© 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. 20491
    Prev
    Next
    INDEX
    Explanations

    disable or prevent actions

    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
    uther
    -0.06
    Åijd
    -0.05
    oor
    -0.05
     convinc
    -0.05
    ocr
    -0.05
    æĢĿè·¯
    -0.05
    ish
    -0.05
    uteur
    -0.05
    uve
    -0.05
     cr
    -0.05
    POSITIVE LOGITS
    æŁ»
    0.06
    æłĩåĩĨè¦ģæ±Ĥ
    0.06
    ä¸įåIJĪ
    0.06
    avanaugh
    0.06
    æ²ģ
    0.05
    ectes
    0.05
    è»
    0.05
    à¹Ģà¸Ĺà¸Ļ
    0.05
    æ¬
    0.05
    gx
    0.05
    Activations Density 0.071%

    No Known Activations