© 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. 49078
    Prev
    Next
    INDEX
    Explanations

    way to describe

    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
    æĭį
    -0.09
    å¥Ĺè·¯
    -0.09
    assy
    -0.09
    é»ijå¤ľ
    -0.09
     воп
    -0.08
    slice
    -0.08
     Baghd
    -0.08
     Fate
    -0.08
     Huff
    -0.08
     СовеÑĤ
    -0.08
    POSITIVE LOGITS
    è¿Ļ个è¯į
    0.18
    称åij¼
    0.14
    æĿ¥å½¢å®¹
    0.14
     term
    0.13
    形容
    0.12
    è¿Ļ个åIJįåŃĹ
    0.12
    çļĦ说æ³ķ
    0.12
    ä¹ĭ称
    0.12
    称ä¹ĭ
    0.11
     word
    0.11
    Activations Density 0.095%

    No Known Activations