© 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. 57624
    Prev
    Next
    INDEX
    Explanations

    in the manner of

    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
    à§į
    -0.10
     зÑĢениÑı
    -0.10
    gers
    -0.10
    andre
    -0.09
    adm
    -0.09
    eler
    -0.09
    çĺ°
    -0.09
    æĢĢæĬ±
    -0.08
    gend
    -0.08
    OrCreate
    -0.08
    POSITIVE LOGITS
    ities
    0.12
    isms
    0.11
    lijke
    0.10
    eturn
    0.09
    lick
    0.09
    osa
    0.09
    iggs
    0.09
    loom
    0.09
    uck
    0.09
    theless
    0.09
    Activations Density 0.015%

    No Known Activations