© 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. 15318
    Prev
    Next
    INDEX
    Explanations

    numbers followed by 'is' or 'runs' or 'has'

    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
     Pav
    -0.23
    balls
    -0.21
    myp
    -0.20
     morally
    -0.18
    ulous
    -0.18
    ent
    -0.18
    é
    -0.18
    etus
    -0.18
    lick
    -0.17
     rib
    -0.17
    POSITIVE LOGITS
    亲åıĭ
    0.22
    æĺ¯éĿŀ
    0.21
    çī¢åĽºæłijç«ĭ
    0.20
    Accessible
    0.19
    ç»Ŀ对
    0.19
    armor
    0.19
    åĽ½éĻħåľ¨çº¿
    0.19
    availability
    0.18
    ENDOR
    0.18
    /tos
    0.18
    Activations Density 0.075%

    No Known Activations