© 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. 4579
    Prev
    Next
    INDEX
    Explanations

    characterizing techniques

    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
    èł¢
    -0.09
    ä¸ĢèĤ¡
    -0.09
    monster
    -0.08
    ç¿ĺ
    -0.08
    ä¸Ń央
    -0.08
    зав
    -0.08
    æݨéĶĢ
    -0.08
     Calibration
    -0.08
    -scrollbar
    -0.08
    åı¯ä¸įæĺ¯
    -0.08
    POSITIVE LOGITS
    æīĭ段
    0.16
     techniques
    0.13
    åħĪè¿ĽæĬĢæľ¯
    0.11
    ä¸ĵä¸ļæĬĢæľ¯
    0.10
     technique
    0.10
    éºŁ
    0.10
    æĬĢæľ¯åĴĮ
    0.09
    åħĪè¿Ľ
    0.09
    çľ¼çľĭ
    0.09
    èĨ³
    0.09
    Activations Density 1.263%

    No Known Activations