© 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. 2668
    Prev
    Next
    INDEX
    Explanations

    fair trial

    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
    ä¹Łä¸įåı¯èĥ½
    -0.09
     ngá»Ļ
    -0.09
    usi
    -0.09
    å®ıè§Ĥ
    -0.09
    Symbols
    -0.08
    符åı·
    -0.08
    æī§æĶ¿
    -0.08
     Symbols
    -0.08
    é£ĺ
    -0.08
    symbols
    -0.08
    POSITIVE LOGITS
    åĽ´è§Ĥ
    0.12
    èĢ»
    0.11
    å¥ĩ
    0.10
    好å¥ĩ
    0.10
     Pit
    0.10
    好å¥ĩå¿ĥ
    0.09
    pit
    0.09
    arness
    0.09
     pit
    0.09
    èĥļèĥİ
    0.09
    Activations Density 0.079%

    No Known Activations