© 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. 34-GEMMASCOPE-RES-131K
    7. 2129
    Prev
    Next
    INDEX
    Explanations

    lauren, laurent, laura names

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

    The neuron fires strongly on tokens containing the substring “Laur” (e.g. names like Laurent, Lauren, Laura).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     settlement
    -1.30
     settling
    -1.10
     settlements
    -1.03
     settles
    -1.01
     fritas
    -1.00
     settle
    -0.96
     inconveniente
    -0.94
    Settlement
    -0.94
     resolving
    -0.92
    結論
    -0.91
    POSITIVE LOGITS
    様が
    0.91
     presque
    0.86
    motion
    0.82
     derfor
    0.80
     LEARNING
    0.78
     ダーク
    0.78
     SAFETY
    0.77
     piti
    0.77
    edges
    0.76
     dioxide
    0.76
    Activations Density 0.011%

    No Known Activations