© 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. 238719
    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
    ╅
    0.48
    ┼
    0.44
    茈
    0.41
    ▏
    0.40
    NavigationView
    0.40
    otropic
    0.39
    Roe
    0.39
    ティー
    0.39
    ጵ
    0.39
    Coe
    0.38
    POSITIVE LOGITS
    irl
    0.43
    nm
    0.40
    nat
    0.39
    explained
    0.39
    fant
    0.38
    nn
    0.37
    iru
    0.37
    nof
    0.36
     myst
    0.36
     imitated
    0.36
    Activations Density 0.006%

    No Known Activations