© 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. 11616
    Prev
    Next
    INDEX
    Explanations

    science and mathematics concepts

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

    The neuron triggers strongly on technical, domain‐specific nouns and measurement terms (e.g. “mass,” “intestinal,” “regression,” “spectrometry”), effectively picking out specialized scientific vocabulary.

    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
     современного
    -1.63
    おわりに
    -1.63
     พ
    -1.57
    ebly
    -1.55
    tempted
    -1.48
    before
    -1.48
     возник
    -1.42
     источники
    -1.41
    mination
    -1.40
    two
    -1.40
    POSITIVE LOGITS
     also
    2.50
     rocha
    1.55
    AMILTON
    1.55
     également
    1.47
     occhiali
    1.46
     пси
    1.44
    這是
    1.44
    ıldığı
    1.43
    一分
    1.43
     pascua
    1.38
    Activations Density 0.504%

    No Known Activations