© 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. 55210
    Prev
    Next
    INDEX
    Explanations

    let's say I have

    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¢ħ
    -0.11
    roid
    -0.10
    ERA
    -0.10
    why
    -0.09
    æ°Ķè´¨
    -0.08
    为ä»Ģä¹Īè¦ģ
    -0.08
    åĬĽäºī
    -0.08
    æĹ¢åı¯ä»¥
    -0.08
    沧
    -0.08
    å²Ĥ
    -0.08
    POSITIVE LOGITS
    ä½łæĥ³
    0.10
    æľīä¸Ģ次
    0.10
    æŁIJ个
    0.09
    ä¸į幸
    0.09
    /start
    0.09
     hypoth
    0.09
    碰
    0.09
    ä½łåľ¨
    0.09
    lah
    0.09
    解æĶ¾
    0.09
    Activations Density 0.176%

    No Known Activations