© 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. 19399
    Prev
    Next
    INDEX
    Explanations

    start of questions

    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
    æĪijä¹Ł
    -0.10
    请èģĶç³»
    -0.10
    æĪijä»¬ä¹Ł
    -0.09
    ä¹Łä¸įæĺ¯
    -0.09
    è´Ł
    -0.09
    ä½Ĩä¹Ł
    -0.09
    (?
    -0.09
    çĮ¾
    -0.09
    UNUSED
    -0.09
    åĪ°åºķ
    -0.09
    POSITIVE LOGITS
    èĤ¢ä½ĵ
    0.10
    ä¸Ĭä¸ĭ
    0.09
    线ä¸Ĭ线ä¸ĭ
    0.09
    åĵĹ
    0.08
    çħ³
    0.08
    Ñģан
    0.08
    erne
    0.08
    åįĹåĮĹ
    0.08
    åł¡
    0.08
    å½ĵåľ°
    0.08
    Activations Density 0.036%

    No Known Activations