© 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. 62797
    Prev
    Next
    INDEX
    Explanations

    stone inscriptions

    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
    èĵģ
    -0.09
    reator
    -0.09
    otron
    -0.09
    bjerg
    -0.09
     ì¤ijìĭ¬
    -0.09
    _NM
    -0.09
    åĽ¢
    -0.09
    ç»ı纬
    -0.08
    å±ĢåĬ¿
    -0.08
    æĶ¾æīĭ
    -0.08
    POSITIVE LOGITS
    éĵŃ
    0.13
    ç¢ij
    0.13
    æłĩè¯Ń
    0.12
     inscription
    0.11
    åĪ»
    0.11
    åĮĸçŁ³
    0.11
    大éĿ¢ç§¯
    0.09
    æĸĩåŃĹ
    0.09
    è®°å½ķ
    0.09
    åħ¬ç¤º
    0.09
    Activations Density 0.042%

    No Known Activations