© 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. 27405
    Prev
    Next
    INDEX
    Explanations

    atmosphere, air, dioxide

    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
    è¾Ħ
    -0.11
     rect
    -0.09
    æ½ľ
    -0.09
    uzzy
    -0.09
    åĨ·èĹı
    -0.09
    çĸijä¼¼
    -0.09
     Goldberg
    -0.09
     NotImplemented
    -0.09
    é«ĺæ°´å¹³
    -0.09
    æŃ£éĿ¢
    -0.09
    POSITIVE LOGITS
    大æ°Ķ
    0.16
    空æ°Ķ
    0.14
     atmosphere
    0.11
     air
    0.11
    æ°§æ°Ķ
    0.11
    èĶļ
    0.10
    æ°Ķä½ĵ
    0.10
     Atmos
    0.10
    äºĮæ°§åĮĸ碳
    0.10
    éĿĽ
    0.09
    Activations Density 0.047%

    No Known Activations