© 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. 136793
    Prev
    Next
    INDEX
    Explanations

    prevent others

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

    magic, denial, forgetting, repetitive

    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
     absorbs
    0.59
     ABSORBED
    0.54
    achtree
    0.53
     изготов
    0.51
     বিটিআই
    0.49
    茘
    0.49
    ក្នុងការ
    0.49
     какую
    0.48
     другую
    0.48
     tejido
    0.47
    POSITIVE LOGITS
    回
    0.47
    珉
    0.45
    의
    0.44
    다
    0.44
    و
    0.43
    อ
    0.43
    Constitution
    0.43
    启
    0.43
    סה
    0.43
    주
    0.43
    Activations Density 0.001%

    No Known Activations