© 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. 35156
    Prev
    Next
    INDEX
    Explanations

    of input, Beer, handicap, resolution, skip, holy, data

    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
    æ¸Ĺ
    -0.10
    é¢IJ
    -0.10
    emin
    -0.10
    çª
    -0.10
    çīĪæĿĥæīĢæľī
    -0.10
    èķĪ
    -0.10
    åĢº
    -0.09
    声éģĵ
    -0.09
    æĽĻ
    -0.09
    大海
    -0.09
    POSITIVE LOGITS
    éģĵçIJĨ
    0.09
    æĪij们ä¸Ģèµ·
    0.08
    饱
    0.08
    人家
    0.08
    鼶ç¢İ
    0.08
    acey
    0.08
    å·¨
    0.08
    xious
    0.08
    åĬĪ
    0.08
    åģļæ³ķ
    0.07
    Activations Density 0.211%

    No Known Activations