© 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-65K
    7. 2941
    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
    ه
    0.52
     thalam
    0.51
    Missense
    0.50
    িনবার্গ
    0.49
     tabBar
    0.49
     全部
    0.48
     wan
    0.47
    мона
    0.47
     />}
    0.46
     lieux
    0.45
    POSITIVE LOGITS
    nd
    1.44
     thirds
    0.75
    thirds
    0.61
    nds
    0.61
     nd
    0.57
    nde
    0.52
    o
    0.52
    ndo
    0.52
    nda
    0.51
    ä
    0.51
    Activations Density 0.672%

    No Known Activations