© 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. 57282
    Prev
    Next
    INDEX
    Explanations

    drift, drifts

    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
    hort
    -0.10
    amental
    -0.10
    ookie
    -0.10
    hots
    -0.10
     Shan
    -0.10
    ounter
    -0.09
    头èĦij
    -0.09
    aments
    -0.09
    ague
    -0.09
    hait
    -0.09
    POSITIVE LOGITS
    é£ĺ
    0.09
    plet
    0.09
    æ¼Ĥ
    0.09
    ships
    0.09
    wood
    0.09
    peare
    0.09
     fancy
    0.08
    sandbox
    0.08
    æ³Ĭ
    0.08
    buster
    0.08
    Activations Density 0.008%

    No Known Activations