© 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. 226186
    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
    u
    0.57
    m
    0.56
    er
    0.55
    o
    0.54
    et
    0.50
    fig
    0.50
    du
    0.50
    of
    0.49
    en
    0.48
    direction
    0.48
    POSITIVE LOGITS
     Уже
    0.58
     apabila
    0.56
     wygl
    0.54
     अगर
    0.53
     gospod
    0.53
     puedan
    0.52
     mengucapkan
    0.52
    阍
    0.52
    땔
    0.51
     Моло
    0.51
    Activations Density 0.007%

    No Known Activations