© 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. 1799
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    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
    )
    0.15
    <0xE3>
    0.15
     ـ
    0.15
    indication
    0.14
    ycle
    0.14
    instruction
    0.14
    ี้ย
    0.14
     نیست
    0.14
     Mereka
    0.14
    ^{\
    0.14
    POSITIVE LOGITS
     getting
    0.37
     creating
    0.36
     finding
    0.36
     making
    0.36
    如何
    0.36
     how
    0.35
     achieving
    0.35
     obtaining
    0.34
     attaining
    0.34
     maintaining
    0.34
    Activations Density 1.290%

    No Known Activations