© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-27B-IT
    5. 40-GEMMASCOPE-2-RES-262K
    7. 4251
    Prev
    Next
    INDEX
    Explanations

    language switching

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

    structural and formatting cues that signal organized, itemized sections—such as headings, subheadings with hyphens, bullets, and emphasized list headers indicating a breakdown or outline.

    oai_token-act-pair · gpt-5Triggered by @vetterc0
    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
    Asie
    0.25
     হইতে
    0.23
    ycop
    0.23
    CCCN
    0.23
    哫
    0.22
    ेंटीना
    0.22
    ζο
    0.21
     columnalign
    0.21
    ্রী
    0.21
    édération
    0.21
    POSITIVE LOGITS
    When
    0.24
     When
    0.23
    Whether
    0.22
    इ
    0.22
     Whether
    0.21
     모든
    0.21
     Using
    0.20
    <h2>
    0.20
     No
    0.20
     During
    0.20
    Activations Density 1.536%

    No Known Activations