© 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. 75641
    Prev
    Next
    INDEX
    Explanations

    lone Indian survivors

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

    The neuron chiefly fires on proper-name tokens (e.g. country, team, trophy, player names)—i.e. named entities in the sports context.

    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
    型の
    -0.75
    zid
    -0.73
    vamos
    -0.71
    itek
    -0.69
     спра
    -0.69
    タン
    -0.69
    ហ
    -0.68
    ⇀
    -0.67
     zak
    -0.67
     waf
    -0.66
    POSITIVE LOGITS
    opley
    0.86
    НС
    0.85
     Indians
    0.85
    clyde
    0.84
    編曲
    0.83
     India
    0.83
     oči
    0.81
    いらっしゃ
    0.81
    ⣿
    0.79
     рождения
    0.77
    Activations Density 0.010%

    No Known Activations