© 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. 53242
    Prev
    Next
    INDEX
    Explanations

    log file configuration

    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
    éģģ
    -0.09
    Registered
    -0.09
    åħ°èĬ±
    -0.09
    è´»
    -0.09
    çķĮéĿ¢
    -0.09
    äºĴèģĶäºĴéĢļ
    -0.09
    _Code
    -0.08
    _interfaces
    -0.08
    空ä¸Ń
    -0.08
     Mattis
    -0.08
    POSITIVE LOGITS
    ä¿ĿåŃĺ
    0.10
    _daily
    0.10
    daily
    0.10
    åį·
    0.10
    åİĭ缩
    0.10
    æ¯ı天
    0.09
    Daily
    0.09
     daily
    0.09
    åĪĩåī²
    0.09
     scrolls
    0.09
    Activations Density 0.296%

    No Known Activations