© 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. 34610
    Prev
    Next
    INDEX
    Explanations

    assuredness and confidence

    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
    æĭĶ
    -0.10
    è¹¼
    -0.08
    èĥŀ
    -0.08
    éĴī
    -0.08
    æıĴ
    -0.07
    å¼±
    -0.07
    enberg
    -0.07
    åŃĢ
    -0.07
     maybe
    -0.07
    emies
    -0.07
    POSITIVE LOGITS
    æĶ¾å¿ĥ
    0.22
    ä¸įçĶ¨æĭħå¿ĥ
    0.16
    å®īå¿ĥ
    0.16
     assured
    0.15
    ä¿Ŀè¯ģ
    0.15
    åĿļä¿¡
    0.15
     assurance
    0.14
    ä¿ĿèŃī
    0.13
     ÑĥвеÑĢен
    0.13
     confident
    0.12
    Activations Density 0.053%

    No Known Activations