© 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. 48605
    Prev
    Next
    INDEX
    Explanations

    acians, otic, eagerly, also, ball, Cleveland

    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
    ç½¢äºĨ
    -0.10
     compart
    -0.09
    两岸
    -0.09
    _play
    -0.08
    彤
    -0.08
    温æ³ī
    -0.08
    Ñijн
    -0.08
    SHARE
    -0.08
    两端
    -0.08
    年代
    -0.08
    POSITIVE LOGITS
    躺çĿĢ
    0.09
    è¿ľæ´ĭ
    0.09
    éħª
    0.09
    èººåľ¨åºĬä¸Ĭ
    0.09
    lein
    0.08
    iant
    0.08
    æĭ¼
    0.08
    lov
    0.08
    éĤĢ请
    0.08
    èĽ¤
    0.08
    Activations Density 0.034%

    No Known Activations