© 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. 10-GEMMASCOPE-RES-131K
    7. 12229
    Prev
    Next
    INDEX
    Explanations

    character id or characterization

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

    The neuron strongly activates on occurrences of the token “Character.”

    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
    栳
    -3.20
    鬭
    -2.73
     icône
    -2.59
     maïs
    -2.53
    嵛
    -2.45
     évolution
    -2.44
    空気
    -2.39
     activité
    -2.39
    (
    -2.36
    覿
    -2.36
    POSITIVE LOGITS
    地说道
    3.03
     беже
    2.70
     需要
    2.64
    地说
    2.59
    ～！
    2.53
    3
    2.52
     zahlreichen
    2.45
    多数
    2.44
     is
    2.44
    行って
    2.42
    Activations Density 0.020%

    No Known Activations