© 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. 23551
    Prev
    Next
    INDEX
    Explanations

    addition, extension, correction

    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
    å¹¿æ³Ľåħ³æ³¨
    -0.09
    çļĭ
    -0.09
    iva
    -0.09
    iblings
    -0.09
    çģĮ
    -0.09
    å«Ĥ
    -0.08
    笳
    -0.08
    æĽ´å¥½
    -0.08
    çĶ¨æĪ·æıIJä¾Ľ
    -0.08
    _FWD
    -0.08
    POSITIVE LOGITS
    è¡¥åħħ
    0.19
    æī©å±ķ
    0.13
    æ·»åĬł
    0.13
    è¡¥
    0.12
     addition
    0.12
     extension
    0.12
    (extension
    0.12
    è£ľ
    0.12
     thêm
    0.12
     дополн
    0.11
    Activations Density 0.131%

    No Known Activations