© 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. 48498
    Prev
    Next
    INDEX
    Explanations

    strip down or remove

    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
    chie
    -0.09
    åĹŁ
    -0.08
    ä¸į容æĺĵ
    -0.08
    éĴ²
    -0.08
    çĢĽ
    -0.08
    enty
    -0.08
    éĹª
    -0.08
    çĶ¨æīĭ
    -0.08
    leÅŁtirme
    -0.08
    une
    -0.08
    POSITIVE LOGITS
    pled
    0.10
    per
    0.10
    pery
    0.09
    ол
    0.09
    æį¢åı¥è¯Ŀ说
    0.09
     Naked
    0.09
    hana
    0.09
    ãĤĤãģĨ
    0.08
     Surrey
    0.08
    away
    0.08
    Activations Density 0.019%

    No Known Activations