© 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. 110637
    Prev
    Next
    INDEX
    Explanations

    Osama bin

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

    The neuron consistently activates on capitalized multi‐letter tokens—proper names (e.g. “Osama”) and technical acronyms or domain-specific terms (e.g. “RPC,” “PPC,” “leech”).

    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
    0
    -3.55
    1
    -3.45
    '
    -3.27
     {
    -3.17
    <td>
    -2.92
     \
    -2.84
    2
    -2.80
    of
    -2.66
    镚
    -2.61
    4
    -2.38
    POSITIVE LOGITS
    鱟
    2.63
    鬭
    2.59
    pecabe
    2.53
    
    2.52
    pulseira
    2.50
     intricacies
    2.50
    تیجه
    2.47
    та
    2.47
    𝐒
    2.38
    ㉓
    2.38
    Activations Density 0.010%

    No Known Activations