© 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. 6385
    Prev
    Next
    INDEX
    Explanations

    variant spelling, variations, versions

    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
    æŃ¢æįŁ
    -0.10
    åĬ¾
    -0.10
    heiro
    -0.09
    笼
    -0.09
    shed
    -0.09
    ä¸ĢåĽ¢
    -0.09
    åħĪéĶĭ
    -0.09
    罡
    -0.09
    åĿ¯
    -0.08
     sö
    -0.08
    POSITIVE LOGITS
     variants
    0.18
     variant
    0.18
     ваÑĢи
    0.16
     variations
    0.16
    variant
    0.15
     Vari
    0.14
     variation
    0.14
    çīĪæľ¬
    0.14
    Variant
    0.13
     versions
    0.13
    Activations Density 0.096%

    No Known Activations