© 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. 9609
    Prev
    Next
    INDEX
    Explanations

    per unit/quantity

    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
    çĬ¶
    -0.08
    èĤĥ
    -0.08
    è¾Ł
    -0.08
    泫
    -0.08
    à¹īà¸ĩ
    -0.08
    pedia
    -0.08
    åľ¨ä¸Ģ
    -0.08
    ä¼ij
    -0.08
    ä¸ĢæĹ¥
    -0.07
    éĺ²èħIJ
    -0.07
    POSITIVE LOGITS
    履约
    0.10
    è¿ĩåħ³
    0.09
    .rm
    0.09
    OID
    0.09
    ä¼ī
    0.09
    vara
    0.09
    mdir
    0.09
    ãģ©ãģĵãĤį
    0.09
     thắng
    0.09
    åIJ»
    0.08
    Activations Density 0.018%

    No Known Activations