© 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. 48943
    Prev
    Next
    INDEX
    Explanations

    copyright or licensed text

    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
    çªĴ
    -0.09
    è¹²
    -0.09
    éű
    -0.09
    ä¹ĥ
    -0.08
    Č
    -0.08
     же
    -0.08
    éĻŁ
    -0.08
     ë¬
    -0.08
    æŀIJ
    -0.08
    åijĬè¯ī她
    -0.08
    POSITIVE LOGITS
    ardon
    0.09
    imid
    0.09
    åħ
    0.09
    åı£å¤´
    0.09
     dr
    0.09
     Marino
    0.09
    glas
    0.08
     Unity
    0.08
    æĬ¥åĪĬ
    0.08
    atis
    0.08
    Activations Density 0.015%

    No Known Activations