© 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. 40-GEMMASCOPE-2-RES-262K
    7. 213021
    Prev
    Next
    INDEX
    Explanations

    database performance

    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
    不
    0.52
    
    0.43
    证据
    0.40
     Закон
    0.39
    行
    0.39
    Billboard
    0.39
    證據
    0.36
    匯
    0.36
    ೧
    0.35
    cached
    0.35
    POSITIVE LOGITS
     quieran
    0.47
     quieren
    0.46
    श्वर
    0.45
     quieres
    0.44
    rän
    0.42
    ulen
    0.41
     necesitan
    0.38
    ःख
    0.38
     Townsend
    0.37
     lecithin
    0.37
    Activations Density 0.000%

    No Known Activations