© 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. 143816
    Prev
    Next
    INDEX
    Explanations

    actually works

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

    preserving text structure and format

    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
     RESERVE
    0.55
    ot
    0.46
    one
    0.44
     reserve
    0.43
    rush
    0.43
    Rush
    0.43
    at
    0.42
    鲇
    0.42
     Reserve
    0.42
    DOWN
    0.42
    POSITIVE LOGITS
     antice
    0.50
     actuación
    0.50
     effet
    0.48
     ض
    0.47
     منت
    0.45
     വരെ
    0.45
    جلس
    0.45
     サン
    0.44
     ставка
    0.44
    ництво
    0.44
    Activations Density 0.001%

    No Known Activations