© 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. 5384
    Prev
    Next
    INDEX
    Explanations

    outskirts 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
    essen
    -0.09
    fully
    -0.09
    为é¦ĸçļĦ
    -0.09
    asis
    -0.08
    为空
    -0.08
    æĶ¾æĺł
    -0.08
    éĿ¢åīį
    -0.08
    å¤©åľ°
    -0.08
    æĮ£
    -0.08
    éŨåīį
    -0.08
    POSITIVE LOGITS
    å¤ĸåĽ´
    0.14
    è¾¹ç¼ĺ
    0.14
    å¾Īä¹ħ
    0.11
    -edge
    0.11
    åľį
    0.11
    è¾¹ä¸Ĭ
    0.10
    .edges
    0.10
    åľĪ
    0.10
    ä¸ĢåľĪ
    0.10
    è¾¹
    0.10
    Activations Density 0.053%

    No Known Activations