© 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. 54201
    Prev
    Next
    INDEX
    Explanations

    service providers and vendors

    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
    åħļæĶ¿
    -0.10
    æķķ
    -0.10
    antine
    -0.10
     Rin
    -0.09
     Jennings
    -0.09
    å¹²éĥ¨éĺŁä¼į
    -0.09
    çĶŁæĦı
    -0.09
    亲æīĭ
    -0.09
    家éķ¿ä»¬
    -0.08
    oland
    -0.08
    POSITIVE LOGITS
    第ä¸īæĸ¹
    0.13
    æľįåĬ¡åķĨ
    0.12
     provider
    0.11
    æıIJä¾ĽåķĨ
    0.11
    åħ¬åı¸å°Ĩ
    0.10
    第ä¸īèĢħ
    0.10
    -caret
    0.09
    .operator
    0.09
    ä¾ĽåºĶåķĨ
    0.09
    æľįåĬ¡æľºæŀĦ
    0.09
    Activations Density 0.077%

    No Known Activations