© 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. 12626
    Prev
    Next
    INDEX
    Explanations

    cumulative exposure, elaboration, supervised

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

    The neuron fires most strongly on low‐frequency, content‐heavy tokens (i.e. uncommon or specialized words).

    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
     mengucap
    -1.31
     beleid
    -1.23
     begren
    -1.23
     mengucapkan
    -1.23
     menyukai
    -1.18
     mengangg
    -1.18
     stoppen
    -1.17
    рки
    -1.17
     menyadari
    -1.17
     laajenta
    -1.16
    POSITIVE LOGITS
     will
    1.80
     and
    1.49
     until
    1.45
     seven
    1.36
     provide
    1.33
     subsequent
    1.33
     next
    1.32
     current
    1.32
     previous
    1.30
     typically
    1.29
    Activations Density 0.213%

    No Known Activations