© 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. 48774
    Prev
    Next
    INDEX
    Explanations

    evidence-based

    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
    å¤ıæĹ¥
    -0.10
    ç¾İæĦŁ
    -0.10
    ç§ĥ
    -0.09
    ÑĤвоÑĢ
    -0.09
    æī©
    -0.09
    俪
    -0.09
    ETS
    -0.09
    _VF
    -0.08
    天涯
    -0.08
    æĿ¾å¼Ľ
    -0.08
    POSITIVE LOGITS
    è¯ģæį®
    0.10
    èŃī
    0.09
    èĽ°
    0.09
    ãī
    0.09
    åĿļæĮģ以
    0.09
    å¤§èµĽ
    0.09
     evidence
    0.08
     Evidence
    0.08
    è¯ģ
    0.08
    uela
    0.08
    Activations Density 0.190%

    No Known Activations