© 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. 22902
    Prev
    Next
    INDEX
    Explanations

    it wasn’t that

    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¸Ģå¦Ĥ
    -0.10
    å¦Ĥä½ķçľĭå¾ħ
    -0.09
    对æĪijæĿ¥è¯´
    -0.09
    è°´
    -0.09
     biç
    -0.09
    åIJĮç±»
    -0.08
    èݹ
    -0.08
    ÑĢеги
    -0.08
     coolest
    -0.08
    OTE
    -0.08
    POSITIVE LOGITS
    èĢĮæĺ¯
    0.09
    ละ
    0.09
    åŀĤ
    0.09
    ä½łä¸į
    0.08
    è¡°
    0.08
    æĹłåħ³
    0.08
    cope
    0.08
    åı£ä¸Ń
    0.08
    主ä¹īæĢĿæĥ³
    0.08
    åīľ
    0.07
    Activations Density 0.010%

    No Known Activations