© 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. 4044
    Prev
    Next
    INDEX
    Explanations

    out of

    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
    æľīä¸Ģ次
    -0.09
    edor
    -0.09
    å¤§åĽ½
    -0.09
    éĤ¹
    -0.08
    éĦĻ
    -0.08
    å¼Ī
    -0.08
    ACY
    -0.08
    oin
    -0.08
    æĭĽå¾ħ
    -0.08
    lookup
    -0.08
    POSITIVE LOGITS
    å½·
    0.10
     пÑĢинÑı
    0.10
    å¾ĺå¾Ĭ
    0.09
    æĺķ
    0.09
    /not
    0.09
    ä¸įåģļ
    0.09
    /bus
    0.09
    åIJ¹
    0.09
    ihil
    0.09
    wish
    0.09
    Activations Density 0.014%

    No Known Activations