© 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. 55736
    Prev
    Next
    INDEX
    Explanations

    other way

    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
    é£Ł
    -0.09
    åIJĦå¼ı
    -0.08
    箬
    -0.08
     Harden
    -0.08
     disappoint
    -0.08
    hoe
    -0.08
    æĮł
    -0.08
     أد
    -0.08
    lv
    -0.08
    çĴĢ
    -0.08
    POSITIVE LOGITS
    åı¦æľī
    0.11
    ä¸ĵåijĺ
    0.10
    deb
    0.10
    åı¦
    0.10
    å¼ĢèįĴ
    0.09
    åħ³äºİæĪij们
    0.09
    æĺİæĺ¾çļĦ
    0.08
    åijĬè¯ī大家
    0.08
    媾
    0.08
    ammo
    0.08
    Activations Density 0.050%

    No Known Activations