© 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. 18715
    Prev
    Next
    INDEX
    Explanations

    completion cycles turns

    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
    عÙĦÙĨت
    -0.07
    amment
    -0.07
    isial
    -0.07
    eksi
    -0.07
    ulong
    -0.07
    çĶ¨æ°´éĩı
    -0.07
    åĪ°ä¸ŃåĽ½
    -0.06
    ç»ıä¹ħ
    -0.06
    isione
    -0.06
    åĪ°ä¸Ń
    -0.06
    POSITIVE LOGITS
    次
    0.06
    äºĮ次
    0.06
    å¤ļ次
    0.05
    ä¸ī次
    0.05
    æĸĻ
    0.05
    olog
    0.05
     cycle
    0.05
     obie
    0.05
    次çļĦ
    0.05
    ä¸Ģ次
    0.05
    Activations Density 0.048%

    No Known Activations