© 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. 55788
    Prev
    Next
    INDEX
    Explanations

    hard to produce or access

    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
    OOM
    -0.09
    èĪģ
    -0.09
    æ´ª
    -0.09
    .semantic
    -0.09
     pil
    -0.09
    lund
    -0.08
     Buen
    -0.08
    è¿ĩåī©
    -0.08
    ä¸īæĿ¡
    -0.08
     Esc
    -0.08
    POSITIVE LOGITS
    éļ¾
    0.14
    çĸijéļ¾
    0.13
    å¤įæĿĤçļĦ
    0.10
    éļ¾é¢ĺ
    0.10
     ÑĢанее
    0.10
    éļ¾åħ³
    0.10
    å°¼åħĭ
    0.10
    éļ¾çĤ¹
    0.10
     difficult
    0.09
    æĮijåīĶ
    0.09
    Activations Density 0.242%

    No Known Activations