© 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. 31-GEMMASCOPE-2-RES-262K
    7. 29257
    Prev
    Next
    INDEX
    Explanations

    most damage or evil

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/resid_post/layer_31_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
    س
    0.50
    不一样
    0.50
     exemplify
    0.49
     باہنی
    0.49
    委员
    0.49
     Veranstaltung
    0.49
    🎋
    0.49
     کرتے
    0.48
     sostiene
    0.48
     pointA
    0.48
    POSITIVE LOGITS
    r
    0.47
    H
    0.40
    v
    0.39
    ERROR
    0.38
    Lock
    0.38
     Dragon
    0.38
     ধার
    0.38
     Lock
    0.37
    h
    0.37
     boundary
    0.37
    Activations Density 0.002%

    No Known Activations