© 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. 19009
    Prev
    Next
    INDEX
    Explanations

    views or view

    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
    unes
    -0.11
    ouns
    -0.11
    ircles
    -0.10
    ä¸Ģå®ļçļĦ
    -0.10
    ä½łçľĭ
    -0.09
    æĪij说
    -0.09
    itude
    -0.09
    ção
    -0.08
    è´¯
    -0.08
     Ñģобой
    -0.08
    POSITIVE LOGITS
    able
    0.15
    eing
    0.14
    shed
    0.13
    .OnClickListener
    0.12
    finder
    0.12
    ings
    0.12
    ably
    0.11
    ables
    0.10
    uale
    0.10
    pager
    0.10
    Activations Density 0.085%

    No Known Activations