© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 91447
    Prev
    Next
    INDEX
    Explanations

    "Sop" or "Soph" prefixes

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

    The neuron fires on any token containing the “soph” (or “sop(h)”) letter‐sequence, e.g. names like Sophus, Sophos, SOPH, 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_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
    倌
    -0.75
    令人
    -0.72
     Force
    -0.72
    原理
    -0.70
     Dead
    -0.70
     歯科
    -0.70
    gered
    -0.69
    olf
    -0.69
     Beam
    -0.69
    hed
    -0.68
    POSITIVE LOGITS
     Sop
    1.26
     sop
    1.19
    Sop
    1.09
     SOP
    1.02
    sop
    1.01
     soph
    1.01
     Soph
    1.00
     Sopho
    0.94
    SOP
    0.94
    Soph
    0.85
    Activations Density 0.014%

    No Known Activations