© 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. 37629
    Prev
    Next
    INDEX
    Explanations

    languages; if your agent ejects a word or phrase that you do not speak, such as wireless or CTA (which are a kind of website mock-up and the phrase for call to

    np_acts-logits-general · gemini-2.5-flash-lite

    adjustable levers and sliders

    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
    magnetic
    0.44
    create
    0.43
    magnet
    0.43
    marketing
    0.42
    營
    0.42
    s
    0.41
    cheap
    0.41
     cheap
    0.41
    GP
    0.41
    k
    0.41
    POSITIVE LOGITS
    窣
    0.49
    𝓑
    0.48
     اتارنا
    0.48
    StarCore
    0.44
    حض
    0.43
    並不
    0.43
    Blu
    0.43
    Anagram
    0.42
     синд
    0.42
     indentation
    0.41
    Activations Density 0.000%

    No Known Activations