© 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. 90355
    Prev
    Next
    INDEX
    Explanations

    names and technical terms

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

    This neuron fires on standalone names or identifiers—proper nouns and code/library identifiers.

    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
     They
    -3.23
     mereka
    -2.59
    狁
    -2.25
     perkara
    -2.22
     Они
    -2.19
    Ꮗ
    -2.16
     Their
    -2.06
    胧
    -2.05
     make
    -2.03
    鲽
    -2.02
    POSITIVE LOGITS
    which
    2.47
     kumaş
    2.44
     ſch
    2.42
    肫
    2.38
     kaplama
    2.38
    
    2.38
     taşlı
    2.34
    or
    2.31
     mágica
    2.23
    叫我
    2.23
    Activations Density 0.003%

    No Known Activations