© 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. 35069
    Prev
    Next
    INDEX
    Explanations

    drug value", "regulatory authorities", "WHO's ATC classification", "pharmacokinetics", "crystallization", "Protein--protein interactions", "Sports", "dataverse"

    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
    éĴī
    -0.13
    çīĽ
    -0.09
    nia
    -0.09
    åĽŀåij³
    -0.09
    tes
    -0.09
    æĮĤåľ¨
    -0.08
    _fa
    -0.08
    è´Ł
    -0.08
    å·¥ç¨ĭæĬĢæľ¯
    -0.08
    çĥŃè¡Ģ
    -0.08
    POSITIVE LOGITS
    èĿł
    0.11
    virt
    0.10
    èį¯ä¸ļ
    0.10
     packaging
    0.10
    èį¯åĵģ
    0.09
    åĮħè£ħ
    0.09
    Rx
    0.09
     паÑĢÑĤи
    0.09
    èį¯å¸Ī
    0.09
     Packaging
    0.08
    Activations Density 0.216%

    No Known Activations