© 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. 20047
    Prev
    Next
    INDEX
    Explanations

    ease or difficulty

    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
    ä¹Łç®Ĺæĺ¯
    -0.09
    ä¾ĽéĶĢ
    -0.08
    éĶij
    -0.08
    滹
    -0.08
    飶
    -0.08
    -hash
    -0.08
    åīįæıIJæĺ¯
    -0.08
     çözüm
    -0.08
    AP
    -0.08
     lat
    -0.07
    POSITIVE LOGITS
    éļ¾
    0.23
    å¾Īéļ¾
    0.23
    éļ¾åº¦
    0.22
    éļ¾ä»¥
    0.21
    ä¸į容æĺĵ
    0.21
    æĽ´å®¹æĺĵ
    0.21
     difficulty
    0.20
    容æĺĵ
    0.19
     Difficulty
    0.19
    ä¸įæĺĵ
    0.19
    Activations Density 0.107%

    No Known Activations