© 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. 52361
    Prev
    Next
    INDEX
    Explanations

    Carey, Cary, Conrad

    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
     Reign
    -0.10
    /her
    -0.09
    иÑĤеÑĤ
    -0.09
    çªĴ
    -0.09
    pery
    -0.09
    енно
    -0.08
    ENA
    -0.08
    inte
    -0.08
    åĮħåĽ´
    -0.08
    itch
    -0.08
    POSITIVE LOGITS
    éĩĮç¨ĭ
    0.09
    nbsp
    0.09
    onne
    0.09
    åİ¢
    0.09
    heet
    0.09
    ëģĶ
    0.09
    ÙĪس
    0.09
    веÑĤ
    0.08
    forth
    0.08
    TURE
    0.08
    Activations Density 0.004%

    No Known Activations