© 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. 27763
    Prev
    Next
    INDEX
    Explanations

    worked out

    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
    CHAIN
    -0.10
    éĹ®æĪij
    -0.09
    åĿļæĮģ以
    -0.09
    åľ¨è¿ĩ渡
    -0.09
    åı®
    -0.09
     Sle
    -0.09
    haar
    -0.09
    zenie
    -0.09
    çĥŁçģ«
    -0.09
    trail
    -0.08
    POSITIVE LOGITS
    大äºĭ
    0.11
    shire
    0.10
    wards
    0.10
    æĿł
    0.09
    ãĤ£
    0.09
    positor
    0.08
     worse
    0.08
    éĢĥ
    0.08
    æĪIJ为ä¸ŃåĽ½
    0.08
     дела
    0.08
    Activations Density 0.008%

    No Known Activations