© 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. 98637
    Prev
    Next
    INDEX
    Explanations

    numbers and commas

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

    This neuron stays inactive on all provided text and thus isn’t picking up any pattern in these examples.

    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
    (ﾟ
    -1.85
     bré
    -1.70
    ｺ
    -1.64
    ﾍ
    -1.64
     conster
    -1.64
    は
    -1.63
    ﾈ
    -1.63
    橚
    -1.59
    鬭
    -1.59
     banane
    -1.58
    POSITIVE LOGITS
    ,
    2.06
     '
    1.77
    ;
    1.64
     neuem
    1.63
    ,'
    1.61
     mehreren
    1.61
     bestimmten
    1.55
    ing
    1.55
     encontrar
    1.53
     irgende
    1.48
    Activations Density 0.025%

    No Known Activations