© 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. 3734
    Prev
    Next
    INDEX
    Explanations

    stripes and bands

    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
    ÑģÑĤÑĥп
    -0.11
    æłĭ
    -0.09
    ly
    -0.09
    lt
    -0.09
    relation
    -0.09
    Ping
    -0.09
    mente
    -0.09
    nd
    -0.09
    bourne
    -0.08
    scape
    -0.08
    POSITIVE LOGITS
    ingly
    0.13
    ãģĹãĤĩãģĨ
    0.12
    æł·çļĦ
    0.11
    çŧ
    0.09
    ategy
    0.09
    à¹Ħล
    0.09
    обÑĢаз
    0.09
    ĭ
    0.09
    ooth
    0.09
    lename
    0.09
    Activations Density 0.006%

    No Known Activations