© 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. 5824
    Prev
    Next
    INDEX
    Explanations

    really bad, fact borne, forget

    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
    unga
    -0.10
    åĬĽè¿ĺæĺ¯
    -0.09
    è¹Ĵ
    -0.09
    ðŁĴĮ
    -0.09
    è¿ĺä¸įæĺ¯
    -0.09
     bè
    -0.08
    æĹłæīĢè°ĵ
    -0.08
    åıŁ
    -0.08
    hort
    -0.08
    OMET
    -0.08
    POSITIVE LOGITS
    ä¸į论
    0.08
     Plat
    0.08
    ä¸ĬæĿ¥
    0.08
    ä¸įä½Ĩ
    0.08
    ä»Ĺ
    0.08
    çĶŁçĶŁ
    0.08
    é«»
    0.08
    å°±å¦ĤåIJĮ
    0.08
    _restrict
    0.08
    alom
    0.07
    Activations Density 0.064%

    No Known Activations