© 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. 10288
    Prev
    Next
    INDEX
    Explanations

    push something forward

    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
     sett
    -0.09
    çªĴ
    -0.08
    躺
    -0.08
    çķĻä¸ĭ
    -0.08
    åĿĩ
    -0.08
    彪
    -0.08
    ç¡ķ
    -0.08
    èįĴ
    -0.08
    olis
    -0.08
    æĢģ度
    -0.08
    POSITIVE LOGITS
    (push
    0.12
    æݨåIJij
    0.10
     pushes
    0.09
     Äijẩy
    0.09
     BUTTON
    0.09
    éĩįå¿ĥ
    0.09
    loys
    0.09
    æĮīéĴ®
    0.09
    åļı
    0.09
    æIJ¡
    0.09
    Activations Density 0.050%

    No Known Activations