© 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. 58432
    Prev
    Next
    INDEX
    Explanations

    making a note, noticing

    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
    ures
    -0.11
    æĪĺèĥľ
    -0.09
     Madness
    -0.09
     Lounge
    -0.09
    ãģĶ
    -0.09
     Vict
    -0.08
    loth
    -0.08
    åĵģæł¼
    -0.08
    æĪĸå¤ļ
    -0.08
    åŃij
    -0.08
    POSITIVE LOGITS
    éĩįè§Ĩ
    0.11
    åħ³æ³¨
    0.10
    /respond
    0.10
    çļĦåħ³æ³¨
    0.10
    çķĻæĦı
    0.10
    å¿Ļ
    0.09
    amma
    0.09
    注æĦıåĪ°
    0.09
    æĹłè§Ĩ
    0.09
     ê´Ģìĭ¬
    0.09
    Activations Density 0.054%

    No Known Activations