© 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. 23647
    Prev
    Next
    INDEX
    Explanations

    seeing or observing

    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
    è¯ķè¯ķ
    -0.09
    поз
    -0.08
    åįı
    -0.08
    天çĦ¶
    -0.08
    æ·±åħ¥äººå¿ĥ
    -0.08
    GLE
    -0.08
    eries
    -0.08
     Majority
    -0.08
    GI
    -0.08
    妥åįı
    -0.08
    POSITIVE LOGITS
    çݰ象
    0.14
    çļĦçݰ象
    0.14
    çľĭåĪ°
    0.13
    çľĭåĪ°äºĨ
    0.12
    åıijçĶŁçļĦ
    0.12
    è§Ĥå¯Ł
    0.11
     observed
    0.11
    çļĦåıijçĶŁ
    0.11
     phenomenon
    0.11
     happening
    0.10
    Activations Density 0.107%

    No Known Activations