© 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. 50680
    Prev
    Next
    INDEX
    Explanations

    alas, behold, beware

    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
    protocol
    -0.09
    ych
    -0.08
    Ñıв
    -0.08
    sex
    -0.08
    é¢Ĭ
    -0.08
    رب
    -0.08
     earliest
    -0.08
     clay
    -0.08
     stress
    -0.08
    ropolis
    -0.08
    POSITIVE LOGITS
    ingly
    0.12
    人éĢł
    0.11
    (er
    0.10
    forth
    0.10
    abox
    0.10
    ëĿ½
    0.09
    åĬ³
    0.09
    å¥ĭæĸĹ
    0.09
    æ¯ĶåĪ©
    0.09
    ãģªãģĮãĤī
    0.09
    Activations Density 0.004%

    No Known Activations