© 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. 58648
    Prev
    Next
    INDEX
    Explanations

    verb plurals

    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
    åIJĦåĽ½
    -0.09
    annon
    -0.08
    alsa
    -0.08
    ä¼Ĺ
    -0.08
     Rover
    -0.08
     Kemp
    -0.08
    á»ĥu
    -0.08
    亥
    -0.08
    apt
    -0.08
    æĬ¤çħ§
    -0.08
    POSITIVE LOGITS
    å¼¢
    0.09
    æĵį
    0.09
    飬
    0.09
    åĭº
    0.09
    олов
    0.09
    ancy
    0.09
    cery
    0.08
    èı°
    0.08
    ault
    0.08
    èµĭ
    0.08
    Activations Density 0.043%

    No Known Activations