© 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. 61842
    Prev
    Next
    INDEX
    Explanations

    matrix and matrices

    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
    åĽ¢ä½ĵ
    -0.09
    ubar
    -0.09
    æĽ²çº¿
    -0.08
    wagon
    -0.08
    çĽĶ
    -0.08
    лом
    -0.07
    å»Ĭ
    -0.07
    æŀĿ
    -0.07
    _BRANCH
    -0.07
    branch
    -0.07
    POSITIVE LOGITS
    磩éĺµ
    0.18
     matrix
    0.17
    matrix
    0.17
     Matrix
    0.15
    	matrix
    0.14
    .matrix
    0.13
    (Matrix
    0.13
    ç½ijæł¼
    0.13
     rows
    0.13
    Matrix
    0.13
    Activations Density 0.392%

    No Known Activations