© 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. 10985
    Prev
    Next
    INDEX
    Explanations

    wide array of

    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
    ä¸Ģç¬ij
    -0.10
    ListOf
    -0.09
    ä¸Ģ大
    -0.09
    utor
    -0.09
    ä¸Ģå¤Ħ
    -0.09
    æ¦ī
    -0.09
    اء
    -0.08
    éĺŁä¼į建设
    -0.08
    ä¸įå°ij人
    -0.08
    æŁIJäºĽ
    -0.08
    POSITIVE LOGITS
    éĢĤåIJĪ
    0.11
    é¢ľèī²
    0.10
    iveness
    0.09
    Hat
    0.09
     ÑĤÑĥ
    0.09
    ä¸įåIJĮ
    0.09
     acc
    0.09
    çĶ¨éĢĶ
    0.08
    积èĵĦ
    0.08
    0.08
    Activations Density 0.046%

    No Known Activations