© 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. 12743
    Prev
    Next
    INDEX
    Explanations

    effects of interventions and treatments

    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
    dsp
    -0.25
    hé
    -0.23
    rawid
    -0.22
    FilterWhere
    -0.21
    çIJĨ解åĴĮ
    -0.20
    许
    -0.20
    تÙĪج
    -0.20
    ä¿ĿèŃ·æĶ¿çŃĸ
    -0.20
    lew
    -0.20
    dÃŃ
    -0.20
    POSITIVE LOGITS
     study
    0.25
    çłĶ究
    0.22
     variables
    0.22
     hypothesis
    0.22
     controlled
    0.21
     иÑģÑģлед
    0.21
     settings
    0.21
     testing
    0.20
     effects
    0.20
     interventions
    0.20
    Activations Density 0.538%

    No Known Activations