© 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. 86826
    Prev
    Next
    INDEX
    Explanations

    triumph

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

    The neuron fires on occurrences of the root “triumph” (in any form, e.g. triumph, triumphal, Triumph).

    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
     Dear
    -0.72
    FLICT
    -0.72
    is
    -0.70
    További
    -0.69
     Drit
    -0.69
     kilogram
    -0.68
     Helpful
    -0.68
    Acknowledgements
    -0.68
    html
    -0.68
     Luxus
    -0.67
    POSITIVE LOGITS
    triumph
    1.59
     triumph
    1.48
    Triumph
    1.45
     Triumph
    1.43
     triumphant
    1.25
     triunfo
    1.02
     ENTRY
    0.96
     triumphed
    0.91
     triumphantly
    0.91
     triun
    0.87
    Activations Density 0.023%

    No Known Activations