© 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. 53966
    Prev
    Next
    INDEX
    Explanations

    To start/be/close/become

    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
    seealso
    -0.08
    bara
    -0.07
     Agenda
    -0.07
    çĽ¸ä¼´
    -0.07
    åĤ¥
    -0.07
     inexp
    -0.07
     p
    -0.07
    é¦
    -0.07
    åĵº
    -0.07
    ties
    -0.07
    POSITIVE LOGITS
    odore
    0.12
    æľĶ
    0.11
    alom
    0.10
    è¿Ļ个éĹ®é¢ĺ
    0.10
    VML
    0.10
    ownik
    0.10
    忽æĤł
    0.10
    å¾ĪæŃ£å¸¸
    0.10
    å¾Īé«ĺåħ´
    0.10
    è¿Ļæĺ¯ä¸Ģ个
    0.09
    Activations Density 0.331%

    No Known Activations