© 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. 13299
    Prev
    Next
    INDEX
    Explanations

    electricity, electrostatic, electronic

    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
    é¥Ńèıľ
    -0.20
    èĭ´
    -0.20
     rag
    -0.18
    家åħ¬åı¸
    -0.18
    name
    -0.18
    yard
    -0.18
    奸
    -0.17
    å£ķ
    -0.17
    kind
    -0.17
    hawk
    -0.17
    POSITIVE LOGITS
    LAS
    0.20
     Gong
    0.19
    æī¶
    0.19
    elves
    0.19
    åı©
    0.18
    /gin
    0.18
    æ¸ħåĩī
    0.18
    DED
    0.18
    efined
    0.18
    æĬļ
    0.17
    Activations Density 0.001%

    No Known Activations