© 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. 63058
    Prev
    Next
    INDEX
    Explanations

    Author, itt, first, every, Half, Linear, have, read

    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
     “
    -0.11
    å²Į
    -0.10
    ãģĮãĤī
    -0.10
    å·´æĸ¯
    -0.10
    éĩįè¦ģåĨħ容
    -0.10
    åįħ
    -0.09
    æİī
    -0.09
    åĽĽä¸ªæĸ¹éĿ¢
    -0.09
    TRL
    -0.09
    “A
    -0.09
    POSITIVE LOGITS
    åįģåĽĽäºĶ
    0.10
    çĸ¯
    0.08
    chia
    0.08
    带走
    0.08
     Verified
    0.08
    9
    0.07
    isser
    0.07
    7
    0.07
    éŁ
    0.07
    æݬ
    0.07
    Activations Density 0.269%

    No Known Activations