© 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. 49811
    Prev
    Next
    INDEX
    Explanations

    roman numeral suffixes

    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¶å±ŀ
    -0.09
    oul
    -0.09
    åĽ¢
    -0.08
    اÙĩ
    -0.08
     krist
    -0.08
    å¸Ń
    -0.08
    é¦Ĩ
    -0.08
    å¤įæ´»
    -0.08
     hon
    -0.08
     ner
    -0.08
    POSITIVE LOGITS
    ä¸Ģä¸ĸ
    0.13
    æĹ¶æľŁçļĦ
    0.10
    æĹ¶æľŁ
    0.10
    继
    0.10
    ä¸ĵé¢ĺ
    0.10
    azzi
    0.09
    æĻĤæľŁ
    0.09
    å¹´éĹ´
    0.09
    æĻļå¹´
    0.09
    ãģķãģ¾
    0.09
    Activations Density 0.085%

    No Known Activations