© 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. 3893
    Prev
    Next
    INDEX
    Explanations

    manual actions or 'i'

    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
    æľĢåŁºæľ¬
    -0.10
    ä¸įåĩº
    -0.10
    è§ħ
    -0.10
    æĭ¾
    -0.09
    å§Ķ
    -0.09
    /mit
    -0.09
    éģĹæĨ¾
    -0.09
    éªĬ
    -0.09
    è±ĨèħIJ
    -0.09
    æĹ»
    -0.09
    POSITIVE LOGITS
    æīĭåĬ¨
    0.10
     manually
    0.09
    _manual
    0.09
     manual
    0.08
     careful
    0.08
     Manual
    0.08
    人为
    0.08
    æīĭå·¥
    0.08
    วà¸Ķ
    0.08
    æľīå¤ļ
    0.08
    Activations Density 0.801%

    No Known Activations