© 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. 39647
    Prev
    Next
    INDEX
    Explanations

    ask me or us

    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
    .instant
    -0.09
    _ulong
    -0.09
     Nguy
    -0.09
    _RF
    -0.09
    igrams
    -0.09
    çĿ¨
    -0.09
    æĥħ人
    -0.08
    _cum
    -0.08
    alles
    -0.08
    iage
    -0.08
    POSITIVE LOGITS
    éĹ®æĪij
    0.12
     khảo
    0.10
    è¿Ļ群
    0.10
    æĪij对
    0.09
     me
    0.09
    éĹ®
    0.09
    çļĦ人éĥ½
    0.09
    æĪij说
    0.09
    åĽ´è§Ĥ
    0.09
    æĪļ
    0.09
    Activations Density 0.021%

    No Known Activations