© 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. 56826
    Prev
    Next
    INDEX
    Explanations

    hyphens or dashes used for lists

    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
    mino
    -0.09
    prit
    -0.09
    grave
    -0.09
    äºĨä¸Ģåľº
    -0.09
    èĪª
    -0.08
    èĶĵ
    -0.08
    æĸ°åįİ社
    -0.08
    æĬ¥ç¤¾
    -0.08
    åĵº
    -0.08
    etsy
    -0.08
    POSITIVE LOGITS
    òng
    0.10
     either
    0.09
    åħ¼
    0.08
    Ëĥ
    0.08
    å¼ķèµ·çļĦ
    0.08
    æļ¨
    0.08
    asher
    0.08
    ral
    0.08
    rtype
    0.08
    inch
    0.08
    Activations Density 0.129%

    No Known Activations