© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 50419
    Prev
    Next
    INDEX
    Explanations

    skills

    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
    LY
    -0.11
    dan
    -0.10
    ly
    -0.09
    nar
    -0.09
    åī©
    -0.09
    eller
    -0.09
    esting
    -0.08
    ãģ¶ãĤĬ
    -0.08
    plete
    -0.08
    PCA
    -0.08
    POSITIVE LOGITS
     Lok
    0.10
    é¼ł
    0.09
     Hib
    0.08
    é£Ĵ
    0.08
    éĵ¾
    0.08
    åľ¨åIJĮä¸Ģ
    0.08
    ä¼ļåľ¨
    0.08
    åľ¨æł¡
    0.08
    _chain
    0.07
     Zone
    0.07
    Activations Density 0.016%

    No Known Activations