© 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. 47817
    Prev
    Next
    INDEX
    Explanations

    later or smaller

    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
    åı¦
    -0.09
    -0.09
    æł¹æľ¬
    -0.09
    æĻŁ
    -0.09
    çĺĹ
    -0.09
     Wizards
    -0.08
    bard
    -0.08
     demographic
    -0.08
    æ£ī
    -0.08
    screen
    -0.08
    POSITIVE LOGITS
    åıªè¦ģæĺ¯
    0.09
    .ptr
    0.09
    æ¸Ĭ
    0.09
    EGA
    0.09
    åŃ«
    0.08
    .dup
    0.08
    为äºĨ让
    0.08
     Gad
    0.08
    åĬŁåĬĽ
    0.08
    身形
    0.08
    Activations Density 0.027%

    No Known Activations