© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 36391
    Prev
    Next
    INDEX
    Explanations

    specific technical terms

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

    The neuron activates on multi‐part scientific or mathematical jargon (complex domain‐specific technical terms and acronyms).

    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
     unicórnio
    -1.61
    providing
    -1.39
     cinquante
    -1.35
    があり
    -1.34
     conster
    -1.32
    が
    -1.31
     geringe
    -1.30
    神
    -1.28
     magnific
    -1.28
     pesso
    -1.28
    POSITIVE LOGITS
     and
    1.68
     it
    1.42
     or
    1.38
    1.36
     With
    1.30
     It
    1.29
     as
    1.28
     This
    1.25
     For
    1.20
    づくり
    1.18
    Activations Density 0.218%

    No Known Activations