© 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. 7560
    Prev
    Next
    INDEX
    Explanations

    then copied, then stored

    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
    ç«łèĬĤ
    -0.10
     inexp
    -0.09
    äºĴ
    -0.09
    ç¾ŀ
    -0.09
    纽
    -0.08
    åĴĮå¹³
    -0.08
    /dist
    -0.08
    Flo
    -0.08
     easily
    -0.07
    误
    -0.07
    POSITIVE LOGITS
    æľīè¶³å¤ŁçļĦ
    0.08
    è¡Ģèħ¥
    0.08
    Its
    0.08
     Äiju
    0.08
     udziaÅĤ
    0.08
    ìĤŃ
    0.08
    ä¸ĢèĤ¡
    0.08
    éŁ³ä¹IJä¼ļ
    0.07
     processing
    0.07
     traversal
    0.07
    Activations Density 1.029%

    No Known Activations