© 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. 56310
    Prev
    Next
    INDEX
    Explanations

    numbers indicating percentage or points

    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
    é¡¿æŶ
    -0.11
    åı¦ä¸ĢåįĬ
    -0.10
    IOUS
    -0.10
    ä¹Łåıªèĥ½
    -0.10
    èī®
    -0.09
    çģĮ
    -0.09
    åĭ¿
    -0.09
    åħ¨æĺ¯
    -0.08
    éĩįä¸Ńä¹ĭéĩį
    -0.08
    ä¸ĢåłĨ
    -0.08
    POSITIVE LOGITS
    å¾®å¦Ļ
    0.10
    irectory
    0.09
    ç¾İ
    0.08
    å°ij许
    0.08
     Seats
    0.08
     fingert
    0.08
    ç½ij游
    0.08
    itto
    0.08
    åı£åı·
    0.08
    è¿ĽåĮĸ
    0.08
    Activations Density 0.055%

    No Known Activations