© 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. 2942
    Prev
    Next
    INDEX
    Explanations

    way to or way of

    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
    å°ıå°ı
    -0.10
    ç¼Ĺ
    -0.10
    饱
    -0.10
    NCY
    -0.09
    è¦ģç´§
    -0.09
    DDL
    -0.09
    åĴĦ
    -0.09
    å¸
    -0.09
    çĥģ
    -0.09
    trag
    -0.08
    POSITIVE LOGITS
    éĢĶå¾Ħ
    0.13
    éĢĶ
    0.10
    å¹³åĿ¦
    0.09
    æĢİä¹Īæł·
    0.09
    way
    0.09
    èĤī
    0.09
    éĢļå¾Ģ
    0.09
     ways
    0.09
     Ways
    0.08
     пÑĥÑĤи
    0.08
    Activations Density 0.066%

    No Known Activations