© 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. 34760
    Prev
    Next
    INDEX
    Explanations

    from a/the

    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
    åĮĹéĥ¨
    -0.10
    lobal
    -0.09
    oter
    -0.09
    AGEMENT
    -0.09
    è¡Ļ
    -0.09
    åĨħå¤ĸ
    -0.09
    agement
    -0.09
    é©·
    -0.08
    éĢ¸
    -0.08
    anno
    -0.08
    POSITIVE LOGITS
    æĢĻ
    0.09
    é²ľ
    0.09
    åģļçļĦäºĭæĥħ
    0.09
    èī²ç´ł
    0.09
    è°ĥåĬ¨
    0.09
    åħļç»Ħ书记
    0.09
    ourke
    0.08
    æ®Ĩ
    0.08
    ORS
    0.08
    åĴļ
    0.08
    Activations Density 0.017%

    No Known Activations