© 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. 16538
    Prev
    Next
    INDEX
    Explanations

    direction of wind

    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
    occo
    -0.20
    æĥĬèī³
    -0.19
    è¿Ļ两年
    -0.19
    çİ°å·²
    -0.18
    ñe
    -0.18
     ramp
    -0.18
    éĥĵ
    -0.18
    inspace
    -0.18
     gre
    -0.18
    pcs
    -0.18
    POSITIVE LOGITS
    æµģåIJij
    0.42
    æĸ¹åIJij
    0.40
    çļĦæĸ¹åIJij
    0.38
    direction
    0.38
     Direction
    0.35
     direction
    0.35
    -direction
    0.33
    æµģåħ¥
    0.33
    Direction
    0.32
     directions
    0.30
    Activations Density 0.086%

    No Known Activations