© 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. 53283
    Prev
    Next
    INDEX
    Explanations

    prepared texts and speeches

    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
     clo
    -0.09
    女æĢ§æľĭåıĭ
    -0.09
    clo
    -0.09
    对ç«ĭ
    -0.09
    奢ä¾Īåĵģ
    -0.09
    ä¸ĢæĬĬ
    -0.09
    ning
    -0.09
    å®ĵ
    -0.09
    å¯Ĩ
    -0.09
    èĴIJ
    -0.09
    POSITIVE LOGITS
     prepared
    0.11
    prepared
    0.11
    ä¸Ģå¥Ĺ
    0.09
    åı°è¯į
    0.09
    led
    0.09
     responses
    0.08
     Prepared
    0.08
    otron
    0.08
    æŃĮ
    0.08
    è¡Ķ
    0.08
    Activations Density 0.060%

    No Known Activations