© 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. 7526
    Prev
    Next
    INDEX
    Explanations

    speaking, reporting, confessing

    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
    6
    -0.09
    7
    -0.09
    0
    -0.09
    ä¸ĵåijĺ
    -0.09
    3
    -0.09
    4
    -0.08
    5
    -0.08
    廪
    -0.08
    快讯
    -0.08
    8
    -0.08
    POSITIVE LOGITS
    说èĩªå·±
    0.14
     admitting
    0.10
    åĿ¦è¨Ģ
    0.09
     admission
    0.09
    åĽŀçŃĶ
    0.09
    è°ĪåıĬ
    0.09
    æĬĬèĩªå·±
    0.09
    çŁ¥éģĵèĩªå·±
    0.09
    éĢıéľ²
    0.09
    reve
    0.09
    Activations Density 0.069%

    No Known Activations