© 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. 38776
    Prev
    Next
    INDEX
    Explanations

    Shepherd variations

    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
    umen
    -0.09
    uar
    -0.09
    opr
    -0.08
    oust
    -0.08
    isi
    -0.08
    åĢ¡å¯¼
    -0.08
    æľīçĤ¹
    -0.08
     bout
    -0.08
    outs
    -0.08
    r
    -0.08
    POSITIVE LOGITS
    amina
    0.11
    à¹Į
    0.11
    å¾Ģä¸ĭ
    0.10
    oot
    0.10
    eur
    0.10
    Ñģон
    0.10
    èģĮä¸ļéģĵå¾·
    0.09
    ви
    0.09
    SCII
    0.09
    ores
    0.09
    Activations Density 0.006%

    No Known Activations