© 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. 55660
    Prev
    Next
    INDEX
    Explanations

    sameness or difference

    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
    .MULT
    -0.09
    å§Ŀ
    -0.09
    lander
    -0.09
    ç¿ķ
    -0.09
    æ°Ķæ°Ľ
    -0.08
     <*>
    -0.08
    èݲèĬ±
    -0.08
    ç£IJ
    -0.08
    sWith
    -0.08
    exas
    -0.08
    POSITIVE LOGITS
    å°ijæķ°
    0.10
    äºĨ好
    0.09
    2
    0.09
    orc
    0.09
     rig
    0.09
     Rae
    0.09
    åIJįæł¡
    0.08
    ailed
    0.08
    æ±Ĺ
    0.08
    é¦ĸ个
    0.08
    Activations Density 0.003%

    No Known Activations