© 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. 2520
    Prev
    Next
    INDEX
    Explanations

    CT followed by specific words

    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
     sắc
    -0.10
    åĭĥåĭĥ
    -0.10
    åŀ²
    -0.10
    éĺĹ
    -0.09
    isto
    -0.09
     اÙĦشرÙĬÙģ
    -0.09
    abin
    -0.09
    iro
    -0.09
    лад
    -0.08
    /callback
    -0.08
    POSITIVE LOGITS
    ively
    0.12
    IVITY
    0.12
    iveness
    0.11
    ivity
    0.11
    ually
    0.10
    ivities
    0.10
    raction
    0.10
    uring
    0.09
    dom
    0.09
    ors
    0.09
    Activations Density 0.085%

    No Known Activations