© 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. 19492
    Prev
    Next
    INDEX
    Explanations

    set

    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
    ä¼ĺç§ĢçļĦ
    -0.09
    åĬ¨æīĭ
    -0.09
    éģĴ
    -0.09
    ä¼ĺç§Ģ
    -0.09
    èĥĮåĮħ
    -0.09
    è°
    -0.09
    å½»
    -0.09
    æĪIJçĨŁçļĦ
    -0.09
    è¶Ĭ好
    -0.09
    åĽłåľ°
    -0.09
    POSITIVE LOGITS
     aside
    0.13
    tlement
    0.12
    aside
    0.12
    tle
    0.11
    åŁºè°ĥ
    0.10
    elah
    0.10
     Aside
    0.09
    uptools
    0.09
    åĸĻ
    0.09
     firmly
    0.09
    Activations Density 0.174%

    No Known Activations