© 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. 34254
    Prev
    Next
    INDEX
    Explanations

    crumble, crumpled, rubble, scrambled

    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
    smith
    -0.10
    çijķ
    -0.09
    åŁĭ
    -0.09
    æ¡ģ
    -0.09
    çĻ¸
    -0.09
    acular
    -0.09
    æīģ
    -0.09
    UMENT
    -0.09
    lag
    -0.09
    /operator
    -0.09
    POSITIVE LOGITS
    认è¯ĨåĪ°
    0.10
    æľµ
    0.10
    羣å®ŀ
    0.09
    æĮĩæĮ¥éĥ¨
    0.09
    .Guid
    0.08
    çļĦæĸ°
    0.08
     SALE
    0.08
     ÐIJлекÑģан
    0.08
    éĤij
    0.08
    .Len
    0.08
    Activations Density 0.002%

    No Known Activations