© 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. 18334
    Prev
    Next
    INDEX
    Explanations

    minced beef

    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
    tery
    -0.10
    nc
    -0.09
    thing
    -0.08
    .OS
    -0.08
    еÑģп
    -0.08
    çĶļä¹Ī
    -0.08
    teenth
    -0.08
    teen
    -0.08
    å·¥å§Ķ
    -0.08
    sons
    -0.08
    POSITIVE LOGITS
    urre
    0.10
    è´»
    0.09
    å¿Į
    0.09
    ä¾£
    0.08
    ä½Ķ
    0.08
    äºĨäºĽ
    0.08
    _Framework
    0.08
    ç¢İ
    0.08
    illas
    0.08
    istring
    0.08
    Activations Density 0.004%

    No Known Activations