© 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. 51214
    Prev
    Next
    INDEX
    Explanations

    step-by-step solution and simplification

    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
     DNS
    0.44
     吃
    0.44
     zs
    0.42
    ÔNG
    0.40
    டன்
    0.40
    DNS
    0.40
     Eat
    0.40
     Consortium
    0.40
     Enemies
    0.40
     غ
    0.39
    POSITIVE LOGITS
    trying
    0.42
    Trying
    0.41
     Bine
    0.40
    числения
    0.40
     해결
    0.39
    ালিয়া
    0.39
    Blank
    0.38
    simplified
    0.38
    Alright
    0.38
    matplotlib
    0.37
    Activations Density 0.004%

    No Known Activations