© 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. 5250
    Prev
    Next
    INDEX
    Explanations

    decided you've heard

    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
    ìłĪ
    -0.11
    åĵī
    -0.10
    帮æĪij
    -0.09
    æĪijåĸľæ¬¢
    -0.09
     forcibly
    -0.09
    æĪijèĩªå·±
    -0.08
    ç»ĵåIJĪ
    -0.08
    opers
    -0.08
    殪
    -0.08
    èģ©
    -0.08
    POSITIVE LOGITS
    åĽłä¸ºä½ł
    0.13
    è¿Ľåħ¥äºĨ
    0.11
    æģ¶
    0.11
    æĪĸ许
    0.11
    æĢİä¹ĪåĬŀ
    0.11
    æĪĸ许æĺ¯
    0.11
    chal
    0.11
     maybe
    0.10
     perhaps
    0.10
    ãģĭãĤĤãģĹãĤĮãģ¾ãģĽãĤĵ
    0.10
    Activations Density 0.147%

    No Known Activations