© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 51510
    Prev
    Next
    INDEX
    Explanations

    disturbing or Bron

    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
    agner
    -0.10
    asma
    -0.09
     univers
    -0.09
    onder
    -0.09
    تب
    -0.09
    赶赴
    -0.09
    adium
    -0.08
    akit
    -0.08
    andbox
    -0.08
    roz
    -0.08
    POSITIVE LOGITS
    ran
    0.10
    teenth
    0.10
    hound
    0.10
    -era
    0.09
    èIJĺ
    0.09
    mousedown
    0.09
    常çĶ¨çļĦ
    0.09
    nü
    0.09
    æĻ°
    0.09
    ly
    0.09
    Activations Density 0.004%

    No Known Activations