© 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. 62604
    Prev
    Next
    INDEX
    Explanations

    all you need

    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
     вмеÑģÑĤ
    -0.09
    æľªæĿ¥çļĦ
    -0.09
    éĥ½æĥ³
    -0.09
    éģĹå¿ĺ
    -0.08
    éĥ½è§īå¾Ĺ
    -0.08
    åħ¬è®¤
    -0.08
    ä¸ĢæĹ¥
    -0.08
     rack
    -0.08
    åıªçŁ¥éģĵ
    -0.08
    éªĮè¯ģçłģ
    -0.08
    POSITIVE LOGITS
    åģļ强
    0.10
    æģ¤
    0.10
     ever
    0.09
    argin
    0.08
    ë²Į
    0.08
    -ever
    0.08
    Drawing
    0.08
    _boost
    0.08
    req
    0.08
     payouts
    0.08
    Activations Density 0.008%

    No Known Activations