© 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. 48267
    Prev
    Next
    INDEX
    Explanations

    text messages and texting

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

    The neuron detects occurrences of the substring “text” (in words like “text,” “texts,” “texting,” etc.).

    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
    寕
    -1.88
     slated
    -1.70
     terwijl
    -1.66
    ​—
    -1.62
     ندار
    -1.59
     trebui
    -1.57
    ࿙
    -1.56
     nôtre
    -1.55
     réglementation
    -1.55
     mantém
    -1.54
    POSITIVE LOGITS
     the
    1.94
     text
    1.73
     number
    1.71
     inicialmente
    1.55
     result
    1.54
    </i>
    1.53
     '
    1.52
    ti
    1.51
     us
    1.47
     induce
    1.46
    Activations Density 0.015%

    No Known Activations