© 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
    5. 31-GEMMASCOPE-2-RES-16K
    7. 13939
    Prev
    Next
    INDEX
    Explanations

    reducing negative outcomes

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

    prevent overfitting

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

    optimization

    np_max-act-logits · claude-4-5-sonnetTriggered by @sripadkarne
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-pt/resid_post/layer_31_width_16k_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
     kontek
    0.51
    电设备
    0.50
    犯罪
    0.49
    绘画
    0.48
    个人
    0.47
    东京
    0.46
     kriminal
    0.46
    贰
    0.46
    ாய்ச்ச
    0.45
    赁
    0.45
    POSITIVE LOGITS
     optimizing
    0.62
     minimizing
    0.61
     maximized
    0.61
     maximizes
    0.59
     minimizes
    0.58
     optimization
    0.57
     minimization
    0.57
     preventing
    0.56
     optimized
    0.56
     avoiding
    0.55
    Activations Density 0.180%

    No Known Activations