© 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-3-27B-IT
    5. 41-GEMMASCOPE-2-RES-262K
    7. 6657
    Prev
    Next
    INDEX
    Explanations

    unusual observation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/resid_post_all/layer_41_width_262k_l0_big
    Prompts (Dashboard)
    238,145 prompts, 512 tokens each
    Dataset (Dashboard)
    lmsys + oasst1
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ায়ক
    0.49
    リーム
    0.48
    জ্জনক
    0.47
    ه‌ها
    0.46
    plication
    0.46
    有人
    0.45
    道路
    0.44
    𝐱
    0.43
     postaje
    0.43
     utilizing
    0.43
    POSITIVE LOGITS
     penchant
    0.91
     knack
    0.76
     fondness
    0.70
     ambitions
    0.64
     pretensions
    0.64
     aversion
    0.59
     Interesse
    0.57
     Verständnis
    0.56
     sympathies
    0.54
     tendency
    0.54
    Activations Density 0.003%

    No Known Activations