© 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. 54910
    Prev
    Next
    INDEX
    Explanations

    Airy disk, bands, light, centrospeckle

    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
    åĴı
    -0.10
    å±ĭ
    -0.09
    жи
    -0.08
    igid
    -0.08
     Cla
    -0.08
    åĪ¬
    -0.08
     stalled
    -0.08
     Oval
    -0.08
    _fixture
    -0.08
    clid
    -0.08
    POSITIVE LOGITS
    çĤī
    0.10
    .energy
    0.09
    åŀĤ
    0.09
    æ´Ħ
    0.09
    ä¸ĢåıĮ
    0.09
    ISR
    0.09
    ammers
    0.09
    дÑĥÑĤ
    0.09
    ä¸įå¿į
    0.08
    没æľī人
    0.08
    Activations Density 0.043%

    No Known Activations