File size: 421 Bytes
1e8d395 e6e5f43 adf023c e6e5f43 adf023c e6e5f43 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 |
# Languages are for reference only, and aren't used in the graph.
languages: [
"ca", "es", "fr", "it", "pt", "ro"
]
# Model has 512 positional embeddings, but only trained on sequences up to 256 tokens
max_length: 256
# Labels that may be predicted before each token
pre_labels: [
"<NULL>",
"¿",
]
# Labels that may be predicted after each token
post_labels: [
"<NULL>",
"<ACRONYM>",
".",
",",
"?",
] |