© 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. 4025
    Prev
    Next
    INDEX
    Explanations

    he/she said

    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
    好åIJ§
    -0.10
    ç¡®å®ŀæĺ¯
    -0.10
    çĥŃæIJľ
    -0.10
    好åIJĹ
    -0.09
     Lal
    -0.09
    å¾Īä¸įéĶĻ
    -0.09
    å®´
    -0.09
    羣çļĦ
    -0.09
    æĺ¯å¤ļä¹Ī
    -0.09
    è¾±
    -0.08
    POSITIVE LOGITS
    认为
    0.13
    èªįçĤº
    0.11
    æĮĩåĩº
    0.11
    è§īå¾Ĺ
    0.11
    覺å¾Ĺ
    0.10
    åħ®
    0.10
    unks
    0.10
    ric
    0.09
    æĮĩçļĦæĺ¯
    0.09
    建议
    0.09
    Activations Density 0.039%

    No Known Activations