© 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. 62290
    Prev
    Next
    INDEX
    Explanations

    descriptive

    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
    æĩĴ
    -0.10
    éı¡
    -0.09
    头æĿ¡
    -0.09
    æĭĨ
    -0.09
     Orth
    -0.09
    Nested
    -0.09
    checker
    -0.08
    ÐĶÐķ
    -0.08
    ORT
    -0.08
    æ¶Ĥ
    -0.08
    POSITIVE LOGITS
    çķ¹
    0.09
    è®®
    0.09
    ooting
    0.09
    æĢ§
    0.08
    pure
    0.08
    åŀĭ
    0.08
    榴
    0.08
    æ³»
    0.08
    anto
    0.08
    orian
    0.08
    Activations Density 0.059%

    No Known Activations