© 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. 9889
    Prev
    Next
    INDEX
    Explanations

    dull, blunt descriptive words

    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
    lyph
    -0.10
    oretical
    -0.10
    綦
    -0.08
    hton
    -0.08
    渰
    -0.08
    ially
    -0.08
    黣
    -0.08
     Ñģли
    -0.08
    /Application
    -0.08
    лÑİ
    -0.08
    POSITIVE LOGITS
    sville
    0.14
    éĴĿ
    0.12
    ness
    0.09
    æĬķèŀįèµĦ
    0.09
    éĹ·
    0.09
    ç§ĥ
    0.09
    æĮ«
    0.09
    ingly
    0.08
    ens
    0.08
    mate
    0.08
    Activations Density 0.004%

    No Known Activations