© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 21788
    Prev
    Next
    INDEX
    Explanations

    get started or subscribe

    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
     Anat
    -0.08
    并ä¸İ
    -0.08
    tero
    -0.08
    repos
    -0.08
    æĵĢ
    -0.08
    ä¹¾åĿ¤
    -0.08
    æīĵçł´äºĨ
    -0.07
    каÑĤ
    -0.07
    ç¥ŀèī²
    -0.07
    bolt
    -0.07
    POSITIVE LOGITS
    _units
    0.11
    å±ŀäºİèĩªå·±
    0.10
     subscribe
    0.10
     доб
    0.10
    å¼ĢæĿ¥
    0.10
     subscri
    0.10
    _subscribe
    0.10
    订éĺħ
    0.10
     subscribing
    0.09
    arf
    0.09
    Activations Density 0.001%

    No Known Activations