Files
gpt2_parfind_en_zh_equal/multilingual_meta.json

20 lines
343 B
JSON
Raw Normal View History

{
"mode": "soft",
"budget_per_language": 16384,
"languages": [
"eng",
"zho"
],
"tokens_per_language": {
"eng": 16384,
"zho": 16231
},
"shared_token_instances_deduped": 93,
"final_vocab_size": 32526,
"space_free_lexicon_languages": [
"zho"
],
"space_free_lexicon_token_counts": {
"zho": 16231
}
}