© 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-262K
    7. 20909
    Prev
    Next
    INDEX
    Explanations

    emerging issue

    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_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
    ра
    0.48
    น
    0.44
     financi
    0.43
    โทร
    0.42
    ร
    0.41
    ро
    0.40
    為了
    0.39
    挀
    0.39
    ړ
    0.39
     ہتھی
    0.38
    POSITIVE LOGITS
    issense
    0.43
     outweighs
    0.43
     negative
    0.42
     Ep
    0.42
    negative
    0.42
     unanswered
    0.41
    incidence
    0.41
     või
    0.40
     objectionable
    0.40
     are
    0.39
    Activations Density 0.013%

    No Known Activations