© 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. 62794
    Prev
    Next
    INDEX
    Explanations

    in-situ or in-

    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
    æīĢåľ¨
    -0.10
     Ñģобой
    -0.09
    åĪĨè¡Į
    -0.09
    PACE
    -0.09
    fab
    -0.09
    stüt
    -0.09
    etical
    -0.09
    asco
    -0.08
    çŀª
    -0.08
    aille
    -0.08
    POSITIVE LOGITS
    /out
    0.17
    ividual
    0.11
    -Out
    0.10
    /on
    0.10
    æ¥Ķ
    0.09
    æݧèĤ¡éĽĨåĽ¢
    0.09
    -out
    0.09
    -Un
    0.08
    æĢ§æł¼
    0.08
    akte
    0.08
    Activations Density 0.145%

    No Known Activations