© 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. 33113
    Prev
    Next
    INDEX
    Explanations

    never followed by past participles or "really"

    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
    åİŁæľ¬
    -0.09
    èµ·åĪĿ
    -0.09
    çİ°åľ¨å·²ç»ı
    -0.09
    åİŁæĿ¥
    -0.09
    æĥ³äºĨæĥ³
    -0.09
    æľĢåĪĿ
    -0.09
    å¾Īåı¯èĥ½
    -0.08
    èĢĦ
    -0.08
    çİ°å·²
    -0.08
    å½ĵå¹´
    -0.08
    POSITIVE LOGITS
    涯
    0.10
     been
    0.09
    -ending
    0.09
    æľīè¿ĩ
    0.08
    DESCRIPTION
    0.08
    çļĵ
    0.08
    ç¿ģ
    0.08
    greens
    0.08
    æĺ¯ä¸ª
    0.08
    lü
    0.08
    Activations Density 0.015%

    No Known Activations