© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 38491
    Prev
    Next
    INDEX
    Explanations

    give a chance

    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¿Ŀé²ľ
    -0.09
    è¶ĬæĿ¥
    -0.09
    è°ĭ
    -0.09
    çĦ±
    -0.08
    è°ľ
    -0.08
    ä¹łæĥ¯äºĨ
    -0.08
    欢åĸľ
    -0.08
    èĦ«
    -0.08
    åı¯æݧ
    -0.08
    ç¾Ĭ
    -0.08
    POSITIVE LOGITS
    ä¿¡ä»»
    0.14
    çļĦä¿¡ä»»
    0.12
    宽容
    0.11
    æľºä¼ļ
    0.11
    èĢIJå¿ĥ
    0.11
    ä¿¡èµĸ
    0.10
    å¹´éĻIJ
    0.10
     kans
    0.10
    åıijå±ķ空éĹ´
    0.10
     trusting
    0.10
    Activations Density 0.045%

    No Known Activations