© 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. 9896
    Prev
    Next
    INDEX
    Explanations

    coverage report or summary

    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
    sWith
    -0.13
    çĦ¶æĺ¯
    -0.10
    sh
    -0.10
    tr
    -0.10
    åĪĨæµģ
    -0.09
    æŃ¤åŁºç¡Ģä¸Ĭ
    -0.09
    ized
    -0.09
     Airways
    -0.09
    tern
    -0.09
    ly
    -0.09
    POSITIVE LOGITS
    åĬĽåº¦
    0.10
    æŃī
    0.10
    ption
    0.10
    ¨
    0.10
    sie
    0.10
    olds
    0.10
    ستاÙĨ
    0.09
    elter
    0.09
    ëģĶ
    0.09
    sar
    0.09
    Activations Density 0.011%

    No Known Activations