© 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. 10013
    Prev
    Next
    INDEX
    Explanations

    how long it takes

    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
     ÑĢазвива
    -0.10
    设计çIJĨ念
    -0.09
    æĬķåħ¥åĪ°
    -0.09
    robat
    -0.09
    åīĸ
    -0.09
    éĥ½æĹłæ³ķ
    -0.09
     Ñģвоего
    -0.09
    åıĤä¸İåĪ°
    -0.09
    æĬ¬å¤´
    -0.08
    éĥ½å·²ç»ı
    -0.08
    POSITIVE LOGITS
    èĢĹ
    0.15
    èĬ±è´¹
    0.14
    èĬ±äºĨ
    0.13
     costs
    0.12
    èĢĹè´¹
    0.12
    çĶ¨äºĨ
    0.12
    代价
    0.12
    éļ¾åº¦
    0.10
    è´¹çĶ¨
    0.10
    è´¹
    0.10
    Activations Density 0.070%

    No Known Activations