© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3.5-4B
    5. 15-RES-MATRYOSHKA-65K
    7. 20087
    Prev
    Next
    INDEX
    Explanations

    touch panel, footman, tempeh, bucket, methyltransferase, singing, red colobus

    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
    Not in Any Lists

    No Comments

    Negative Logits
    daf
    -0.06
     wid
    -0.06
    alc
    -0.06
     })).
    -0.06
    TON
    -0.05
    è±IJ
    -0.05
    aker
    -0.05
    ë¶Ģ
    -0.05
    èĨ
    -0.05
    stv
    -0.05
    POSITIVE LOGITS
    صÙĦØ©
    0.05
     ocas
    0.05
    为好
    0.05
    大人ãģ®
    0.05
    ulut
    0.05
    å¦ĤæĦı
    0.05
    岸边
    0.05
    pth
    0.05
    estig
    0.05
    轧
    0.05
    Activations Density 0.167%

    No Known Activations