© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-27B
    5. 31-GEMMASCOPE-2-RES-262K
    7. 216269
    Prev
    Next
    INDEX
    Explanations

    proxy software, learn from, can turn, this one, always wins, died from, Table 5

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

    ordering, tokenize, proactive mitigation

    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
    ло
    0.48
    常规
    0.46
    購入
    0.46
    硚
    0.44
    憑
    0.44
    刈
    0.44
    調
    0.43
    ਮ
    0.42
    ধা
    0.42
     playthrough
    0.42
    POSITIVE LOGITS
    Gilbert
    0.54
    that
    0.53
    Gib
    0.52
    Upp
    0.52
    ÈRES
    0.51
    PanelVisual
    0.50
     Caledonia
    0.50
     روسی
    0.49
    codewords
    0.49
     materna
    0.49
    Activations Density 0.000%

    No Known Activations