© 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. 35002
    Prev
    Next
    INDEX
    Explanations

    most likely

    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
    åħļå·¥å§Ķ
    -0.09
    å§ĭç»ĪåĿļæĮģ
    -0.09
    å®īå®ģ
    -0.09
    æ¹Ľ
    -0.08
    æķħæĦı
    -0.08
     Proto
    -0.08
    ä¸ĢçĤ¹ä¹Łä¸į
    -0.08
    éļį
    -0.08
    æľīæľºä¼ļ
    -0.08
    atics
    -0.08
    POSITIVE LOGITS
     likely
    0.18
    likely
    0.16
     Likely
    0.16
    åı¯èĥ½æĺ¯
    0.16
     probably
    0.15
    ãģĬãģĿãĤīãģı
    0.14
    probably
    0.14
    æĺ¯åĽłä¸º
    0.13
     Probably
    0.12
    大æ¦Ĥæĺ¯
    0.12
    Activations Density 0.057%

    No Known Activations