© 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. 19183
    Prev
    Next
    INDEX
    Explanations

    marc names

    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
    lined
    -0.10
    brace
    -0.10
    辫
    -0.09
    enga
    -0.09
    ваÑĢ
    -0.09
    ged
    -0.09
     dÃłng
    -0.09
    mite
    -0.08
    gear
    -0.08
    lix
    -0.08
    POSITIVE LOGITS
    ussen
    0.11
    çģĮ
    0.11
    ony
    0.09
    åĽŀäºĭ
    0.09
    illo
    0.09
     Ñģобой
    0.09
    inus
    0.09
    eur
    0.09
    svm
    0.08
    itecture
    0.08
    Activations Density 0.005%

    No Known Activations