© 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. 1054
    Prev
    Next
    INDEX
    Explanations

    compare

    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
     widest
    -0.23
     simplest
    -0.23
     WHETHER
    -0.22
     sublic
    -0.22
    =<?=$
    -0.21
    =<?=
    -0.21
    enment
    -0.20
     weakest
    -0.20
    acen
    -0.19
    ä»ħ次
    -0.19
    POSITIVE LOGITS
    æĽ´åĬł
    0.55
     more
    0.52
    æĽ´é«ĺçļĦ
    0.51
    æĽ´
    0.50
     better
    0.49
    æĽ´å¤ļçļĦ
    0.48
     æĽ´
    0.48
    æĽ´å¤ļ
    0.46
    æĽ´å¥½çļĦ
    0.46
     fewer
    0.45
    Activations Density 4.981%

    No Known Activations