© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 910
    Prev
    Next
    INDEX
    Explanations

    glass panels

    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
    å¯Ħ
    -0.09
    æĨ©
    -0.09
    ä¸Ģ缴æĺ¯
    -0.09
    çĻ¾èĬ±
    -0.09
    à¹Ĥà¸Ļ
    -0.08
     Nec
    -0.08
     над
    -0.08
    æĺ¯ä¸Ģ座
    -0.08
    åµĺ
    -0.08
     cute
    -0.08
    POSITIVE LOGITS
    çİ»çĴĥ
    0.15
     glass
    0.14
    Glass
    0.12
    glass
    0.12
     Glass
    0.12
    çĴĥ
    0.11
    à¸ģระà¸Ī
    0.11
    glas
    0.10
    æ¤Ĵ
    0.10
    .getWindow
    0.09
    Activations Density 0.077%

    No Known Activations