© 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. 31-GEMMASCOPE-2-RES-65K
    7. 51157
    Prev
    Next
    INDEX
    Explanations

    possible research

    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_31_width_65k_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
    一开始
    0.46
    出力
    0.45
     गिरी
    0.43
    也可
    0.43
    開始
    0.42
    下来
    0.42
    なっている
    0.42
    要
    0.42
    ones
    0.42
    花
    0.42
    POSITIVE LOGITS
     Relationship
    0.48
     liabilities
    0.42
     ámbito
    0.42
     sabot
    0.41
     subplot
    0.41
     День
    0.41
     नोटबंदी
    0.41
     Preheat
    0.41
     nostalgia
    0.41
     Expenditures
    0.40
    Activations Density 0.014%

    No Known Activations