© 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. 55253
    Prev
    Next
    INDEX
    Explanations

    limits or boundaries

    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
    .inline
    -0.10
    illed
    -0.09
    å½ķ
    -0.09
    åķľ
    -0.08
    sic
    -0.08
    ERGY
    -0.08
     Jung
    -0.08
    åIJĦç§į
    -0.08
     Loft
    -0.08
    åĬ¢
    -0.08
    POSITIVE LOGITS
    太ä¹ħ
    0.09
    åĨħçļĦ
    0.09
    éĻĦè¿ijçļĦ
    0.09
    ogl
    0.09
    æ¸ħæĻ°
    0.09
    åİŁæľīçļĦ
    0.09
    ür
    0.08
     late
    0.08
    åĿIJçĿĢ
    0.08
    æĹº
    0.08
    Activations Density 0.031%

    No Known Activations