© 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. Qwen3-14B
    5. 20-RESID-BATCHTOPK-65K__L0-80
    7. 14303
    Prev
    Next
    INDEX
    Explanations

    policies, fairness, safety, ethics

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-14b-saes/saes_Qwen_Qwen3-14B_batch_top_k/resid_post_layer_20
    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
    çĨ¬
    -0.22
    èĢĥ
    -0.22
     bib
    -0.21
     penet
    -0.20
     timed
    -0.20
     litt
    -0.20
    ĵåIJį
    -0.19
    SAT
    -0.19
    è°ĵ
    -0.19
    ä¸įåľ¨ä¹İ
    -0.19
    POSITIVE LOGITS
    å®īåħ¨
    0.27
    å®īåħ¨æĢ§
    0.22
    éĺ²æŃ¢
    0.22
    éĻIJåĪ¶
    0.22
     censorship
    0.22
     safety
    0.22
    åIJĪè§Ħ
    0.22
    人èĦ¸
    0.21
    .Safe
    0.21
     ethical
    0.21
    Activations Density 0.021%

    No Known Activations