Commit ·
bf586e1
1
Parent(s): 3a020da
updated tokenizer.
Browse files
onnx/lilylet-tokenizer.json → tokenizer.json
RENAMED
|
@@ -1,11 +1,6 @@
|
|
| 1 |
{
|
| 2 |
"version": 1,
|
| 3 |
"type": "byte-level-bpe",
|
| 4 |
-
"source": {
|
| 5 |
-
"corpusDir": "tests/output/notagenx-from-abc",
|
| 6 |
-
"serializer": "source/lilylet/serializer.ts",
|
| 7 |
-
"frequencyTsv": "tests/output/bpe-tokenizer/token-frequencies.tsv"
|
| 8 |
-
},
|
| 9 |
"config": {
|
| 10 |
"vocabSize": 256,
|
| 11 |
"minFrequency": 2,
|
|
|
|
| 1 |
{
|
| 2 |
"version": 1,
|
| 3 |
"type": "byte-level-bpe",
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 4 |
"config": {
|
| 5 |
"vocabSize": 256,
|
| 6 |
"minFrequency": 2,
|