© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3.5-4B
    5. 15-RES-MATRYOSHKA-65K
    7. 21597
    Prev
    Next
    INDEX
    Explanations

    help with

    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
    أ
    -0.05
    欣
    -0.05
    ÑĢеа
    -0.05
     supr
    -0.05
    é¢ľ
    -0.05
    ä½ĨåıĪ
    -0.05
    æĪijåľ¨è¿ĻéĩĮ
    -0.05
    å½¹
    -0.05
    standing
    -0.05
    reten
    -0.05
    POSITIVE LOGITS
    istel
    0.06
    å»ĵ
    0.06
    寻常
    0.06
    adamente
    0.06
    rohr
    0.06
    곡
    0.06
    大éĻĨçļĦ
    0.06
    åĻ¨çŃī
    0.06
    alts
    0.05
    anh
    0.05
    Activations Density 0.006%

    No Known Activations