© 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. 12598
    Prev
    Next
    INDEX
    Explanations

    nothing, something, everything, things

    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
    ettel
    -0.20
    çķĻ
    -0.20
    ä»İä¸Ń
    -0.19
    笼罩
    -0.19
    çķĻæĦı
    -0.19
    ä¸ī人
    -0.18
    ç»IJ
    -0.18
    ниÑĨа
    -0.18
    æŶåĪ»
    -0.18
    ÙģÙĬ
    -0.18
    POSITIVE LOGITS
     else
    0.23
    /from
    0.21
     everything
    0.21
    ä»Ģä¹Īåij¢
    0.21
     things
    0.21
     nothing
    0.20
    things
    0.19
    ä»Ģä¹Īéĥ½ä¸į
    0.19
    nothing
    0.19
     something
    0.19
    Activations Density 0.045%

    No Known Activations