© 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. 121400
    Prev
    Next
    INDEX
    Explanations

    embrace dark, strike tuned, rights or

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

    verbs for actions or states

    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
    땄
    0.47
     nič
    0.45
    。“
    0.45
    😧
    0.45
    '،
    0.43
    `
    0.42
     যদিও
    0.42
     ponieważ
    0.41
     pretože
    0.41
     obwohl
    0.41
    POSITIVE LOGITS
    es
    0.48
     reformulation
    0.46
    хе
    0.45
    ાલુ
    0.45
    felter
    0.44
    зим
    0.43
     қол
    0.43
    зге
    0.43
    ueness
    0.43
    粼
    0.43
    Activations Density 0.001%

    No Known Activations