© 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. 12148
    Prev
    Next
    INDEX
    Explanations

    stealing, theft, burglary, or phone related crimes

    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
    çľĭ书
    -0.20
     Bott
    -0.20
     reint
    -0.19
    ture
    -0.19
    çIJī
    -0.19
    èł¹
    -0.18
    ä¼´éļıçĿĢ
    -0.18
     rele
    -0.18
    usher
    -0.18
    Ùħرا
    -0.18
    POSITIVE LOGITS
    使çĶ¨æĿĥ
    0.26
     stolen
    0.23
    ä»ĸ人
    0.22
    夺åıĸ
    0.21
     Synopsis
    0.21
    æĶ¹è£ħ
    0.20
    红æĹĹ
    0.20
     hostname
    0.20
    åĪ©çĶ¨çİĩ
    0.20
    æīĢæľīæĿĥ
    0.20
    Activations Density 0.013%

    No Known Activations