© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 50391
    Prev
    Next
    INDEX
    Explanations

    advantage, shortcomings, issues, common

    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
    Means
    -0.08
    agan
    -0.08
    atin
    -0.08
    alus
    -0.08
    æķħ
    -0.08
    yt
    -0.08
    åĽºçĦ¶
    -0.08
    æľ¬è´¨ä¸Ĭ
    -0.07
    æĦıåij³çĿĢ
    -0.07
    ugin
    -0.07
    POSITIVE LOGITS
    ä»Ģä¹Īåij¢
    0.11
    :
    0.11
    é¢ĩ为
    0.10
    åŃ³
    0.09
    å®ŀåľ¨å¤ª
    0.09
    ä¹Łæ¯Ķè¾ĥ
    0.09
    宥
    0.09
    ä¹ŁéĿŀ常
    0.09
    éĤ£å°±æĺ¯
    0.09
    æĿ¥åİĨ
    0.09
    Activations Density 0.642%

    No Known Activations