© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 36625
    Prev
    Next
    INDEX
    Explanations

    presumably

    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
    çīĪæĿĥå½Ĵ
    -0.10
    avo
    -0.10
    çī¹æ´¾
    -0.09
    æĽ²æĬĺ
    -0.09
     bart
    -0.09
    ric
    -0.08
    æŦ
    -0.08
    ration
    -0.08
    åģļåĪ°
    -0.08
    åģļå¾Ĺ
    -0.08
    POSITIVE LOGITS
    æį®è¯´
    0.23
     presumably
    0.23
     reportedly
    0.23
     probably
    0.23
    probably
    0.21
    æį®
    0.20
    ä¼¼ä¹İ
    0.19
     apparently
    0.19
    ä¼¼ä¹İæĺ¯
    0.18
    æĥ³å¿ħ
    0.18
    Activations Density 0.821%

    No Known Activations