© 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-3-27B-IT
    5. 40-GEMMASCOPE-2-RES-262K
    7. 98690
    Prev
    Next
    INDEX
    Explanations

    humane or clean endings

    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
    embedded
    0.40
    өрд
    0.40
     balans
    0.40
    ходи
    0.38
    驅
    0.37
     угро
    0.37
    僮
    0.36
    棰
    0.36
    AppComponent
    0.36
    jesi
    0.36
    POSITIVE LOGITS
     painless
    0.77
     deaths
    0.64
     graceful
    0.64
     dignified
    0.64
     gracefully
    0.64
     cleanly
    0.64
     manner
    0.62
     abrupt
    0.60
    优雅
    0.59
    manner
    0.59
    Activations Density 0.038%

    No Known Activations