© 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. 19473
    Prev
    Next
    INDEX
    Explanations

    the + organization name

    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
    legate
    -0.10
     ÑĢоли
    -0.09
    éĽ
    -0.09
    auge
    -0.08
    ZH
    -0.08
    èĽī
    -0.08
    è¦ĭ
    -0.08
    sted
    -0.08
     Advisor
    -0.08
    ylum
    -0.08
    POSITIVE LOGITS
     gr
    0.09
    å¾Ĺ太
    0.09
    å°»
    0.08
    åħ¨å¤©
    0.08
    fully
    0.08
    éĶ²
    0.08
    æ´ĩ
    0.08
    æį¢æĪIJ
    0.08
    æİĸ
    0.08
    erring
    0.08
    Activations Density 0.136%

    No Known Activations