© 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. 1210
    Prev
    Next
    INDEX
    Explanations

    so choose, so much, so happens

    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
    éĵ¢
    -0.08
    渤
    -0.08
    ç
    -0.08
    å¸Ĥä¸Ńå¿ĥ
    -0.08
    ça
    -0.08
    éģ´éĢī
    -0.08
    aug
    -0.08
    ä¸įçĶ±å¾Ĺ
    -0.08
    _ATTRIBUTES
    -0.08
    crets
    -0.08
    POSITIVE LOGITS
    æĶ¯åĩº
    0.10
    è¿Ļæĸ¹éĿ¢
    0.09
    PUTE
    0.09
    .pointer
    0.09
    讳
    0.09
    igram
    0.09
    ients
    0.09
    æĭ¿åĩºæĿ¥
    0.09
    åį³å°Ĩ
    0.09
    PointerType
    0.08
    Activations Density 0.020%

    No Known Activations