© 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. 19649
    Prev
    Next
    INDEX
    Explanations

    no control

    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¿İ
    -0.08
     symbolic
    -0.08
    容
    -0.08
     Darwin
    -0.08
    owered
    -0.08
     nbytes
    -0.08
    åIJı
    -0.08
    ishi
    -0.08
     umb
    -0.07
     vein
    -0.07
    POSITIVE LOGITS
    æĦĨ
    0.12
    æĹłè¾ľ
    0.10
    å±ħ
    0.10
    ä»ħä»ħæĺ¯
    0.10
    æĹłåħ³
    0.10
    被åĬ¨
    0.10
    ä¸Ģè½®
    0.09
    éĺµéĺµ
    0.09
    icks
    0.09
    ças
    0.09
    Activations Density 0.073%

    No Known Activations