© 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. 26789
    Prev
    Next
    INDEX
    Explanations

    pandas, react

    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
    ç§įç§į
    -0.09
    çĺķ
    -0.09
    ing
    -0.08
    缴è¾ĸå¸Ĥ
    -0.07
    ides
    -0.07
    æĮģ
    -0.07
    è½²
    -0.07
     Tob
    -0.07
    å¿Į
    -0.07
    ouver
    -0.07
    POSITIVE LOGITS
    è¿Ļä½į
    0.10
    LEY
    0.09
    çľĭå®Į
    0.09
    DED
    0.09
     [âĢ¦
    0.08
     advert
    0.08
    å®Ĺ
    0.08
    åIJ¬å®Į
    0.08
     [âĢ¦]
    0.08
    æľĢåĸľæ¬¢
    0.08
    Activations Density 1.255%

    No Known Activations