© 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. 11230
    Prev
    Next
    INDEX
    Explanations

    ball grid array, socket joint

    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
    éĻ¤æѤ
    -0.10
    çľ¼è¢ĭ
    -0.09
     challenger
    -0.09
    ç²
    -0.09
    æį¢åı¥è¯Ŀ说
    -0.09
    caps
    -0.09
    sic
    -0.08
    -ending
    -0.08
    解æĶ¾
    -0.08
    åħ¶ä»ĸåĽ½å®¶
    -0.08
    POSITIVE LOGITS
     compromise
    0.09
     Tess
    0.08
    yte
    0.08
    ç¥Ń
    0.08
     convers
    0.08
    çĤĬ
    0.08
     Kov
    0.08
    oku
    0.08
    姻
    0.08
    {"
    0.08
    Activations Density 0.002%

    No Known Activations