© 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. 76403
    Prev
    Next
    INDEX
    Explanations

    Fabrication, Size, AL, Raf

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

    The neuron strongly activates on tokens in news‐wire style political reports—particularly mentions of parties, rallies, dates, and locations in Bangladeshi politics.

    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
     skall
    -1.00
     Vám
    -0.88
     सम्
    -0.86
    一抹
    -0.78
    鬘
    -0.74
     lige
    -0.74
    pizz
    -0.74
     trả
    -0.73
    プロフィール
    -0.73
     desmotiv
    -0.71
    POSITIVE LOGITS
     Hef
    0.92
     кілька
    0.89
     היתה
    0.85
     sued
    0.85
    OMEM
    0.78
     Cocktail
    0.78
     tasked
    0.77
     vicepresidente
    0.77
    ykjav
    0.76
     wack
    0.76
    Activations Density 0.010%

    No Known Activations