© 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. 3492
    Prev
    Next
    INDEX
    Explanations

    meals, ticket, passengers, aircraft, damaged

    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
    orus
    -0.09
    zd
    -0.09
    ick
    -0.09
    çĿĢä»ĸ
    -0.08
    anni
    -0.08
     hire
    -0.08
     texas
    -0.08
    æīĵæĸŃ
    -0.08
    loh
    -0.08
    ä¸ĵä¸ļæĬĢæľ¯
    -0.08
    POSITIVE LOGITS
    å·§åIJĪ
    0.11
     вÑĭÑĢажен
    0.10
    缼
    0.10
    æĭĵ
    0.10
     Davidson
    0.09
    æĮģæľī
    0.09
    otts
    0.09
     miá»ħn
    0.09
    oto
    0.09
    åħ¨çIJĥ
    0.09
    Activations Density 0.074%

    No Known Activations