© 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. 3375
    Prev
    Next
    INDEX
    Explanations

    performance benchmark

    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
    æĥ´
    -0.10
    佬
    -0.10
    å°Ĩèĩªå·±çļĦ
    -0.09
    ä¸Ģ个好çļĦ
    -0.09
    IÃĵN
    -0.09
    ä»»ä½ķæŶåĢĻ
    -0.09
    æĬĬèĩªå·±
    -0.09
    геÑĢ
    -0.08
    éĻĤ
    -0.08
    èĮĥåĽ´åĨħ
    -0.08
    POSITIVE LOGITS
    æĢ§èĥ½
    0.14
     performance
    0.12
    åIJŀ
    0.11
     Performance
    0.10
    çĪĨåıij
    0.10
    Performance
    0.10
    _performance
    0.10
    .performance
    0.09
    performance
    0.09
     PERFORMANCE
    0.08
    Activations Density 0.128%

    No Known Activations