© 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. 238190
    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
    સભા
    0.50
    ádiz
    0.46
    AvlTree
    0.44
    phalodes
    0.43
    玾
    0.42
    ότερα
    0.41
    áne
    0.40
    Trajectories
    0.40
    ͖
    0.40
    ","",
    0.39
    POSITIVE LOGITS
     enormes
    0.43
     gave
    0.39
     puiss
    0.38
     mów
    0.38
     extrêmement
    0.38
     erset
    0.38
     enorme
    0.38
    substitution
    0.37
     tremendous
    0.37
     falar
    0.36
    Activations Density 0.001%

    No Known Activations