© 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. 30242
    Prev
    Next
    INDEX
    Explanations

    professional before open, at, of, who, or punctuation

    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
    angent
    -0.11
    æł·çļĦ
    -0.10
    lings
    -0.10
    OrUpdate
    -0.10
    strap
    -0.09
     пÑĢоиÑģ
    -0.09
    breaker
    -0.09
    bose
    -0.09
    åĽŀäºĭ
    -0.09
    æ¼±
    -0.09
    POSITIVE LOGITS
    åĮĸçļĦ
    0.12
    人士
    0.11
    -looking
    0.11
    -minded
    0.10
    ität
    0.10
    /company
    0.09
    nels
    0.09
    ized
    0.09
    isme
    0.09
    iling
    0.09
    Activations Density 0.023%

    No Known Activations