© 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. 17631
    Prev
    Next
    INDEX
    Explanations

    punctuations

    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
    erval
    -0.10
    sWith
    -0.10
    empre
    -0.10
    ä¹ĭ家
    -0.09
    EP
    -0.08
    eenth
    -0.08
    ä»ĬçĶŁ
    -0.08
    ep
    -0.08
     Fran
    -0.08
    äºĨä¸įå°ij
    -0.08
    POSITIVE LOGITS
    cpy
    0.11
    iscard
    0.09
    anford
    0.09
     ÑĢазв
    0.09
    avax
    0.09
    ufact
    0.09
    ystate
    0.09
    óg
    0.09
    ãģ£ãģįãĤĬ
    0.09
    .addItem
    0.09
    Activations Density 0.114%

    No Known Activations