© 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. 34166
    Prev
    Next
    INDEX
    Explanations

    Macross, Avast, Crytek

    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
    kaar
    -0.10
    çļĦä¸ĢåĪĩ
    -0.10
    åĮ£
    -0.10
    èĵĸ
    -0.09
    åĵĩ
    -0.09
    æ¯ı天éĥ½
    -0.09
    çīĮ
    -0.09
    èĤ¢
    -0.09
    HEET
    -0.09
    è³´
    -0.09
    POSITIVE LOGITS
    趣åij³
    0.11
    èĭ¦èĭ¦
    0.09
    çĭĹ
    0.08
    affer
    0.08
    å£ħ
    0.08
    æµİ
    0.08
     back
    0.08
    æĻ¯
    0.08
    èĭ±éĽĦ
    0.08
    éħ¸å¥¶
    0.08
    Activations Density 0.027%

    No Known Activations