© 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. 22061
    Prev
    Next
    INDEX
    Explanations

    passion for

    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
    çļĦéģĵçIJĨ
    -0.10
    rust
    -0.09
    though
    -0.09
    éģĵçIJĨ
    -0.09
    rem
    -0.09
    èĢ³æľµ
    -0.09
     naked
    -0.08
    رة
    -0.08
    roe
    -0.08
    .gradle
    -0.08
    POSITIVE LOGITS
    æĬĺèħ¾
    0.10
    vard
    0.10
    äºĭçī©
    0.09
    åģļçļĦäºĭæĥħ
    0.09
    0.09
    ãĤĴä¸İ
    0.09
    /of
    0.09
    detail
    0.09
    ÙĪاÙĨ
    0.08
    ಸ
    0.08
    Activations Density 0.028%

    No Known Activations