© 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. 6631
    Prev
    Next
    INDEX
    Explanations

    physical shapes or forms

    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
    cedes
    -0.10
    弦
    -0.09
    hift
    -0.09
    ÑĨеп
    -0.09
    ooks
    -0.09
    idth
    -0.09
    cken
    -0.09
    oran
    -0.09
    åħ¨èº«
    -0.09
    upply
    -0.09
    POSITIVE LOGITS
     shape
    0.14
    shape
    0.14
    -shaped
    0.13
    (shape
    0.12
     shaped
    0.12
    å½¢çĬ¶
    0.12
     shapes
    0.12
    .shape
    0.11
    çĭĢ
    0.11
    åIJĦå¼Ĥ
    0.11
    Activations Density 0.064%

    No Known Activations