© 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. 21406
    Prev
    Next
    INDEX
    Explanations

    keep the clear

    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
    egn
    -0.07
    æĹłå¤Ħ
    -0.06
    åħ®
    -0.06
    æľ¬åľºæ¯ĶèµĽ
    -0.05
    jum
    -0.05
    acu
    -0.05
    gara
    -0.05
     dat
    -0.05
    æĪijæ¯ı天
    -0.05
    ê°IJ
    -0.05
    POSITIVE LOGITS
    adoras
    0.07
    ä¸ĩä¸Ģ
    0.05
    adores
    0.05
     manch
    0.05
    çİ°å®ŀ
    0.05
     indeb
    0.05
     YEARS
    0.05
    aderas
    0.05
     możliwo
    0.05
    istica
    0.05
    Activations Density 0.001%

    No Known Activations