© 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. 33418
    Prev
    Next
    INDEX
    Explanations

    key drivers and factors

    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
    æĹ¶å¸¸
    -0.10
    æıIJåIJį
    -0.10
    вÑıз
    -0.10
    áŀĦ
    -0.09
    [ii
    -0.09
    æĤ²è§Ĥ
    -0.09
    é¹ij
    -0.09
    luck
    -0.09
    è¿ĩçļĦ
    -0.09
    å®ĿèĹı
    -0.09
    POSITIVE LOGITS
    åĽłç´ł
    0.16
     factor
    0.14
    åĽłåŃIJ
    0.13
     factors
    0.13
    factor
    0.12
    åİŁåĽłä¹ĭä¸Ģ
    0.12
    驱åĬ¨
    0.12
     driving
    0.12
    主è¦ģåİŁåĽł
    0.12
    çļĦåĽłç´ł
    0.11
    Activations Density 0.077%

    No Known Activations