© 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. 48590
    Prev
    Next
    INDEX
    Explanations

    preposition + noun/time/verb

    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
     zwar
    -0.17
    ãģĦãģ¾ãģĻãģĮ
    -0.12
    ãģ§ãģĻãģĹ
    -0.12
    ìĿ´ë©°
    -0.11
    ãģ§ãģĹãģŁãģĮ
    -0.11
     ìŀĪìľ¼ë©°
    -0.11
    èĻ½çĦ¶
    -0.10
     ìŀĪëĬĶëį°
    -0.10
    ä½Ĩæĺ¯åľ¨
    -0.10
    ãģĹãģ¦ãģĬãĤĬ
    -0.10
    POSITIVE LOGITS
    å®ŀåľ¨æĺ¯å¤ª
    0.09
    æķ¬è¯·
    0.08
    çĿĢå®ŀ
    0.08
    ).
    0.08
    åıĪä¸Ģ次
    0.08
    为æŃ¢
    0.08
    åı¯è°ĵæĺ¯
    0.08
     Rao
    0.08
    ">
    0.08
    åı¯è°ĵ
    0.08
    Activations Density 3.124%

    No Known Activations