© 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. 18524
    Prev
    Next
    INDEX
    Explanations

    auto compounds

    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
    hips
    -0.10
    头åıij
    -0.10
    uez
    -0.10
    ãģĴ
    -0.09
    ç©Ĺ
    -0.09
    abr
    -0.09
    otec
    -0.09
    ipel
    -0.09
    aber
    -0.08
    ault
    -0.08
    POSITIVE LOGITS
    /manual
    0.12
    иÑĤеÑĤ
    0.11
    åĭĥåĭĥ
    0.11
    entic
    0.11
    onomous
    0.11
    à¹Ģลà¸Ĥ
    0.11
    verständlich
    0.10
    ies
    0.10
    ick
    0.10
    istic
    0.10
    Activations Density 0.101%

    No Known Activations