© 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. 12728
    Prev
    Next
    INDEX
    Explanations

    sentence endings followed by introductions

    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
    likely
    -0.19
    åľ¨æĪijçļĦ
    -0.18
    ot
    -0.18
    els
    -0.18
    ie
    -0.18
    tie
    -0.18
    åľ¨åĮĹ京
    -0.18
    ke
    -0.18
    ry
    -0.18
    wood
    -0.18
    POSITIVE LOGITS
    好å¥ĩå¿ĥ
    0.20
    .same
    0.19
    æģļ
    0.19
    âĪİ
    0.19
    ennie
    0.18
     taxp
    0.18
    ç´«èī²
    0.18
    arth
    0.18
    æŃ£å®Ĺ
    0.18
    åŃ½
    0.18
    Activations Density 0.482%

    No Known Activations