© 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. 20030
    Prev
    Next
    INDEX
    Explanations

    promotional group, label, conference

    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
    æĹ¶æľº
    -0.09
    æģ¯
    -0.09
    èĬ±èįī
    -0.08
    æ¹¾åĮº
    -0.08
    OwnProperty
    -0.08
    åĩºçİ°åľ¨
    -0.08
    UME
    -0.08
    ufe
    -0.08
    åĪĨå±Ģ
    -0.08
    omanip
    -0.08
    POSITIVE LOGITS
    wide
    0.13
    -wide
    0.13
    -specific
    0.12
    pecific
    0.12
    ALAR
    0.11
    ë³Ħ
    0.10
    åĴĮåľ°åĮº
    0.10
    åĨħçļĦ
    0.10
     wide
    0.10
    -level
    0.09
    Activations Density 0.337%

    No Known Activations