© 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. 5774
    Prev
    Next
    INDEX
    Explanations

    tough pill to swallow

    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
    onn
    -0.10
    èIJĮ
    -0.09
    amac
    -0.09
    æ´Ĺæīĭ
    -0.08
    æİ¥æīĭ
    -0.08
    æįķæįī
    -0.08
     Dane
    -0.08
    åĢĻ
    -0.08
    æĢ§å¼º
    -0.08
    èŁĴ
    -0.08
    POSITIVE LOGITS
    wick
    0.10
    æĪijåİ»
    0.10
    常åĬ¡åī¯
    0.10
    ÐļÑĥÑĢÑģ
    0.09
    ä¸įæħİ
    0.09
    _encoding
    0.09
    æ²ĥ
    0.09
     Ventura
    0.08
    ä¿İ
    0.08
    rement
    0.08
    Activations Density 0.032%

    No Known Activations