© 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. 49296
    Prev
    Next
    INDEX
    Explanations

    i was reading

    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
    ayas
    -0.10
    æĹĥ
    -0.09
    vang
    -0.09
    erson
    -0.09
     Bett
    -0.09
    缦
    -0.09
    宫
    -0.09
    .vs
    -0.08
    ä¾ł
    -0.08
    好åIJ§
    -0.08
    POSITIVE LOGITS
     wondering
    0.10
    æĥ³èµ·äºĨ
    0.09
    ç´«
    0.08
    æĢĿèĢĥ
    0.08
    è¿ĺ以为
    0.08
    åĽŀ顾
    0.08
    ä¸ŃåĽ½
    0.08
    ä¸įçĶ±
    0.08
    çľĭåĪ°äºĨ
    0.08
    æĥ³åĪ°äºĨ
    0.08
    Activations Density 0.096%

    No Known Activations