© 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. 57174
    Prev
    Next
    INDEX
    Explanations

    at the heart of

    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
    apot
    -0.09
    edom
    -0.09
    æĭį
    -0.09
    好è¿IJ
    -0.09
    unicorn
    -0.09
    å½ķ
    -0.09
    iphy
    -0.08
    erca
    -0.08
    OrFail
    -0.08
    åIJ¯èĴĻ
    -0.08
    POSITIVE LOGITS
    æł¸å¿ĥ
    0.16
     core
    0.13
    çļĦæł¸å¿ĥ
    0.12
    ä¸ºæł¸å¿ĥ
    0.11
    (core
    0.11
    ä¸ºæł¸å¿ĥçļĦ
    0.11
     centerpiece
    0.10
    core
    0.10
    æĻĥ
    0.10
     CORE
    0.10
    Activations Density 0.046%

    No Known Activations