© 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. 37726
    Prev
    Next
    INDEX
    Explanations

    hallucinations, delusions, distort

    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
    _inactive
    -0.10
    åĿĿ
    -0.10
    æĦł
    -0.09
    irement
    -0.09
    ä¸įå°ı
    -0.08
    åģļ大
    -0.08
    æĹłå£°
    -0.08
    殡
    -0.08
    VERBOSE
    -0.08
    æµİ
    -0.08
    POSITIVE LOGITS
    æīŃæĽ²
    0.14
    å¹»
    0.13
     halluc
    0.12
    æ¢Ń
    0.11
     illusion
    0.11
     illusions
    0.10
    èľĥ
    0.10
     ÐŁÑĢе
    0.10
     distort
    0.10
    梦幻
    0.10
    Activations Density 0.099%

    No Known Activations