© 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. 30794
    Prev
    Next
    INDEX
    Explanations

    first and then

    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
    æľ¬æ¬¡
    -0.09
     timeline
    -0.09
    thood
    -0.09
    æŃ¤æ¬¡
    -0.09
    timeline
    -0.09
    çĪ¶
    -0.08
    è¿Ļ次
    -0.08
    èĩĨ
    -0.08
    plan
    -0.08
    æݧèĤ¡
    -0.08
    POSITIVE LOGITS
    åľ¨è¿Ļæĸ¹éĿ¢
    0.11
    å©´
    0.10
    æį¢äºĨ
    0.09
    å¹³åĿĩæ¯ı
    0.09
    æİĴ
    0.09
    åľºåĿĩ
    0.08
    ä¸Ĭæ¦ľ
    0.08
     restraint
    0.08
    å¹³åĿĩæ°´å¹³
    0.08
    èĩªå¾ĭ
    0.08
    Activations Density 0.025%

    No Known Activations