© 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. 40-GEMMASCOPE-2-RES-262K
    7. 24495
    Prev
    Next
    INDEX
    Explanations

    final answer in a box

    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
     \[
    0.48
    az
    0.43
    amp
    0.42
     (
    0.40
     $\
    0.39
     CO
    0.38
    ck
    0.38
     Al
    0.38
     <
    0.37
    rt
    0.37
    POSITIVE LOGITS
    マト
    0.46
    бно
    0.45
    ოლოგი
    0.45
    ิล
    0.42
    ților
    0.42
     GoName
    0.42
    बर
    0.42
    ποίη
    0.42
    ेल
    0.42
     barra
    0.41
    Activations Density 0.002%

    No Known Activations