© 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. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 10014
    Prev
    Next
    INDEX
    Explanations

    .com

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-32b-saes/saes_Qwen_Qwen3-32B_batch_top_k/resid_post_layer_32
    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´ĭæ´ĭ
    -0.09
    untime
    -0.09
    èĤī
    -0.09
    åĪ©äºļ
    -0.08
    grav
    -0.08
    gom
    -0.08
    chet
    -0.08
    oute
    -0.08
    prit
    -0.08
    rieve
    -0.08
    POSITIVE LOGITS
    odore
    0.11
    æĸ¹å¼ıè¿Ľè¡Į
    0.10
    çĶŁåij½çļĦ
    0.09
    åįı
    0.09
    çİ®
    0.09
    åŃIJéĩĮ
    0.09
    羣çIJĨ
    0.09
    æł¸åĩĨ
    0.09
    etheless
    0.09
    å¹³åĿĩæ°´å¹³
    0.09
    Activations Density 0.051%

    No Known Activations