© 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. 56587
    Prev
    Next
    INDEX
    Explanations

    supervised supervision

    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
    åĬŀå®ŀäºĭ
    -0.10
    annah
    -0.10
    ToWorld
    -0.09
    çĦ¦
    -0.09
    让å®ĥ
    -0.08
    èį¡
    -0.08
    iets
    -0.08
    é¢ĺæĿIJ
    -0.08
    ä¸įä¸Ĭ
    -0.08
    ç´¯
    -0.08
    POSITIVE LOGITS
    æĮĩ导ä¸ĭ
    0.17
    æĮĩ导
    0.15
     supervision
    0.14
     guidance
    0.13
    -guid
    0.12
    çĽijçĿ£
    0.12
    æĮĩå°İ
    0.12
    ä¸ĵä¸ļ人士
    0.11
     Guidance
    0.11
    çĽij管
    0.11
    Activations Density 0.040%

    No Known Activations