© 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. 67490
    Prev
    Next
    INDEX
    Explanations

    names, categories, specific terms

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

    The neuron fires on short, standalone title‐ or heading‐style phrases (e.g. product names or document titles) rather than running sentence text.

    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
     cuma
    -1.14
     Reino
    -1.10
     diamantes
    -1.05
     pantal
    -1.05
     éter
    -1.05
     Diciembre
    -1.04
    *********/
    -1.03
     nena
    -1.03
     bayi
    -1.02
     septem
    -1.02
    POSITIVE LOGITS
     enthält
    1.04
     dwie
    0.99
    athered
    0.97
    就可以了
    0.94
     pojawia
    0.90
     weiteren
    0.90
    或者是
    0.89
     überprü
    0.88
     السياس
    0.86
     позволяет
    0.86
    Activations Density 0.000%

    No Known Activations