© 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. 34-GEMMASCOPE-RES-131K
    7. 102635
    Prev
    Next
    INDEX
    Explanations

    numbers and dates

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

    The neuron detects transliterated Arabic personal names (with diacritics).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    acté
    -0.80
     Ottoman
    -0.79
     entré
    -0.78
    ≀
    -0.77
    kane
    -0.77
    VV
    -0.77
    Assets
    -0.75
     kral
    -0.74
     gide
    -0.74
    vereins
    -0.74
    POSITIVE LOGITS
     Uma
    0.87
    Uma
    0.79
    ond
    0.75
     Iraqi
    0.75
    Bates
    0.74
     Gubernur
    0.73
     pasirink
    0.73
     envelope
    0.72
     Ku
    0.71
     squares
    0.71
    Activations Density 0.014%

    No Known Activations