© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 22378
    Prev
    Next
    INDEX
    Explanations

    served as

    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
    è°ģèĥ½
    -0.10
    OST
    -0.09
    çİĭçīĮ
    -0.09
    çĮĿ
    -0.09
     пÑĢименениÑı
    -0.09
    廪
    -0.09
    è§Ħæł¼
    -0.09
    æĮģãģ¡
    -0.09
    è¿ĻäºĽéĹ®é¢ĺ
    -0.08
    åīįè¿°
    -0.08
    POSITIVE LOGITS
    Desk
    0.10
    ç§ij室
    0.10
    .appspot
    0.09
    ä¸įæĩĪ
    0.08
    åĢľ
    0.08
    åĽ½ç«ĭ
    0.08
    @student
    0.08
     каÑĩеÑģÑĤве
    0.08
    为
    0.08
    erves
    0.08
    Activations Density 0.030%

    No Known Activations