© 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. 58628
    Prev
    Next
    INDEX
    Explanations

    PUR, cul, waves

    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
    æľĶ
    -0.10
    åĵªä¸Ģ个
    -0.09
    没éĶĻ
    -0.09
    inox
    -0.09
    讯
    -0.09
    astr
    -0.09
    以èī²
    -0.09
    eeper
    -0.09
    à®´
    -0.09
    rone
    -0.08
    POSITIVE LOGITS
    PUR
    0.08
    -kind
    0.08
    oul
    0.08
    æ´»è·ĥ
    0.08
    æ²»æĦĪ
    0.08
    çģ¼
    0.08
    bek
    0.08
    pur
    0.08
    cul
    0.08
    ides
    0.08
    Activations Density 0.118%

    No Known Activations