© 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. 62392
    Prev
    Next
    INDEX
    Explanations

    already or in the past

    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
    èĤī
    -0.10
    æľĢç»Ī
    -0.09
    æľĢåIJİä¸Ģ次
    -0.09
    忽
    -0.09
     idi
    -0.09
     Woodward
    -0.08
    ivas
    -0.08
    yah
    -0.08
    .FILL
    -0.08
    icas
    -0.08
    POSITIVE LOGITS
    ä¸Ģå¼Ģå§ĭå°±
    0.20
    就已ç»ı
    0.16
    åħĪ天
    0.14
    æľ¬èº«å°±
    0.14
    æĻĤçĤ¹ãģ§
    0.12
     already
    0.12
    åŁºè°ĥ
    0.12
    注å®ļ
    0.12
     outset
    0.11
     Already
    0.11
    Activations Density 0.085%

    No Known Activations