© 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. 39072
    Prev
    Next
    INDEX
    Explanations

    'an' or 'en' followed by common endings

    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
    åĵĩ
    -0.16
    teen
    -0.15
    een
    -0.14
    gun
    -0.13
    ters
    -0.13
    down
    -0.13
    ta
    -0.13
    den
    -0.12
    gen
    -0.12
    te
    -0.12
    POSITIVE LOGITS
    åijIJ
    0.22
    ning
    0.21
    etwork
    0.19
    avigator
    0.17
    alysis
    0.17
    ned
    0.17
    ormal
    0.16
    exus
    0.16
     dolayı
    0.16
    egative
    0.16
    Activations Density 2.215%

    No Known Activations