© 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. 6342
    Prev
    Next
    INDEX
    Explanations

    viscolin, pangolin, iPod, Dropbox

    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
    åĨī
    -0.10
    çĮ´åŃIJ
    -0.09
    èµ°å¾Ĺ
    -0.09
    åĺīå¹´
    -0.09
    å¸ļ
    -0.09
    коÑģ
    -0.08
    åIJ¦è®¤
    -0.08
    istant
    -0.08
     khÄĥn
    -0.08
    转æĬĺ
    -0.08
    POSITIVE LOGITS
    ä»°
    0.11
    atron
    0.10
    æĺĤ
    0.09
    éĽĦ
    0.09
    çĽij管éĥ¨éŨ
    0.09
    oley
    0.09
    çŃĶ
    0.09
    AFP
    0.09
    ç»Ŀ
    0.08
    olo
    0.08
    Activations Density 0.047%

    No Known Activations