© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 19677
    Prev
    Next
    INDEX
    Explanations

    high average statistics

    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
    ä¸Ģ件
    -0.10
    rone
    -0.10
    ÙĤاد
    -0.10
    iges
    -0.10
    åºŃ
    -0.10
    ä¸Ģéĥ¨
    -0.10
    .af
    -0.09
    ä»Ģä¹Īä¸ľè¥¿
    -0.09
    ä¸ĵéŨ
    -0.09
    åħļçļĦ建设
    -0.09
    POSITIVE LOGITS
    å¢ĥçķĮ
    0.09
    -holder
    0.08
    çŁŃä¿¡
    0.08
    iate
    0.08
    ç®ķ
    0.08
    iating
    0.08
    layan
    0.08
    accom
    0.08
    è¡Įäºĭ
    0.08
    客
    0.07
    Activations Density 0.002%

    No Known Activations