© 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. 49126
    Prev
    Next
    INDEX
    Explanations

    oily, very, -like, suffix -y, suffix -ody

    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
    åĴĮåıijå±ķ
    -0.11
    黯
    -0.10
    åIJĪæ³ķ
    -0.09
    è¿ŀå¤ľ
    -0.09
    æĢİä¹Īçľĭ
    -0.09
    äºĨä¸Ģ个
    -0.09
    (concat
    -0.08
    åĬ¾
    -0.08
    .FontStyle
    -0.08
    patches
    -0.08
    POSITIVE LOGITS
    -kind
    0.10
     Winn
    0.09
     seg
    0.09
    羣åĪĩ
    0.09
    rray
    0.08
    chen
    0.08
    ino
    0.08
    INO
    0.08
    为äºĨ让
    0.08
     вÑħод
    0.08
    Activations Density 0.077%

    No Known Activations