© 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. 31136
    Prev
    Next
    INDEX
    Explanations

    attaching files or tags

    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
    tings
    -0.09
    ä¼ģä¸ļæıIJä¾Ľ
    -0.09
    tier
    -0.08
    æľªçŁ¥
    -0.08
    izia
    -0.08
    èłĸ
    -0.08
     Poz
    -0.08
     vá»ı
    -0.08
    è¿ģ移
    -0.08
    ooting
    -0.08
    POSITIVE LOGITS
     attach
    0.14
    _attached
    0.14
     attaching
    0.13
     ë¶
    0.13
     attached
    0.13
     attaches
    0.13
     Attached
    0.12
    éĻĦ
    0.12
    attached
    0.12
    Attached
    0.12
    Activations Density 0.111%

    No Known Activations