© 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. 49631
    Prev
    Next
    INDEX
    Explanations

    respective owners/rightful owners

    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
    å®ŀåľ¨æĺ¯å¤ª
    -0.08
    æĦŁè§¦
    -0.08
    Ñģан
    -0.08
    æ²¾
    -0.08
    ush
    -0.08
    丼
    -0.08
    æ½¢
    -0.08
    å¤Ħå¤Ħ
    -0.08
    æľīåĩłä¸ª
    -0.08
    kop
    -0.08
    POSITIVE LOGITS
    åIJĦèĩªçļĦ
    0.12
     dish
    0.10
    åIJĦèĩª
    0.10
    对åºĶçļĦ
    0.10
    èį·
    0.09
    æŀķ
    0.09
    IgnoreCase
    0.09
    å±
    0.09
    efore
    0.09
    vä
    0.09
    Activations Density 0.006%

    No Known Activations