© 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. 57984
    Prev
    Next
    INDEX
    Explanations

    for details:

    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
    .cum
    -0.10
    etty
    -0.10
    ahren
    -0.10
    áĢļ
    -0.09
    çĿĢä»ĸ
    -0.09
    antha
    -0.09
    ä»ĸ认为
    -0.09
    ITOR
    -0.09
     ngá»Ŀ
    -0.09
    eer
    -0.09
    POSITIVE LOGITS
    acre
    0.09
    ves
    0.09
     extents
    0.09
    vale
    0.08
    lich
    0.08
    ç¡®åĪĩ
    0.08
    å¸ļ
    0.08
     Frag
    0.08
    شت
    0.08
    AMB
    0.08
    Activations Density 0.027%

    No Known Activations