© 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. 18927
    Prev
    Next
    INDEX
    Explanations

    if + subject + verb

    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
    æŃ£åĢ¼
    -0.09
    пÑĥÑģка
    -0.09
    åģĩ
    -0.09
    åĽ¾çĶ»
    -0.09
    é¢ĨæĤŁ
    -0.08
    èĭ¥æĤ¨
    -0.08
    oka
    -0.08
     hoe
    -0.08
    èĴ¿
    -0.08
    å¯Ĵ
    -0.08
    POSITIVE LOGITS
    æĪĸå¤ļ
    0.09
    æī°
    0.09
    åħ±åIJĮåıijå±ķ
    0.09
    æ¢Ĺ
    0.09
    ä¸İæѤ
    0.09
    ä¸ĵ家åѦèĢħ
    0.09
    ugu
    0.09
    idel
    0.08
    æħĬ
    0.08
     Singer
    0.08
    Activations Density 0.027%

    No Known Activations