© 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. 35571
    Prev
    Next
    INDEX
    Explanations

    forced action and coercion

    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
     agony
    -0.09
    è··
    -0.09
    æ¶Į
    -0.08
     phenomen
    -0.08
    æĺ¼
    -0.08
     tá»Ļi
    -0.08
    æķ¬ä¸ļ
    -0.08
     agon
    -0.08
     modelBuilder
    -0.08
    reated
    -0.08
    POSITIVE LOGITS
    æѦè£ħ
    0.18
    æļ´åĬĽ
    0.16
    æīĭ段
    0.13
    åĬĽéĩı
    0.13
    åĨĽäºĭ
    0.12
     armed
    0.12
    æѦ
    0.11
    çļĦåĬĽéĩı
    0.11
    æĭ³å¤´
    0.11
     Ñģил
    0.10
    Activations Density 0.039%

    No Known Activations