© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 68884
    Prev
    Next
    INDEX
    Explanations

    questions and explanations

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

    The neuron activates on numeric tokens (digits or numbers), i.e. it detects numbers and numeric values in the 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
     السياس
    -1.58
    ὕ
    -1.38
    begins
    -1.34
    ਖ਼
    -1.31
     miembros
    -1.29
    ύ
    -1.29
    下一步
    -1.27
    Dzień
    -1.27
    varios
    -1.27
    
    -1.27
    POSITIVE LOGITS
     his
    1.62
     every
    1.55
     you
    1.54
     not
    1.52
     your
    1.49
     function
    1.38
    //
    1.38
     at
    1.34
     which
    1.32
     first
    1.31
    Activations Density 0.005%

    No Known Activations