© 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. 49377
    Prev
    Next
    INDEX
    Explanations

    exclamations and informal reactions

    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
    ühl
    -0.08
    åĸ½
    -0.07
    ä¸įåIJĮç¨ĭ度
    -0.07
    ission
    -0.07
    ston
    -0.07
     Alb
    -0.07
     bile
    -0.07
    çħ¦
    -0.07
     IDM
    -0.07
     consequat
    -0.07
    POSITIVE LOGITS
    åĮ®
    0.13
    ÑĢеги
    0.10
    lan
    0.10
    LAN
    0.09
    å¥ł
    0.09
    ç¼
    0.09
    å²·
    0.09
    åħ°
    0.09
     lanc
    0.09
    ONTAL
    0.09
    Activations Density 0.167%

    No Known Activations