© 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. 10-GEMMASCOPE-RES-131K
    7. 13674
    Prev
    Next
    INDEX
    Explanations

    won competition or award

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

    The neuron detects tokens referring to competition results and placements (e.g. “won,” “fourth,” “champion,” medal colors).

    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
    ↵↵↵
    -1.74
    {
    -1.57
    sistent
    -1.52
    ，
    -1.48
    ]
    -1.46
     they
    -1.45
    hauled
    -1.44
    「
    -1.43
     ">=
    -1.42
    <?>>
    -1.42
    POSITIVE LOGITS
     of
    1.81
     a
    1.67
     pensamiento
    1.47
    𞤢
    1.40
     velocità
    1.40
     his
    1.36
     böyle
    1.36
    organizer
    1.34
     sentito
    1.34
    stiker
    1.34
    Activations Density 0.029%

    No Known Activations