© 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. 82386
    Prev
    Next
    INDEX
    Explanations

    digits ofdishes inall Kind) followed by canpressure can, followed by shouldI followed by should

    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
     generative
    0.44
     ਜਾਂ
    0.43
     అనేది
    0.41
    くなった
    0.41
    गात
    0.40
    🌈
    0.40
     உலகில்
    0.39
    咩
    0.38
     acontece
    0.38
     বছর
    0.38
    POSITIVE LOGITS
     superintend
    0.45
     should
    0.44
    Phys
    0.42
     kitchenette
    0.40
     $>
    0.40
    leyball
    0.39
     typhoid
    0.39
     dokt
    0.38
    蟎
    0.38
    ylamine
    0.38
    Activations Density 0.014%

    No Known Activations