© 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. 14659
    Prev
    Next
    INDEX
    Explanations

    does not

    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
    çĤ¹äºĨ
    -0.21
    ly
    -0.19
    ä¸ĢæŦ
    -0.18
    移åĪ°
    -0.17
    åİĨç¨ĭ
    -0.17
    è°İè¨Ģ
    -0.17
    rox
    -0.17
    issen
    -0.16
    hole
    -0.16
    rio
    -0.16
    POSITIVE LOGITS
    ctype
    0.23
    ctors
    0.21
    å¾Ĺå¾Ī
    0.21
    vided
    0.20
    ctest
    0.19
    xygen
    0.19
    omorphic
    0.19
    ouble
    0.19
     not
    0.19
    /do
    0.18
    Activations Density 0.173%

    No Known Activations