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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Andy Arditi · GPT-OSS BatchTopK SAEs
    3. GPT-OSS-20B
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 113975
    Prev
    Next
    INDEX
    Explanations

    puzzles with multiple solutions

    np_max-act · gemini-2.0-flash

    attends to tokens related to darts from tokens related to individuals throwing them.

    oai_attention-head · gpt-4oTriggered by @mlfinancebook
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    andyrdt/saes-gpt-oss-20b/resid_post_layer_11/trainer_0
    Dataset (Dashboard)
    Various
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    /sn
    -0.09
     Jap
    -0.08
    antia
    -0.08
     ben
    -0.08
     praises
    -0.08
    .fn
    -0.08
    Resumen
    -0.08
    ORIA
    -0.08
     Festivals
    -0.08
    buk
    -0.08
    POSITIVE LOGITS
     ambiguity
    0.10
     uniqueness
    0.10
     ambiguous
    0.10
     решений
    0.09
     solution
    0.09
    ambigu
    0.09
     exist
    0.09
     dependence
    0.09
     متعددة
    0.09
     meerdere
    0.09
    Activations Density 0.051%

    No Known Activations