© 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. 136205
    Prev
    Next
    INDEX
    Explanations

    variety and strategy

    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
    =/
    0.41
    oxygène
    0.41
    いつ
    0.41
     deceiving
    0.40
    ({"
    0.39
    ίες
    0.39
    做
    0.38
     Políticas
    0.38
     diciamo
    0.37
    ر
    0.37
    POSITIVE LOGITS
    ktional
    0.48
     Martine
    0.46
    elic
    0.45
    勳
    0.45
    phasis
    0.45
    jsdel
    0.45
    lıkla
    0.44
     innervation
    0.44
     emphas
    0.44
    getClient
    0.44
    Activations Density 0.002%

    No Known Activations