© 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-262K
    7. 194433
    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
    噎
    0.39
     അക
    0.37
    venir
    0.37
     latéral
    0.36
    δικ
    0.35
    Korean
    0.34
    腌
    0.34
    贩
    0.34
    オシャレ
    0.34
     ಭಾಗ
    0.34
    POSITIVE LOGITS
    ždy
    0.42
     siempre
    0.42
     preserv
    0.40
     constantly
    0.40
     shift
    0.39
     tetrachloride
    0.39
    uidado
    0.39
    үм
    0.38
     save
    0.38
     klin
    0.37
    Activations Density 0.000%

    No Known Activations