© 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. 10261
    Prev
    Next
    INDEX
    Explanations

    mole, moling, molester, molar

    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
     gloss
    -0.09
    ฤษ
    -0.09
    _NOW
    -0.09
    æĭŃ
    -0.09
    proved
    -0.08
    åIJĮæľŁ
    -0.08
    rap
    -0.08
    é«ĺä½į
    -0.08
     pace
    -0.08
    ill
    -0.08
    POSITIVE LOGITS
    cul
    0.11
    itud
    0.10
    æĦĨ
    0.10
    stones
    0.09
    HEET
    0.09
    ä¸ºæľŁ
    0.09
    itudes
    0.09
    eson
    0.08
    gram
    0.08
    åºķéĥ¨
    0.08
    Activations Density 0.019%

    No Known Activations