© 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
    5. 31-GEMMASCOPE-2-RES-262K
    7. 238905
    Prev
    Next
    INDEX
    Explanations

    distinct rogue call magic originally capturing

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-pt/resid_post/layer_31_width_262k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    apps
    0.46
    i
    0.45
    ior
    0.44
    iii
    0.44
    software
    0.42
     hashed
    0.42
    C
    0.41
     partners
    0.41
    des
    0.41
     software
    0.41
    POSITIVE LOGITS
    ົບ
    0.57
     ім
    0.51
    ımızı
    0.50
    ımıza
    0.50
     وا
    0.49
     ஓ
    0.49
     ഉ
    0.48
    ımızın
    0.48
     ഏറ്റവും
    0.48
    онер
    0.48
    Activations Density 0.000%

    No Known Activations