© 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. 33424
    Prev
    Next
    INDEX
    Explanations

    Brock or Brooke

    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
    esto
    -0.13
    ities
    -0.12
    atics
    -0.10
    eger
    -0.09
     stret
    -0.09
    teen
    -0.09
    teenth
    -0.09
    meric
    -0.09
    ality
    -0.09
    åŃIJéĩĮ
    -0.09
    POSITIVE LOGITS
    éĻIJ度
    0.12
    shaw
    0.10
    ãĥ¯ãĤ¤
    0.09
    mediate
    0.09
    æľīæĿ¡ä»¶
    0.09
    hausen
    0.08
    ä¸Ģä¼ļ
    0.08
    è¦ģæĺ¯
    0.08
     Papua
    0.08
    ioni
    0.08
    Activations Density 0.005%

    No Known Activations