© 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. 22-GEMMASCOPE-RES-131K
    7. 35603
    Prev
    Next
    INDEX
    Explanations

    end of sentence or clause

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

    The neuron strongly activates on e-commerce–style product descriptions—text detailing items’ features, colors, sizes, and marketing/promotional language.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     a
    -1.51
    ;
    -1.42
    topRight
    -1.34
    shall
    -1.31
    usions
    -1.30
    usive
    -1.30
    的气息
    -1.30
     mathvariant
    -1.29
     будь
    -1.27
    BitConverter
    -1.26
    POSITIVE LOGITS
     they
    1.92
     people
    1.75
     because
    1.74
     różne
    1.71
     there
    1.71
     when
    1.69
     you
    1.64
     najdete
    1.63
     chcete
    1.63
    就會
    1.57
    Activations Density 0.214%

    No Known Activations