© 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. 53354
    Prev
    Next
    INDEX
    Explanations

    time spans ago

    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
    两类
    -0.10
    æĥħåĨµè¿Ľè¡Į
    -0.10
    æĽ´å¥½åľ°
    -0.09
    åħĪè¡Į
    -0.09
    å·¥ä½ľæľºåĪ¶
    -0.09
    æł¹æľ¬ä¸į
    -0.08
    æĸ¹å¼ıè¿Ľè¡Į
    -0.08
    ä¸įä½İäºİ
    -0.08
    çĦ¶åIJİåĨį
    -0.08
    åīij
    -0.08
    POSITIVE LOGITS
    yles
    0.10
    è¿ĻçĤ¹
    0.09
    æľĽåİ»
    0.09
    istically
    0.09
    èµĦæĸĻæĺ¾ç¤º
    0.09
    _YEAR
    0.09
    _NT
    0.09
    èĢĮå·²
    0.09
    aty
    0.09
    êľ
    0.09
    Activations Density 0.039%

    No Known Activations