© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 10382
    Prev
    Next
    INDEX
    Explanations

    door, doors, doorstep

    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
     عÙĨÙĩ
    -0.09
    .fm
    -0.09
     жид
    -0.09
    )::
    -0.09
    ipy
    -0.09
    éĤij
    -0.09
    举
    -0.09
     Oaks
    -0.09
    å®ĮäºĨ
    -0.09
    çĮĹ
    -0.09
    POSITIVE LOGITS
     doorstep
    0.19
    éŨåīį
    0.17
    éĹ¨æ§Ľ
    0.16
    éŨåı£
    0.16
     door
    0.16
    ä¸ĬéŨ
    0.16
     doors
    0.15
    大éŨ
    0.15
    éŨ
    0.14
    door
    0.14
    Activations Density 0.073%

    No Known Activations