© 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. 61275
    Prev
    Next
    INDEX
    Explanations

    fiction, fantasy, reality, dreams

    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
    åħ¨ç¤¾ä¼ļ
    -0.10
    .Localization
    -0.10
    æķ´ä¸ªäºº
    -0.09
    ëıĪ
    -0.09
    çģµæ°Ķ
    -0.09
    aniu
    -0.09
    ição
    -0.09
     Surveillance
    -0.09
    å±ıèĶ½
    -0.09
    .fml
    -0.09
    POSITIVE LOGITS
    奢
    0.13
    åĵį
    0.11
    梦
    0.10
     possibility
    0.09
    夢
    0.09
    çļĦåı¯èĥ½æĢ§
    0.09
    çIJĨæĥ³
    0.08
     mine
    0.08
    èħ¹
    0.08
     Dream
    0.08
    Activations Density 0.056%

    No Known Activations