© 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. Qwen2.5-7B-IT
    5. 20-MATRYOSHKA-65K
    7. 4321
    Prev
    Next
    INDEX
    Explanations

    unexpected characters

    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
    Negative Logits
     AuÃŁerdem
    -0.23
    elay
    -0.23
    ollapsed
    -0.22
    ]={↵
    -0.21
    twor
    -0.21
    UniqueId
    -0.21
    aload
    -0.21
    '].$
    -0.20
    ::.
    -0.20
    á»ĥu
    -0.20
    POSITIVE LOGITS
    ç»ĵæŀľæĺ¾ç¤º
    0.50
    ç»ĵæŀľ
    0.40
    çļĦç»ĵæŀľ
    0.38
     resulted
    0.37
    çµIJæŀľ
    0.35
    æŀľçĦ¶
    0.33
     results
    0.31
     Result
    0.31
     resulting
    0.30
    å¾Ĺåĩº
    0.30
    Activations Density 1.056%

    No Known Activations