© 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. Qwen3.5-4B
    5. 15-RES-MATRYOSHKA-65K
    7. 22477
    Prev
    Next
    INDEX
    Explanations

    article cited

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    decoderesearch/qwen-3.5-saes/qwen-3.5-4b
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Negative Logits
    ç»Ŀ
    -0.06
    å½ĵ天
    -0.06
    åĿŀ
    -0.05
    ç»ĦæĪIJ
    -0.05
     acronym
    -0.05
    æ³ķåĪĻ
    -0.05
    æ¸Ŀ
    -0.05
    æĪĺäºī
    -0.05
    æĺİ
    -0.05
    imenti
    -0.05
    POSITIVE LOGITS
    éĹ®åĮ»çĶŁ
    0.06
    мееÑĤÑģÑı
    0.06
     |=
    0.06
     surre
    0.06
    åŃĺçļĦ
    0.06
    ä¼ļæľīä¸Ģ个
    0.06
    è®°ä¸ĭ
    0.05
    istringstream
    0.05
    åŃĺåľ¨çļĦ
    0.05
    çļĦæĶ¶åħ¥
    0.05
    Activations Density 0.001%

    No Known Activations