© 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. 39791
    Prev
    Next
    INDEX
    Explanations

    short-term effects

    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
    丰满
    -0.10
    裾
    -0.09
    å½ĵåĪĿ
    -0.09
    罩
    -0.09
    çĽĤ
    -0.09
    åīįåįģ
    -0.08
    æł·çļĦ
    -0.08
    åĪļæīį
    -0.08
    æľ¬äºĭ
    -0.08
    ım
    -0.08
    POSITIVE LOGITS
    é©·
    0.11
    (short
    0.10
    SHORT
    0.10
    åºĶä»ĺ
    0.09
    -focused
    0.09
    ÙĦØŃ
    0.09
     thôi
    0.09
    -short
    0.09
    aneously
    0.09
    ismo
    0.09
    Activations Density 0.030%

    No Known Activations