© 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-IT
    5. 40-GEMMASCOPE-2-RES-262K
    7. 1644
    Prev
    Next
    INDEX
    Explanations

    Programs and optional features

    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/layer_40_width_262k_l0_medium
    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
     Namely
    0.50
     Moch
    0.48
     Fir
    0.43
     Siv
    0.42
     SM
    0.42
    ানের
    0.40
     Mans
    0.40
    __))
    0.40
     وه
    0.39
     सिल
    0.39
    POSITIVE LOGITS
     goalie
    0.48
     exorc
    0.43
     дыха
    0.43
    悯
    0.43
     youngsters
    0.43
    保持
    0.42
     испол
    0.42
     राक्ष
    0.42
     музыка
    0.41
     holidays
    0.40
    Activations Density 0.000%

    No Known Activations