© 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. 56795
    Prev
    Next
    INDEX
    Explanations

    programming and technical terms

    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
    orem
    -0.11
    een
    -0.10
    aus
    -0.10
     лиÑĪÑĮ
    -0.09
    eer
    -0.09
    ev
    -0.09
    Ŀ
    -0.09
    ean
    -0.09
    ÑģÑı
    -0.08
    ãĤº
    -0.08
    POSITIVE LOGITS
    iture
    0.08
    è¾Ī
    0.08
    upro
    0.08
    encoded
    0.08
     perc
    0.08
    èıĮ
    0.08
    3
    0.07
    空æ°Ķä¸Ń
    0.07
    ä¸Ģ对ä¸Ģ
    0.07
    8
    0.07
    Activations Density 1.036%

    No Known Activations