© 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. 7657
    Prev
    Next
    INDEX
    Explanations

    are ... really

    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
    è¿Ľé¡¹
    -0.09
    .EX
    -0.08
    çļĦåīįæıIJ
    -0.08
    imm
    -0.08
    elen
    -0.08
    æ§İ
    -0.08
     either
    -0.08
     Hogan
    -0.08
    anes
    -0.08
    PBS
    -0.08
    POSITIVE LOGITS
    羣çļĦ
    0.11
    羣çļĦæĺ¯
    0.11
    çī¹åĪ«æĺ¯
    0.10
    éļ¾éģĵ
    0.10
    opr
    0.10
     Lig
    0.10
    å¿ħè¦ģ
    0.09
     really
    0.09
     unthinkable
    0.09
    nym
    0.09
    Activations Density 0.085%

    No Known Activations