© 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. 56868
    Prev
    Next
    INDEX
    Explanations

    names/titles followed by verbs

    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
    å¹´è½»
    -0.10
    å·¡åĽŀ
    -0.10
    åł¤
    -0.10
    ç»ķ
    -0.09
    诰
    -0.09
    æĹĹä¸ĭ
    -0.08
    åĩºè½¨
    -0.08
    åĦ¿ç«¥
    -0.08
    è¿Ļä¹Īå¤ļå¹´
    -0.08
    èĢĥçĤ¹
    -0.08
    POSITIVE LOGITS
    ç¼¢
    0.11
    çıŃ级
    0.10
    女çĶŁ
    0.10
    çıŃéķ¿
    0.10
    室åıĭ
    0.10
     classmates
    0.09
    æ»´
    0.09
    æįº
    0.09
    å¯¹æľªæĿ¥
    0.09
    -finals
    0.09
    Activations Density 0.065%

    No Known Activations