© 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. 5004
    Prev
    Next
    INDEX
    Explanations

    suspend and suspension

    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
    èµ°åĪ°
    -0.09
    åĪĽå§ĭ
    -0.09
    hiba
    -0.09
    hardt
    -0.09
    idend
    -0.09
    íĮħ
    -0.09
    anos
    -0.09
    çłº
    -0.08
    andom
    -0.08
    ä½łåºĶ该
    -0.08
    POSITIVE LOGITS
     suspended
    0.13
     susp
    0.13
    æĤ¬
    0.12
     suspension
    0.11
     Susp
    0.11
     suspend
    0.11
    Susp
    0.11
    åIJĬ
    0.10
     Suspension
    0.10
    suspend
    0.10
    Activations Density 0.025%

    No Known Activations