© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-27B
    5. 31-GEMMASCOPE-2-RES-262K
    7. 99199
    Prev
    Next
    INDEX
    Explanations

    second-biggest, fraction, keys, recover, game, sun, stances, LT

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

    explaining importance or simple actions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-pt/resid_post/layer_31_width_262k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     overexpression
    0.52
     absenteeism
    0.50
     थिंक
    0.49
    🥐
    0.46
     epigenetic
    0.46
    यम
    0.45
     breaks
    0.45
     ಅಥವಾ
    0.45
    broke
    0.45
     ubiquitin
    0.43
    POSITIVE LOGITS
    넬
    0.45
     FOD
    0.45
     счет
    0.44
    เพื่อให้
    0.43
    язы
    0.42
    ды
    0.42
     GMC
    0.41
     Кра
    0.40
    html
    0.40
     помещения
    0.40
    Activations Density 0.001%

    No Known Activations