© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 38098
    Prev
    Next
    INDEX
    Explanations

    "my arm"

    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
    ÑĤÑı
    -0.09
    erton
    -0.09
    arro
    -0.09
    гиб
    -0.09
    èľī
    -0.09
    èĬĻèĵī
    -0.09
     anonymously
    -0.09
    ë¸Ķ
    -0.09
    rys
    -0.09
    arov
    -0.08
    POSITIVE LOGITS
    mond
    0.11
    shaw
    0.10
    atorio
    0.10
    nings
    0.09
    abin
    0.09
     rÃłng
    0.09
    eous
    0.09
    castle
    0.09
    leigh
    0.08
    oise
    0.08
    Activations Density 0.737%

    No Known Activations