© 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. 21358
    Prev
    Next
    INDEX
    Explanations

    acceleration

    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
    .annot
    -0.09
    é¼ł
    -0.09
    èĩªåıij
    -0.09
    books
    -0.08
     Kaplan
    -0.08
    coded
    -0.08
    è¾¹çķĮ
    -0.08
    edly
    -0.08
    boundary
    -0.08
    æIJºå¸¦
    -0.08
    POSITIVE LOGITS
    稳妥
    0.11
    ÑĤан
    0.09
    èģĶç½ij
    0.09
    ä¸įèµ·
    0.09
    çªĥ
    0.09
    .mj
    0.09
    Registrar
    0.08
    ÑĢоÑģÑĤ
    0.08
     Jer
    0.08
    æ®ĭçĸ¾
    0.08
    Activations Density 0.051%

    No Known Activations