© 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. 31287
    Prev
    Next
    INDEX
    Explanations

    sch names and beginnings

    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
    eous
    -0.10
    cean
    -0.10
     лиÑĪÑĮ
    -0.10
    orary
    -0.09
    åħħæ²Ľ
    -0.09
    ted
    -0.09
    eenth
    -0.09
    idad
    -0.09
    ãģĬãĤĬãģ¾ãģĻ
    -0.09
    ssue
    -0.09
    POSITIVE LOGITS
    neider
    0.12
    izophren
    0.12
    ultz
    0.11
    ×ķת
    0.11
    itecture
    0.11
    illing
    0.11
    pedia
    0.11
    midt
    0.11
    intosh
    0.10
    enery
    0.10
    Activations Density 0.053%

    No Known Activations