© 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. 23924
    Prev
    Next
    INDEX
    Explanations

    "is" followed by a question or statement

    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
    åĨĴ
    -0.10
    éļį
    -0.09
    åħ¹
    -0.09
    åľ°ä¸ĭ
    -0.09
    ä»ĺåĩº
    -0.09
    ASON
    -0.09
    åľ°ä¸ĭ室
    -0.08
    ä¸Ģå®ļèĥ½
    -0.08
    æIJĶ
    -0.08
    plies
    -0.08
    POSITIVE LOGITS
    伤çĹħ
    0.08
    entifier
    0.08
    ç©°
    0.08
     Sabb
    0.08
    prech
    0.08
    _Man
    0.08
    æĬ¡
    0.08
    oms
    0.07
    åİŁæĿ¥
    0.07
    clr
    0.07
    Activations Density 0.042%

    No Known Activations