© 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. 54171
    Prev
    Next
    INDEX
    Explanations

    submarine

    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
    eton
    -0.11
    anden
    -0.09
    à¹Ĥà¸ģ
    -0.09
    andro
    -0.09
    LEVEL
    -0.09
    çĵ¢
    -0.09
     airborne
    -0.09
    amina
    -0.08
    é£ĺ
    -0.08
     Kuwait
    -0.08
    POSITIVE LOGITS
    æįŀ
    0.11
    erule
    0.10
     depths
    0.10
    .Depth
    0.10
    åijĺ
    0.09
    /oct
    0.09
    orna
    0.09
    ä¸ĸ人
    0.09
    æ¾Ī
    0.09
    æµ·åºķ
    0.08
    Activations Density 0.036%

    No Known Activations