© 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. 56869
    Prev
    Next
    INDEX
    Explanations

    subtle, subtle changes, subtle way

    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
    ëĭ¨
    -0.09
    æĢ»ä¹ĭ
    -0.09
    sWith
    -0.09
    ä¿Ĺ称
    -0.09
    dr
    -0.09
    缼
    -0.08
    nung
    -0.08
    ya
    -0.08
    æĸ¤
    -0.08
    atr
    -0.08
    POSITIVE LOGITS
    -minded
    0.12
    æģĻ
    0.11
     minded
    0.11
    ä¹ĭå¤Ħ
    0.11
    ãģĮãĤī
    0.10
     à¹Ĩ
    0.10
    åķĨæľº
    0.10
    à¹Ĩ
    0.10
    NESS
    0.10
    çļĦåıĺåĮĸ
    0.09
    Activations Density 0.005%

    No Known Activations