© 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. 19956
    Prev
    Next
    INDEX
    Explanations

    thunder, lightning, scrolling, convective, envelope, scroll

    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
    çģµæ°Ķ
    -0.11
    åĴĮæľįåĬ¡
    -0.10
    edium
    -0.09
    èĤ¡
    -0.09
    RAR
    -0.09
    ignum
    -0.09
    ä¸įä¿¡
    -0.09
    pie
    -0.09
    çĻ¾ç§ij
    -0.09
    uding
    -0.09
    POSITIVE LOGITS
    quan
    0.09
    /thumb
    0.08
    æĻĹ
    0.08
    æī¼
    0.08
    _passed
    0.08
    aison
    0.08
    天æīį
    0.08
    æªĢ
    0.08
    夬
    0.08
    GetString
    0.08
    Activations Density 0.033%

    No Known Activations