© 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. 49979
    Prev
    Next
    INDEX
    Explanations

    sickening, sickened, sick children, sick persons

    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
    ãģ£ãģįãĤĬ
    -0.09
    cpy
    -0.09
    ç»Ĩå¾®
    -0.09
    кÑĢÑĭÑĤ
    -0.09
    HTTPHeader
    -0.09
    éĺĹ
    -0.09
     sân
    -0.09
    å°ıå¹ħ
    -0.09
    輩
    -0.09
    ize
    -0.09
    POSITIVE LOGITS
    ening
    0.20
    ened
    0.16
    ly
    0.13
    ler
    0.12
    iest
    0.12
    ometer
    0.10
    -minded
    0.10
     minded
    0.10
    opy
    0.09
    sville
    0.09
    Activations Density 0.017%

    No Known Activations