© 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. 10850
    Prev
    Next
    INDEX
    Explanations

    HTML public declarations

    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
    Hierarchy
    -0.19
    heim
    -0.18
    Hora
    -0.18
    ´
    -0.18
    hole
    -0.17
     horm
    -0.17
     hypothetical
    -0.17
    éļIJæĤ£
    -0.17
    hist
    -0.17
    houses
    -0.16
    POSITIVE LOGITS
    izontal
    0.30
    essian
    0.25
    ancock
    0.24
    .itemView
    0.24
    izontally
    0.23
    archical
    0.20
    itage
    0.20
    icap
    0.19
    ometown
    0.19
    forth
    0.19
    Activations Density 2.709%

    No Known Activations