© 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. 55425
    Prev
    Next
    INDEX
    Explanations

    free subscription

    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
    贷款
    -0.09
    çĤ¬
    -0.09
    ä¹ĭæĺŁ
    -0.09
    éĨĭ
    -0.08
    inality
    -0.08
    draft
    -0.08
    RAFT
    -0.08
    åĢŁçĶ¨
    -0.08
    åİ»åĵªåĦ¿
    -0.08
     Loan
    -0.08
    POSITIVE LOGITS
    订éĺħ
    0.11
    çĽĪ
    0.10
    adle
    0.09
     sust
    0.09
    кÑĥÑģ
    0.09
    åºĸ
    0.09
    éĻĢ
    0.09
    _subscription
    0.09
    istani
    0.08
     subscriber
    0.08
    Activations Density 0.040%

    No Known Activations