© 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. Qwen3-14B
    5. 20-RESID-BATCHTOPK-65K__L0-80
    7. 15607
    Prev
    Next
    INDEX
    Explanations

    sister or sisters

    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
    /she
    -0.25
    ìį¨
    -0.22
    ¤
    -0.18
    stered
    -0.18
    ä¸ĩä¸Ģ
    -0.17
    æѤäºĭ
    -0.15
    æĭŃ
    -0.15
    shield
    -0.15
    URNS
    -0.15
    annon
    -0.15
    POSITIVE LOGITS
    hood
    0.27
    们çļĦ
    0.23
    å¼Ł
    0.21
    们
    0.20
    åĢij
    0.20
    ÙĪات
    0.17
    chaft
    0.17
    hip
    0.17
    ency
    0.16
    gram
    0.16
    Activations Density 0.006%

    No Known Activations