© 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. 34-GEMMASCOPE-RES-131K
    7. 58957
    Prev
    Next
    INDEX
    Explanations

    biological and technical fields

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

    The neuron identifies hyphenated compound words or technical terms, particularly in scientific and academic contexts.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron strongly activates on the dash character and the subword pieces immediately around it in hyphenated technical or compound terms.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    ENSE
    -0.83
    uales
    -0.80
     Système
    -0.78
    Nonnull
    -0.77
    quierda
    -0.77
     bru
    -0.73
    ariste
    -0.72
     понадоби
    -0.72
    изнь
    -0.70
    Indep
    -0.70
    POSITIVE LOGITS
    ственно
    0.93
     عرو
    0.86
     этих
    0.78
     rhinestone
    0.77
    ांकि
    0.77
    쁘
    0.75
    льно
    0.75
     Bessel
    0.75
     abd
    0.75
     Nieren
    0.73
    Activations Density 0.038%

    No Known Activations