© 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. 14304
    Prev
    Next
    INDEX
    Explanations

    won awards or championships

    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
    /Instruction
    -0.18
    áb
    -0.18
    ornado
    -0.17
    onne
    -0.15
    inia
    -0.15
    aus
    -0.15
     Warfare
    -0.15
    amento
    -0.15
    ople
    -0.15
    iline
    -0.15
    POSITIVE LOGITS
    ä¸įäºĨ
    0.25
    nable
    0.21
    ners
    0.21
    å¾Ĺèµ·
    0.21
    èĢħçļĦ
    0.20
    nen
    0.19
    äºĨä¸Ģ
    0.19
    ows
    0.18
    fully
    0.18
    ning
    0.18
    Activations Density 0.048%

    No Known Activations