© 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. 58575
    Prev
    Next
    INDEX
    Explanations

    codes followed by numbers or specific suffixes

    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
    åıĸèĥľ
    -0.09
    åĥ»
    -0.08
    tics
    -0.08
    yth
    -0.08
    iny
    -0.08
    avra
    -0.08
    رÙĨ
    -0.07
    URED
    -0.07
     Walker
    -0.07
    wner
    -0.07
    POSITIVE LOGITS
    ä»Ģä¹Īåij¢
    0.13
    ese
    0.09
    igram
    0.08
    /goto
    0.08
    ä¹İ
    0.08
    ä¼ĺèī¯
    0.08
     booster
    0.08
    ç³»åĪĹ产åĵģ
    0.08
    ä¸įå®Į
    0.08
    -google
    0.08
    Activations Density 0.341%

    No Known Activations