© 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. 40-GEMMASCOPE-2-RES-262K
    7. 168648
    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_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
    ДУ
    0.42
    땃
    0.42
    yecto
    0.41
    ЛЬ
    0.41
     elicited
    0.41
     র
    0.40
    owsky
    0.40
    bonded
    0.40
     ге
    0.39
     braid
    0.39
    POSITIVE LOGITS
    偷
    0.46
     advisers
    0.39
    첩
    0.38
     Taal
    0.38
     London
    0.38
    大战
    0.38
     অত
    0.37
     Osman
    0.36
     जगहों
    0.36
    iyasi
    0.36
    Activations Density 0.000%

    No Known Activations

    This feature has no known activations.