© 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. 49197
    Prev
    Next
    INDEX
    Explanations

    how it is

    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
    çī¹æľĹ
    -0.09
    åį°èĬ±
    -0.08
    ä¸ĭçıŃ
    -0.08
    磫
    -0.08
    举
    -0.08
     somehow
    -0.08
    пад
    -0.08
    èij©
    -0.08
    idon
    -0.07
    线索
    -0.07
    POSITIVE LOGITS
    ç®Ģ称
    0.10
    æŃ£å¸¸çļĦ
    0.10
    ä¸Ģ缴éĥ½æĺ¯
    0.09
    gó
    0.09
    __/
    0.09
    amespace
    0.08
    gan
    0.08
    aveled
    0.08
    bara
    0.08
    嵩
    0.08
    Activations Density 0.161%

    No Known Activations