© 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. Andy Arditi · GPT-OSS BatchTopK SAEs
    5. GPT-OSS-20B
    7. Resid Post - 131k
    9. 19-RESID-POST-AA
    11. 18342
    Prev
    Next
    INDEX
    Explanations

    word counting

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    andyrdt/saes-gpt-oss-20b/resid_post_layer_19/trainer_0
    Dataset (Dashboard)
    Various
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    因此
    -0.09
     yields
    -0.08
     daardoor
    -0.08
    Other
    -0.08
     thus
    -0.08
     따라서
    -0.08
     velocities
    -0.08
     તેથી
    -0.08
     therefore
    -0.08
    ഇ
    -0.07
    POSITIVE LOGITS
     gjithë
    0.09
     específicamente
    0.09
     exclusivamente
    0.09
     aproximadamente
    0.09
     ausschließlich
    0.09
     especificamente
    0.09
    -use
    0.08
     mindestens
    0.08
    "ב
    0.08
    ‑
    0.08
    Activations Density 0.014%

    No Known Activations