© 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. 38697
    Prev
    Next
    INDEX
    Explanations

    douglas fir, wood, gregor, fisher

    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
    åĦ¿
    -0.09
    说æĺ¯
    -0.08
    å·´å·´
    -0.08
    çIJĨ论ä¸Ĭ
    -0.08
    ار
    -0.08
    el
    -0.08
    .office
    -0.08
    exus
    -0.08
    sis
    -0.08
    sole
    -0.08
    POSITIVE LOGITS
    mente
    0.10
    athy
    0.10
    baugh
    0.09
    azzi
    0.09
    rim
    0.09
    loon
    0.09
    thur
    0.09
    åĽ½éĻħæľºåľº
    0.09
    lah
    0.09
    anoia
    0.09
    Activations Density 0.006%

    No Known Activations