© 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. 252094
    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_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
    ك
    0.54
     l
    0.47
    ͎
    0.43
    чні
    0.41
     Popov
    0.41
    <0xAE>
    0.40
    نه
    0.40
    َه
    0.40
    صل
    0.40
    ق
    0.40
    POSITIVE LOGITS
     सक्षम
    0.47
     मजेदार
    0.47
     Casualty
    0.47
    Access
    0.47
     новое
    0.47
     tenancy
    0.47
     жиз
    0.46
     रोचक
    0.45
     उपलब्
    0.45
     исключением
    0.44
    Activations Density 0.007%

    No Known Activations