© 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. Qwen2.5-7B-IT
    5. 20-MATRYOSHKA-65K
    7. 1090
    Prev
    Next
    INDEX
    Explanations

    process or algorithm

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    chanind/qwen2.5-7B-it-layer-20-saes/pile/matryoshka/k-100
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     enthus
    -0.40
     safezone
    -0.32
    éĶ¦æłĩ
    -0.31
    建档ç«ĭ
    -0.30
     LENG
    -0.27
     misog
    -0.27
     Kardashian
    -0.26
    segue
    -0.25
    -UA
    -0.25
    /Internal
    -0.25
    POSITIVE LOGITS
    çļĦ带é¢Ĩ
    0.28
    çĹ¼
    0.28
    å¿ĥå¾Ĺä½ĵä¼ļ
    0.27
    çŀŁ
    0.27
    è¿Ļ个èģĮä¸ļ
    0.26
     lineWidth
    0.25
    åĻĺ
    0.25
    NamedQuery
    0.24
     each
    0.24
    çļĦç¼ĺ
    0.23
    Activations Density 1.286%

    No Known Activations