© 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. 37558
    Prev
    Next
    INDEX
    Explanations

    passage or access

    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
    飬
    -0.09
    åīĮ
    -0.08
    inned
    -0.08
    plib
    -0.08
    éĨĴäºĨ
    -0.08
    åľ¨ç¾İåĽ½
    -0.08
    票
    -0.08
     Eph
    -0.08
    ä¸ĢéĶ®
    -0.08
     Hide
    -0.08
    POSITIVE LOGITS
    è¿Ľåĩº
    0.17
    åĩºåħ¥
    0.15
    æŃ£å¸¸
    0.13
    éĢļè¡Į
    0.12
    è¿Ľåİ»
    0.11
    åį³ä¾¿æĺ¯
    0.11
    åīįå¾Ģ
    0.11
    å¾ĢæĿ¥
    0.11
    èĩªå¦Ĥ
    0.11
    åĭī强
    0.10
    Activations Density 0.206%

    No Known Activations