© 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. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 35416
    Prev
    Next
    INDEX
    Explanations

    loops, cords, trim, pouch

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-32b-saes/saes_Qwen_Qwen3-32B_batch_top_k/resid_post_layer_32
    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
    åijĬè¯īä»ĸ
    -0.10
    criminal
    -0.09
     criminal
    -0.09
    alic
    -0.08
    lander
    -0.08
    éĢĢ
    -0.08
    ä¹Łæĺ¯ä¸Ģ个
    -0.08
    ntp
    -0.08
     clo
    -0.08
    åĪģ
    -0.08
    POSITIVE LOGITS
    åŀĤ
    0.12
    èĢ·
    0.11
    ordo
    0.10
    ikt
    0.10
    åĪ¶ä½ľ
    0.10
    -Feb
    0.10
    éĴ¡
    0.10
    è£½ä½ľ
    0.09
    æĮĤåľ¨
    0.09
    大èĥĨ
    0.09
    Activations Density 0.060%

    No Known Activations