© 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. 11049
    Prev
    Next
    INDEX
    Explanations

    better understanding

    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
    çĶ¬
    -0.09
    è°ĵ
    -0.08
    æIJº
    -0.08
    yll
    -0.08
    igraphy
    -0.08
    ç®Ģ缴æĺ¯
    -0.08
    éĥ½ä¸įæĺ¯
    -0.08
    碰
    -0.08
    èĩªåįij
    -0.08
    å®ŀçī©
    -0.08
    POSITIVE LOGITS
    çIJĨ解
    0.38
     understanding
    0.33
    çIJĨ解åĴĮ
    0.30
     understand
    0.28
    çļĦçIJĨ解
    0.28
    äºĨ解
    0.27
     Understanding
    0.26
    Understanding
    0.26
     understood
    0.24
     understands
    0.23
    Activations Density 0.193%

    No Known Activations