© 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. 37579
    Prev
    Next
    INDEX
    Explanations

    guilt and conscience

    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
    æ°¸ä¸į
    -0.10
    èIJ½åľ°
    -0.09
    èĦĨ
    -0.09
     اÙĦÙĥر
    -0.09
    imum
    -0.09
    _FALL
    -0.09
    éľ²å¤©
    -0.09
    ä¿¡ä»»
    -0.09
    æĪijåĸľæ¬¢
    -0.09
    çľĭæ³ķ
    -0.08
    POSITIVE LOGITS
     guilt
    0.19
    æĦ§
    0.14
     guilty
    0.14
    çĸļ
    0.12
    罪
    0.12
     conscience
    0.11
    æĥŃ
    0.11
    åĴİ
    0.11
    arness
    0.10
     justified
    0.09
    Activations Density 0.052%

    No Known Activations