© 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. 49430
    Prev
    Next
    INDEX
    Explanations

    what type of

    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
    ä¸į论
    -0.09
     generic
    -0.08
    æľīéĻIJ
    -0.08
    å¼¹æĢ§
    -0.08
    “
    -0.08
     عاÙħ
    -0.08
     Generic
    -0.08
    åĨ²
    -0.07
    Generic
    -0.07
    thr
    -0.07
    POSITIVE LOGITS
    åºĶçĶ¨åľºæĻ¯
    0.12
    åħ·ä½ĵæĥħåĨµ
    0.12
    楮
    0.11
    desired
    0.11
    ળ
    0.10
    å±±èĦī
    0.10
    ç»ıèIJ¥æ´»åĬ¨
    0.10
    缸åħ³è§Ħå®ļ
    0.10
    æ®Ħ
    0.10
    å¼Ħå¾Ĺ
    0.10
    Activations Density 0.070%

    No Known Activations