© 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. 21884
    Prev
    Next
    INDEX
    Explanations

    gratification, gratings, gratifying

    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
    涯
    -0.11
    /or
    -0.09
    -0.09
    åĴĮå¹³
    -0.09
    ×ķת
    -0.08
    ever
    -0.08
    маг
    -0.08
    ering
    -0.08
    isle
    -0.08
    òa
    -0.08
    POSITIVE LOGITS
    fully
    0.10
    سÙħ
    0.10
    ÑĨен
    0.09
    eous
    0.09
    æĹĸ
    0.09
    icia
    0.09
    空åīį
    0.09
    ively
    0.09
    illions
    0.09
    夯
    0.08
    Activations Density 0.004%

    No Known Activations