© 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. 38879
    Prev
    Next
    INDEX
    Explanations

    soft things

    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
    å¯ĤéĿĻ
    -0.10
    åŃIJå¼Ł
    -0.09
    çħ½
    -0.09
    åħ´æĹº
    -0.08
    çİĭåħĪçĶŁ
    -0.08
     part
    -0.08
    ç³Ĭ
    -0.08
    饯
    -0.08
    åĩĿèģļ
    -0.08
    ç²¾ç¥ŀæĸĩæĺİ
    -0.08
    POSITIVE LOGITS
    niest
    0.10
    sponsor
    0.09
    ·»
    0.09
    ä¹Ĥ
    0.09
    ail
    0.09
    spoken
    0.09
    ening
    0.09
     Rapids
    0.09
    ensed
    0.09
    ael
    0.08
    Activations Density 0.006%

    No Known Activations