© 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. 11062
    Prev
    Next
    INDEX
    Explanations

    speaking, talking, and words

    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
    è§Ĩéĩİ
    -0.20
     Goodman
    -0.20
    åıĤ
    -0.19
    bers
    -0.19
    avid
    -0.19
    æīĭæ³ķ
    -0.19
    æıĨ
    -0.18
    asString
    -0.18
    æĬ±
    -0.18
    çļĦçľ¼åħī
    -0.18
    POSITIVE LOGITS
     mouth
    0.24
    -speaking
    0.24
     spoken
    0.23
    åı£
    0.23
     boca
    0.22
    ä¸Ģåı¥
    0.22
    说åĩº
    0.22
    浯
    0.22
    mouth
    0.21
     words
    0.21
    Activations Density 0.110%

    No Known Activations