© 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. Qwen3-14B
    5. 20-RESID-BATCHTOPK-65K__L0-80
    7. 12197
    Prev
    Next
    INDEX
    Explanations

    prep tea, prep races, prep watches, prep boarding

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-14b-saes/saes_Qwen_Qwen3-14B_batch_top_k/resid_post_layer_20
    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
    hood
    -0.19
    heits
    -0.18
    ht
    -0.18
    implicitly
    -0.17
    cation
    -0.17
    éĥĬ
    -0.17
    stag
    -0.17
    atics
    -0.17
    Mui
    -0.17
    ality
    -0.16
    POSITIVE LOGITS
    ressed
    0.17
    ched
    0.17
    nas
    0.17
     Mahar
    0.16
    na
    0.16
    и
    0.15
    erna
    0.15
    åīĮ
    0.15
    ä¾Ŀæ³ķ追究
    0.15
    edly
    0.15
    Activations Density 0.006%

    No Known Activations