© 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. 47643
    Prev
    Next
    INDEX
    Explanations

    amyloidosis, book, or log

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

    The neuron selectively fires on longer, domain‐specific multi-syllabic content words (often rare technical terms or proper nouns).

    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
    recently
    -1.97
    퐼
    -1.66
     広場
    -1.66
     getNome
    -1.60
     GETTING
    -1.59
    鹀
    -1.58
     westfalen
    -1.55
     CONSTANT
    -1.53
    ものである
    -1.53
     "")
    -1.52
    POSITIVE LOGITS
     for
    2.13
     a
    1.93
     with
    1.85
     does
    1.81
     more
    1.81
     any
    1.80
     may
    1.80
     only
    1.80
     on
    1.80
     can
    1.76
    Activations Density 0.141%

    No Known Activations