© 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. 34750
    Prev
    Next
    INDEX
    Explanations

    media services and content

    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
    à¸Ħว
    -0.09
    aldo
    -0.09
     воÑģ
    -0.09
    æIJºç¨ĭ
    -0.09
    èĴĻ
    -0.08
    éģIJ
    -0.08
    translated
    -0.08
    Realm
    -0.08
    å¼±
    -0.08
    æ°Ķæģ¯
    -0.08
    POSITIVE LOGITS
    æĦ§
    0.11
    ä¸įéĶĻ
    0.10
    èĩªæĿ¥
    0.09
    åĨĽå·¥
    0.09
    éļıæŶéļı
    0.09
    çĴĩ
    0.09
    çģ¿
    0.09
    Lifetime
    0.09
    emic
    0.08
    åĨĴçĿĢ
    0.08
    Activations Density 0.020%

    No Known Activations