© 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. 27783
    Prev
    Next
    INDEX
    Explanations

    reservation or reserve

    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
    -0.14
     mẽ
    -0.11
    ening
    -0.10
    аÑĤÑĥÑĢ
    -0.10
    aces
    -0.09
    åŃ¦å®¶
    -0.09
    ronic
    -0.09
    ushman
    -0.09
    orrh
    -0.09
    çģ¼
    -0.09
    POSITIVE LOGITS
    ĭ
    0.10
    edly
    0.10
    ataires
    0.09
    /release
    0.09
    prung
    0.09
    idue
    0.09
    zm
    0.09
    ending
    0.09
    aphael
    0.09
    bie
    0.09
    Activations Density 0.025%

    No Known Activations