© 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. 22-GEMMASCOPE-RES-131K
    7. 93367
    Prev
    Next
    INDEX
    Explanations

    Occurs, group, education, degrees, double, continue

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

    This neuron fires on unusual or domain-specific tokens—that is, low-frequency or “rare” words (like metadata, maxOccurs, homoscedasticity, Rectangle, 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_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
     también
    -1.07
     hefty
    -1.02
    及其
    -1.01
     และ
    -0.98
     그리고
    -0.98
     amd
    -0.98
     alcune
    -0.96
    又要
    -0.96
    URAL
    -0.95
     Также
    -0.95
    POSITIVE LOGITS
     if
    1.08
     pasan
    1.05
     because
    1.05
     when
    1.02
    总之
    1.00
     using
    0.96
     Prompt
    0.96
     with
    0.90
    gunakan
    0.89
     prevenir
    0.88
    Activations Density 0.028%

    No Known Activations