© 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. 52447
    Prev
    Next
    INDEX
    Explanations

    seemingly

    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
    кÑĢов
    -0.10
    otropic
    -0.09
    åħ¶å®ŀ
    -0.08
    ishing
    -0.08
    ixon
    -0.08
    Fabric
    -0.08
    åħ¶å¯¦
    -0.08
    åį³ä½¿
    -0.08
    åĪ°åºķ
    -0.08
     truth
    -0.08
    POSITIVE LOGITS
    ä¹Łä¸įæķ¢
    0.10
    éĺī
    0.10
    å·²ä¹ħ
    0.09
    åĩĨå¤ĩå·¥ä½ľ
    0.09
    lessly
    0.09
    hood
    0.09
    ä¸Ģæīĭ
    0.09
    å·²ä¹ħçļĦ
    0.09
    åľ¨éĩĮéĿ¢
    0.09
     аб
    0.09
    Activations Density 0.007%

    No Known Activations