© 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. 61007
    Prev
    Next
    INDEX
    Explanations

    it is possible that

    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
    èİ«æĸ¯
    -0.09
    amat
    -0.09
    å·®ä¸įå¤ļ
    -0.09
    天河
    -0.09
    èĶº
    -0.08
    ROTO
    -0.08
    éļĹ
    -0.08
    èŃī
    -0.08
    害
    -0.08
    ä»Ķ
    -0.08
    POSITIVE LOGITS
     possibility
    0.17
    åı¯èĥ½æĢ§
    0.15
    ä¸įæİĴéĻ¤
    0.15
    çļĦåı¯èĥ½æĢ§
    0.13
     possibilité
    0.12
     possibilities
    0.12
    æľīåı¯èĥ½
    0.12
     possibilitÃł
    0.10
    åı¯èĥ½æĺ¯
    0.10
    .bs
    0.10
    Activations Density 0.025%

    No Known Activations