© 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. 4811
    Prev
    Next
    INDEX
    Explanations

    local, regional, global

    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
     hsv
    -0.09
     iceberg
    -0.08
     ster
    -0.08
    æĤ¬
    -0.08
     Straw
    -0.08
    ECT
    -0.08
    ä¸ĢéŨ
    -0.08
     Shim
    -0.08
    	mv
    -0.08
    avr
    -0.08
    POSITIVE LOGITS
    -local
    0.14
    åĮºåŁŁæĢ§
    0.14
    /global
    0.14
    /local
    0.13
    åľ°åĮºçļĦ
    0.13
     local
    0.12
     regional
    0.12
    local
    0.12
    -global
    0.12
    æľ¬åľ°
    0.12
    Activations Density 0.086%

    No Known Activations