© 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. 53503
    Prev
    Next
    INDEX
    Explanations

    nearly as / quite as / half bad / remotely true

    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
     mere
    -0.10
    estival
    -0.09
    åºı
    -0.09
    æĻ°
    -0.08
    etro
    -0.08
    åĮĻ
    -0.08
    éĻ£
    -0.08
    æīĢåľ¨çļĦ
    -0.08
    tributes
    -0.08
    arme
    -0.08
    POSITIVE LOGITS
    ä¹Łä¸įåı¯èĥ½
    0.11
    ä¸įåıĬ
    0.10
    å¦ĤæľŁ
    0.10
    lette
    0.09
    è¿ĺä¸įå¦Ĥ
    0.09
    å¤ł
    0.09
     nor
    0.08
    åīįèĢħ
    0.08
    åı¯è§Ĥ
    0.08
     ÑĤакого
    0.08
    Activations Density 0.003%

    No Known Activations