© 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. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 26440
    Prev
    Next
    INDEX
    Explanations

    `into`, `less`, `more`, `instead`, `towards`

    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
    ä¼ijéĹ²
    -0.10
    æĹ»
    -0.09
     báºŃt
    -0.09
    åıĪæĺ¯
    -0.09
    éªŀ
    -0.09
    ARK
    -0.09
    æħµ
    -0.08
    amac
    -0.08
    ูà¹Ī
    -0.08
    ä¼¼ä¹İæĺ¯
    -0.08
    POSITIVE LOGITS
    ä¸İä¸ŃåĽ½
    0.10
    ãģŁãĤĬ
    0.09
    ATUS
    0.09
    æĬĬèĩªå·±çļĦ
    0.08
    æľīèĩªå·±çļĦ
    0.08
     ÑģобÑģÑĤвенно
    0.08
    ville
    0.08
    æĬĬèĩªå·±
    0.08
    aul
    0.08
    éĺ²èħIJ
    0.08
    Activations Density 0.007%

    No Known Activations