© 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. 2843
    Prev
    Next
    INDEX
    Explanations

    flag or flagged content

    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
    loth
    -0.10
    íĴĢ
    -0.10
    æĭĬ
    -0.09
    OUGH
    -0.09
    бÑĭ
    -0.09
    omas
    -0.09
    æŃ¤åŁºç¡Ģä¸Ĭ
    -0.09
    ads
    -0.08
    weet
    -0.08
    èī¯å¿ĥ
    -0.08
    POSITIVE LOGITS
    ging
    0.15
    pped
    0.13
    ï¸ı
    0.13
    ged
    0.13
    gable
    0.12
    GED
    0.12
    pole
    0.12
    .FLAG
    0.11
    eted
    0.11
    mith
    0.11
    Activations Density 0.027%

    No Known Activations