Upload README.md with huggingface_hub
Browse files
README.md
ADDED
|
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# Super Vocabulary
|
| 2 |
+
|
| 3 |
+
A merged super-vocabulary built from 11 tokenizer(s).
|
| 4 |
+
|
| 5 |
+
**Vocab size:** 163711
|
| 6 |
+
|
| 7 |
+
## Tokenizers
|
| 8 |
+
|
| 9 |
+
- `flexitok/bpe_script_Arab_16000`
|
| 10 |
+
- `flexitok/bpe_script_CmJp_16000`
|
| 11 |
+
- `flexitok/bpe_ltr_ell_Grek_8000_v2`
|
| 12 |
+
- `flexitok/bpe_ltr_fw_edu_32000_v2`
|
| 13 |
+
- `flexitok/bpe_ltr_hun_Latn_8000_v2`
|
| 14 |
+
- `flexitok/bpe_ltr_rus_Cyrl_16000_v2`
|
| 15 |
+
- `flexitok/bpe_ltr_tur_Latn_8000_v2`
|
| 16 |
+
- `flexitok/bpe_script_Germ_32000`
|
| 17 |
+
- `flexitok/bpe_script_Roma_32000`
|
| 18 |
+
- `flexitok/bpe_script_SEAS_16000`
|
| 19 |
+
- `flexitok/bpe_script_Slav_16000`
|
| 20 |
+
|
| 21 |
+
## Files
|
| 22 |
+
|
| 23 |
+
- `super_vocab.json` — merged vocabulary mapping token string → super index
|
| 24 |
+
- `config.yaml` — model config with `vocab_size`
|
| 25 |
+
- `participating_tokenizers.json` — list of tokenizer names included
|
| 26 |
+
- `<tokenizer>_super_mapping.json` — per-tokenizer index → super index mapping
|
| 27 |
+
- `<tokenizer>_vocab.json` — per-tokenizer vocabulary
|
| 28 |
+
- `<tokenizer>_info.json` / `.yaml` — tokenizer metadata
|