© 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. 26520
    Prev
    Next
    INDEX
    Explanations

    going into or deeper

    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
    为ä¹ĭ
    -0.12
    天空
    -0.09
    åĩºåı£
    -0.09
    ä¸įçŁ¥ä¸įè§ī
    -0.09
    ÑĪиб
    -0.09
    æŶè£ħ
    -0.08
    _fid
    -0.08
    ãģ¤ãģ¤
    -0.08
    à¸Ħา
    -0.08
    çľ¼åīį
    -0.08
    POSITIVE LOGITS
    æ·±å±Ĥ
    0.15
    æ·±å¤Ħ
    0.14
    æºIJ头
    0.12
    å¾Īæ·±
    0.12
    æ·±æ·±åľ°
    0.12
    æ·±å±Ĥ次
    0.12
    触åıĬ
    0.11
    究竣
    0.11
    深深
    0.11
    深度
    0.11
    Activations Density 0.061%

    No Known Activations