© 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. 11873
    Prev
    Next
    INDEX
    Explanations

    seated

    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
    æ´¾
    -0.11
    nels
    -0.10
     Hud
    -0.10
    çĥĪ
    -0.09
    åłµ
    -0.09
    ERIC
    -0.09
     Rooney
    -0.09
    çĽ¸æľº
    -0.09
    obuf
    -0.09
    éĽī
    -0.09
    POSITIVE LOGITS
    æĸ¹å¼ı
    0.10
    åĭĭ
    0.09
    ews
    0.08
    龸çİĭ
    0.08
    å½¢å¼ı
    0.08
    ull
    0.08
    为æŃ¢
    0.08
    4
    0.07
    æĥħ绪
    0.07
    eker
    0.07
    Activations Density 0.006%

    No Known Activations