© 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. 53379
    Prev
    Next
    INDEX
    Explanations

    introduces following code or examples

    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
     trợ
    -0.10
    åIJ»
    -0.09
    ä»ĸå¦Ī
    -0.09
    çªģ
    -0.08
    éĴī
    -0.08
    æľ¬èº«
    -0.08
    otic
    -0.08
    æľ¬èº«çļĦ
    -0.08
    æľĢåħĪ
    -0.08
    ç¼Ŀ
    -0.08
    POSITIVE LOGITS
    ç´łæĿIJ
    0.10
    è¿Ļ段
    0.09
    è¿Ļä¸įæĺ¯
    0.09
    éıĸ
    0.09
    æ³½
    0.08
    æĺ¯æĪijçļĦ
    0.08
    娣
    0.08
    ä¸ĭåİ»
    0.08
    æħİéĩį
    0.08
     serious
    0.08
    Activations Density 0.295%

    No Known Activations