© 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. 23730
    Prev
    Next
    INDEX
    Explanations

    competitors

    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
    eken
    -0.11
    èĩªæĿ¥
    -0.10
    éĢĿ
    -0.09
    ãģĻãģĶ
    -0.09
    поз
    -0.09
    agus
    -0.09
    éľĩèį¡
    -0.09
    ESH
    -0.09
    ä¸Ńç¾İ
    -0.09
    éĽĨåĽ¢èĤ¡ä»½
    -0.09
    POSITIVE LOGITS
    çļĦ产åĵģ
    0.10
    ç¬ĥ
    0.10
    ely
    0.09
    usi
    0.09
    纷纷
    0.08
    apa
    0.08
    å¼ĢæĿ¥
    0.08
     Dmit
    0.08
    ape
    0.08
     Bail
    0.08
    Activations Density 0.037%

    No Known Activations