© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 6245
    Prev
    Next
    INDEX
    Explanations

    most or highest of

    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
    Negative Logits
    çļĦæľĢä½³
    -0.11
    aight
    -0.08
    亲çĪ±
    -0.08
    æľĢä½³
    -0.08
    æĻ°
    -0.08
    akit
    -0.08
    å°±æĦıåij³çĿĢ
    -0.08
    ç¥Ĺ
    -0.08
    ughty
    -0.08
    borough
    -0.08
    POSITIVE LOGITS
    /all
    0.09
    å¼±çĤ¹
    0.08
    é¢Ĩ
    0.08
    ä¸įæĺĵ
    0.08
     amp
    0.08
    çĶ¨å¿ĥ
    0.08
    äºī论
    0.08
     brag
    0.08
    erville
    0.08
    amp
    0.08
    Activations Density 0.011%

    No Known Activations