© 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. 13066
    Prev
    Next
    INDEX
    Explanations

    old story / new story

    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
    åĨħçļĦ
    -0.19
    å½ĵä»Ĭ
    -0.18
    kle
    -0.16
     dioxide
    -0.16
    æīĢåľ¨çļĦ
    -0.15
    acks
    -0.15
    odb
    -0.15
    è¾±
    -0.15
    adian
    -0.14
    eya
    -0.14
    POSITIVE LOGITS
    -fashioned
    0.48
     fashioned
    0.33
    most
    0.26
    /new
    0.23
    è¯Ŀ说
    0.23
    liness
    0.22
    enburg
    0.21
    -school
    0.21
    ness
    0.21
    à¹Ĩ
    0.20
    Activations Density 0.026%

    No Known Activations