© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 8854
    Prev
    Next
    INDEX
    Explanations

    rooftops and tops

    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
    anke
    -0.10
    èł²
    -0.10
    èIJ¥ä¸ļ
    -0.09
    åIJĿ
    -0.09
    æ·ĺ
    -0.09
    éĿ¢èĨľ
    -0.09
    è¿Ľé¡¹
    -0.09
    berger
    -0.09
    èIJ¥ä¸ļæĶ¶åħ¥
    -0.09
    ç¨Ĺ
    -0.09
    POSITIVE LOGITS
    å¹³åı°ä¸Ĭ
    0.11
    æĶĢçĻ»
    0.10
    顶端
    0.10
     rooft
    0.10
    å¹³åı°
    0.10
     atop
    0.10
    è§Ĥæµĭ
    0.09
    é«ĺãģı
    0.09
    çĪ¬ä¸Ĭ
    0.09
    å·ħ
    0.09
    Activations Density 0.042%

    No Known Activations