© 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. 31151
    Prev
    Next
    INDEX
    Explanations

    positive and negative evaluations

    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
    -than
    -0.11
    é»İæĺİ
    -0.11
    纰
    -0.10
     ÑģоÑģ
    -0.09
    åĨµ
    -0.09
    è®®äºĭ
    -0.09
    å°¤
    -0.09
    éĹ³
    -0.09
    æī£éĻ¤
    -0.08
    ta
    -0.08
    POSITIVE LOGITS
    æĽ´å¤ļ
    0.17
    æĽ´åĬł
    0.17
    æĽ´ä¸º
    0.16
     more
    0.16
    æĽ´å¤ļçļĦ
    0.15
    æĽ´èĥ½
    0.15
    æĽ´
    0.15
    æĽ´å¥½åľ°
    0.15
    æľīæĽ´å¤ļçļĦ
    0.15
    æĽ´å®¹æĺĵ
    0.14
    Activations Density 0.252%

    No Known Activations