© 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. 2308
    Prev
    Next
    INDEX
    Explanations

    Sal, Orthodox Christian, Ialin, Rosetti, conquer Ganja

    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
    ä¾Ĺ
    -0.09
    zimmer
    -0.09
    æĥ³çŁ¥éģĵ
    -0.09
    Ideal
    -0.09
    踢
    -0.08
    åľ¨è¿ĩ渡
    -0.08
    çIJ¯
    -0.08
     coerc
    -0.08
     Stern
    -0.08
     Cald
    -0.08
    POSITIVE LOGITS
    ä¸į失
    0.10
    ä¸įæĸŃåľ°
    0.09
    ä½Ľ
    0.09
     Bod
    0.08
    Metric
    0.08
    èĪī
    0.08
    ä¸įæĸŃæıIJé«ĺ
    0.08
    ä¿Ħ
    0.08
    éĵº
    0.08
     thaw
    0.08
    Activations Density 0.080%

    No Known Activations