© 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. 10-GEMMASCOPE-RES-131K
    7. 41504
    Prev
    Next
    INDEX
    Explanations

    matching multiple indices or sorting lists

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

    The neuron fires on non-English tokens—i.e. words or characters from other languages (Chinese, Japanese, Portuguese, etc.).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     -
    -2.83
    躋
    -2.52
    u
    -2.45
     sogenannten
    -2.44
     kleinen
    -2.42
     großen
    -2.42
     Want
    -2.41
     Ironically
    -2.41
     Initially
    -2.39
     crux
    -2.39
    POSITIVE LOGITS
    蟊
    3.08
     herren
    2.77
    CURLOPT
    2.72
    骜
    2.69
    猡
    2.67
     hini
    2.67
     entrando
    2.64
    蹀
    2.63
     rangs
    2.63
    AdvEx
    2.52
    Activations Density 0.002%

    No Known Activations