© 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. 35702
    Prev
    Next
    INDEX
    Explanations

    aur/ur followed by specific endings

    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
    opa
    -0.10
    们çļĦ
    -0.10
    Ùħا
    -0.10
    idos
    -0.09
    usk
    -0.09
    æĬĬæīĭ
    -0.09
    ion
    -0.09
    upt
    -0.09
    odos
    -0.09
    .cum
    -0.09
    POSITIVE LOGITS
    /or
    0.13
    gether
    0.10
    ian
    0.09
    /OR
    0.09
    ÑĤоп
    0.09
    大æĸ¹
    0.08
    ầy
    0.08
     Arabian
    0.08
    è°ģèĥ½
    0.08
    comma
    0.08
    Activations Density 0.064%

    No Known Activations