© 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. 27705
    Prev
    Next
    INDEX
    Explanations

    movement prepositions

    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
    绫
    -0.09
    ankan
    -0.09
    æĮ½
    -0.09
    éĴ°
    -0.09
    ç¼Ł
    -0.08
     IT
    -0.08
    èķŀ
    -0.08
     Ches
    -0.08
    èĬ¯
    -0.08
    æªĶ
    -0.08
    POSITIVE LOGITS
    ç£Ķ
    0.09
    åĩºåľº
    0.09
    è¾±
    0.09
    åIJ»
    0.09
     Crane
    0.09
    èĻĵ
    0.09
    æ¿¡
    0.09
    éĽĦ
    0.08
    prise
    0.08
     medi
    0.08
    Activations Density 0.052%

    No Known Activations