© 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. 15215
    Prev
    Next
    INDEX
    Explanations

    correctly or right

    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
    æ·Ŀ
    -0.19
    cene
    -0.19
    REA
    -0.19
    ulis
    -0.18
    á»IJ
    -0.18
    ãĤŃãĥ£ãĥ³ãĥļ
    -0.18
    ily
    -0.18
    annes
    -0.18
    æīĭç»Ń
    -0.18
    åĪĽå§ĭ
    -0.18
    POSITIVE LOGITS
    è¿Ļåı¥è¯Ŀ
    0.28
    è¿Ļè¯Ŀ
    0.23
    ä¸Ģéģį
    0.22
    'gc
    0.20
     ton
    0.19
    å°¼æĸ¯
    0.19
    åı¥è¯Ŀ
    0.19
    :
    0.19
     sentence
    0.18
    äºĨä¸Ģéģį
    0.18
    Activations Density 0.025%

    No Known Activations