© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-27B-IT
    5. 40-GEMMASCOPE-2-RES-262K
    7. 10336
    Prev
    Next
    INDEX
    Explanations

    rate, quantity, or comparison

    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
    他說
    0.55
    တွေကို
    0.52
    DummyView
    0.50
    萂
    0.49
    checkFor
    0.48
     أعمال
    0.46
    㮔
    0.46
    訒
    0.46
     машиналарын
    0.46
     infrastrukt
    0.46
    POSITIVE LOGITS
    0.56
     <=
    0.53
     ≥
    0.52
     ≤
    0.51
    level
    0.51
    ≤
    0.51
    5
    0.50
     reaches
    0.49
    7
    0.49
     level
    0.49
    Activations Density 0.012%

    No Known Activations