© 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. 1813
    Prev
    Next
    INDEX
    Explanations

    when followed by pronoun or article

    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
    说å®Į
    -0.12
    没æľīåĬŀæ³ķ
    -0.10
    èį·
    -0.10
     zwar
    -0.09
     Ñĥме
    -0.09
    íķ´ì£¼
    -0.09
    è¦ģ好好
    -0.09
    ä»ĸèĩªå·±
    -0.09
    帮ä»ĸ
    -0.09
    èĪįä¸įå¾Ĺ
    -0.08
    POSITIVE LOGITS
    çªģçĦ¶
    0.14
    çªģ
    0.11
    åıijçĶŁäºĨ
    0.10
    åıijçİ°äºĨ
    0.10
    éģĩåĪ°äºĨ
    0.10
    åıijçİ°
    0.10
    çªģåıij
    0.10
    忽çĦ¶
    0.10
    è¿İæĿ¥äºĨ
    0.09
    åıĹåĪ°äºĨ
    0.09
    Activations Density 0.080%

    No Known Activations