© 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. 31-GEMMASCOPE-2-RES-262K
    7. 81927
    Prev
    Next
    INDEX
    Explanations

    wider open

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/resid_post/layer_31_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
     देखील
    0.52
    ؟!
    0.52
     యొక్క
    0.49
     మరి
    0.48
    ütfen
    0.46
    ೀ
    0.46
    惬
    0.46
    ?!
    0.45
    willReturn
    0.45
    ﭔ
    0.45
    POSITIVE LOGITS
     initially
    0.71
     inicialmente
    0.65
     zunächst
    0.64
     முதலில்
    0.62
     Initially
    0.60
     first
    0.58
     began
    0.57
     まず
    0.57
     devised
    0.56
     because
    0.55
    Activations Density 0.022%

    No Known Activations