© 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. 27697
    Prev
    Next
    INDEX
    Explanations

    Then I, maybe someone, feel obligated

    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
     lob
    -0.09
    atetime
    -0.09
    å¼Ģåħĥ
    -0.08
     HAL
    -0.08
    ç»Ī
    -0.08
    ç½Ħ
    -0.08
     famously
    -0.08
    IFY
    -0.08
     vain
    -0.08
    ify
    -0.08
    POSITIVE LOGITS
    尤为éĩįè¦ģ
    0.10
    çļĦç¡®
    0.10
    åı«åģļ
    0.09
    é¢ĩ为
    0.09
    åį´æĺ¯
    0.09
     {}:
    0.08
    æĪIJè´¥
    0.08
    管çIJĨå·¥ä½ľ
    0.08
    åIJįåı«
    0.08
    å¾Ħ
    0.08
    Activations Density 0.378%

    No Known Activations