© 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. 35072
    Prev
    Next
    INDEX
    Explanations

    breakdown/decomposition

    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
    åħµé©¬
    -0.09
    æīŃ转
    -0.09
    hya
    -0.09
    ắc
    -0.09
    akte
    -0.09
    æĸ°ä¸ī
    -0.09
    (PropertyName
    -0.09
    æİĴ污
    -0.08
    leo
    -0.08
    antino
    -0.08
    POSITIVE LOGITS
    åĪĨ解
    0.13
    æĭĨ
    0.11
    ç¢İ
    0.10
    å¼ĢäºĨ
    0.10
     decomposition
    0.10
     breakdown
    0.09
    楼ä¸Ĭ
    0.09
    åĪĨè£Ĥ
    0.09
    å¼ĢæĿ¥
    0.09
    åķIJ
    0.09
    Activations Density 0.254%

    No Known Activations