© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 22377
    Prev
    Next
    INDEX
    Explanations

    that edge, little, most, one, luxurious

    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
    ç«łç¨ĭ
    -0.11
    adden
    -0.10
    åįľ
    -0.10
    æ¨Ĭ
    -0.09
    ful
    -0.09
    å¿ĥæĥ³
    -0.09
    ç»Ħå§Ķä¼ļ
    -0.09
    lä
    -0.09
    çļĦåľ°
    -0.08
    indsay
    -0.08
    POSITIVE LOGITS
    éĤ£ç§į
    0.16
    éĤ£ä»½
    0.11
    çĢĽ
    0.09
    çļĦéĤ£ç§į
    0.09
    æµĵæµĵ
    0.09
    ç´¯
    0.09
    itu
    0.09
    称ä¹ĭ
    0.09
    æµĵæµĵçļĦ
    0.09
    ä½ĵä¼ļåĪ°
    0.09
    Activations Density 0.049%

    No Known Activations