© 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. 5901
    Prev
    Next
    INDEX
    Explanations

    objective, measurement, quality, morality, decision

    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
    æĸ°çļĦä¸Ģ
    -0.09
    оÑģÑĮ
    -0.09
     sad
    -0.09
    åı£æ°´
    -0.09
    sad
    -0.09
    çļĦ羣å®ŀ
    -0.09
    fol
    -0.08
    羣æŃ£åģļåĪ°
    -0.08
    注æĦı
    -0.08
    çĿĢéĩį
    -0.08
    POSITIVE LOGITS
    -minded
    0.13
    客è§Ĥ
    0.12
     minded
    0.10
    -looking
    0.10
    ive
    0.10
    主ä¹īèĢħ
    0.10
    å¥ĭæĸĹ缮æłĩ
    0.09
    åľ°è¯´
    0.09
    obs
    0.09
    åħ¬æŃ£
    0.09
    Activations Density 0.023%

    No Known Activations