© 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. 13802
    Prev
    Next
    INDEX
    Explanations

    character names followed by actions or descriptors

    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
    æݬ
    -0.20
    åĹ½
    -0.19
     itself
    -0.19
    ully
    -0.18
     puss
    -0.18
    noon
    -0.18
     nostalg
    -0.18
    åķIJ
    -0.18
    Capture
    -0.18
    éĢļä¿Ĺ
    -0.18
    POSITIVE LOGITS
     compartment
    0.21
    ghan
    0.20
    ocratic
    0.20
    主治
    0.19
    INET
    0.18
    .Roles
    0.18
    è¿Ļå¼ł
    0.18
    çļĦè§Ĵèī²
    0.18
    gué
    0.18
     |_|
    0.17
    Activations Density 0.022%

    No Known Activations