© 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. 50811
    Prev
    Next
    INDEX
    Explanations

    computational and painstaking descriptions

    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
    teenth
    -0.10
    wyn
    -0.09
    Ĵ
    -0.09
    裳
    -0.08
    ãģŁãģ¡
    -0.08
    ì¦Ī
    -0.08
    ever
    -0.08
    един
    -0.08
    thing
    -0.08
    çĹĽçĤ¹
    -0.08
    POSITIVE LOGITS
    burgh
    0.11
    à¹Į
    0.09
    arily
    0.09
    éĻ¤æѤ
    0.09
    ively
    0.09
    ãģĦãĤĭ
    0.09
    ary
    0.09
    漫éķ¿
    0.09
    imeline
    0.09
     trùng
    0.09
    Activations Density 0.012%

    No Known Activations