© 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. 5818
    Prev
    Next
    INDEX
    Explanations

    language words

    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
     Hir
    -0.09
    楼梯
    -0.08
    å¸ĤæĶ¿
    -0.08
    读èĢħ
    -0.08
    åį¿
    -0.08
    大åݦ
    -0.08
    æı´
    -0.08
    çĴ§
    -0.08
    庸
    -0.08
    æŀ¢
    -0.08
    POSITIVE LOGITS
    åħ¨æĿij
    0.18
    æĿijåŃIJ
    0.14
    æĿijéĩĮ
    0.14
    æĿijæ°ij
    0.13
    æĿijèIJ½
    0.13
     village
    0.13
    å½ĵåľ°äºº
    0.13
    æĿijæ°ij们
    0.13
    æĿijåºĦ
    0.13
    éĥ¨èIJ½
    0.13
    Activations Density 0.267%

    No Known Activations