© 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. 12643
    Prev
    Next
    INDEX
    Explanations

    version or variant

    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
     analogy
    -0.19
    arsers
    -0.19
    ctal
    -0.19
    yas
    -0.19
    .ACTION
    -0.19
    ç½ijé¦ĸ页
    -0.18
    imat
    -0.17
    Eval
    -0.17
    _children
    -0.17
     hail
    -0.17
    POSITIVE LOGITS
    çīĪæľ¬
    0.55
     version
    0.47
    çīĪ
    0.45
     versions
    0.44
     variant
    0.41
     variants
    0.40
     VERSION
    0.38
     edition
    0.37
    -version
    0.36
     Variant
    0.36
    Activations Density 0.384%

    No Known Activations