© 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. 10283
    Prev
    Next
    INDEX
    Explanations

    policy guidelines

    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
    ç¨Ģ
    -0.10
    è¡Įæĥħ
    -0.09
    utas
    -0.09
    æĭĽåij¼
    -0.09
    æİĮ声
    -0.09
    éĤ£ä»½
    -0.09
    adar
    -0.09
    unan
    -0.08
    esser
    -0.08
    æĬĵç´§
    -0.08
    POSITIVE LOGITS
    æĸ¹éĴĪ
    0.16
    .policy
    0.13
     policies
    0.13
     policy
    0.12
    (policy
    0.12
    ubo
    0.11
    æĶ¿çŃĸ
    0.10
    policy
    0.10
     Policy
    0.10
     Policies
    0.10
    Activations Density 0.037%

    No Known Activations