© 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. 52922
    Prev
    Next
    INDEX
    Explanations

    thinking you

    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
    æİł
    -0.10
    æĿ¥çļĦ
    -0.09
    æŀ¥
    -0.08
     Gerr
    -0.08
    enes
    -0.08
    çŃij
    -0.08
    Patterns
    -0.08
    rary
    -0.08
    夺
    -0.08
    gsub
    -0.08
    POSITIVE LOGITS
    /Set
    0.10
    æľīå¤ļå°ij
    0.10
    åħ¬çĶ¨
    0.09
     tán
    0.09
    æ²¾
    0.09
    è¿Ļè¾ĪåŃIJ
    0.08
    éĤ£ä½į
    0.08
    æĺ±
    0.08
     discrimin
    0.08
    æľīå¤ļ大
    0.08
    Activations Density 0.018%

    No Known Activations