© 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. 7101
    Prev
    Next
    INDEX
    Explanations

    why do / why does

    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
     воÑĤ
    -0.10
    æ¯Ķå¦Ĥ
    -0.09
    好å¤Ħ
    -0.08
    ä¸Ńæĸ°
    -0.08
    èĶ¬
    -0.08
    å°ģ
    -0.08
    lem
    -0.08
    宥
    -0.08
    åĪİ
    -0.08
    arend
    -0.08
    POSITIVE LOGITS
    nable
    0.09
    etter
    0.09
    åĩij
    0.09
    onds
    0.08
    强çĥĪ
    0.08
    溯æºIJ
    0.08
    ió
    0.08
    深度
    0.08
    ç»Ŀ大å¤ļæķ°
    0.08
    fully
    0.08
    Activations Density 0.015%

    No Known Activations