© 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-65K
    7. 69
    Prev
    Next
    INDEX
    Explanations

    ensures better outcomes

    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_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
     depends
    0.18
    '
    0.18
     celles
    0.18
    ns
    0.18
    osi
    0.18
    nx
    0.17
    ↵↵↵↵↵
    0.16
     If
    0.16
    cn
    0.16
    enes
    0.16
    POSITIVE LOGITS
     increased
    0.34
     greater
    0.29
     unprecedented
    0.29
     heightened
    0.28
     постепен
    0.27
    increased
    0.27
     gradual
    0.27
     peningkatan
    0.27
     повышен
    0.27
     дальней
    0.26
    Activations Density 1.481%

    No Known Activations