Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Andy Arditi · GPT-OSS BatchTopK SAEs
    3. GPT-OSS-20B
    4. Resid Post - 131k
    5. 23-RESID-POST-AA
    6. 98617
    Prev
    Next
    INDEX
    Explanations

    say never

    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_23/trainer_0
    Dataset (Dashboard)
    Various
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Must
    -0.10
     Certainly
    -0.10
     Definitely
    -0.10
     deben
    -0.10
    Must
    -0.10
     definitely
    -0.09
    Certainly
    -0.09
     unbedingt
    -0.09
     Typically
    -0.09
     Obviously
    -0.08
    POSITIVE LOGITS
     never
    0.21
     никогда
    0.20
     nunca
    0.19
     aldrig
    0.19
     aldri
    0.19
     nooit
    0.18
    never
    0.17
     nikdy
    0.16
     Never
    0.15
     jamás
    0.14
    Activations Density 0.039%

    No Known Activations