© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 20027
    Prev
    Next
    INDEX
    Explanations

    VP, V, or vb followed by numbers or specific codes

    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
    æĿijæ°ij
    -0.13
    åħļæĶ¯éĥ¨
    -0.13
     воÑĤ
    -0.13
     вÑģÑĤÑĢеÑĩа
    -0.09
    virt
    -0.09
    /videos
    -0.09
     visc
    -0.09
    viz
    -0.09
     vastly
    -0.09
    vpn
    -0.09
    POSITIVE LOGITS
    atile
    0.15
    LOCITY
    0.15
    quez
    0.13
    ä¹İ
    0.13
    OLUME
    0.13
    yards
    0.13
    olatile
    0.13
    atility
    0.12
    osity
    0.12
    odafone
    0.12
    Activations Density 1.253%

    No Known Activations