© 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. 22601
    Prev
    Next
    INDEX
    Explanations

    file, glucose, designs, information, evidence, points, results, data

    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
    aktu
    -0.06
    atch
    -0.06
     kontakt
    -0.06
    ibro
    -0.06
    part
    -0.06
     '{$
    -0.06
    ılarak
    -0.06
    atching
    -0.06
    etan
    -0.05
    æŁIJ
    -0.05
    POSITIVE LOGITS
     Forget
    0.06
     rather
    0.06
    æĪIJæŀľ
    0.06
    æĪĸèĢħ说
    0.05
    æīįèĥ½çľŁæŃ£
    0.05
     sheer
    0.05
    oly
    0.05
    fø
    0.05
    idades
    0.05
     piuttosto
    0.05
    Activations Density 0.038%

    No Known Activations