© 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
    5. 31-GEMMASCOPE-2-RES-262K
    7. 128410
    Prev
    Next
    INDEX
    Explanations

    God, sacred, safe concepts

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

    the cost accuracy corner safety

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-pt/resid_post/layer_31_width_262k_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
    ﺖ
    0.50
     મહે
    0.46
     deles
    0.45
     bulging
    0.45
     folklore
    0.45
     shielding
    0.44
     सदर
    0.44
     неожидан
    0.44
     dread
    0.43
     allegiance
    0.43
    POSITIVE LOGITS
    范围内
    0.63
    濩
    0.49
    范围
    0.47
    约
    0.47
    ボル
    0.46
     UpdateWindow
    0.46
    各类
    0.46
    周
    0.45
    有不少
    0.45
     ISE
    0.44
    Activations Density 0.001%

    No Known Activations