
    嘉宾
    1.11
     совмести
    1.09
     únicamente
    1.07
     cuenca
    1.01
     colección
    0.99
     vegetales
    0.98
     всеки
    0.98
    
    0.97
     bosques
    0.96
    Activations Density 0.013%

    No Known Activations