© 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. 49733
    Prev
    Next
    INDEX
    Explanations

    checked out, checkout from

    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
    çĺ´
    -0.09
    åĤį
    -0.09
     Buf
    -0.09
    æĿĤ
    -0.08
    溺
    -0.08
    éĹªåħī
    -0.08
    alia
    -0.08
    å¯Ĥ
    -0.08
    çĪ¬
    -0.08
    ç²ī
    -0.08
    POSITIVE LOGITS
    .borrow
    0.10
    åĢŁãĤĬ
    0.10
     дом
    0.10
     overdue
    0.10
    Locker
    0.09
    çĶ³
    0.09
    æľŁéĻIJ
    0.09
    æ³®
    0.09
    éĴİ
    0.09
    åĽ¾ä¹¦
    0.09
    Activations Density 0.057%

    No Known Activations