© 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. 50355
    Prev
    Next
    INDEX
    Explanations

    design

    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
    æĬľ
    -0.09
    ustain
    -0.09
    burgh
    -0.09
    !--
    -0.09
    argv
    -0.08
     Supern
    -0.08
    æĹ¥æĬ¥è®°èĢħ
    -0.08
    æĥħåĨµè¿Ľè¡Į
    -0.08
    ony
    -0.08
    vÃŃ
    -0.08
    POSITIVE LOGITS
    éŁŃ
    0.10
    åĩĽ
    0.09
    并åIJij
    0.09
    èĬ¹
    0.08
    æºħ
    0.08
    Fabric
    0.08
    idian
    0.08
    Èĥ
    0.08
    dsl
    0.08
    Coverage
    0.08
    Activations Density 0.027%

    No Known Activations