© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 21298
    Prev
    Next
    INDEX
    Explanations

    words ending in ed or ion

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

    The neuron is primarily detecting hyphens (dash tokens) in compound or hyphenated terms.

    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
     "
    -2.44
     two
    -2.22
     it
    -2.02
     of
    -1.99
     
    -1.98
     The
    -1.87
     most
    -1.84
     three
    -1.80
     by
    -1.80
     you
    -1.77
    POSITIVE LOGITS
     faciliter
    1.90
     deň
    1.85
     constituer
    1.84
    ścioła
    1.80
    giveness
    1.72
     favoriser
    1.72
    当我
    1.71
     individuelles
    1.70
     هستند
    1.68
     桥
    1.67
    Activations Density 0.283%

    No Known Activations