© 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. 11483
    Prev
    Next
    INDEX
    Explanations

    short specific terms

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

    The neuron is picking out Chinese content words (multi-character Chinese tokens) rather than English or punctuation.

    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
    袛
    -1.92
    ᾿
    -1.73
     săptăm
    -1.66
     ligeramente
    -1.59
     ᾿
    -1.58
    骉
    -1.52
     ส์
    -1.52
    LIP
    -1.48
     tecnologías
    -1.48
     SMALL
    -1.47
    POSITIVE LOGITS
     –
    2.11
     have
    1.73
    </i>
    1.66
     truly
    1.63
     $-
    1.59
     more
    1.59
     increased
    1.49
     can
    1.46
     will
    1.45
    
    1.45
    Activations Density 0.005%

    No Known Activations