© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 9862
    Prev
    Next
    INDEX
    Explanations

    commenting on threads

    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
    baugh
    -0.12
     ears
    -0.09
    æĪijçĪ±ä½ł
    -0.09
    å¨ĺ
    -0.09
    åѦåĪ°
    -0.09
    èĢ³æľµ
    -0.09
    çĸĹ
    -0.08
    ä¸įå¾Ĺä¸į
    -0.08
    æĮŀ
    -0.08
    gaard
    -0.08
    POSITIVE LOGITS
    aries
    0.10
    aires
    0.09
    åĬ¨æľº
    0.09
    eting
    0.09
    mate
    0.08
    ariat
    0.08
    stre
    0.08
    alt
    0.08
    mas
    0.08
     Canonical
    0.08
    Activations Density 0.048%

    No Known Activations