© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 53111
    Prev
    Next
    INDEX
    Explanations

    discussion events

    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
    iston
    -0.09
    ifes
    -0.08
    mux
    -0.08
    issy
    -0.07
    rah
    -0.07
    ÄĻ
    -0.07
     Toy
    -0.07
     долго
    -0.07
    èĬŃ
    -0.07
     spar
    -0.07
    POSITIVE LOGITS
    енно
    0.11
     ÑģоÑģед
    0.11
    zilla
    0.10
    æī¾åĩº
    0.10
    å¾ģ
    0.10
    eted
    0.09
    _codegen
    0.09
    RequestMethod
    0.09
    èĪŀ
    0.09
     góp
    0.09
    Activations Density 0.003%

    No Known Activations