© 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. 7559
    Prev
    Next
    INDEX
    Explanations

    no longer just a matter

    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
    åĿ¦
    -0.10
    ä¸į论
    -0.10
    ä¸į管æĺ¯
    -0.09
    æĹłè®º
    -0.09
    åIJ¦
    -0.09
    æĹłè®ºæĺ¯
    -0.09
    ä¸į管
    -0.08
    主è¦ģæľī
    -0.08
    çĶŃ
    -0.08
     mostly
    -0.08
    POSITIVE LOGITS
    èĢĮæĺ¯
    0.15
    éĤ£ä¹Īç®Ģåįķ
    0.11
     sondern
    0.11
    ä¹Łä¸įæĺ¯
    0.10
    è¿Ļå¥Ĺ
    0.10
     nor
    0.10
    ç¬Ķ
    0.09
    å¾ħ
    0.09
    è¿Ļåıª
    0.09
    æ²»çIJĨä½ĵç³»
    0.09
    Activations Density 0.023%

    No Known Activations