© 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. 17397
    Prev
    Next
    INDEX
    Explanations

    PG, FTA, S, SA, SI, H, F, SEL, LAN abbreviations

    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
    æ¡Ĩ
    -0.08
    ura
    -0.08
     Panc
    -0.08
    ãĥĻãĥ³
    -0.08
    eller
    -0.07
    ioc
    -0.07
    ä¸įå°ij
    -0.07
    quina
    -0.07
    许å¤ļ
    -0.07
    ä¾Ŀèµĸ
    -0.07
    POSITIVE LOGITS
    è¾¹éĻħ
    0.10
    OWL
    0.10
    åıĪæľī
    0.09
    çļĦ人æĿ¥è¯´
    0.09
    å½ĵåľ°æŶéĹ´
    0.09
    ä¹°äºĨ
    0.09
    è²Ĥ
    0.09
    å°±å¦ĤåIJĮ
    0.09
    _CREAT
    0.09
    åij¨äºĶ
    0.09
    Activations Density 0.753%

    No Known Activations